2022-12-22 10:29:13,704 INFO [train.py:966] (2/4) Training started
2022-12-22 10:29:13,704 INFO [train.py:976] (2/4) Device: cuda:2
2022-12-22 10:29:13,707 INFO [train.py:985] (2/4) {'best_train_loss': inf, 'best_valid_loss': inf, 'best_train_epoch': -1, 'best_valid_epoch': -1, 'batch_idx_train': 0, 'log_interval': 50, 'reset_interval': 200, 'valid_interval': 3000, 'feature_dim': 80, 'subsampling_factor': 4, 'warm_step': 2000, 'env_info': {'k2-version': '1.23.2', 'k2-build-type': 'Release', 'k2-with-cuda': True, 'k2-git-sha1': 'efd83642a940dc7db08688cc0791985bed1fafcd', 'k2-git-date': 'Sun Nov 27 19:12:00 2022', 'lhotse-version': '1.4.0.dev+git.21b5b3f.dirty', 'torch-version': '1.10.0+cu102', 'torch-cuda-available': True, 'torch-cuda-version': '10.2', 'python-version': '3.8', 'icefall-git-branch': 'zipformer_streaming', 'icefall-git-sha1': '2f75b00-dirty', 'icefall-git-date': 'Wed Dec 21 22:15:21 2022', 'icefall-path': '/ceph-zw/workspace/zipformer/icefall_streaming', 'k2-path': '/ceph-zw/workspace/share/k2-last/k2/python/k2/__init__.py', 'lhotse-path': '/ceph-jb/yaozengwei/workspace/rnnt/lhotse/lhotse/__init__.py', 'hostname': 'de-74279-k2-train-2-1216192652-5bcf7587b4-n6q9m', 'IP address': '10.177.74.211'}, 'world_size': 4, 'master_port': 12345, 'tensorboard': True, 'num_epochs': 30, 'start_epoch': 1, 'start_batch': 0, 'exp_dir': PosixPath('pruned_transducer_stateless7_streaming/exp-full-dynamic-chunk'), 'bpe_model': 'data/lang_bpe_500/bpe.model', 'base_lr': 0.05, 'lr_batches': 5000, 'lr_epochs': 3.5, 'context_size': 2, 'prune_range': 5, 'lm_scale': 0.25, 'am_scale': 0.0, 'simple_loss_scale': 0.5, 'seed': 42, 'print_diagnostics': False, 'inf_check': False, 'save_every_n': 2000, 'keep_last_k': 30, 'average_period': 200, 'use_fp16': True, 'num_encoder_layers': '2,4,3,2,4', 'feedforward_dims': '1024,1024,2048,2048,1024', 'nhead': '8,8,8,8,8', 'encoder_dims': '384,384,384,384,384', 'attention_dims': '192,192,192,192,192', 'encoder_unmasked_dims': '256,256,256,256,256', 'zipformer_downsampling_factors': '1,2,4,8,2', 'cnn_module_kernels': '31,31,31,31,31', 'decoder_dim': 512, 'joiner_dim': 512, 'short_chunk_size': 50, 'num_left_chunks': 4, 'full_libri': True, 'manifest_dir': PosixPath('data/fbank'), 'max_duration': 750, 'bucketing_sampler': True, 'num_buckets': 30, 'concatenate_cuts': False, 'duration_factor': 1.0, 'gap': 1.0, 'on_the_fly_feats': False, 'shuffle': True, 'drop_last': True, 'return_cuts': True, 'num_workers': 2, 'enable_spec_aug': True, 'spec_aug_time_warp_factor': 80, 'enable_musan': True, 'input_strategy': 'PrecomputedFeatures', 'blank_id': 0, 'vocab_size': 500}
2022-12-22 10:29:13,707 INFO [train.py:987] (2/4) About to create model
2022-12-22 10:29:14,196 INFO [zipformer.py:185] (2/4) At encoder stack 4, which has downsampling_factor=2, we will combine the outputs of layers 1 and 3, with downsampling_factors=2 and 8.
2022-12-22 10:29:14,211 INFO [train.py:991] (2/4) Number of model parameters: 70369391
2022-12-22 10:29:18,823 INFO [train.py:1006] (2/4) Using DDP
2022-12-22 10:29:19,178 INFO [asr_datamodule.py:398] (2/4) About to get train-clean-100 cuts
2022-12-22 10:29:19,188 INFO [asr_datamodule.py:405] (2/4) About to get train-clean-360 cuts
2022-12-22 10:29:19,197 INFO [asr_datamodule.py:412] (2/4) About to get train-other-500 cuts
2022-12-22 10:29:19,207 INFO [asr_datamodule.py:224] (2/4) Enable MUSAN
2022-12-22 10:29:19,208 INFO [asr_datamodule.py:225] (2/4) About to get Musan cuts
2022-12-22 10:29:21,482 INFO [asr_datamodule.py:249] (2/4) Enable SpecAugment
2022-12-22 10:29:21,483 INFO [asr_datamodule.py:250] (2/4) Time warp factor: 80
2022-12-22 10:29:21,483 INFO [asr_datamodule.py:260] (2/4) Num frame mask: 10
2022-12-22 10:29:21,483 INFO [asr_datamodule.py:273] (2/4) About to create train dataset
2022-12-22 10:29:21,483 INFO [asr_datamodule.py:300] (2/4) Using DynamicBucketingSampler.
2022-12-22 10:29:25,257 WARNING [train.py:1060] (2/4) Exclude cut with ID 7859-102521-0017-7548-0_sp1.1 from training. Duration: 22.2954375
2022-12-22 10:29:26,373 INFO [asr_datamodule.py:315] (2/4) About to create train dataloader
2022-12-22 10:29:26,374 INFO [asr_datamodule.py:429] (2/4) About to get dev-clean cuts
2022-12-22 10:29:26,376 INFO [asr_datamodule.py:436] (2/4) About to get dev-other cuts
2022-12-22 10:29:26,377 INFO [asr_datamodule.py:346] (2/4) About to create dev dataset
2022-12-22 10:29:26,622 INFO [asr_datamodule.py:363] (2/4) About to create dev dataloader
2022-12-22 10:29:26,622 INFO [train.py:1206] (2/4) Sanity check -- see if any of the batches in epoch 1 would cause OOM.
2022-12-22 10:29:30,333 WARNING [train.py:1060] (2/4) Exclude cut with ID 7859-102521-0017-7548-0_sp1.1 from training. Duration: 22.2954375
2022-12-22 10:29:35,242 WARNING [train.py:1060] (2/4) Exclude cut with ID 7859-102521-0017-7548-0_sp1.1 from training. Duration: 22.2954375
2022-12-22 10:29:38,736 WARNING [train.py:1060] (2/4) Exclude cut with ID 298-126791-0067-24026-0_sp0.9 from training. Duration: 21.438875
2022-12-22 10:29:38,908 WARNING [train.py:1060] (2/4) Exclude cut with ID 5652-39938-0025-23684-0_sp0.9 from training. Duration: 22.2055625
2022-12-22 10:29:45,231 WARNING [train.py:1060] (2/4) Exclude cut with ID 7859-102521-0017-7548-0 from training. Duration: 24.525
2022-12-22 10:29:46,909 WARNING [train.py:1060] (2/4) Exclude cut with ID 3699-47246-0007-3408-0_sp0.9 from training. Duration: 20.26675
2022-12-22 10:29:47,336 WARNING [train.py:1060] (2/4) Exclude cut with ID 7859-102521-0017-7548-0_sp0.9 from training. Duration: 27.25
2022-12-22 10:29:50,264 WARNING [train.py:1060] (2/4) Exclude cut with ID 6426-64292-0017-15984-0 from training. Duration: 21.68
2022-12-22 10:29:50,664 WARNING [train.py:1060] (2/4) Exclude cut with ID 4278-13270-0007-59342-0 from training. Duration: 21.6300625
2022-12-22 10:29:51,397 WARNING [train.py:1060] (2/4) Exclude cut with ID 4278-13270-0007-59342-0_sp0.9 from training. Duration: 24.033375
2022-12-22 10:29:53,254 WARNING [train.py:1060] (2/4) Exclude cut with ID 4278-13270-0009-59344-0 from training. Duration: 22.905
2022-12-22 10:29:53,297 WARNING [train.py:1060] (2/4) Exclude cut with ID 5622-44585-0006-90525-0_sp1.1 from training. Duration: 23.4318125
2022-12-22 10:29:57,642 WARNING [train.py:1060] (2/4) Exclude cut with ID 4278-13270-0009-59344-0_sp1.1 from training. Duration: 20.82275
2022-12-22 10:29:57,687 WARNING [train.py:1060] (2/4) Exclude cut with ID 4278-13270-0009-59344-0_sp0.9 from training. Duration: 25.45
2022-12-22 10:30:00,216 WARNING [train.py:1060] (2/4) Exclude cut with ID 5622-44585-0006-90525-0 from training. Duration: 25.775
2022-12-22 10:30:00,923 WARNING [train.py:1060] (2/4) Exclude cut with ID 2929-85685-0071-62375-0_sp0.9 from training. Duration: 22.25
2022-12-22 10:30:01,789 WARNING [train.py:1060] (2/4) Exclude cut with ID 3972-170212-0014-23379-0 from training. Duration: 26.205
2022-12-22 10:30:02,931 WARNING [train.py:1060] (2/4) Exclude cut with ID 5239-32139-0047-9341-0_sp0.9 from training. Duration: 30.1555625
2022-12-22 10:30:03,105 WARNING [train.py:1060] (2/4) Exclude cut with ID 1265-135635-0050-6781-0_sp0.9 from training. Duration: 21.8333125
2022-12-22 10:30:03,401 WARNING [train.py:1060] (2/4) Exclude cut with ID 1914-133440-0024-94914-0_sp1.1 from training. Duration: 20.6545625
2022-12-22 10:30:04,769 WARNING [train.py:1060] (2/4) Exclude cut with ID 7395-89880-0045-39920-0_sp0.9 from training. Duration: 20.52225
2022-12-22 10:30:05,406 WARNING [train.py:1060] (2/4) Exclude cut with ID 3972-170212-0014-23379-0_sp0.9 from training. Duration: 29.1166875
2022-12-22 10:30:07,853 WARNING [train.py:1060] (2/4) Exclude cut with ID 543-133211-0007-59831-0_sp0.9 from training. Duration: 21.388875
2022-12-22 10:30:09,025 WARNING [train.py:1060] (2/4) Exclude cut with ID 1914-133440-0024-94914-0 from training. Duration: 22.72
2022-12-22 10:30:09,069 WARNING [train.py:1060] (2/4) Exclude cut with ID 1914-133440-0031-94921-0_sp0.9 from training. Duration: 22.7444375
2022-12-22 10:30:10,471 WARNING [train.py:1060] (2/4) Exclude cut with ID 4133-6541-0027-40495-0_sp1.1 from training. Duration: 0.9681875
2022-12-22 10:30:10,590 WARNING [train.py:1060] (2/4) Exclude cut with ID 6330-62851-0022-91297-0_sp0.9 from training. Duration: 22.3166875
2022-12-22 10:30:11,360 WARNING [train.py:1060] (2/4) Exclude cut with ID 543-133212-0015-59917-0_sp0.9 from training. Duration: 21.8166875
2022-12-22 10:30:14,630 WARNING [train.py:1060] (2/4) Exclude cut with ID 4957-30119-0041-23990-0_sp0.9 from training. Duration: 20.22775
2022-12-22 10:30:16,639 WARNING [train.py:1060] (2/4) Exclude cut with ID 5239-32139-0047-9341-0_sp1.1 from training. Duration: 24.67275
2022-12-22 10:30:17,542 WARNING [train.py:1060] (2/4) Exclude cut with ID 3082-165428-0081-50734-0_sp0.9 from training. Duration: 21.8055625
2022-12-22 10:30:18,761 WARNING [train.py:1060] (2/4) Exclude cut with ID 3340-169293-0054-76830-0_sp0.9 from training. Duration: 22.6666875
2022-12-22 10:30:21,356 WARNING [train.py:1060] (2/4) Exclude cut with ID 2411-132532-0017-82279-0_sp1.1 from training. Duration: 0.9681875
2022-12-22 10:30:22,199 WARNING [train.py:1060] (2/4) Exclude cut with ID 6330-62850-0007-91323-0 from training. Duration: 22.485
2022-12-22 10:30:24,024 WARNING [train.py:1060] (2/4) Exclude cut with ID 3972-170212-0014-23379-0_sp1.1 from training. Duration: 23.82275
2022-12-22 10:30:24,464 WARNING [train.py:1060] (2/4) Exclude cut with ID 4860-13185-0032-76709-0 from training. Duration: 20.77
2022-12-22 10:30:24,719 WARNING [train.py:1060] (2/4) Exclude cut with ID 6426-64292-0017-15984-0_sp0.9 from training. Duration: 24.088875
2022-12-22 10:30:25,677 WARNING [train.py:1060] (2/4) Exclude cut with ID 6330-62850-0007-91323-0_sp1.1 from training. Duration: 20.4409375
2022-12-22 10:30:28,707 WARNING [train.py:1060] (2/4) Exclude cut with ID 2929-85685-0062-62366-0_sp0.9 from training. Duration: 22.511125
2022-12-22 10:30:28,728 WARNING [train.py:1060] (2/4) Exclude cut with ID 7395-89880-0031-39906-0 from training. Duration: 20.675
2022-12-22 10:30:31,982 WARNING [train.py:1060] (2/4) Exclude cut with ID 6330-62850-0007-91323-0_sp0.9 from training. Duration: 24.9833125
2022-12-22 10:30:33,559 WARNING [train.py:1060] (2/4) Exclude cut with ID 5239-32139-0047-9341-0 from training. Duration: 27.14
2022-12-22 10:30:34,086 WARNING [train.py:1060] (2/4) Exclude cut with ID 2929-85685-0044-62348-0 from training. Duration: 22.44
2022-12-22 10:30:37,240 WARNING [train.py:1060] (2/4) Exclude cut with ID 2929-85685-0060-62364-0_sp0.9 from training. Duration: 21.361125
2022-12-22 10:30:37,436 WARNING [train.py:1060] (2/4) Exclude cut with ID 2929-85685-0079-62383-0_sp1.1 from training. Duration: 27.0318125
2022-12-22 10:30:37,739 WARNING [train.py:1060] (2/4) Exclude cut with ID 5622-44585-0006-90525-0_sp0.9 from training. Duration: 28.638875
2022-12-22 10:30:38,239 WARNING [train.py:1060] (2/4) Exclude cut with ID 3340-169293-0054-76830-0 from training. Duration: 20.4
2022-12-22 10:30:39,279 WARNING [train.py:1060] (2/4) Exclude cut with ID 2929-85685-0071-62375-0 from training. Duration: 20.025
2022-12-22 10:30:39,287 WARNING [train.py:1060] (2/4) Exclude cut with ID 2364-131735-0112-64612-0_sp0.9 from training. Duration: 20.488875
2022-12-22 10:30:39,478 WARNING [train.py:1060] (2/4) Exclude cut with ID 2929-85685-0079-62383-0 from training. Duration: 29.735
2022-12-22 10:30:42,701 WARNING [train.py:1060] (2/4) Exclude cut with ID 7276-92427-0014-12983-0_sp0.9 from training. Duration: 21.3055625
2022-12-22 10:30:42,749 WARNING [train.py:1060] (2/4) Exclude cut with ID 1025-75365-0008-79168-0_sp0.9 from training. Duration: 22.0666875
2022-12-22 10:30:46,780 WARNING [train.py:1060] (2/4) Exclude cut with ID 2929-85685-0062-62366-0 from training. Duration: 20.26
2022-12-22 10:30:47,219 WARNING [train.py:1060] (2/4) Exclude cut with ID 5239-32139-0030-9324-0_sp0.9 from training. Duration: 21.3444375
2022-12-22 10:30:49,704 WARNING [train.py:1060] (2/4) Exclude cut with ID 497-129325-0061-62254-0_sp1.1 from training. Duration: 0.97725
2022-12-22 10:30:51,058 WARNING [train.py:1060] (2/4) Exclude cut with ID 7395-89880-0031-39906-0_sp0.9 from training. Duration: 22.97225
2022-12-22 10:30:52,070 WARNING [train.py:1060] (2/4) Exclude cut with ID 7395-89880-0047-39922-0_sp0.9 from training. Duration: 21.97775
2022-12-22 10:30:52,500 WARNING [train.py:1060] (2/4) Exclude cut with ID 1112-1043-0006-89194-0_sp0.9 from training. Duration: 21.8333125
2022-12-22 10:30:52,863 WARNING [train.py:1060] (2/4) Exclude cut with ID 1914-133440-0031-94921-0 from training. Duration: 20.47
2022-12-22 10:30:55,549 WARNING [train.py:1060] (2/4) Exclude cut with ID 7395-89880-0037-39912-0_sp0.9 from training. Duration: 20.67225
2022-12-22 10:30:56,177 WARNING [train.py:1060] (2/4) Exclude cut with ID 1914-133440-0024-94914-0_sp0.9 from training. Duration: 25.2444375
2022-12-22 10:30:56,951 WARNING [train.py:1060] (2/4) Exclude cut with ID 3340-169293-0021-76797-0_sp0.9 from training. Duration: 21.1445
2022-12-22 10:30:59,883 WARNING [train.py:1060] (2/4) Exclude cut with ID 2929-85685-0079-62383-0_sp0.9 from training. Duration: 33.038875
2022-12-22 10:31:01,117 WARNING [train.py:1060] (2/4) Exclude cut with ID 6426-64291-0000-16059-0_sp0.9 from training. Duration: 20.0944375
2022-12-22 10:31:02,167 WARNING [train.py:1060] (2/4) Exclude cut with ID 2929-85685-0044-62348-0_sp1.1 from training. Duration: 20.4
2022-12-22 10:31:02,409 WARNING [train.py:1060] (2/4) Exclude cut with ID 6330-62851-0022-91297-0 from training. Duration: 20.085
2022-12-22 10:31:02,758 WARNING [train.py:1060] (2/4) Exclude cut with ID 4860-13185-0032-76709-0_sp0.9 from training. Duration: 23.07775
2022-12-22 10:31:04,219 WARNING [train.py:1060] (2/4) Exclude cut with ID 2929-85685-0044-62348-0_sp0.9 from training. Duration: 24.9333125
2022-12-22 10:31:05,640 WARNING [train.py:1060] (2/4) Exclude cut with ID 5118-111612-0016-124680-0_sp0.9 from training. Duration: 20.388875
2022-12-22 10:31:06,413 WARNING [train.py:1060] (2/4) Exclude cut with ID 432-122774-0017-62487-0_sp1.1 from training. Duration: 20.3590625
2022-12-22 10:31:08,603 WARNING [train.py:1060] (2/4) Exclude cut with ID 3557-8342-0013-54691-0_sp1.1 from training. Duration: 0.836375
2022-12-22 10:31:10,526 WARNING [train.py:1060] (2/4) Exclude cut with ID 8565-290391-0049-67394-0_sp0.9 from training. Duration: 21.3166875
2022-12-22 10:31:11,477 WARNING [train.py:1060] (2/4) Exclude cut with ID 6533-399-0029-104863-0_sp0.9 from training. Duration: 22.1055625
2022-12-22 10:31:11,818 WARNING [train.py:1060] (2/4) Exclude cut with ID 7699-105389-0094-26379-0_sp1.1 from training. Duration: 21.77725
2022-12-22 10:31:12,446 WARNING [train.py:1060] (2/4) Exclude cut with ID 7255-291500-0005-134304-0_sp0.9 from training. Duration: 27.8166875
2022-12-22 10:31:13,238 WARNING [train.py:1060] (2/4) Exclude cut with ID 7357-94126-0021-15852-0_sp1.1 from training. Duration: 22.5090625
2022-12-22 10:31:13,430 WARNING [train.py:1060] (2/4) Exclude cut with ID 7255-291500-0005-134304-0 from training. Duration: 25.035
2022-12-22 10:31:14,565 WARNING [train.py:1060] (2/4) Exclude cut with ID 774-127930-0014-10412-0_sp1.1 from training. Duration: 0.95
2022-12-22 10:31:15,154 WARNING [train.py:1060] (2/4) Exclude cut with ID 3033-130750-0096-55598-0_sp0.9 from training. Duration: 0.92225
2022-12-22 10:31:15,975 WARNING [train.py:1060] (2/4) Exclude cut with ID 4511-76322-0006-80011-0 from training. Duration: 21.97
2022-12-22 10:31:16,549 WARNING [train.py:1060] (2/4) Exclude cut with ID 7492-105653-0055-62765-0_sp0.9 from training. Duration: 21.97225
2022-12-22 10:31:16,582 WARNING [train.py:1060] (2/4) Exclude cut with ID 453-131332-0000-47844-0_sp0.9 from training. Duration: 25.3333125
2022-12-22 10:31:16,942 WARNING [train.py:1060] (2/4) Exclude cut with ID 5172-29468-0015-19128-0_sp0.9 from training. Duration: 21.5055625
2022-12-22 10:31:17,248 WARNING [train.py:1060] (2/4) Exclude cut with ID 453-131332-0000-47844-0_sp1.1 from training. Duration: 20.72725
2022-12-22 10:31:18,903 WARNING [train.py:1060] (2/4) Exclude cut with ID 8631-249866-0030-130156-0_sp0.9 from training. Duration: 26.32775
2022-12-22 10:31:19,884 WARNING [train.py:1060] (2/4) Exclude cut with ID 3867-173237-0077-144769-0 from training. Duration: 20.025
2022-12-22 10:31:20,051 WARNING [train.py:1060] (2/4) Exclude cut with ID 6709-74022-0004-86860-0_sp1.1 from training. Duration: 0.9409375
2022-12-22 10:31:20,057 WARNING [train.py:1060] (2/4) Exclude cut with ID 4757-1811-0023-62229-0_sp0.9 from training. Duration: 21.37775
2022-12-22 10:31:20,705 WARNING [train.py:1060] (2/4) Exclude cut with ID 1250-135782-0004-25974-0_sp0.9 from training. Duration: 21.17225
2022-12-22 10:31:20,718 WARNING [train.py:1060] (2/4) Exclude cut with ID 7357-94126-0021-15852-0_sp0.9 from training. Duration: 27.511125
2022-12-22 10:31:21,738 WARNING [train.py:1060] (2/4) Exclude cut with ID 453-131332-0000-47844-0 from training. Duration: 22.8
2022-12-22 10:31:22,470 WARNING [train.py:1060] (2/4) Exclude cut with ID 4964-30587-0040-44509-0 from training. Duration: 22.585
2022-12-22 10:31:23,441 WARNING [train.py:1060] (2/4) Exclude cut with ID 6978-92210-0001-146967-0_sp0.9 from training. Duration: 22.0166875
2022-12-22 10:31:23,973 WARNING [train.py:1060] (2/4) Exclude cut with ID 7255-291500-0003-134302-0_sp1.1 from training. Duration: 24.395375
2022-12-22 10:31:24,187 WARNING [train.py:1060] (2/4) Exclude cut with ID 3488-85273-0017-41203-0_sp0.9 from training. Duration: 27.47775
2022-12-22 10:31:24,321 WARNING [train.py:1060] (2/4) Exclude cut with ID 432-122774-0017-62487-0_sp0.9 from training. Duration: 24.8833125
2022-12-22 10:31:24,408 WARNING [train.py:1060] (2/4) Exclude cut with ID 6758-72288-0033-108368-0 from training. Duration: 23.39
2022-12-22 10:31:24,598 WARNING [train.py:1060] (2/4) Exclude cut with ID 8291-282929-0007-12994-0_sp0.9 from training. Duration: 28.72225
2022-12-22 10:31:24,911 WARNING [train.py:1060] (2/4) Exclude cut with ID 585-294811-0110-133686-0_sp0.9 from training. Duration: 20.8944375
2022-12-22 10:31:25,387 WARNING [train.py:1060] (2/4) Exclude cut with ID 5796-66357-0007-116447-0_sp0.9 from training. Duration: 23.8444375
2022-12-22 10:31:26,711 WARNING [train.py:1060] (2/4) Exclude cut with ID 8291-282929-0007-12994-0 from training. Duration: 25.85
2022-12-22 10:31:26,720 WARNING [train.py:1060] (2/4) Exclude cut with ID 8291-282929-0023-13010-0 from training. Duration: 21.39
2022-12-22 10:31:27,056 WARNING [train.py:1060] (2/4) Exclude cut with ID 7357-94126-0014-15845-0 from training. Duration: 27.92
2022-12-22 10:31:27,944 WARNING [train.py:1060] (2/4) Exclude cut with ID 8631-249866-0039-130165-0_sp0.9 from training. Duration: 20.661125
2022-12-22 10:31:28,679 WARNING [train.py:1060] (2/4) Exclude cut with ID 7357-94126-0043-15874-0_sp0.9 from training. Duration: 20.07225
2022-12-22 10:31:28,925 WARNING [train.py:1060] (2/4) Exclude cut with ID 1085-156170-0017-128270-0 from training. Duration: 21.01
2022-12-22 10:31:31,709 WARNING [train.py:1060] (2/4) Exclude cut with ID 2195-150901-0045-59933-0 from training. Duration: 20.65
2022-12-22 10:31:31,961 WARNING [train.py:1060] (2/4) Exclude cut with ID 5796-66357-0007-116447-0 from training. Duration: 21.46
2022-12-22 10:31:33,421 WARNING [train.py:1060] (2/4) Exclude cut with ID 3557-8342-0013-54691-0 from training. Duration: 0.92
2022-12-22 10:31:33,640 WARNING [train.py:1060] (2/4) Exclude cut with ID 8291-282929-0023-13010-0_sp0.9 from training. Duration: 23.7666875
2022-12-22 10:31:35,404 WARNING [train.py:1060] (2/4) Exclude cut with ID 8544-281189-0060-101339-0_sp0.9 from training. Duration: 20.861125
2022-12-22 10:31:35,731 WARNING [train.py:1060] (2/4) Exclude cut with ID 3488-65654-0031-41259-0_sp0.9 from training. Duration: 22.711125
2022-12-22 10:31:37,021 WARNING [train.py:1060] (2/4) Exclude cut with ID 6951-79737-0043-132310-0_sp1.1 from training. Duration: 22.986375
2022-12-22 10:31:37,537 WARNING [train.py:1060] (2/4) Exclude cut with ID 8040-260924-0003-80960-0_sp0.9 from training. Duration: 22.07225
2022-12-22 10:31:37,687 WARNING [train.py:1060] (2/4) Exclude cut with ID 7699-105389-0045-26330-0_sp0.9 from training. Duration: 20.3055625
2022-12-22 10:31:37,777 WARNING [train.py:1060] (2/4) Exclude cut with ID 6356-271890-0060-94317-0_sp0.9 from training. Duration: 20.72225
2022-12-22 10:31:38,927 WARNING [train.py:1060] (2/4) Exclude cut with ID 3488-85273-0017-41203-0_sp1.1 from training. Duration: 22.4818125
2022-12-22 10:31:39,718 WARNING [train.py:1060] (2/4) Exclude cut with ID 4964-30587-0040-44509-0_sp0.9 from training. Duration: 25.0944375
2022-12-22 10:31:39,836 WARNING [train.py:1060] (2/4) Exclude cut with ID 6533-399-0047-104881-0 from training. Duration: 21.515
2022-12-22 10:31:40,032 WARNING [train.py:1060] (2/4) Exclude cut with ID 7357-94126-0009-15840-0_sp0.9 from training. Duration: 27.02225
2022-12-22 10:31:40,199 WARNING [train.py:1060] (2/4) Exclude cut with ID 432-122774-0010-62480-0_sp0.9 from training. Duration: 22.22225
2022-12-22 10:31:40,430 WARNING [train.py:1060] (2/4) Exclude cut with ID 4964-30587-0085-44554-0_sp0.9 from training. Duration: 20.85
2022-12-22 10:31:41,486 WARNING [train.py:1060] (2/4) Exclude cut with ID 4295-39940-0007-92567-0 from training. Duration: 21.54
2022-12-22 10:31:41,631 WARNING [train.py:1060] (2/4) Exclude cut with ID 4964-30587-0040-44509-0_sp1.1 from training. Duration: 20.5318125
2022-12-22 10:31:41,925 WARNING [train.py:1060] (2/4) Exclude cut with ID 7255-291500-0012-134311-0_sp0.9 from training. Duration: 21.9333125
2022-12-22 10:31:44,005 WARNING [train.py:1060] (2/4) Exclude cut with ID 8631-249866-0025-130151-0_sp0.9 from training. Duration: 21.7944375
2022-12-22 10:31:44,404 WARNING [train.py:1060] (2/4) Exclude cut with ID 8291-282929-0002-12989-0_sp0.9 from training. Duration: 22.4666875
2022-12-22 10:31:44,633 WARNING [train.py:1060] (2/4) Exclude cut with ID 6121-9014-0076-24124-0 from training. Duration: 21.635
2022-12-22 10:31:44,815 WARNING [train.py:1060] (2/4) Exclude cut with ID 6121-9014-0076-24124-0_sp0.9 from training. Duration: 24.038875
2022-12-22 10:31:46,173 WARNING [train.py:1060] (2/4) Exclude cut with ID 7255-291500-0009-134308-0_sp1.1 from training. Duration: 21.786375
2022-12-22 10:31:47,122 WARNING [train.py:1060] (2/4) Exclude cut with ID 8291-282929-0002-12989-0 from training. Duration: 20.22
2022-12-22 10:31:51,500 WARNING [train.py:1060] (2/4) Exclude cut with ID 6951-79737-0043-132310-0 from training. Duration: 25.285
2022-12-22 10:31:53,528 WARNING [train.py:1060] (2/4) Exclude cut with ID 811-130148-0001-63453-0_sp0.9 from training. Duration: 20.861125
2022-12-22 10:31:54,220 WARNING [train.py:1060] (2/4) Exclude cut with ID 6010-56788-0055-90261-0 from training. Duration: 20.88
2022-12-22 10:31:55,819 WARNING [train.py:1060] (2/4) Exclude cut with ID 7357-94126-0045-15876-0_sp0.9 from training. Duration: 23.4166875
2022-12-22 10:31:58,220 WARNING [train.py:1060] (2/4) Exclude cut with ID 7255-291500-0006-134305-0 from training. Duration: 21.24
2022-12-22 10:31:58,815 WARNING [train.py:1060] (2/4) Exclude cut with ID 6533-399-0047-104881-0_sp0.9 from training. Duration: 23.9055625
2022-12-22 10:31:59,873 WARNING [train.py:1060] (2/4) Exclude cut with ID 6758-72288-0033-108368-0_sp0.9 from training. Duration: 25.988875
2022-12-22 10:32:00,143 WARNING [train.py:1060] (2/4) Exclude cut with ID 7255-291500-0001-134300-0_sp0.9 from training. Duration: 20.67225
2022-12-22 10:32:01,638 WARNING [train.py:1060] (2/4) Exclude cut with ID 3488-85273-0038-41224-0 from training. Duration: 20.34
2022-12-22 10:32:04,562 WARNING [train.py:1060] (2/4) Exclude cut with ID 7357-94126-0026-15857-0_sp0.9 from training. Duration: 25.061125
2022-12-22 10:32:04,913 WARNING [train.py:1060] (2/4) Exclude cut with ID 3033-130750-0096-55598-0 from training. Duration: 0.83
2022-12-22 10:32:05,716 WARNING [train.py:1060] (2/4) Exclude cut with ID 3488-85273-0017-41203-0 from training. Duration: 24.73
2022-12-22 10:32:06,125 WARNING [train.py:1060] (2/4) Exclude cut with ID 7255-291500-0009-134308-0 from training. Duration: 23.965
2022-12-22 10:32:06,980 WARNING [train.py:1060] (2/4) Exclude cut with ID 6978-92210-0030-146996-0_sp0.9 from training. Duration: 22.088875
2022-12-22 10:32:07,492 WARNING [train.py:1060] (2/4) Exclude cut with ID 7255-291500-0006-134305-0_sp0.9 from training. Duration: 23.6
2022-12-22 10:32:11,267 WARNING [train.py:1060] (2/4) Exclude cut with ID 8291-282929-0024-13011-0 from training. Duration: 23.795
2022-12-22 10:32:11,782 WARNING [train.py:1060] (2/4) Exclude cut with ID 8631-249866-0030-130156-0_sp1.1 from training. Duration: 21.5409375
2022-12-22 10:32:11,875 WARNING [train.py:1060] (2/4) Exclude cut with ID 6978-92210-0019-146985-0_sp0.9 from training. Duration: 24.97775
2022-12-22 10:32:12,231 WARNING [train.py:1060] (2/4) Exclude cut with ID 1085-156170-0017-128270-0_sp0.9 from training. Duration: 23.3444375
2022-12-22 10:32:13,089 WARNING [train.py:1060] (2/4) Exclude cut with ID 6010-56788-0055-90261-0_sp0.9 from training. Duration: 23.2
2022-12-22 10:32:13,275 WARNING [train.py:1060] (2/4) Exclude cut with ID 5653-46179-0060-117930-0_sp0.9 from training. Duration: 21.17225
2022-12-22 10:32:13,884 WARNING [train.py:1060] (2/4) Exclude cut with ID 7255-291500-0008-134307-0_sp0.9 from training. Duration: 24.6555625
2022-12-22 10:32:16,231 WARNING [train.py:1060] (2/4) Exclude cut with ID 3488-65654-0031-41259-0 from training. Duration: 20.44
2022-12-22 10:32:16,718 WARNING [train.py:1060] (2/4) Exclude cut with ID 6951-79737-0018-132285-0_sp0.9 from training. Duration: 23.45
2022-12-22 10:32:17,668 WARNING [train.py:1060] (2/4) Exclude cut with ID 6945-60535-0076-12784-0_sp0.9 from training. Duration: 20.52225
2022-12-22 10:32:17,911 WARNING [train.py:1060] (2/4) Exclude cut with ID 7255-291500-0008-134307-0 from training. Duration: 22.19
2022-12-22 10:32:18,817 WARNING [train.py:1060] (2/4) Exclude cut with ID 7357-94126-0014-15845-0_sp1.1 from training. Duration: 25.3818125
2022-12-22 10:32:19,368 WARNING [train.py:1060] (2/4) Exclude cut with ID 6951-79737-0043-132310-0_sp0.9 from training. Duration: 28.0944375
2022-12-22 10:32:19,582 WARNING [train.py:1060] (2/4) Exclude cut with ID 2195-150901-0045-59933-0_sp0.9 from training. Duration: 22.9444375
2022-12-22 10:32:19,837 WARNING [train.py:1060] (2/4) Exclude cut with ID 8291-282929-0024-13011-0_sp1.1 from training. Duration: 21.6318125
2022-12-22 10:32:20,411 WARNING [train.py:1060] (2/4) Exclude cut with ID 8631-249866-0030-130156-0 from training. Duration: 23.695
2022-12-22 10:32:21,269 WARNING [train.py:1060] (2/4) Exclude cut with ID 7699-105389-0094-26379-0 from training. Duration: 23.955
2022-12-22 10:32:22,351 WARNING [train.py:1060] (2/4) Exclude cut with ID 8291-282929-0024-13011-0_sp0.9 from training. Duration: 26.438875
2022-12-22 10:32:24,340 WARNING [train.py:1060] (2/4) Exclude cut with ID 7699-105389-0021-26306-0_sp0.9 from training. Duration: 21.2444375
2022-12-22 10:32:24,360 WARNING [train.py:1060] (2/4) Exclude cut with ID 7357-94126-0014-15845-0_sp0.9 from training. Duration: 31.02225
2022-12-22 10:32:24,698 WARNING [train.py:1060] (2/4) Exclude cut with ID 432-122774-0017-62487-0 from training. Duration: 22.395
2022-12-22 10:32:25,231 WARNING [train.py:1060] (2/4) Exclude cut with ID 7357-94126-0045-15876-0 from training. Duration: 21.075
2022-12-22 10:32:25,401 WARNING [train.py:1060] (2/4) Exclude cut with ID 6482-98857-0025-147532-0_sp0.9 from training. Duration: 20.0055625
2022-12-22 10:32:25,410 WARNING [train.py:1060] (2/4) Exclude cut with ID 6951-79737-0037-132304-0_sp0.9 from training. Duration: 22.05
2022-12-22 10:32:25,421 WARNING [train.py:1060] (2/4) Exclude cut with ID 7255-291500-0003-134302-0 from training. Duration: 26.8349375
2022-12-22 10:32:25,520 WARNING [train.py:1060] (2/4) Exclude cut with ID 7357-94126-0009-15840-0_sp1.1 from training. Duration: 22.1090625
2022-12-22 10:32:25,750 WARNING [train.py:1060] (2/4) Exclude cut with ID 7699-105389-0094-26379-0_sp0.9 from training. Duration: 26.6166875
2022-12-22 10:32:26,309 WARNING [train.py:1060] (2/4) Exclude cut with ID 2046-178027-0000-53705-0_sp0.9 from training. Duration: 20.3055625
2022-12-22 10:32:27,630 WARNING [train.py:1060] (2/4) Exclude cut with ID 7205-50138-0008-5373-0_sp0.9 from training. Duration: 20.7
2022-12-22 10:32:29,024 WARNING [train.py:1060] (2/4) Exclude cut with ID 6978-92210-0019-146985-0 from training. Duration: 22.48
2022-12-22 10:32:29,527 WARNING [train.py:1060] (2/4) Exclude cut with ID 7255-291500-0003-134302-0_sp0.9 from training. Duration: 29.816625
2022-12-22 10:32:30,171 WARNING [train.py:1060] (2/4) Exclude cut with ID 7255-291500-0005-134304-0_sp1.1 from training. Duration: 22.7590625
2022-12-22 10:32:30,343 WARNING [train.py:1060] (2/4) Exclude cut with ID 7357-94126-0026-15857-0 from training. Duration: 22.555
2022-12-22 10:32:31,576 WARNING [train.py:1060] (2/4) Exclude cut with ID 1250-135782-0005-25975-0_sp0.9 from training. Duration: 21.688875
2022-12-22 10:32:32,687 WARNING [train.py:1060] (2/4) Exclude cut with ID 3488-85273-0038-41224-0_sp0.9 from training. Duration: 22.6
2022-12-22 10:32:33,883 WARNING [train.py:1060] (2/4) Exclude cut with ID 7357-94126-0009-15840-0 from training. Duration: 24.32
2022-12-22 10:32:36,316 WARNING [train.py:1060] (2/4) Exclude cut with ID 8291-276745-0093-13116-0_sp0.9 from training. Duration: 21.061125
2022-12-22 10:32:36,783 WARNING [train.py:1060] (2/4) Exclude cut with ID 7357-94126-0024-15855-0_sp0.9 from training. Duration: 20.32225
2022-12-22 10:32:37,183 WARNING [train.py:1060] (2/4) Exclude cut with ID 3033-130750-0096-55598-0_sp1.1 from training. Duration: 0.7545625
2022-12-22 10:32:37,680 WARNING [train.py:1060] (2/4) Exclude cut with ID 4295-39940-0007-92567-0_sp0.9 from training. Duration: 23.9333125
2022-12-22 10:32:39,321 WARNING [train.py:1060] (2/4) Exclude cut with ID 7255-291500-0008-134307-0_sp1.1 from training. Duration: 20.17275
2022-12-22 10:32:39,503 WARNING [train.py:1060] (2/4) Exclude cut with ID 6978-92210-0019-146985-0_sp1.1 from training. Duration: 20.436375
2022-12-22 10:32:42,082 WARNING [train.py:1060] (2/4) Exclude cut with ID 4234-40345-0022-142709-0_sp0.9 from training. Duration: 23.1055625
2022-12-22 10:32:42,153 WARNING [train.py:1060] (2/4) Exclude cut with ID 8291-282929-0007-12994-0_sp1.1 from training. Duration: 23.5
2022-12-22 10:32:42,494 WARNING [train.py:1060] (2/4) Exclude cut with ID 7255-291500-0009-134308-0_sp0.9 from training. Duration: 26.62775
2022-12-22 10:32:43,479 WARNING [train.py:1060] (2/4) Exclude cut with ID 6951-79737-0018-132285-0 from training. Duration: 21.105
2022-12-22 10:32:43,623 WARNING [train.py:1060] (2/4) Exclude cut with ID 4511-76322-0006-80011-0_sp0.9 from training. Duration: 24.411125
2022-12-22 10:32:44,476 WARNING [train.py:1060] (2/4) Exclude cut with ID 6758-72288-0033-108368-0_sp1.1 from training. Duration: 21.263625
2022-12-22 10:32:45,504 WARNING [train.py:1060] (2/4) Exclude cut with ID 4234-40345-0022-142709-0 from training. Duration: 20.795
2022-12-22 10:32:45,839 WARNING [train.py:1060] (2/4) Exclude cut with ID 7357-94126-0021-15852-0 from training. Duration: 24.76
2022-12-22 10:32:45,861 WARNING [train.py:1060] (2/4) Exclude cut with ID 3867-173237-0077-144769-0_sp0.9 from training. Duration: 22.25
2022-12-22 10:32:47,224 WARNING [train.py:1060] (2/4) Exclude cut with ID 7357-94126-0026-15857-0_sp1.1 from training. Duration: 20.5045625
2022-12-22 10:33:08,508 INFO [train.py:1234] (2/4) Maximum memory allocated so far is 20231MB
2022-12-22 10:33:11,368 INFO [train.py:1234] (2/4) Maximum memory allocated so far is 20231MB
2022-12-22 10:33:15,384 INFO [train.py:1234] (2/4) Maximum memory allocated so far is 20231MB
2022-12-22 10:33:18,697 INFO [train.py:1234] (2/4) Maximum memory allocated so far is 20231MB
2022-12-22 10:33:23,719 INFO [train.py:1234] (2/4) Maximum memory allocated so far is 22212MB
2022-12-22 10:33:27,333 INFO [train.py:1234] (2/4) Maximum memory allocated so far is 22490MB
2022-12-22 10:33:38,265 WARNING [train.py:1060] (2/4) Exclude cut with ID 7859-102521-0017-7548-0_sp1.1 from training. Duration: 22.2954375
2022-12-22 10:33:43,146 INFO [train.py:894] (2/4) Epoch 1, batch 0, loss[loss=7.405, simple_loss=6.709, pruned_loss=6.951, over 18696.00 frames. ], tot_loss[loss=7.405, simple_loss=6.709, pruned_loss=6.951, over 18696.00 frames. ], batch size: 46, lr: 2.50e-02, grad_scale: 2.0
2022-12-22 10:33:43,146 INFO [train.py:919] (2/4) Computing validation loss
2022-12-22 10:33:55,185 INFO [train.py:928] (2/4) Epoch 1, validation: loss=6.913, simple_loss=6.239, pruned_loss=6.726, over 944034.00 frames. 
2022-12-22 10:33:55,186 INFO [train.py:929] (2/4) Maximum memory allocated so far is 22490MB
2022-12-22 10:33:56,608 INFO [scaling.py:679] (2/4) Whitening: num_groups=8, num_channels=96, metric=3.24 vs. limit=2.0
2022-12-22 10:33:59,096 INFO [zipformer.py:660] (2/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=5.0, num_to_drop=2, layers_to_drop={0, 1}
2022-12-22 10:34:01,173 INFO [scaling.py:679] (2/4) Whitening: num_groups=8, num_channels=96, metric=4.67 vs. limit=2.0
2022-12-22 10:34:08,638 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([5.0705, 4.8521, 5.1115, 5.1068, 5.1205, 5.1101, 5.0901, 5.0902],
       device='cuda:2'), covar=tensor([0.0071, 0.0050, 0.0175, 0.0088, 0.0071, 0.0082, 0.0036, 0.0078],
       device='cuda:2'), in_proj_covar=tensor([0.0009, 0.0009, 0.0009, 0.0009, 0.0009, 0.0009, 0.0009, 0.0009],
       device='cuda:2'), out_proj_covar=tensor([9.0315e-06, 9.0522e-06, 8.9825e-06, 8.8194e-06, 9.0207e-06, 8.8961e-06,
        8.7820e-06, 8.9276e-06], device='cuda:2')
2022-12-22 10:34:17,401 INFO [zipformer.py:660] (2/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=23.0, num_to_drop=1, layers_to_drop={0}
2022-12-22 10:34:25,696 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([4.5480, 4.5508, 4.4833, 4.4230, 4.5490, 4.5485, 4.5080, 4.4955],
       device='cuda:2'), covar=tensor([0.0023, 0.0013, 0.0014, 0.0016, 0.0025, 0.0014, 0.0016, 0.0017],
       device='cuda:2'), in_proj_covar=tensor([0.0009, 0.0009, 0.0009, 0.0009, 0.0009, 0.0009, 0.0009, 0.0009],
       device='cuda:2'), out_proj_covar=tensor([9.0208e-06, 8.9320e-06, 9.2049e-06, 8.9907e-06, 9.2164e-06, 9.0287e-06,
        9.1265e-06, 9.1130e-06], device='cuda:2')
2022-12-22 10:34:31,890 WARNING [train.py:1060] (2/4) Exclude cut with ID 298-126791-0067-24026-0_sp0.9 from training. Duration: 21.438875
2022-12-22 10:34:34,415 WARNING [train.py:1060] (2/4) Exclude cut with ID 5652-39938-0025-23684-0_sp0.9 from training. Duration: 22.2055625
2022-12-22 10:34:39,248 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([6.0589, 6.0258, 6.0081, 6.0442, 6.0515, 6.0581, 6.0554, 6.0142],
       device='cuda:2'), covar=tensor([0.0005, 0.0006, 0.0008, 0.0008, 0.0005, 0.0009, 0.0008, 0.0007],
       device='cuda:2'), in_proj_covar=tensor([0.0009, 0.0009, 0.0009, 0.0009, 0.0009, 0.0009, 0.0009, 0.0009],
       device='cuda:2'), out_proj_covar=tensor([8.9683e-06, 9.0940e-06, 8.9387e-06, 9.1167e-06, 8.9277e-06, 9.0123e-06,
        9.0311e-06, 9.1188e-06], device='cuda:2')
2022-12-22 10:34:43,579 INFO [train.py:894] (2/4) Epoch 1, batch 50, loss[loss=1.417, simple_loss=1.255, pruned_loss=1.448, over 18585.00 frames. ], tot_loss[loss=2.155, simple_loss=1.95, pruned_loss=1.969, over 836836.59 frames. ], batch size: 69, lr: 2.75e-02, grad_scale: 1.0
2022-12-22 10:34:46,932 INFO [scaling.py:679] (2/4) Whitening: num_groups=1, num_channels=384, metric=47.73 vs. limit=5.0
2022-12-22 10:35:05,905 INFO [scaling.py:679] (2/4) Whitening: num_groups=8, num_channels=96, metric=3.77 vs. limit=2.0
2022-12-22 10:35:15,744 INFO [zipformer.py:660] (2/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=83.0, num_to_drop=1, layers_to_drop={0}
2022-12-22 10:35:29,304 INFO [scaling.py:679] (2/4) Whitening: num_groups=1, num_channels=384, metric=49.95 vs. limit=5.0
2022-12-22 10:35:32,558 INFO [train.py:894] (2/4) Epoch 1, batch 100, loss[loss=1.117, simple_loss=0.9525, pruned_loss=1.297, over 18414.00 frames. ], tot_loss[loss=1.628, simple_loss=1.448, pruned_loss=1.614, over 1474761.69 frames. ], batch size: 48, lr: 3.00e-02, grad_scale: 0.0625
2022-12-22 10:35:38,778 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 6.402e+01 1.918e+02 5.720e+02 1.444e+03 1.131e+05, threshold=1.144e+03, percent-clipped=0.0
2022-12-22 10:36:05,381 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([2.2645, 2.2644, 2.2645, 2.2645, 2.2645, 2.2644, 2.2645, 2.2642],
       device='cuda:2'), covar=tensor([0.0021, 0.0019, 0.0019, 0.0018, 0.0019, 0.0020, 0.0017, 0.0022],
       device='cuda:2'), in_proj_covar=tensor([0.0009, 0.0009, 0.0009, 0.0009, 0.0009, 0.0009, 0.0009, 0.0009],
       device='cuda:2'), out_proj_covar=tensor([8.9194e-06, 9.0101e-06, 8.9692e-06, 8.9141e-06, 9.2357e-06, 9.0197e-06,
        9.0543e-06, 9.0380e-06], device='cuda:2')
2022-12-22 10:36:16,929 INFO [zipformer.py:660] (2/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=144.0, num_to_drop=2, layers_to_drop={0, 3}
2022-12-22 10:36:23,422 INFO [train.py:894] (2/4) Epoch 1, batch 150, loss[loss=0.9355, simple_loss=0.7913, pruned_loss=1.036, over 18548.00 frames. ], tot_loss[loss=1.393, simple_loss=1.221, pruned_loss=1.439, over 1972605.03 frames. ], batch size: 47, lr: 3.25e-02, grad_scale: 0.0625
2022-12-22 10:36:29,783 WARNING [train.py:1060] (2/4) Exclude cut with ID 7859-102521-0017-7548-0 from training. Duration: 24.525
2022-12-22 10:36:48,293 INFO [scaling.py:679] (2/4) Whitening: num_groups=8, num_channels=192, metric=2.41 vs. limit=2.0
2022-12-22 10:36:53,887 WARNING [train.py:1060] (2/4) Exclude cut with ID 3699-47246-0007-3408-0_sp0.9 from training. Duration: 20.26675
2022-12-22 10:36:55,562 INFO [scaling.py:679] (2/4) Whitening: num_groups=8, num_channels=96, metric=1.76 vs. limit=2.0
2022-12-22 10:36:58,301 INFO [scaling.py:679] (2/4) Whitening: num_groups=1, num_channels=384, metric=14.98 vs. limit=5.0
2022-12-22 10:37:02,447 WARNING [train.py:1060] (2/4) Exclude cut with ID 7859-102521-0017-7548-0_sp0.9 from training. Duration: 27.25
2022-12-22 10:37:03,593 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([3.7231, 3.7232, 3.7232, 3.7230, 3.7232, 3.7231, 3.7232, 3.7231],
       device='cuda:2'), covar=tensor([0.0016, 0.0020, 0.0015, 0.0018, 0.0020, 0.0014, 0.0016, 0.0017],
       device='cuda:2'), in_proj_covar=tensor([0.0009, 0.0009, 0.0009, 0.0009, 0.0009, 0.0009, 0.0009, 0.0009],
       device='cuda:2'), out_proj_covar=tensor([8.9249e-06, 9.0457e-06, 8.9882e-06, 8.9077e-06, 9.2267e-06, 9.0199e-06,
        9.0547e-06, 9.0280e-06], device='cuda:2')
2022-12-22 10:37:13,469 INFO [train.py:894] (2/4) Epoch 1, batch 200, loss[loss=0.9107, simple_loss=0.7671, pruned_loss=0.9561, over 18398.00 frames. ], tot_loss[loss=1.25, simple_loss=1.086, pruned_loss=1.304, over 2358774.51 frames. ], batch size: 46, lr: 3.50e-02, grad_scale: 0.125
2022-12-22 10:37:19,084 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 8.110e+01 1.446e+02 1.990e+02 2.489e+02 9.125e+02, threshold=3.981e+02, percent-clipped=0.0
2022-12-22 10:37:45,258 INFO [scaling.py:679] (2/4) Whitening: num_groups=1, num_channels=384, metric=6.35 vs. limit=5.0
2022-12-22 10:37:50,328 WARNING [train.py:1060] (2/4) Exclude cut with ID 6426-64292-0017-15984-0 from training. Duration: 21.68
2022-12-22 10:37:57,854 WARNING [train.py:1060] (2/4) Exclude cut with ID 4278-13270-0007-59342-0 from training. Duration: 21.6300625
2022-12-22 10:38:03,581 INFO [train.py:894] (2/4) Epoch 1, batch 250, loss[loss=0.9539, simple_loss=0.801, pruned_loss=0.954, over 18577.00 frames. ], tot_loss[loss=1.164, simple_loss=1.003, pruned_loss=1.204, over 2659535.82 frames. ], batch size: 51, lr: 3.75e-02, grad_scale: 0.125
2022-12-22 10:38:13,595 WARNING [train.py:1060] (2/4) Exclude cut with ID 4278-13270-0007-59342-0_sp0.9 from training. Duration: 24.033375
2022-12-22 10:38:48,878 INFO [zipformer.py:660] (2/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=296.0, num_to_drop=1, layers_to_drop={1}
2022-12-22 10:38:51,352 WARNING [train.py:1060] (2/4) Exclude cut with ID 4278-13270-0009-59344-0 from training. Duration: 22.905
2022-12-22 10:38:52,223 WARNING [train.py:1060] (2/4) Exclude cut with ID 5622-44585-0006-90525-0_sp1.1 from training. Duration: 23.4318125
2022-12-22 10:38:52,375 INFO [zipformer.py:660] (2/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=300.0, num_to_drop=2, layers_to_drop={0, 3}
2022-12-22 10:38:53,077 INFO [train.py:894] (2/4) Epoch 1, batch 300, loss[loss=0.889, simple_loss=0.7388, pruned_loss=0.8742, over 18540.00 frames. ], tot_loss[loss=1.098, simple_loss=0.9402, pruned_loss=1.123, over 2893120.39 frames. ], batch size: 47, lr: 4.00e-02, grad_scale: 0.25
2022-12-22 10:38:58,485 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 1.051e+02 1.633e+02 2.114e+02 2.790e+02 8.264e+02, threshold=4.229e+02, percent-clipped=8.0
2022-12-22 10:39:41,577 INFO [train.py:894] (2/4) Epoch 1, batch 350, loss[loss=0.8947, simple_loss=0.7418, pruned_loss=0.8449, over 18396.00 frames. ], tot_loss[loss=1.057, simple_loss=0.8986, pruned_loss=1.063, over 3075800.35 frames. ], batch size: 46, lr: 4.25e-02, grad_scale: 0.25
2022-12-22 10:39:47,513 INFO [zipformer.py:660] (2/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=357.0, num_to_drop=2, layers_to_drop={1, 3}
2022-12-22 10:40:09,209 WARNING [train.py:1060] (2/4) Exclude cut with ID 4278-13270-0009-59344-0_sp1.1 from training. Duration: 20.82275
2022-12-22 10:40:10,150 WARNING [train.py:1060] (2/4) Exclude cut with ID 4278-13270-0009-59344-0_sp0.9 from training. Duration: 25.45
2022-12-22 10:40:17,625 INFO [zipformer.py:660] (2/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=387.0, num_to_drop=1, layers_to_drop={0}
2022-12-22 10:40:31,697 INFO [train.py:894] (2/4) Epoch 1, batch 400, loss[loss=0.8907, simple_loss=0.7439, pruned_loss=0.7894, over 18456.00 frames. ], tot_loss[loss=1.026, simple_loss=0.8684, pruned_loss=1.01, over 3216684.37 frames. ], batch size: 50, lr: 4.50e-02, grad_scale: 0.5
2022-12-22 10:40:37,773 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 9.684e+01 1.284e+02 1.583e+02 1.935e+02 3.190e+02, threshold=3.166e+02, percent-clipped=0.0
2022-12-22 10:40:46,851 WARNING [train.py:1060] (2/4) Exclude cut with ID 5622-44585-0006-90525-0 from training. Duration: 25.775
2022-12-22 10:41:02,183 WARNING [train.py:1060] (2/4) Exclude cut with ID 2929-85685-0071-62375-0_sp0.9 from training. Duration: 22.25
2022-12-22 10:41:06,061 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([2.5888, 2.4476, 2.0739, 1.8009, 2.2774, 2.2388, 2.0790, 2.1494],
       device='cuda:2'), covar=tensor([0.1552, 0.4107, 0.1641, 0.1776, 0.3126, 0.2330, 0.3375, 0.1674],
       device='cuda:2'), in_proj_covar=tensor([0.0011, 0.0013, 0.0013, 0.0012, 0.0012, 0.0012, 0.0012, 0.0012],
       device='cuda:2'), out_proj_covar=tensor([1.0758e-05, 1.2250e-05, 1.1263e-05, 1.1089e-05, 1.1496e-05, 1.1333e-05,
        1.1530e-05, 1.1023e-05], device='cuda:2')
2022-12-22 10:41:09,529 INFO [zipformer.py:660] (2/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=439.0, num_to_drop=2, layers_to_drop={1, 2}
2022-12-22 10:41:18,120 INFO [zipformer.py:660] (2/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=448.0, num_to_drop=2, layers_to_drop={0, 2}
2022-12-22 10:41:18,768 WARNING [train.py:1060] (2/4) Exclude cut with ID 3972-170212-0014-23379-0 from training. Duration: 26.205
2022-12-22 10:41:20,481 INFO [train.py:894] (2/4) Epoch 1, batch 450, loss[loss=0.8812, simple_loss=0.7357, pruned_loss=0.7541, over 18673.00 frames. ], tot_loss[loss=1.002, simple_loss=0.8457, pruned_loss=0.9583, over 3327844.02 frames. ], batch size: 48, lr: 4.75e-02, grad_scale: 0.5
2022-12-22 10:41:29,734 WARNING [train.py:1060] (2/4) Exclude cut with ID 5239-32139-0047-9341-0_sp0.9 from training. Duration: 30.1555625
2022-12-22 10:41:34,470 WARNING [train.py:1060] (2/4) Exclude cut with ID 1265-135635-0050-6781-0_sp0.9 from training. Duration: 21.8333125
2022-12-22 10:41:41,167 WARNING [train.py:1060] (2/4) Exclude cut with ID 1914-133440-0024-94914-0_sp1.1 from training. Duration: 20.6545625
2022-12-22 10:42:05,205 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.3843, 1.9162, 1.3384, 1.4477, 2.6054, 1.2706, 2.0562, 1.2388],
       device='cuda:2'), covar=tensor([0.3884, 0.6968, 0.8296, 0.4015, 0.3144, 0.8670, 0.6118, 0.8259],
       device='cuda:2'), in_proj_covar=tensor([0.0019, 0.0023, 0.0025, 0.0020, 0.0018, 0.0024, 0.0022, 0.0023],
       device='cuda:2'), out_proj_covar=tensor([1.7327e-05, 2.0769e-05, 2.1193e-05, 1.7962e-05, 1.7600e-05, 2.3020e-05,
        2.0805e-05, 2.1275e-05], device='cuda:2')
2022-12-22 10:42:08,597 WARNING [train.py:1060] (2/4) Exclude cut with ID 7395-89880-0045-39920-0_sp0.9 from training. Duration: 20.52225
2022-12-22 10:42:10,413 INFO [train.py:894] (2/4) Epoch 1, batch 500, loss[loss=0.8982, simple_loss=0.7581, pruned_loss=0.7223, over 18531.00 frames. ], tot_loss[loss=0.972, simple_loss=0.8204, pruned_loss=0.8999, over 3413043.98 frames. ], batch size: 55, lr: 4.99e-02, grad_scale: 1.0
2022-12-22 10:42:15,914 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 1.309e+02 2.085e+02 3.413e+02 4.603e+02 1.381e+03, threshold=6.826e+02, percent-clipped=53.0
2022-12-22 10:42:22,162 WARNING [train.py:1060] (2/4) Exclude cut with ID 3972-170212-0014-23379-0_sp0.9 from training. Duration: 29.1166875
2022-12-22 10:42:59,354 WARNING [train.py:1060] (2/4) Exclude cut with ID 543-133211-0007-59831-0_sp0.9 from training. Duration: 21.388875
2022-12-22 10:43:00,257 INFO [train.py:894] (2/4) Epoch 1, batch 550, loss[loss=0.7684, simple_loss=0.6624, pruned_loss=0.5669, over 18439.00 frames. ], tot_loss[loss=0.9426, simple_loss=0.7972, pruned_loss=0.8416, over 3480784.23 frames. ], batch size: 48, lr: 4.98e-02, grad_scale: 1.0
2022-12-22 10:43:12,084 INFO [zipformer.py:660] (2/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=562.0, num_to_drop=1, layers_to_drop={1}
2022-12-22 10:43:17,522 INFO [zipformer.py:660] (2/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=568.0, num_to_drop=1, layers_to_drop={0}
2022-12-22 10:43:23,276 INFO [scaling.py:679] (2/4) Whitening: num_groups=1, num_channels=384, metric=5.58 vs. limit=5.0
2022-12-22 10:43:23,653 WARNING [train.py:1060] (2/4) Exclude cut with ID 1914-133440-0024-94914-0 from training. Duration: 22.72
2022-12-22 10:43:24,751 WARNING [train.py:1060] (2/4) Exclude cut with ID 1914-133440-0031-94921-0_sp0.9 from training. Duration: 22.7444375
2022-12-22 10:43:38,703 INFO [zipformer.py:660] (2/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=590.0, num_to_drop=1, layers_to_drop={1}
2022-12-22 10:43:48,736 INFO [zipformer.py:660] (2/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=600.0, num_to_drop=2, layers_to_drop={0, 2}
2022-12-22 10:43:49,534 INFO [train.py:894] (2/4) Epoch 1, batch 600, loss[loss=0.7179, simple_loss=0.6219, pruned_loss=0.5113, over 18615.00 frames. ], tot_loss[loss=0.9045, simple_loss=0.7679, pruned_loss=0.7783, over 3532204.62 frames. ], batch size: 45, lr: 4.98e-02, grad_scale: 1.0
2022-12-22 10:43:52,502 WARNING [train.py:1060] (2/4) Exclude cut with ID 4133-6541-0027-40495-0_sp1.1 from training. Duration: 0.9681875
2022-12-22 10:43:55,944 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 3.031e+02 4.953e+02 7.048e+02 1.005e+03 3.491e+03, threshold=1.410e+03, percent-clipped=51.0
2022-12-22 10:43:55,980 WARNING [train.py:1060] (2/4) Exclude cut with ID 6330-62851-0022-91297-0_sp0.9 from training. Duration: 22.3166875
2022-12-22 10:43:58,845 WARNING [train.py:1060] (2/4) Exclude cut with ID 543-133212-0015-59917-0_sp0.9 from training. Duration: 21.8166875
2022-12-22 10:44:03,407 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([0.5965, 0.4312, 0.5193, 0.5881, 0.5345, 0.7785, 0.7254, 0.5505],
       device='cuda:2'), covar=tensor([1.1821, 1.8376, 1.3286, 1.1636, 1.1063, 0.9277, 1.0125, 1.3237],
       device='cuda:2'), in_proj_covar=tensor([0.0033, 0.0036, 0.0034, 0.0033, 0.0032, 0.0032, 0.0032, 0.0038],
       device='cuda:2'), out_proj_covar=tensor([2.8583e-05, 3.5432e-05, 2.9179e-05, 2.8325e-05, 3.0450e-05, 2.8031e-05,
        2.8295e-05, 3.2890e-05], device='cuda:2')
2022-12-22 10:44:12,229 INFO [zipformer.py:660] (2/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=623.0, num_to_drop=2, layers_to_drop={0, 2}
2022-12-22 10:44:18,482 INFO [zipformer.py:660] (2/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=629.0, num_to_drop=2, layers_to_drop={0, 3}
2022-12-22 10:44:34,017 INFO [scaling.py:679] (2/4) Whitening: num_groups=1, num_channels=384, metric=5.92 vs. limit=5.0
2022-12-22 10:44:35,817 INFO [zipformer.py:660] (2/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=648.0, num_to_drop=1, layers_to_drop={1}
2022-12-22 10:44:38,610 INFO [train.py:894] (2/4) Epoch 1, batch 650, loss[loss=0.7952, simple_loss=0.693, pruned_loss=0.5469, over 18549.00 frames. ], tot_loss[loss=0.8662, simple_loss=0.7389, pruned_loss=0.7175, over 3573460.58 frames. ], batch size: 55, lr: 4.98e-02, grad_scale: 1.0
2022-12-22 10:44:38,843 INFO [zipformer.py:660] (2/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=651.0, num_to_drop=2, layers_to_drop={1, 2}
2022-12-22 10:44:39,600 INFO [zipformer.py:660] (2/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=652.0, num_to_drop=2, layers_to_drop={2, 3}
2022-12-22 10:45:05,370 WARNING [train.py:1060] (2/4) Exclude cut with ID 4957-30119-0041-23990-0_sp0.9 from training. Duration: 20.22775
2022-12-22 10:45:26,782 INFO [train.py:894] (2/4) Epoch 1, batch 700, loss[loss=0.6454, simple_loss=0.569, pruned_loss=0.4238, over 18579.00 frames. ], tot_loss[loss=0.8314, simple_loss=0.7131, pruned_loss=0.663, over 3604087.19 frames. ], batch size: 49, lr: 4.98e-02, grad_scale: 1.0
2022-12-22 10:45:31,262 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.2014, 1.7758, 2.1157, 1.2665, 1.2369, 2.5169, 1.8920, 1.3978],
       device='cuda:2'), covar=tensor([2.8368, 1.4250, 0.8034, 1.4830, 1.1566, 0.9023, 1.1616, 1.0292],
       device='cuda:2'), in_proj_covar=tensor([0.0057, 0.0047, 0.0043, 0.0046, 0.0050, 0.0051, 0.0045, 0.0046],
       device='cuda:2'), out_proj_covar=tensor([5.4882e-05, 4.8464e-05, 3.9014e-05, 4.4926e-05, 4.5621e-05, 4.7749e-05,
        4.7051e-05, 4.4185e-05], device='cuda:2')
2022-12-22 10:45:32,842 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 3.871e+02 5.831e+02 8.351e+02 1.310e+03 8.148e+03, threshold=1.670e+03, percent-clipped=20.0
2022-12-22 10:45:33,774 WARNING [train.py:1060] (2/4) Exclude cut with ID 5239-32139-0047-9341-0_sp1.1 from training. Duration: 24.67275
2022-12-22 10:45:52,876 WARNING [train.py:1060] (2/4) Exclude cut with ID 3082-165428-0081-50734-0_sp0.9 from training. Duration: 21.8055625
2022-12-22 10:46:03,907 INFO [zipformer.py:660] (2/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=739.0, num_to_drop=2, layers_to_drop={0, 2}
2022-12-22 10:46:07,520 INFO [zipformer.py:660] (2/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=743.0, num_to_drop=2, layers_to_drop={2, 3}
2022-12-22 10:46:14,945 INFO [train.py:894] (2/4) Epoch 1, batch 750, loss[loss=0.6581, simple_loss=0.5911, pruned_loss=0.4063, over 18531.00 frames. ], tot_loss[loss=0.7951, simple_loss=0.6862, pruned_loss=0.6108, over 3629547.61 frames. ], batch size: 58, lr: 4.97e-02, grad_scale: 1.0
2022-12-22 10:46:15,019 WARNING [train.py:1060] (2/4) Exclude cut with ID 3340-169293-0054-76830-0_sp0.9 from training. Duration: 22.6666875
2022-12-22 10:46:46,316 INFO [scaling.py:679] (2/4) Whitening: num_groups=1, num_channels=384, metric=5.22 vs. limit=5.0
2022-12-22 10:46:50,897 INFO [zipformer.py:660] (2/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=787.0, num_to_drop=1, layers_to_drop={1}
2022-12-22 10:46:56,439 WARNING [train.py:1060] (2/4) Exclude cut with ID 2411-132532-0017-82279-0_sp1.1 from training. Duration: 0.9681875
2022-12-22 10:47:04,406 INFO [train.py:894] (2/4) Epoch 1, batch 800, loss[loss=0.6425, simple_loss=0.5811, pruned_loss=0.386, over 18670.00 frames. ], tot_loss[loss=0.7625, simple_loss=0.6623, pruned_loss=0.5649, over 3648535.19 frames. ], batch size: 78, lr: 4.97e-02, grad_scale: 2.0
2022-12-22 10:47:10,444 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 3.426e+02 6.087e+02 8.345e+02 1.128e+03 4.122e+03, threshold=1.669e+03, percent-clipped=11.0
2022-12-22 10:47:13,292 WARNING [train.py:1060] (2/4) Exclude cut with ID 6330-62850-0007-91323-0 from training. Duration: 22.485
2022-12-22 10:47:38,748 WARNING [train.py:1060] (2/4) Exclude cut with ID 3972-170212-0014-23379-0_sp1.1 from training. Duration: 23.82275
2022-12-22 10:47:46,957 WARNING [train.py:1060] (2/4) Exclude cut with ID 4860-13185-0032-76709-0 from training. Duration: 20.77
2022-12-22 10:47:49,124 INFO [zipformer.py:660] (2/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=847.0, num_to_drop=1, layers_to_drop={0}
2022-12-22 10:47:52,633 INFO [train.py:894] (2/4) Epoch 1, batch 850, loss[loss=0.6557, simple_loss=0.5908, pruned_loss=0.3946, over 18726.00 frames. ], tot_loss[loss=0.7356, simple_loss=0.6428, pruned_loss=0.5265, over 3663225.62 frames. ], batch size: 52, lr: 4.96e-02, grad_scale: 2.0
2022-12-22 10:47:53,297 WARNING [train.py:1060] (2/4) Exclude cut with ID 6426-64292-0017-15984-0_sp0.9 from training. Duration: 24.088875
2022-12-22 10:48:13,413 WARNING [train.py:1060] (2/4) Exclude cut with ID 6330-62850-0007-91323-0_sp1.1 from training. Duration: 20.4409375
2022-12-22 10:48:27,438 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.2795, 1.0195, 0.9539, 0.9920, 1.1469, 0.9268, 1.2441, 0.9151],
       device='cuda:2'), covar=tensor([1.3976, 1.9201, 1.6537, 1.6786, 1.3895, 1.8295, 3.0117, 1.8490],
       device='cuda:2'), in_proj_covar=tensor([0.0027, 0.0024, 0.0026, 0.0025, 0.0024, 0.0030, 0.0031, 0.0028],
       device='cuda:2'), out_proj_covar=tensor([2.3342e-05, 2.2151e-05, 2.2286e-05, 2.1503e-05, 2.2275e-05, 2.6354e-05,
        2.9937e-05, 2.4709e-05], device='cuda:2')
2022-12-22 10:48:41,501 INFO [train.py:894] (2/4) Epoch 1, batch 900, loss[loss=0.574, simple_loss=0.5212, pruned_loss=0.3367, over 18689.00 frames. ], tot_loss[loss=0.7096, simple_loss=0.6238, pruned_loss=0.4919, over 3673352.55 frames. ], batch size: 46, lr: 4.96e-02, grad_scale: 2.0
2022-12-22 10:48:42,953 INFO [scaling.py:679] (2/4) Whitening: num_groups=1, num_channels=384, metric=4.71 vs. limit=5.0
2022-12-22 10:48:47,838 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 3.705e+02 7.532e+02 9.958e+02 1.427e+03 3.845e+03, threshold=1.992e+03, percent-clipped=19.0
2022-12-22 10:48:49,082 INFO [zipformer.py:660] (2/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=908.0, num_to_drop=2, layers_to_drop={0, 2}
2022-12-22 10:48:58,767 INFO [zipformer.py:660] (2/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=918.0, num_to_drop=2, layers_to_drop={0, 1}
2022-12-22 10:49:02,072 WARNING [train.py:1060] (2/4) Exclude cut with ID 2929-85685-0062-62366-0_sp0.9 from training. Duration: 22.511125
2022-12-22 10:49:02,366 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.6599, 1.1817, 1.3359, 1.4651, 1.5502, 1.3159, 1.4980, 1.3450],
       device='cuda:2'), covar=tensor([1.0968, 1.7745, 1.4511, 1.4316, 1.1909, 1.4688, 2.4510, 1.3953],
       device='cuda:2'), in_proj_covar=tensor([0.0027, 0.0025, 0.0026, 0.0025, 0.0025, 0.0030, 0.0031, 0.0028],
       device='cuda:2'), out_proj_covar=tensor([2.3193e-05, 2.2676e-05, 2.2892e-05, 2.1677e-05, 2.2590e-05, 2.6896e-05,
        2.9685e-05, 2.5360e-05], device='cuda:2')
2022-12-22 10:49:03,725 WARNING [train.py:1060] (2/4) Exclude cut with ID 7395-89880-0031-39906-0 from training. Duration: 20.675
2022-12-22 10:49:05,642 INFO [zipformer.py:660] (2/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=924.0, num_to_drop=2, layers_to_drop={0, 1}
2022-12-22 10:49:25,724 INFO [zipformer.py:660] (2/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=946.0, num_to_drop=2, layers_to_drop={1, 2}
2022-12-22 10:49:30,613 INFO [train.py:894] (2/4) Epoch 1, batch 950, loss[loss=0.545, simple_loss=0.495, pruned_loss=0.3177, over 18627.00 frames. ], tot_loss[loss=0.684, simple_loss=0.6058, pruned_loss=0.4587, over 3682458.96 frames. ], batch size: 45, lr: 4.96e-02, grad_scale: 2.0
2022-12-22 10:49:31,827 INFO [zipformer.py:660] (2/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=952.0, num_to_drop=2, layers_to_drop={0, 2}
2022-12-22 10:50:08,447 WARNING [train.py:1060] (2/4) Exclude cut with ID 6330-62850-0007-91323-0_sp0.9 from training. Duration: 24.9833125
2022-12-22 10:50:17,260 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.8788, 1.2893, 0.9609, 1.0364, 0.9014, 1.4582, 1.2541, 1.2494],
       device='cuda:2'), covar=tensor([0.7092, 1.5969, 1.0042, 1.1339, 0.9962, 0.7279, 1.1737, 1.5037],
       device='cuda:2'), in_proj_covar=tensor([0.0050, 0.0059, 0.0056, 0.0058, 0.0047, 0.0048, 0.0052, 0.0077],
       device='cuda:2'), out_proj_covar=tensor([4.4223e-05, 5.5619e-05, 4.9316e-05, 5.2081e-05, 4.4071e-05, 4.3285e-05,
        4.7534e-05, 7.2544e-05], device='cuda:2')
2022-12-22 10:50:19,377 INFO [zipformer.py:660] (2/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=1000.0, num_to_drop=1, layers_to_drop={1}
2022-12-22 10:50:20,153 INFO [train.py:894] (2/4) Epoch 1, batch 1000, loss[loss=0.5234, simple_loss=0.4869, pruned_loss=0.2877, over 18529.00 frames. ], tot_loss[loss=0.6623, simple_loss=0.5902, pruned_loss=0.4316, over 3688243.90 frames. ], batch size: 44, lr: 4.95e-02, grad_scale: 2.0
2022-12-22 10:50:26,434 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 4.198e+02 6.579e+02 9.009e+02 1.268e+03 5.908e+03, threshold=1.802e+03, percent-clipped=7.0
2022-12-22 10:50:28,391 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.5600, 1.1434, 0.9467, 1.1543, 1.4599, 1.2346, 1.1469, 1.2287],
       device='cuda:2'), covar=tensor([1.1607, 1.8677, 2.0983, 3.5587, 1.3304, 1.5206, 2.2821, 1.5081],
       device='cuda:2'), in_proj_covar=tensor([0.0027, 0.0030, 0.0039, 0.0050, 0.0029, 0.0033, 0.0032, 0.0032],
       device='cuda:2'), out_proj_covar=tensor([2.4204e-05, 2.8782e-05, 3.2784e-05, 4.8939e-05, 2.7200e-05, 3.0605e-05,
        3.1217e-05, 3.0109e-05], device='cuda:2')
2022-12-22 10:50:29,078 WARNING [train.py:1060] (2/4) Exclude cut with ID 5239-32139-0047-9341-0 from training. Duration: 27.14
2022-12-22 10:50:39,666 WARNING [train.py:1060] (2/4) Exclude cut with ID 2929-85685-0044-62348-0 from training. Duration: 22.44
2022-12-22 10:51:04,338 INFO [zipformer.py:660] (2/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=1043.0, num_to_drop=2, layers_to_drop={0, 2}
2022-12-22 10:51:12,175 INFO [train.py:894] (2/4) Epoch 1, batch 1050, loss[loss=0.6198, simple_loss=0.5631, pruned_loss=0.3573, over 18724.00 frames. ], tot_loss[loss=0.6432, simple_loss=0.577, pruned_loss=0.4075, over 3693424.28 frames. ], batch size: 52, lr: 4.95e-02, grad_scale: 2.0
2022-12-22 10:51:27,086 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.3813, 1.2399, 0.8955, 1.3662, 1.5625, 1.0614, 1.1856, 1.1918],
       device='cuda:2'), covar=tensor([0.9791, 1.2174, 1.9880, 2.8189, 1.0180, 1.3839, 1.7636, 1.2603],
       device='cuda:2'), in_proj_covar=tensor([0.0025, 0.0029, 0.0038, 0.0050, 0.0029, 0.0033, 0.0031, 0.0031],
       device='cuda:2'), out_proj_covar=tensor([2.3427e-05, 2.7263e-05, 3.3009e-05, 4.8438e-05, 2.7219e-05, 3.0864e-05,
        3.0108e-05, 3.0073e-05], device='cuda:2')
2022-12-22 10:51:36,138 WARNING [train.py:1060] (2/4) Exclude cut with ID 2929-85685-0060-62364-0_sp0.9 from training. Duration: 21.361125
2022-12-22 10:51:40,698 WARNING [train.py:1060] (2/4) Exclude cut with ID 2929-85685-0079-62383-0_sp1.1 from training. Duration: 27.0318125
2022-12-22 10:51:48,448 WARNING [train.py:1060] (2/4) Exclude cut with ID 5622-44585-0006-90525-0_sp0.9 from training. Duration: 28.638875
2022-12-22 10:51:54,219 INFO [zipformer.py:660] (2/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=1091.0, num_to_drop=1, layers_to_drop={1}
2022-12-22 10:51:58,225 WARNING [train.py:1060] (2/4) Exclude cut with ID 3340-169293-0054-76830-0 from training. Duration: 20.4
2022-12-22 10:52:04,448 INFO [train.py:894] (2/4) Epoch 1, batch 1100, loss[loss=0.4908, simple_loss=0.4708, pruned_loss=0.2512, over 18683.00 frames. ], tot_loss[loss=0.6234, simple_loss=0.563, pruned_loss=0.3848, over 3697414.70 frames. ], batch size: 46, lr: 4.94e-02, grad_scale: 2.0
2022-12-22 10:52:10,276 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 4.217e+02 6.903e+02 9.658e+02 1.239e+03 4.082e+03, threshold=1.932e+03, percent-clipped=8.0
2022-12-22 10:52:19,795 WARNING [train.py:1060] (2/4) Exclude cut with ID 2929-85685-0071-62375-0 from training. Duration: 20.025
2022-12-22 10:52:19,803 WARNING [train.py:1060] (2/4) Exclude cut with ID 2364-131735-0112-64612-0_sp0.9 from training. Duration: 20.488875
2022-12-22 10:52:24,889 WARNING [train.py:1060] (2/4) Exclude cut with ID 2929-85685-0079-62383-0 from training. Duration: 29.735
2022-12-22 10:52:36,395 INFO [scaling.py:679] (2/4) Whitening: num_groups=1, num_channels=384, metric=5.58 vs. limit=5.0
2022-12-22 10:52:57,055 INFO [train.py:894] (2/4) Epoch 1, batch 1150, loss[loss=0.5083, simple_loss=0.4802, pruned_loss=0.2691, over 18701.00 frames. ], tot_loss[loss=0.6053, simple_loss=0.5505, pruned_loss=0.3644, over 3701448.97 frames. ], batch size: 46, lr: 4.94e-02, grad_scale: 2.0
2022-12-22 10:52:59,434 INFO [zipformer.py:660] (2/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=1153.0, num_to_drop=1, layers_to_drop={0}
2022-12-22 10:53:20,529 WARNING [train.py:1060] (2/4) Exclude cut with ID 7276-92427-0014-12983-0_sp0.9 from training. Duration: 21.3055625
2022-12-22 10:53:21,426 WARNING [train.py:1060] (2/4) Exclude cut with ID 1025-75365-0008-79168-0_sp0.9 from training. Duration: 22.0666875
2022-12-22 10:53:49,545 INFO [train.py:894] (2/4) Epoch 1, batch 1200, loss[loss=0.5425, simple_loss=0.5107, pruned_loss=0.289, over 18528.00 frames. ], tot_loss[loss=0.5892, simple_loss=0.5393, pruned_loss=0.3469, over 3703102.79 frames. ], batch size: 52, lr: 4.93e-02, grad_scale: 4.0
2022-12-22 10:53:51,643 INFO [zipformer.py:660] (2/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=1203.0, num_to_drop=2, layers_to_drop={0, 3}
2022-12-22 10:53:55,438 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 4.627e+02 7.086e+02 8.753e+02 1.222e+03 5.052e+03, threshold=1.751e+03, percent-clipped=6.0
2022-12-22 10:54:02,993 INFO [zipformer.py:660] (2/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=1214.0, num_to_drop=2, layers_to_drop={0, 1}
2022-12-22 10:54:07,663 INFO [zipformer.py:660] (2/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=1218.0, num_to_drop=2, layers_to_drop={1, 2}
2022-12-22 10:54:14,035 INFO [zipformer.py:660] (2/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=1224.0, num_to_drop=1, layers_to_drop={0}
2022-12-22 10:54:34,886 WARNING [train.py:1060] (2/4) Exclude cut with ID 2929-85685-0062-62366-0 from training. Duration: 20.26
2022-12-22 10:54:36,268 INFO [zipformer.py:660] (2/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=1246.0, num_to_drop=1, layers_to_drop={1}
2022-12-22 10:54:40,926 INFO [train.py:894] (2/4) Epoch 1, batch 1250, loss[loss=0.5391, simple_loss=0.4999, pruned_loss=0.2949, over 18643.00 frames. ], tot_loss[loss=0.5737, simple_loss=0.5293, pruned_loss=0.3301, over 3704816.04 frames. ], batch size: 98, lr: 4.92e-02, grad_scale: 4.0
2022-12-22 10:54:43,654 WARNING [train.py:1060] (2/4) Exclude cut with ID 5239-32139-0030-9324-0_sp0.9 from training. Duration: 21.3444375
2022-12-22 10:54:56,279 INFO [zipformer.py:660] (2/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=1266.0, num_to_drop=1, layers_to_drop={1}
2022-12-22 10:55:03,197 INFO [zipformer.py:660] (2/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=1272.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 10:55:25,085 WARNING [train.py:1060] (2/4) Exclude cut with ID 497-129325-0061-62254-0_sp1.1 from training. Duration: 0.97725
2022-12-22 10:55:27,071 INFO [zipformer.py:660] (2/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=1294.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 10:55:33,991 INFO [train.py:894] (2/4) Epoch 1, batch 1300, loss[loss=0.5072, simple_loss=0.4882, pruned_loss=0.2593, over 18588.00 frames. ], tot_loss[loss=0.5606, simple_loss=0.5203, pruned_loss=0.3167, over 3707079.15 frames. ], batch size: 51, lr: 4.92e-02, grad_scale: 4.0
2022-12-22 10:55:39,952 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 3.931e+02 6.636e+02 8.494e+02 1.247e+03 3.025e+03, threshold=1.699e+03, percent-clipped=9.0
2022-12-22 10:55:53,630 WARNING [train.py:1060] (2/4) Exclude cut with ID 7395-89880-0031-39906-0_sp0.9 from training. Duration: 22.97225
2022-12-22 10:56:18,621 WARNING [train.py:1060] (2/4) Exclude cut with ID 7395-89880-0047-39922-0_sp0.9 from training. Duration: 21.97775
2022-12-22 10:56:27,899 INFO [train.py:894] (2/4) Epoch 1, batch 1350, loss[loss=0.4594, simple_loss=0.4568, pruned_loss=0.2216, over 18558.00 frames. ], tot_loss[loss=0.55, simple_loss=0.5139, pruned_loss=0.3051, over 3709009.47 frames. ], batch size: 49, lr: 4.91e-02, grad_scale: 4.0
2022-12-22 10:56:27,976 WARNING [train.py:1060] (2/4) Exclude cut with ID 1112-1043-0006-89194-0_sp0.9 from training. Duration: 21.8333125
2022-12-22 10:56:37,043 WARNING [train.py:1060] (2/4) Exclude cut with ID 1914-133440-0031-94921-0 from training. Duration: 20.47
2022-12-22 10:57:23,824 INFO [train.py:894] (2/4) Epoch 1, batch 1400, loss[loss=0.5306, simple_loss=0.5052, pruned_loss=0.2769, over 18634.00 frames. ], tot_loss[loss=0.538, simple_loss=0.506, pruned_loss=0.2936, over 3709571.81 frames. ], batch size: 53, lr: 4.91e-02, grad_scale: 4.0
2022-12-22 10:57:26,009 WARNING [train.py:1060] (2/4) Exclude cut with ID 7395-89880-0037-39912-0_sp0.9 from training. Duration: 20.67225
2022-12-22 10:57:30,217 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 3.771e+02 6.721e+02 8.735e+02 1.207e+03 3.329e+03, threshold=1.747e+03, percent-clipped=8.0
2022-12-22 10:57:40,222 WARNING [train.py:1060] (2/4) Exclude cut with ID 1914-133440-0024-94914-0_sp0.9 from training. Duration: 25.2444375
2022-12-22 10:57:52,476 INFO [scaling.py:679] (2/4) Whitening: num_groups=1, num_channels=384, metric=4.84 vs. limit=5.0
2022-12-22 10:57:59,666 WARNING [train.py:1060] (2/4) Exclude cut with ID 3340-169293-0021-76797-0_sp0.9 from training. Duration: 21.1445
2022-12-22 10:58:02,164 INFO [zipformer.py:660] (2/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=1434.0, num_to_drop=1, layers_to_drop={0}
2022-12-22 10:58:20,583 INFO [train.py:894] (2/4) Epoch 1, batch 1450, loss[loss=0.505, simple_loss=0.467, pruned_loss=0.2755, over 18491.00 frames. ], tot_loss[loss=0.5289, simple_loss=0.5004, pruned_loss=0.2847, over 3709954.81 frames. ], batch size: 43, lr: 4.90e-02, grad_scale: 4.0
2022-12-22 10:58:54,426 WARNING [train.py:1060] (2/4) Exclude cut with ID 2929-85685-0079-62383-0_sp0.9 from training. Duration: 33.038875
2022-12-22 10:59:11,428 INFO [zipformer.py:660] (2/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=1495.0, num_to_drop=2, layers_to_drop={1, 3}
2022-12-22 10:59:17,562 INFO [train.py:894] (2/4) Epoch 1, batch 1500, loss[loss=0.4752, simple_loss=0.4724, pruned_loss=0.2321, over 18460.00 frames. ], tot_loss[loss=0.521, simple_loss=0.4955, pruned_loss=0.277, over 3710176.11 frames. ], batch size: 50, lr: 4.89e-02, grad_scale: 4.0
2022-12-22 10:59:20,029 INFO [zipformer.py:660] (2/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=1503.0, num_to_drop=1, layers_to_drop={1}
2022-12-22 10:59:21,597 WARNING [train.py:1060] (2/4) Exclude cut with ID 6426-64291-0000-16059-0_sp0.9 from training. Duration: 20.0944375
2022-12-22 10:59:24,964 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 3.791e+02 8.013e+02 9.741e+02 1.259e+03 2.345e+03, threshold=1.948e+03, percent-clipped=8.0
2022-12-22 10:59:27,306 INFO [zipformer.py:660] (2/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=1509.0, num_to_drop=2, layers_to_drop={2, 3}
2022-12-22 10:59:32,771 WARNING [train.py:1060] (2/4) Exclude cut with ID 2929-85685-0044-62348-0_sp1.1 from training. Duration: 20.4
2022-12-22 10:59:38,377 WARNING [train.py:1060] (2/4) Exclude cut with ID 6330-62851-0022-91297-0 from training. Duration: 20.085
2022-12-22 10:59:48,177 WARNING [train.py:1060] (2/4) Exclude cut with ID 4860-13185-0032-76709-0_sp0.9 from training. Duration: 23.07775
2022-12-22 10:59:50,043 INFO [scaling.py:679] (2/4) Whitening: num_groups=8, num_channels=192, metric=1.84 vs. limit=2.0
2022-12-22 11:00:15,526 INFO [train.py:894] (2/4) Epoch 1, batch 1550, loss[loss=0.3865, simple_loss=0.3941, pruned_loss=0.182, over 18615.00 frames. ], tot_loss[loss=0.5114, simple_loss=0.4901, pruned_loss=0.2681, over 3712131.86 frames. ], batch size: 41, lr: 4.89e-02, grad_scale: 4.0
2022-12-22 11:00:15,688 INFO [zipformer.py:660] (2/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=1551.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 11:00:23,663 WARNING [train.py:1060] (2/4) Exclude cut with ID 2929-85685-0044-62348-0_sp0.9 from training. Duration: 24.9333125
2022-12-22 11:00:32,886 INFO [zipformer.py:660] (2/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=1566.0, num_to_drop=1, layers_to_drop={1}
2022-12-22 11:00:57,539 WARNING [train.py:1060] (2/4) Exclude cut with ID 5118-111612-0016-124680-0_sp0.9 from training. Duration: 20.388875
2022-12-22 11:00:59,242 INFO [scaling.py:679] (2/4) Whitening: num_groups=8, num_channels=192, metric=2.24 vs. limit=2.0
2022-12-22 11:01:01,002 WARNING [train.py:1060] (2/4) Exclude cut with ID 432-122774-0017-62487-0_sp1.1 from training. Duration: 20.3590625
2022-12-22 11:01:13,471 INFO [train.py:894] (2/4) Epoch 1, batch 1600, loss[loss=0.6046, simple_loss=0.5544, pruned_loss=0.3316, over 18464.00 frames. ], tot_loss[loss=0.5033, simple_loss=0.4847, pruned_loss=0.2615, over 3712037.99 frames. ], batch size: 64, lr: 4.88e-02, grad_scale: 8.0
2022-12-22 11:01:20,612 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 4.215e+02 7.636e+02 9.611e+02 1.412e+03 2.370e+03, threshold=1.922e+03, percent-clipped=7.0
2022-12-22 11:01:25,145 INFO [zipformer.py:660] (2/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=1611.0, num_to_drop=1, layers_to_drop={1}
2022-12-22 11:01:43,991 INFO [zipformer.py:660] (2/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=1627.0, num_to_drop=2, layers_to_drop={1, 3}
2022-12-22 11:01:48,755 INFO [scaling.py:679] (2/4) Whitening: num_groups=8, num_channels=96, metric=1.61 vs. limit=2.0
2022-12-22 11:01:54,271 WARNING [train.py:1060] (2/4) Exclude cut with ID 3557-8342-0013-54691-0_sp1.1 from training. Duration: 0.836375
2022-12-22 11:02:10,608 INFO [train.py:894] (2/4) Epoch 1, batch 1650, loss[loss=0.5338, simple_loss=0.476, pruned_loss=0.3017, over 18536.00 frames. ], tot_loss[loss=0.4994, simple_loss=0.4821, pruned_loss=0.2582, over 3711630.84 frames. ], batch size: 44, lr: 4.87e-02, grad_scale: 8.0
2022-12-22 11:02:25,730 WARNING [train.py:1060] (2/4) Exclude cut with ID 8565-290391-0049-67394-0_sp0.9 from training. Duration: 21.3166875
2022-12-22 11:02:35,840 INFO [zipformer.py:660] (2/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=1672.0, num_to_drop=2, layers_to_drop={0, 2}
2022-12-22 11:02:45,303 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([2.3310, 2.0303, 1.5672, 2.2392, 3.4325, 1.8840, 1.5539, 1.1257],
       device='cuda:2'), covar=tensor([0.1648, 0.2438, 0.2238, 0.1788, 0.0880, 0.2337, 0.4295, 0.3654],
       device='cuda:2'), in_proj_covar=tensor([0.0049, 0.0048, 0.0049, 0.0044, 0.0044, 0.0048, 0.0055, 0.0052],
       device='cuda:2'), out_proj_covar=tensor([4.6824e-05, 4.4265e-05, 4.2932e-05, 4.1103e-05, 3.7018e-05, 4.5275e-05,
        5.1763e-05, 4.8376e-05], device='cuda:2')
2022-12-22 11:02:50,713 WARNING [train.py:1060] (2/4) Exclude cut with ID 6533-399-0029-104863-0_sp0.9 from training. Duration: 22.1055625
2022-12-22 11:02:52,714 INFO [scaling.py:679] (2/4) Whitening: num_groups=1, num_channels=384, metric=4.56 vs. limit=5.0
2022-12-22 11:02:58,821 WARNING [train.py:1060] (2/4) Exclude cut with ID 7699-105389-0094-26379-0_sp1.1 from training. Duration: 21.77725
2022-12-22 11:03:09,738 INFO [train.py:894] (2/4) Epoch 1, batch 1700, loss[loss=0.4746, simple_loss=0.4504, pruned_loss=0.2492, over 18376.00 frames. ], tot_loss[loss=0.4962, simple_loss=0.4799, pruned_loss=0.2556, over 3711709.32 frames. ], batch size: 46, lr: 4.86e-02, grad_scale: 8.0
2022-12-22 11:03:13,201 WARNING [train.py:1060] (2/4) Exclude cut with ID 7255-291500-0005-134304-0_sp0.9 from training. Duration: 27.8166875
2022-12-22 11:03:17,159 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 4.788e+02 7.530e+02 9.640e+02 1.219e+03 2.819e+03, threshold=1.928e+03, percent-clipped=4.0
2022-12-22 11:03:33,995 WARNING [train.py:1060] (2/4) Exclude cut with ID 7357-94126-0021-15852-0_sp1.1 from training. Duration: 22.5090625
2022-12-22 11:03:38,380 WARNING [train.py:1060] (2/4) Exclude cut with ID 7255-291500-0005-134304-0 from training. Duration: 25.035
2022-12-22 11:03:53,553 WARNING [train.py:1060] (2/4) Exclude cut with ID 774-127930-0014-10412-0_sp1.1 from training. Duration: 0.95
2022-12-22 11:04:08,692 WARNING [train.py:1060] (2/4) Exclude cut with ID 3033-130750-0096-55598-0_sp0.9 from training. Duration: 0.92225
2022-12-22 11:04:09,843 INFO [train.py:894] (2/4) Epoch 1, batch 1750, loss[loss=0.4963, simple_loss=0.4684, pruned_loss=0.2623, over 18712.00 frames. ], tot_loss[loss=0.4975, simple_loss=0.4806, pruned_loss=0.2565, over 3713038.00 frames. ], batch size: 50, lr: 4.86e-02, grad_scale: 8.0
2022-12-22 11:04:29,854 WARNING [train.py:1060] (2/4) Exclude cut with ID 4511-76322-0006-80011-0 from training. Duration: 21.97
2022-12-22 11:04:44,010 WARNING [train.py:1060] (2/4) Exclude cut with ID 7492-105653-0055-62765-0_sp0.9 from training. Duration: 21.97225
2022-12-22 11:04:45,256 WARNING [train.py:1060] (2/4) Exclude cut with ID 453-131332-0000-47844-0_sp0.9 from training. Duration: 25.3333125
2022-12-22 11:04:53,016 WARNING [train.py:1060] (2/4) Exclude cut with ID 5172-29468-0015-19128-0_sp0.9 from training. Duration: 21.5055625
2022-12-22 11:04:56,685 INFO [zipformer.py:660] (2/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=1790.0, num_to_drop=1, layers_to_drop={2}
2022-12-22 11:05:00,934 WARNING [train.py:1060] (2/4) Exclude cut with ID 453-131332-0000-47844-0_sp1.1 from training. Duration: 20.72725
2022-12-22 11:05:08,949 INFO [train.py:894] (2/4) Epoch 1, batch 1800, loss[loss=0.4928, simple_loss=0.4737, pruned_loss=0.2553, over 18669.00 frames. ], tot_loss[loss=0.5015, simple_loss=0.4819, pruned_loss=0.26, over 3712352.18 frames. ], batch size: 48, lr: 4.85e-02, grad_scale: 8.0
2022-12-22 11:05:16,281 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 5.367e+02 7.495e+02 9.120e+02 1.278e+03 3.607e+03, threshold=1.824e+03, percent-clipped=3.0
2022-12-22 11:05:18,611 INFO [zipformer.py:660] (2/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=1809.0, num_to_drop=1, layers_to_drop={2}
2022-12-22 11:05:24,757 INFO [scaling.py:679] (2/4) Whitening: num_groups=1, num_channels=384, metric=4.08 vs. limit=5.0
2022-12-22 11:05:27,597 WARNING [train.py:1060] (2/4) Exclude cut with ID 8631-249866-0030-130156-0_sp0.9 from training. Duration: 26.32775
2022-12-22 11:05:51,130 WARNING [train.py:1060] (2/4) Exclude cut with ID 3867-173237-0077-144769-0 from training. Duration: 20.025
2022-12-22 11:05:53,674 INFO [zipformer.py:660] (2/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=1838.0, num_to_drop=1, layers_to_drop={1}
2022-12-22 11:05:55,593 WARNING [train.py:1060] (2/4) Exclude cut with ID 6709-74022-0004-86860-0_sp1.1 from training. Duration: 0.9409375
2022-12-22 11:05:55,599 WARNING [train.py:1060] (2/4) Exclude cut with ID 4757-1811-0023-62229-0_sp0.9 from training. Duration: 21.37775
2022-12-22 11:06:08,807 INFO [train.py:894] (2/4) Epoch 1, batch 1850, loss[loss=0.4401, simple_loss=0.4303, pruned_loss=0.2239, over 18607.00 frames. ], tot_loss[loss=0.502, simple_loss=0.482, pruned_loss=0.2605, over 3712908.18 frames. ], batch size: 45, lr: 4.84e-02, grad_scale: 8.0
2022-12-22 11:06:12,295 WARNING [train.py:1060] (2/4) Exclude cut with ID 1250-135782-0004-25974-0_sp0.9 from training. Duration: 21.17225
2022-12-22 11:06:12,308 WARNING [train.py:1060] (2/4) Exclude cut with ID 7357-94126-0021-15852-0_sp0.9 from training. Duration: 27.511125
2022-12-22 11:06:15,906 INFO [zipformer.py:660] (2/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=1857.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 11:06:16,131 INFO [zipformer.py:660] (2/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=1857.0, num_to_drop=1, layers_to_drop={0}
2022-12-22 11:06:32,159 INFO [zipformer.py:660] (2/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=1870.0, num_to_drop=1, layers_to_drop={0}
2022-12-22 11:06:38,685 WARNING [train.py:1060] (2/4) Exclude cut with ID 453-131332-0000-47844-0 from training. Duration: 22.8
2022-12-22 11:06:44,467 WARNING [train.py:1060] (2/4) Exclude cut with ID 4964-30587-0040-44509-0 from training. Duration: 22.585
2022-12-22 11:07:08,335 INFO [zipformer.py:660] (2/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=1899.0, num_to_drop=2, layers_to_drop={1, 3}
2022-12-22 11:07:09,403 WARNING [train.py:1060] (2/4) Exclude cut with ID 6978-92210-0001-146967-0_sp0.9 from training. Duration: 22.0166875
2022-12-22 11:07:10,542 INFO [train.py:894] (2/4) Epoch 1, batch 1900, loss[loss=0.5156, simple_loss=0.4801, pruned_loss=0.276, over 18532.00 frames. ], tot_loss[loss=0.5029, simple_loss=0.4828, pruned_loss=0.261, over 3713694.41 frames. ], batch size: 47, lr: 4.83e-02, grad_scale: 8.0
2022-12-22 11:07:13,642 INFO [scaling.py:679] (2/4) Whitening: num_groups=8, num_channels=192, metric=2.12 vs. limit=2.0
2022-12-22 11:07:17,481 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 4.376e+02 8.337e+02 1.120e+03 1.422e+03 4.063e+03, threshold=2.239e+03, percent-clipped=10.0
2022-12-22 11:07:22,989 WARNING [train.py:1060] (2/4) Exclude cut with ID 7255-291500-0003-134302-0_sp1.1 from training. Duration: 24.395375
2022-12-22 11:07:27,655 WARNING [train.py:1060] (2/4) Exclude cut with ID 3488-85273-0017-41203-0_sp0.9 from training. Duration: 27.47775
2022-12-22 11:07:30,881 WARNING [train.py:1060] (2/4) Exclude cut with ID 432-122774-0017-62487-0_sp0.9 from training. Duration: 24.8833125
2022-12-22 11:07:31,252 INFO [zipformer.py:660] (2/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=1918.0, num_to_drop=2, layers_to_drop={0, 2}
2022-12-22 11:07:32,218 WARNING [train.py:1060] (2/4) Exclude cut with ID 6758-72288-0033-108368-0 from training. Duration: 23.39
2022-12-22 11:07:36,573 INFO [zipformer.py:660] (2/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=1922.0, num_to_drop=1, layers_to_drop={3}
2022-12-22 11:07:37,537 WARNING [train.py:1060] (2/4) Exclude cut with ID 8291-282929-0007-12994-0_sp0.9 from training. Duration: 28.72225
2022-12-22 11:07:43,609 WARNING [train.py:1060] (2/4) Exclude cut with ID 585-294811-0110-133686-0_sp0.9 from training. Duration: 20.8944375
2022-12-22 11:07:47,946 INFO [zipformer.py:660] (2/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=1931.0, num_to_drop=2, layers_to_drop={1, 2}
2022-12-22 11:07:55,753 WARNING [train.py:1060] (2/4) Exclude cut with ID 5796-66357-0007-116447-0_sp0.9 from training. Duration: 23.8444375
2022-12-22 11:08:11,714 INFO [train.py:894] (2/4) Epoch 1, batch 1950, loss[loss=0.4478, simple_loss=0.4359, pruned_loss=0.2296, over 18682.00 frames. ], tot_loss[loss=0.4988, simple_loss=0.4803, pruned_loss=0.2582, over 3714396.21 frames. ], batch size: 46, lr: 4.83e-02, grad_scale: 8.0
2022-12-22 11:08:14,353 WARNING [train.py:1060] (2/4) Exclude cut with ID 8291-282929-0007-12994-0 from training. Duration: 25.85
2022-12-22 11:08:14,363 WARNING [train.py:1060] (2/4) Exclude cut with ID 8291-282929-0023-13010-0 from training. Duration: 21.39
2022-12-22 11:08:23,975 WARNING [train.py:1060] (2/4) Exclude cut with ID 7357-94126-0014-15845-0 from training. Duration: 27.92
2022-12-22 11:08:30,707 INFO [zipformer.py:660] (2/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=1967.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 11:08:46,685 WARNING [train.py:1060] (2/4) Exclude cut with ID 8631-249866-0039-130165-0_sp0.9 from training. Duration: 20.661125
2022-12-22 11:09:05,262 WARNING [train.py:1060] (2/4) Exclude cut with ID 7357-94126-0043-15874-0_sp0.9 from training. Duration: 20.07225
2022-12-22 11:09:11,382 WARNING [train.py:1060] (2/4) Exclude cut with ID 1085-156170-0017-128270-0 from training. Duration: 21.01
2022-12-22 11:09:16,247 INFO [train.py:894] (2/4) Epoch 1, batch 2000, loss[loss=0.4669, simple_loss=0.4385, pruned_loss=0.2477, over 18545.00 frames. ], tot_loss[loss=0.4969, simple_loss=0.479, pruned_loss=0.257, over 3714059.82 frames. ], batch size: 41, lr: 4.82e-02, grad_scale: 8.0
2022-12-22 11:09:24,808 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 4.889e+02 7.270e+02 9.125e+02 1.253e+03 3.242e+03, threshold=1.825e+03, percent-clipped=1.0
2022-12-22 11:10:17,835 WARNING [train.py:1060] (2/4) Exclude cut with ID 2195-150901-0045-59933-0 from training. Duration: 20.65
2022-12-22 11:10:24,436 INFO [train.py:894] (2/4) Epoch 1, batch 2050, loss[loss=0.4801, simple_loss=0.4645, pruned_loss=0.2478, over 18669.00 frames. ], tot_loss[loss=0.489, simple_loss=0.4747, pruned_loss=0.2514, over 3713603.32 frames. ], batch size: 48, lr: 4.81e-02, grad_scale: 8.0
2022-12-22 11:10:24,444 WARNING [train.py:1060] (2/4) Exclude cut with ID 5796-66357-0007-116447-0 from training. Duration: 21.46
2022-12-22 11:11:06,265 WARNING [train.py:1060] (2/4) Exclude cut with ID 3557-8342-0013-54691-0 from training. Duration: 0.92
2022-12-22 11:11:11,543 WARNING [train.py:1060] (2/4) Exclude cut with ID 8291-282929-0023-13010-0_sp0.9 from training. Duration: 23.7666875
2022-12-22 11:11:17,603 INFO [zipformer.py:660] (2/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=2090.0, num_to_drop=2, layers_to_drop={0, 1}
2022-12-22 11:11:20,229 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.4624, 1.4898, 1.6190, 1.5418, 1.3476, 1.4914, 1.2776, 2.1409],
       device='cuda:2'), covar=tensor([0.3238, 0.1977, 0.3474, 0.2388, 0.3633, 0.3036, 0.2703, 0.1856],
       device='cuda:2'), in_proj_covar=tensor([0.0066, 0.0061, 0.0103, 0.0064, 0.0078, 0.0088, 0.0062, 0.0071],
       device='cuda:2'), out_proj_covar=tensor([7.5606e-05, 7.0701e-05, 1.0711e-04, 7.1609e-05, 8.7229e-05, 9.4891e-05,
        7.2676e-05, 7.5402e-05], device='cuda:2')
2022-12-22 11:11:27,256 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([2.6306, 2.9727, 3.8772, 1.8931, 3.7845, 2.1935, 1.0721, 2.3006],
       device='cuda:2'), covar=tensor([0.1398, 0.0897, 0.0936, 0.2356, 0.0648, 0.1587, 0.4186, 0.2031],
       device='cuda:2'), in_proj_covar=tensor([0.0081, 0.0066, 0.0092, 0.0076, 0.0061, 0.0061, 0.0103, 0.0075],
       device='cuda:2'), out_proj_covar=tensor([9.1923e-05, 7.1030e-05, 1.0739e-04, 7.8875e-05, 6.9633e-05, 6.8908e-05,
        1.0100e-04, 8.3287e-05], device='cuda:2')
2022-12-22 11:11:27,722 INFO [scaling.py:679] (2/4) Whitening: num_groups=8, num_channels=192, metric=1.92 vs. limit=2.0
2022-12-22 11:11:32,146 INFO [train.py:894] (2/4) Epoch 1, batch 2100, loss[loss=0.4925, simple_loss=0.4854, pruned_loss=0.2498, over 18476.00 frames. ], tot_loss[loss=0.4808, simple_loss=0.4699, pruned_loss=0.2456, over 3713140.78 frames. ], batch size: 54, lr: 4.80e-02, grad_scale: 16.0
2022-12-22 11:11:39,983 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 4.126e+02 6.724e+02 7.895e+02 9.819e+02 2.004e+03, threshold=1.579e+03, percent-clipped=3.0
2022-12-22 11:11:44,956 WARNING [train.py:1060] (2/4) Exclude cut with ID 8544-281189-0060-101339-0_sp0.9 from training. Duration: 20.861125
2022-12-22 11:11:55,672 WARNING [train.py:1060] (2/4) Exclude cut with ID 3488-65654-0031-41259-0_sp0.9 from training. Duration: 22.711125
2022-12-22 11:12:20,898 INFO [zipformer.py:660] (2/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=2138.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 11:12:30,708 WARNING [train.py:1060] (2/4) Exclude cut with ID 6951-79737-0043-132310-0_sp1.1 from training. Duration: 22.986375
2022-12-22 11:12:39,160 INFO [train.py:894] (2/4) Epoch 1, batch 2150, loss[loss=0.4492, simple_loss=0.4611, pruned_loss=0.2186, over 18547.00 frames. ], tot_loss[loss=0.4774, simple_loss=0.4679, pruned_loss=0.2432, over 3713587.03 frames. ], batch size: 55, lr: 4.79e-02, grad_scale: 16.0
2022-12-22 11:12:44,446 WARNING [train.py:1060] (2/4) Exclude cut with ID 8040-260924-0003-80960-0_sp0.9 from training. Duration: 22.07225
2022-12-22 11:12:47,560 INFO [scaling.py:679] (2/4) Whitening: num_groups=8, num_channels=96, metric=1.93 vs. limit=2.0
2022-12-22 11:12:50,048 WARNING [train.py:1060] (2/4) Exclude cut with ID 7699-105389-0045-26330-0_sp0.9 from training. Duration: 20.3055625
2022-12-22 11:12:52,554 WARNING [train.py:1060] (2/4) Exclude cut with ID 6356-271890-0060-94317-0_sp0.9 from training. Duration: 20.72225
2022-12-22 11:13:06,255 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.8793, 1.8636, 1.5380, 1.6100, 1.4124, 1.5739, 1.3570, 2.7313],
       device='cuda:2'), covar=tensor([0.2692, 0.1678, 0.3324, 0.2188, 0.3396, 0.2998, 0.2931, 0.1213],
       device='cuda:2'), in_proj_covar=tensor([0.0069, 0.0062, 0.0107, 0.0067, 0.0080, 0.0093, 0.0065, 0.0073],
       device='cuda:2'), out_proj_covar=tensor([7.9815e-05, 7.3430e-05, 1.1319e-04, 7.6244e-05, 9.1226e-05, 1.0081e-04,
        7.7692e-05, 7.9534e-05], device='cuda:2')
2022-12-22 11:13:08,498 WARNING [train.py:1060] (2/4) Exclude cut with ID 3488-85273-0017-41203-0_sp1.1 from training. Duration: 22.4818125
2022-12-22 11:13:29,165 WARNING [train.py:1060] (2/4) Exclude cut with ID 4964-30587-0040-44509-0_sp0.9 from training. Duration: 25.0944375
2022-12-22 11:13:32,226 WARNING [train.py:1060] (2/4) Exclude cut with ID 6533-399-0047-104881-0 from training. Duration: 21.515
2022-12-22 11:13:37,452 WARNING [train.py:1060] (2/4) Exclude cut with ID 7357-94126-0009-15840-0_sp0.9 from training. Duration: 27.02225
2022-12-22 11:13:37,685 INFO [zipformer.py:660] (2/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=2194.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 11:13:42,804 WARNING [train.py:1060] (2/4) Exclude cut with ID 432-122774-0010-62480-0_sp0.9 from training. Duration: 22.22225
2022-12-22 11:13:46,887 INFO [train.py:894] (2/4) Epoch 1, batch 2200, loss[loss=0.4239, simple_loss=0.4427, pruned_loss=0.2026, over 18725.00 frames. ], tot_loss[loss=0.471, simple_loss=0.4645, pruned_loss=0.2386, over 3714360.58 frames. ], batch size: 52, lr: 4.78e-02, grad_scale: 16.0
2022-12-22 11:13:50,191 WARNING [train.py:1060] (2/4) Exclude cut with ID 4964-30587-0085-44554-0_sp0.9 from training. Duration: 20.85
2022-12-22 11:13:55,273 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 3.692e+02 7.033e+02 8.615e+02 1.094e+03 2.363e+03, threshold=1.723e+03, percent-clipped=9.0
2022-12-22 11:14:03,735 INFO [zipformer.py:660] (2/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=2213.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 11:14:15,512 INFO [zipformer.py:660] (2/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=2222.0, num_to_drop=2, layers_to_drop={1, 2}
2022-12-22 11:14:20,311 WARNING [train.py:1060] (2/4) Exclude cut with ID 4295-39940-0007-92567-0 from training. Duration: 21.54
2022-12-22 11:14:20,454 INFO [zipformer.py:660] (2/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=2226.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 11:14:24,633 WARNING [train.py:1060] (2/4) Exclude cut with ID 4964-30587-0040-44509-0_sp1.1 from training. Duration: 20.5318125
2022-12-22 11:14:33,789 WARNING [train.py:1060] (2/4) Exclude cut with ID 7255-291500-0012-134311-0_sp0.9 from training. Duration: 21.9333125
2022-12-22 11:14:53,776 INFO [train.py:894] (2/4) Epoch 1, batch 2250, loss[loss=0.3631, simple_loss=0.3849, pruned_loss=0.1707, over 18683.00 frames. ], tot_loss[loss=0.4636, simple_loss=0.4601, pruned_loss=0.2335, over 3714194.96 frames. ], batch size: 46, lr: 4.77e-02, grad_scale: 16.0
2022-12-22 11:15:15,722 WARNING [train.py:1060] (2/4) Exclude cut with ID 8631-249866-0025-130151-0_sp0.9 from training. Duration: 21.7944375
2022-12-22 11:15:15,965 INFO [zipformer.py:660] (2/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=2267.0, num_to_drop=1, layers_to_drop={2}
2022-12-22 11:15:19,875 INFO [zipformer.py:660] (2/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=2270.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 11:15:28,269 WARNING [train.py:1060] (2/4) Exclude cut with ID 8291-282929-0002-12989-0_sp0.9 from training. Duration: 22.4666875
2022-12-22 11:15:34,096 WARNING [train.py:1060] (2/4) Exclude cut with ID 6121-9014-0076-24124-0 from training. Duration: 21.635
2022-12-22 11:15:39,949 WARNING [train.py:1060] (2/4) Exclude cut with ID 6121-9014-0076-24124-0_sp0.9 from training. Duration: 24.038875
2022-12-22 11:16:01,172 INFO [train.py:894] (2/4) Epoch 1, batch 2300, loss[loss=0.3797, simple_loss=0.402, pruned_loss=0.1786, over 18460.00 frames. ], tot_loss[loss=0.4583, simple_loss=0.4567, pruned_loss=0.2299, over 3714226.02 frames. ], batch size: 50, lr: 4.77e-02, grad_scale: 16.0
2022-12-22 11:16:08,972 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 4.591e+02 7.252e+02 9.061e+02 1.161e+03 2.853e+03, threshold=1.812e+03, percent-clipped=4.0
2022-12-22 11:16:17,711 WARNING [train.py:1060] (2/4) Exclude cut with ID 7255-291500-0009-134308-0_sp1.1 from training. Duration: 21.786375
2022-12-22 11:16:20,584 INFO [zipformer.py:660] (2/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=2315.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 11:16:27,677 WARNING [train.py:1060] (2/4) Exclude cut with ID 8291-282929-0002-12989-0 from training. Duration: 20.22
2022-12-22 11:16:36,181 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.7988, 1.2582, 1.0834, 1.9124, 1.7587, 2.2057, 1.9902, 1.2513],
       device='cuda:2'), covar=tensor([0.2041, 0.2568, 0.3407, 0.2209, 0.1679, 0.1504, 0.1559, 0.3260],
       device='cuda:2'), in_proj_covar=tensor([0.0070, 0.0062, 0.0082, 0.0071, 0.0063, 0.0074, 0.0061, 0.0073],
       device='cuda:2'), out_proj_covar=tensor([8.5102e-05, 7.4540e-05, 9.6135e-05, 8.9095e-05, 7.3853e-05, 9.4112e-05,
        7.4785e-05, 8.2946e-05], device='cuda:2')
2022-12-22 11:16:54,427 INFO [zipformer.py:660] (2/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=2340.0, num_to_drop=1, layers_to_drop={1}
2022-12-22 11:17:09,286 INFO [train.py:894] (2/4) Epoch 1, batch 2350, loss[loss=0.4664, simple_loss=0.4683, pruned_loss=0.2323, over 18692.00 frames. ], tot_loss[loss=0.4552, simple_loss=0.4545, pruned_loss=0.2279, over 3714581.70 frames. ], batch size: 69, lr: 4.76e-02, grad_scale: 16.0
2022-12-22 11:18:16,102 WARNING [train.py:1060] (2/4) Exclude cut with ID 6951-79737-0043-132310-0 from training. Duration: 25.285
2022-12-22 11:18:17,757 INFO [train.py:894] (2/4) Epoch 1, batch 2400, loss[loss=0.3983, simple_loss=0.4001, pruned_loss=0.1982, over 18418.00 frames. ], tot_loss[loss=0.452, simple_loss=0.4534, pruned_loss=0.2253, over 3714234.45 frames. ], batch size: 42, lr: 4.75e-02, grad_scale: 16.0
2022-12-22 11:18:18,204 INFO [zipformer.py:660] (2/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=2401.0, num_to_drop=2, layers_to_drop={1, 2}
2022-12-22 11:18:25,231 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 5.700e+02 7.707e+02 9.733e+02 1.236e+03 3.171e+03, threshold=1.947e+03, percent-clipped=7.0
2022-12-22 11:19:15,071 WARNING [train.py:1060] (2/4) Exclude cut with ID 811-130148-0001-63453-0_sp0.9 from training. Duration: 20.861125
2022-12-22 11:19:25,848 INFO [train.py:894] (2/4) Epoch 1, batch 2450, loss[loss=0.4325, simple_loss=0.4377, pruned_loss=0.2137, over 18598.00 frames. ], tot_loss[loss=0.4498, simple_loss=0.452, pruned_loss=0.2238, over 3712994.79 frames. ], batch size: 45, lr: 4.74e-02, grad_scale: 16.0
2022-12-22 11:19:33,841 WARNING [train.py:1060] (2/4) Exclude cut with ID 6010-56788-0055-90261-0 from training. Duration: 20.88
2022-12-22 11:20:04,220 WARNING [train.py:1060] (2/4) Exclude cut with ID 7357-94126-0045-15876-0_sp0.9 from training. Duration: 23.4166875
2022-12-22 11:20:24,699 INFO [zipformer.py:660] (2/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=2494.0, num_to_drop=1, layers_to_drop={2}
2022-12-22 11:20:34,033 INFO [train.py:894] (2/4) Epoch 1, batch 2500, loss[loss=0.5152, simple_loss=0.4908, pruned_loss=0.2698, over 18582.00 frames. ], tot_loss[loss=0.443, simple_loss=0.4476, pruned_loss=0.2191, over 3712645.12 frames. ], batch size: 56, lr: 4.73e-02, grad_scale: 16.0
2022-12-22 11:20:41,410 INFO [scaling.py:679] (2/4) Whitening: num_groups=8, num_channels=192, metric=2.05 vs. limit=2.0
2022-12-22 11:20:41,986 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 3.819e+02 7.053e+02 8.815e+02 1.099e+03 3.368e+03, threshold=1.763e+03, percent-clipped=3.0
2022-12-22 11:20:50,388 INFO [zipformer.py:660] (2/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=2513.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 11:21:08,528 INFO [zipformer.py:660] (2/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=2526.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 11:21:12,473 WARNING [train.py:1060] (2/4) Exclude cut with ID 7255-291500-0006-134305-0 from training. Duration: 21.24
2022-12-22 11:21:12,488 WARNING [train.py:1060] (2/4) Exclude cut with ID 6533-399-0047-104881-0_sp0.9 from training. Duration: 23.9055625
2022-12-22 11:21:31,254 INFO [zipformer.py:660] (2/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=2542.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 11:21:42,363 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([0.9082, 1.6402, 2.0472, 1.3441, 2.1242, 2.0953, 1.6878, 1.4311],
       device='cuda:2'), covar=tensor([0.2304, 0.1460, 0.0719, 0.1791, 0.0826, 0.0526, 0.1079, 0.1081],
       device='cuda:2'), in_proj_covar=tensor([0.0079, 0.0077, 0.0058, 0.0089, 0.0066, 0.0055, 0.0075, 0.0065],
       device='cuda:2'), out_proj_covar=tensor([8.2365e-05, 7.3698e-05, 5.4519e-05, 8.4441e-05, 6.9936e-05, 5.1049e-05,
        7.2411e-05, 6.1122e-05], device='cuda:2')
2022-12-22 11:21:43,232 INFO [train.py:894] (2/4) Epoch 1, batch 2550, loss[loss=0.3756, simple_loss=0.3919, pruned_loss=0.1796, over 18449.00 frames. ], tot_loss[loss=0.4418, simple_loss=0.4473, pruned_loss=0.2181, over 3712570.01 frames. ], batch size: 43, lr: 4.72e-02, grad_scale: 16.0
2022-12-22 11:21:43,308 WARNING [train.py:1060] (2/4) Exclude cut with ID 6758-72288-0033-108368-0_sp0.9 from training. Duration: 25.988875
2022-12-22 11:21:51,052 WARNING [train.py:1060] (2/4) Exclude cut with ID 7255-291500-0001-134300-0_sp0.9 from training. Duration: 20.67225
2022-12-22 11:21:57,782 INFO [zipformer.py:660] (2/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=2561.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 11:22:15,421 INFO [zipformer.py:660] (2/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=2574.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 11:22:35,017 WARNING [train.py:1060] (2/4) Exclude cut with ID 3488-85273-0038-41224-0 from training. Duration: 20.34
2022-12-22 11:22:37,845 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([4.4656, 3.6598, 3.6989, 3.9684, 3.9295, 3.8782, 4.4230, 1.3860],
       device='cuda:2'), covar=tensor([0.0328, 0.0524, 0.0614, 0.0425, 0.1114, 0.0497, 0.0426, 0.3943],
       device='cuda:2'), in_proj_covar=tensor([0.0073, 0.0084, 0.0087, 0.0062, 0.0107, 0.0074, 0.0091, 0.0140],
       device='cuda:2'), out_proj_covar=tensor([7.2660e-05, 8.5702e-05, 9.0442e-05, 6.1242e-05, 9.7910e-05, 7.8135e-05,
        9.5266e-05, 1.2304e-04], device='cuda:2')
2022-12-22 11:22:52,224 INFO [train.py:894] (2/4) Epoch 1, batch 2600, loss[loss=0.4465, simple_loss=0.4511, pruned_loss=0.221, over 18634.00 frames. ], tot_loss[loss=0.4394, simple_loss=0.4457, pruned_loss=0.2165, over 3712676.46 frames. ], batch size: 69, lr: 4.71e-02, grad_scale: 16.0
2022-12-22 11:23:00,099 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 4.621e+02 7.074e+02 9.179e+02 1.186e+03 2.558e+03, threshold=1.836e+03, percent-clipped=4.0
2022-12-22 11:23:43,968 WARNING [train.py:1060] (2/4) Exclude cut with ID 7357-94126-0026-15857-0_sp0.9 from training. Duration: 25.061125
2022-12-22 11:23:52,030 WARNING [train.py:1060] (2/4) Exclude cut with ID 3033-130750-0096-55598-0 from training. Duration: 0.83
2022-12-22 11:24:01,858 INFO [train.py:894] (2/4) Epoch 1, batch 2650, loss[loss=0.4254, simple_loss=0.4454, pruned_loss=0.2027, over 18572.00 frames. ], tot_loss[loss=0.435, simple_loss=0.4432, pruned_loss=0.2134, over 3713760.73 frames. ], batch size: 56, lr: 4.70e-02, grad_scale: 8.0
2022-12-22 11:24:15,257 WARNING [train.py:1060] (2/4) Exclude cut with ID 3488-85273-0017-41203-0 from training. Duration: 24.73
2022-12-22 11:24:26,861 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.5442, 1.7005, 1.8254, 1.6670, 1.8255, 1.8225, 1.0049, 2.3546],
       device='cuda:2'), covar=tensor([0.2663, 0.2128, 0.2222, 0.3506, 0.2370, 0.1746, 0.3762, 0.0843],
       device='cuda:2'), in_proj_covar=tensor([0.0138, 0.0106, 0.0117, 0.0146, 0.0119, 0.0107, 0.0120, 0.0097],
       device='cuda:2'), out_proj_covar=tensor([1.3034e-04, 1.0548e-04, 1.1665e-04, 1.3750e-04, 1.1677e-04, 1.0226e-04,
        1.1483e-04, 9.1979e-05], device='cuda:2')
2022-12-22 11:24:27,698 WARNING [train.py:1060] (2/4) Exclude cut with ID 7255-291500-0009-134308-0 from training. Duration: 23.965
2022-12-22 11:24:35,992 WARNING [train.py:1060] (2/4) Exclude cut with ID 6978-92210-0030-146996-0_sp0.9 from training. Duration: 22.088875
2022-12-22 11:24:50,457 INFO [zipformer.py:660] (2/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=2685.0, num_to_drop=1, layers_to_drop={1}
2022-12-22 11:24:53,039 WARNING [train.py:1060] (2/4) Exclude cut with ID 7255-291500-0006-134305-0_sp0.9 from training. Duration: 23.6
2022-12-22 11:25:05,257 INFO [zipformer.py:660] (2/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=2696.0, num_to_drop=1, layers_to_drop={2}
2022-12-22 11:25:12,218 INFO [train.py:894] (2/4) Epoch 1, batch 2700, loss[loss=0.3982, simple_loss=0.3983, pruned_loss=0.1991, over 18613.00 frames. ], tot_loss[loss=0.4295, simple_loss=0.4397, pruned_loss=0.2097, over 3713714.16 frames. ], batch size: 41, lr: 4.69e-02, grad_scale: 8.0
2022-12-22 11:25:22,295 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 3.615e+02 7.385e+02 8.951e+02 1.108e+03 2.259e+03, threshold=1.790e+03, percent-clipped=4.0
2022-12-22 11:26:17,110 INFO [zipformer.py:660] (2/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=2746.0, num_to_drop=2, layers_to_drop={0, 2}
2022-12-22 11:26:23,447 INFO [train.py:894] (2/4) Epoch 1, batch 2750, loss[loss=0.3799, simple_loss=0.4076, pruned_loss=0.1761, over 18447.00 frames. ], tot_loss[loss=0.4272, simple_loss=0.4385, pruned_loss=0.2079, over 3714001.43 frames. ], batch size: 48, lr: 4.68e-02, grad_scale: 8.0
2022-12-22 11:26:23,484 WARNING [train.py:1060] (2/4) Exclude cut with ID 8291-282929-0024-13011-0 from training. Duration: 23.795
2022-12-22 11:26:38,586 WARNING [train.py:1060] (2/4) Exclude cut with ID 8631-249866-0030-130156-0_sp1.1 from training. Duration: 21.5409375
2022-12-22 11:26:41,256 WARNING [train.py:1060] (2/4) Exclude cut with ID 6978-92210-0019-146985-0_sp0.9 from training. Duration: 24.97775
2022-12-22 11:26:51,948 WARNING [train.py:1060] (2/4) Exclude cut with ID 1085-156170-0017-128270-0_sp0.9 from training. Duration: 23.3444375
2022-12-22 11:27:16,502 WARNING [train.py:1060] (2/4) Exclude cut with ID 6010-56788-0055-90261-0_sp0.9 from training. Duration: 23.2
2022-12-22 11:27:22,951 WARNING [train.py:1060] (2/4) Exclude cut with ID 5653-46179-0060-117930-0_sp0.9 from training. Duration: 21.17225
2022-12-22 11:27:33,911 INFO [train.py:894] (2/4) Epoch 1, batch 2800, loss[loss=0.3866, simple_loss=0.3999, pruned_loss=0.1867, over 18685.00 frames. ], tot_loss[loss=0.4266, simple_loss=0.4384, pruned_loss=0.2074, over 3714782.44 frames. ], batch size: 41, lr: 4.67e-02, grad_scale: 8.0
2022-12-22 11:27:39,277 WARNING [train.py:1060] (2/4) Exclude cut with ID 7255-291500-0008-134307-0_sp0.9 from training. Duration: 24.6555625
2022-12-22 11:27:43,869 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 4.105e+02 7.234e+02 8.983e+02 1.149e+03 3.907e+03, threshold=1.797e+03, percent-clipped=2.0
2022-12-22 11:28:06,976 INFO [zipformer.py:660] (2/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=2824.0, num_to_drop=1, layers_to_drop={0}
2022-12-22 11:28:28,500 WARNING [train.py:1060] (2/4) Exclude cut with ID 3488-65654-0031-41259-0 from training. Duration: 20.44
2022-12-22 11:28:31,730 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([2.1454, 2.5397, 2.5888, 1.5880, 2.2785, 2.6682, 1.1228, 3.6317],
       device='cuda:2'), covar=tensor([0.3688, 0.2426, 0.3473, 0.5613, 0.2784, 0.2082, 0.3988, 0.0689],
       device='cuda:2'), in_proj_covar=tensor([0.0142, 0.0112, 0.0127, 0.0149, 0.0125, 0.0112, 0.0130, 0.0102],
       device='cuda:2'), out_proj_covar=tensor([1.3636e-04, 1.1412e-04, 1.2834e-04, 1.4231e-04, 1.2400e-04, 1.0892e-04,
        1.2466e-04, 9.9123e-05], device='cuda:2')
2022-12-22 11:28:42,889 WARNING [train.py:1060] (2/4) Exclude cut with ID 6951-79737-0018-132285-0_sp0.9 from training. Duration: 23.45
2022-12-22 11:28:44,166 INFO [train.py:894] (2/4) Epoch 1, batch 2850, loss[loss=0.4067, simple_loss=0.4062, pruned_loss=0.2036, over 18590.00 frames. ], tot_loss[loss=0.4259, simple_loss=0.4388, pruned_loss=0.2064, over 3714325.18 frames. ], batch size: 41, lr: 4.66e-02, grad_scale: 8.0
2022-12-22 11:29:09,090 WARNING [train.py:1060] (2/4) Exclude cut with ID 6945-60535-0076-12784-0_sp0.9 from training. Duration: 20.52225
2022-12-22 11:29:15,876 WARNING [train.py:1060] (2/4) Exclude cut with ID 7255-291500-0008-134307-0 from training. Duration: 22.19
2022-12-22 11:29:25,864 WARNING [train.py:1060] (2/4) Exclude cut with ID 7357-94126-0014-15845-0_sp1.1 from training. Duration: 25.3818125
2022-12-22 11:29:32,118 INFO [zipformer.py:660] (2/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=2885.0, num_to_drop=2, layers_to_drop={1, 3}
2022-12-22 11:29:41,119 WARNING [train.py:1060] (2/4) Exclude cut with ID 6951-79737-0043-132310-0_sp0.9 from training. Duration: 28.0944375
2022-12-22 11:29:48,304 WARNING [train.py:1060] (2/4) Exclude cut with ID 2195-150901-0045-59933-0_sp0.9 from training. Duration: 22.9444375
2022-12-22 11:29:53,523 INFO [train.py:894] (2/4) Epoch 1, batch 2900, loss[loss=0.4325, simple_loss=0.4522, pruned_loss=0.2064, over 18459.00 frames. ], tot_loss[loss=0.4231, simple_loss=0.4366, pruned_loss=0.2048, over 3714889.04 frames. ], batch size: 54, lr: 4.65e-02, grad_scale: 8.0
2022-12-22 11:29:53,573 WARNING [train.py:1060] (2/4) Exclude cut with ID 8291-282929-0024-13011-0_sp1.1 from training. Duration: 21.6318125
2022-12-22 11:30:04,352 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 3.831e+02 7.832e+02 9.330e+02 1.210e+03 2.996e+03, threshold=1.866e+03, percent-clipped=5.0
2022-12-22 11:30:09,802 WARNING [train.py:1060] (2/4) Exclude cut with ID 8631-249866-0030-130156-0 from training. Duration: 23.695
2022-12-22 11:30:33,907 WARNING [train.py:1060] (2/4) Exclude cut with ID 7699-105389-0094-26379-0 from training. Duration: 23.955
2022-12-22 11:30:45,583 INFO [zipformer.py:660] (2/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=2937.0, num_to_drop=1, layers_to_drop={1}
2022-12-22 11:31:04,252 INFO [train.py:894] (2/4) Epoch 1, batch 2950, loss[loss=0.4337, simple_loss=0.4609, pruned_loss=0.2033, over 18475.00 frames. ], tot_loss[loss=0.4197, simple_loss=0.4344, pruned_loss=0.2025, over 3714184.01 frames. ], batch size: 64, lr: 4.64e-02, grad_scale: 8.0
2022-12-22 11:31:04,482 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([2.5000, 4.8319, 4.7251, 2.5268, 4.6989, 3.2338, 0.9717, 2.8576],
       device='cuda:2'), covar=tensor([0.1579, 0.0459, 0.0995, 0.2680, 0.0842, 0.1571, 0.5440, 0.2150],
       device='cuda:2'), in_proj_covar=tensor([0.0091, 0.0061, 0.0114, 0.0083, 0.0069, 0.0075, 0.0126, 0.0087],
       device='cuda:2'), out_proj_covar=tensor([1.1119e-04, 7.3144e-05, 1.4556e-04, 9.2157e-05, 9.0106e-05, 9.1118e-05,
        1.2571e-04, 1.0481e-04], device='cuda:2')
2022-12-22 11:31:04,569 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([5.6859, 4.8061, 4.8603, 5.1034, 5.0863, 4.8664, 5.1060, 1.3687],
       device='cuda:2'), covar=tensor([0.0319, 0.0413, 0.0510, 0.0323, 0.1068, 0.0466, 0.0327, 0.4661],
       device='cuda:2'), in_proj_covar=tensor([0.0081, 0.0092, 0.0095, 0.0068, 0.0121, 0.0082, 0.0092, 0.0151],
       device='cuda:2'), out_proj_covar=tensor([8.4638e-05, 9.7834e-05, 1.0288e-04, 7.1458e-05, 1.1606e-04, 9.0563e-05,
        9.9012e-05, 1.3209e-04], device='cuda:2')
2022-12-22 11:31:05,548 WARNING [train.py:1060] (2/4) Exclude cut with ID 8291-282929-0024-13011-0_sp0.9 from training. Duration: 26.438875
2022-12-22 11:31:46,315 WARNING [train.py:1060] (2/4) Exclude cut with ID 7699-105389-0021-26306-0_sp0.9 from training. Duration: 21.2444375
2022-12-22 11:31:46,340 WARNING [train.py:1060] (2/4) Exclude cut with ID 7357-94126-0014-15845-0_sp0.9 from training. Duration: 31.02225
2022-12-22 11:31:57,530 WARNING [train.py:1060] (2/4) Exclude cut with ID 432-122774-0017-62487-0 from training. Duration: 22.395
2022-12-22 11:32:07,033 INFO [zipformer.py:660] (2/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=2996.0, num_to_drop=1, layers_to_drop={0}
2022-12-22 11:32:09,687 INFO [zipformer.py:660] (2/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=2998.0, num_to_drop=2, layers_to_drop={1, 3}
2022-12-22 11:32:10,094 INFO [scaling.py:679] (2/4) Whitening: num_groups=1, num_channels=384, metric=4.20 vs. limit=5.0
2022-12-22 11:32:10,515 WARNING [train.py:1060] (2/4) Exclude cut with ID 7357-94126-0045-15876-0 from training. Duration: 21.075
2022-12-22 11:32:14,379 INFO [train.py:894] (2/4) Epoch 1, batch 3000, loss[loss=0.4132, simple_loss=0.4435, pruned_loss=0.1915, over 18592.00 frames. ], tot_loss[loss=0.4183, simple_loss=0.4339, pruned_loss=0.2013, over 3713938.78 frames. ], batch size: 56, lr: 4.63e-02, grad_scale: 8.0
2022-12-22 11:32:14,379 INFO [train.py:919] (2/4) Computing validation loss
2022-12-22 11:32:25,520 INFO [train.py:928] (2/4) Epoch 1, validation: loss=0.3402, simple_loss=0.4207, pruned_loss=0.1299, over 944034.00 frames. 
2022-12-22 11:32:25,520 INFO [train.py:929] (2/4) Maximum memory allocated so far is 22490MB
2022-12-22 11:32:28,927 WARNING [train.py:1060] (2/4) Exclude cut with ID 6482-98857-0025-147532-0_sp0.9 from training. Duration: 20.0055625
2022-12-22 11:32:28,937 WARNING [train.py:1060] (2/4) Exclude cut with ID 6951-79737-0037-132304-0_sp0.9 from training. Duration: 22.05
2022-12-22 11:32:30,384 WARNING [train.py:1060] (2/4) Exclude cut with ID 7255-291500-0003-134302-0 from training. Duration: 26.8349375
2022-12-22 11:32:33,098 WARNING [train.py:1060] (2/4) Exclude cut with ID 7357-94126-0009-15840-0_sp1.1 from training. Duration: 22.1090625
2022-12-22 11:32:35,854 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 3.958e+02 7.073e+02 9.090e+02 1.113e+03 2.521e+03, threshold=1.818e+03, percent-clipped=3.0
2022-12-22 11:32:40,431 WARNING [train.py:1060] (2/4) Exclude cut with ID 7699-105389-0094-26379-0_sp0.9 from training. Duration: 26.6166875
2022-12-22 11:32:51,732 INFO [zipformer.py:660] (2/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=3019.0, num_to_drop=1, layers_to_drop={1}
2022-12-22 11:32:56,399 WARNING [train.py:1060] (2/4) Exclude cut with ID 2046-178027-0000-53705-0_sp0.9 from training. Duration: 20.3055625
2022-12-22 11:33:18,448 WARNING [train.py:1060] (2/4) Exclude cut with ID 7205-50138-0008-5373-0_sp0.9 from training. Duration: 20.7
2022-12-22 11:33:23,986 INFO [zipformer.py:660] (2/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=3041.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 11:33:28,135 INFO [zipformer.py:660] (2/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=3044.0, num_to_drop=1, layers_to_drop={0}
2022-12-22 11:33:38,545 INFO [train.py:894] (2/4) Epoch 1, batch 3050, loss[loss=0.4311, simple_loss=0.4444, pruned_loss=0.2089, over 18652.00 frames. ], tot_loss[loss=0.4168, simple_loss=0.4328, pruned_loss=0.2004, over 3713387.87 frames. ], batch size: 62, lr: 4.62e-02, grad_scale: 8.0
2022-12-22 11:33:59,982 WARNING [train.py:1060] (2/4) Exclude cut with ID 6978-92210-0019-146985-0 from training. Duration: 22.48
2022-12-22 11:34:14,193 WARNING [train.py:1060] (2/4) Exclude cut with ID 7255-291500-0003-134302-0_sp0.9 from training. Duration: 29.816625
2022-12-22 11:34:20,691 INFO [zipformer.py:660] (2/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=3080.0, num_to_drop=2, layers_to_drop={1, 3}
2022-12-22 11:34:30,387 INFO [scaling.py:679] (2/4) Whitening: num_groups=1, num_channels=384, metric=3.97 vs. limit=5.0
2022-12-22 11:34:32,411 WARNING [train.py:1060] (2/4) Exclude cut with ID 7255-291500-0005-134304-0_sp1.1 from training. Duration: 22.7590625
2022-12-22 11:34:37,962 WARNING [train.py:1060] (2/4) Exclude cut with ID 7357-94126-0026-15857-0 from training. Duration: 22.555
2022-12-22 11:34:39,917 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.1568, 1.2510, 0.8035, 1.4300, 1.3090, 1.0528, 1.1803, 1.2377],
       device='cuda:2'), covar=tensor([0.2710, 0.0944, 0.2428, 0.1417, 0.1323, 0.1638, 0.1558, 0.1486],
       device='cuda:2'), in_proj_covar=tensor([0.0069, 0.0039, 0.0059, 0.0066, 0.0047, 0.0055, 0.0050, 0.0050],
       device='cuda:2'), out_proj_covar=tensor([6.8662e-05, 3.7118e-05, 5.3417e-05, 5.8876e-05, 4.6541e-05, 4.9981e-05,
        4.6649e-05, 4.8871e-05], device='cuda:2')
2022-12-22 11:34:48,852 INFO [train.py:894] (2/4) Epoch 1, batch 3100, loss[loss=0.4963, simple_loss=0.4882, pruned_loss=0.2522, over 18601.00 frames. ], tot_loss[loss=0.4149, simple_loss=0.4319, pruned_loss=0.1989, over 3714273.47 frames. ], batch size: 57, lr: 4.61e-02, grad_scale: 8.0
2022-12-22 11:34:58,944 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 3.605e+02 6.687e+02 8.494e+02 1.046e+03 2.071e+03, threshold=1.699e+03, percent-clipped=1.0
2022-12-22 11:34:58,968 WARNING [train.py:1060] (2/4) Exclude cut with ID 1250-135782-0005-25975-0_sp0.9 from training. Duration: 21.688875
2022-12-22 11:35:31,287 WARNING [train.py:1060] (2/4) Exclude cut with ID 3488-85273-0038-41224-0_sp0.9 from training. Duration: 22.6
2022-12-22 11:35:58,873 INFO [train.py:894] (2/4) Epoch 1, batch 3150, loss[loss=0.3248, simple_loss=0.3576, pruned_loss=0.146, over 18543.00 frames. ], tot_loss[loss=0.4128, simple_loss=0.4304, pruned_loss=0.1976, over 3713661.08 frames. ], batch size: 44, lr: 4.60e-02, grad_scale: 8.0
2022-12-22 11:36:03,767 WARNING [train.py:1060] (2/4) Exclude cut with ID 7357-94126-0009-15840-0 from training. Duration: 24.32
2022-12-22 11:36:40,606 INFO [zipformer.py:660] (2/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=3180.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 11:36:43,464 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.5969, 1.4920, 1.2293, 1.4079, 1.3873, 1.2551, 1.2794, 2.2045],
       device='cuda:2'), covar=tensor([0.2431, 0.1911, 0.3185, 0.2174, 0.3090, 0.2677, 0.2432, 0.1517],
       device='cuda:2'), in_proj_covar=tensor([0.0076, 0.0082, 0.0121, 0.0082, 0.0092, 0.0100, 0.0079, 0.0085],
       device='cuda:2'), out_proj_covar=tensor([0.0001, 0.0001, 0.0001, 0.0001, 0.0001, 0.0001, 0.0001, 0.0001],
       device='cuda:2')
2022-12-22 11:36:58,487 WARNING [train.py:1060] (2/4) Exclude cut with ID 8291-276745-0093-13116-0_sp0.9 from training. Duration: 21.061125
2022-12-22 11:37:09,802 INFO [train.py:894] (2/4) Epoch 1, batch 3200, loss[loss=0.4255, simple_loss=0.4478, pruned_loss=0.2016, over 18500.00 frames. ], tot_loss[loss=0.4099, simple_loss=0.4281, pruned_loss=0.1958, over 3714329.80 frames. ], batch size: 52, lr: 4.59e-02, grad_scale: 8.0
2022-12-22 11:37:12,464 WARNING [train.py:1060] (2/4) Exclude cut with ID 7357-94126-0024-15855-0_sp0.9 from training. Duration: 20.32225
2022-12-22 11:37:16,849 INFO [scaling.py:679] (2/4) Whitening: num_groups=1, num_channels=384, metric=3.51 vs. limit=5.0
2022-12-22 11:37:19,187 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 5.026e+02 7.855e+02 1.051e+03 1.351e+03 2.916e+03, threshold=2.101e+03, percent-clipped=7.0
2022-12-22 11:37:22,086 WARNING [train.py:1060] (2/4) Exclude cut with ID 3033-130750-0096-55598-0_sp1.1 from training. Duration: 0.7545625
2022-12-22 11:37:36,530 WARNING [train.py:1060] (2/4) Exclude cut with ID 4295-39940-0007-92567-0_sp0.9 from training. Duration: 23.9333125
2022-12-22 11:38:06,648 WARNING [train.py:1060] (2/4) Exclude cut with ID 7255-291500-0008-134307-0_sp1.1 from training. Duration: 20.17275
2022-12-22 11:38:11,070 WARNING [train.py:1060] (2/4) Exclude cut with ID 6978-92210-0019-146985-0_sp1.1 from training. Duration: 20.436375
2022-12-22 11:38:18,736 INFO [train.py:894] (2/4) Epoch 1, batch 3250, loss[loss=0.3607, simple_loss=0.3914, pruned_loss=0.165, over 18696.00 frames. ], tot_loss[loss=0.4088, simple_loss=0.4275, pruned_loss=0.195, over 3715002.25 frames. ], batch size: 46, lr: 4.58e-02, grad_scale: 8.0
2022-12-22 11:39:07,672 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.2298, 1.4981, 1.9162, 1.6248, 1.4227, 2.8108, 1.4699, 2.3746],
       device='cuda:2'), covar=tensor([0.4912, 0.3414, 0.1789, 0.3346, 0.3001, 0.0481, 0.2667, 0.1206],
       device='cuda:2'), in_proj_covar=tensor([0.0117, 0.0101, 0.0101, 0.0103, 0.0115, 0.0063, 0.0107, 0.0089],
       device='cuda:2'), out_proj_covar=tensor([1.4100e-04, 1.2703e-04, 1.2832e-04, 1.3215e-04, 1.5079e-04, 8.7606e-05,
        1.4040e-04, 1.2319e-04], device='cuda:2')
2022-12-22 11:39:10,459 INFO [zipformer.py:660] (2/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=3288.0, num_to_drop=1, layers_to_drop={0}
2022-12-22 11:39:17,236 INFO [zipformer.py:660] (2/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=3293.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 11:39:28,601 INFO [train.py:894] (2/4) Epoch 1, batch 3300, loss[loss=0.3304, simple_loss=0.3714, pruned_loss=0.1447, over 18535.00 frames. ], tot_loss[loss=0.4046, simple_loss=0.4248, pruned_loss=0.1922, over 3715334.05 frames. ], batch size: 47, lr: 4.57e-02, grad_scale: 8.0
2022-12-22 11:39:28,657 WARNING [train.py:1060] (2/4) Exclude cut with ID 4234-40345-0022-142709-0_sp0.9 from training. Duration: 23.1055625
2022-12-22 11:39:31,191 WARNING [train.py:1060] (2/4) Exclude cut with ID 8291-282929-0007-12994-0_sp1.1 from training. Duration: 23.5
2022-12-22 11:39:37,652 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 4.762e+02 7.922e+02 9.750e+02 1.198e+03 1.925e+03, threshold=1.950e+03, percent-clipped=0.0
2022-12-22 11:39:40,574 WARNING [train.py:1060] (2/4) Exclude cut with ID 7255-291500-0009-134308-0_sp0.9 from training. Duration: 26.62775
2022-12-22 11:39:46,269 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.6087, 1.3637, 0.9755, 1.7650, 1.8505, 2.6407, 1.8491, 1.5508],
       device='cuda:2'), covar=tensor([0.1728, 0.2169, 0.2546, 0.1888, 0.1473, 0.0711, 0.1394, 0.2643],
       device='cuda:2'), in_proj_covar=tensor([0.0086, 0.0077, 0.0094, 0.0088, 0.0087, 0.0079, 0.0080, 0.0088],
       device='cuda:2'), out_proj_covar=tensor([0.0001, 0.0001, 0.0001, 0.0001, 0.0001, 0.0001, 0.0001, 0.0001],
       device='cuda:2')
2022-12-22 11:39:52,404 WARNING [train.py:1060] (2/4) Exclude cut with ID 6951-79737-0018-132285-0 from training. Duration: 21.105
2022-12-22 11:39:56,912 WARNING [train.py:1060] (2/4) Exclude cut with ID 4511-76322-0006-80011-0_sp0.9 from training. Duration: 24.411125
2022-12-22 11:40:00,042 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([2.0832, 1.3198, 1.2638, 1.7432, 1.6730, 1.5635, 1.8181, 1.1017],
       device='cuda:2'), covar=tensor([0.0995, 0.1468, 0.2946, 0.1227, 0.1044, 0.1496, 0.1053, 0.1927],
       device='cuda:2'), in_proj_covar=tensor([0.0051, 0.0053, 0.0065, 0.0045, 0.0047, 0.0056, 0.0044, 0.0055],
       device='cuda:2'), out_proj_covar=tensor([4.7015e-05, 4.1601e-05, 5.9045e-05, 3.8765e-05, 3.9204e-05, 4.8058e-05,
        3.6697e-05, 4.9146e-05], device='cuda:2')
2022-12-22 11:40:15,611 INFO [scaling.py:679] (2/4) Whitening: num_groups=8, num_channels=192, metric=1.85 vs. limit=2.0
2022-12-22 11:40:22,053 WARNING [train.py:1060] (2/4) Exclude cut with ID 6758-72288-0033-108368-0_sp1.1 from training. Duration: 21.263625
2022-12-22 11:40:25,095 INFO [zipformer.py:660] (2/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=3341.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 11:40:37,036 INFO [zipformer.py:660] (2/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=3349.0, num_to_drop=2, layers_to_drop={1, 3}
2022-12-22 11:40:39,251 INFO [train.py:894] (2/4) Epoch 1, batch 3350, loss[loss=0.3802, simple_loss=0.4131, pruned_loss=0.1737, over 18599.00 frames. ], tot_loss[loss=0.4034, simple_loss=0.4242, pruned_loss=0.1913, over 3714903.34 frames. ], batch size: 56, lr: 4.56e-02, grad_scale: 8.0
2022-12-22 11:40:51,668 WARNING [train.py:1060] (2/4) Exclude cut with ID 4234-40345-0022-142709-0 from training. Duration: 20.795
2022-12-22 11:41:02,111 WARNING [train.py:1060] (2/4) Exclude cut with ID 7357-94126-0021-15852-0 from training. Duration: 24.76
2022-12-22 11:41:03,289 WARNING [train.py:1060] (2/4) Exclude cut with ID 3867-173237-0077-144769-0_sp0.9 from training. Duration: 22.25
2022-12-22 11:41:13,741 INFO [zipformer.py:660] (2/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=3375.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 11:41:24,760 WARNING [train.py:1060] (2/4) Exclude cut with ID 7357-94126-0026-15857-0_sp1.1 from training. Duration: 20.5045625
2022-12-22 11:41:34,526 INFO [zipformer.py:660] (2/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=3389.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 11:41:51,775 INFO [train.py:894] (2/4) Epoch 1, batch 3400, loss[loss=0.4906, simple_loss=0.4839, pruned_loss=0.2486, over 18494.00 frames. ], tot_loss[loss=0.4032, simple_loss=0.4245, pruned_loss=0.191, over 3715375.76 frames. ], batch size: 78, lr: 4.55e-02, grad_scale: 8.0
2022-12-22 11:42:01,145 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 4.101e+02 7.385e+02 8.854e+02 1.159e+03 2.009e+03, threshold=1.771e+03, percent-clipped=1.0
2022-12-22 11:42:45,393 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([2.2192, 1.6613, 1.7429, 2.4052, 2.1288, 2.9553, 1.8687, 1.8393],
       device='cuda:2'), covar=tensor([0.1511, 0.2382, 0.2372, 0.1479, 0.1356, 0.1198, 0.1647, 0.2441],
       device='cuda:2'), in_proj_covar=tensor([0.0090, 0.0081, 0.0099, 0.0092, 0.0091, 0.0083, 0.0085, 0.0090],
       device='cuda:2'), out_proj_covar=tensor([0.0001, 0.0001, 0.0001, 0.0001, 0.0001, 0.0001, 0.0001, 0.0001],
       device='cuda:2')
2022-12-22 11:42:58,272 INFO [train.py:894] (2/4) Epoch 1, batch 3450, loss[loss=0.3577, simple_loss=0.3971, pruned_loss=0.1591, over 18456.00 frames. ], tot_loss[loss=0.4032, simple_loss=0.4239, pruned_loss=0.1912, over 3715646.93 frames. ], batch size: 50, lr: 4.54e-02, grad_scale: 4.0
2022-12-22 11:43:37,993 INFO [zipformer.py:660] (2/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=3480.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 11:43:54,317 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.9892, 2.2712, 1.9583, 2.0161, 1.9903, 2.1333, 1.9825, 2.4278],
       device='cuda:2'), covar=tensor([0.1979, 0.1320, 0.2472, 0.1708, 0.2531, 0.1888, 0.1753, 0.1341],
       device='cuda:2'), in_proj_covar=tensor([0.0080, 0.0087, 0.0129, 0.0085, 0.0098, 0.0104, 0.0081, 0.0091],
       device='cuda:2'), out_proj_covar=tensor([0.0001, 0.0001, 0.0002, 0.0001, 0.0001, 0.0001, 0.0001, 0.0001],
       device='cuda:2')
2022-12-22 11:44:07,894 INFO [train.py:894] (2/4) Epoch 1, batch 3500, loss[loss=0.3784, simple_loss=0.4049, pruned_loss=0.176, over 18678.00 frames. ], tot_loss[loss=0.3992, simple_loss=0.4214, pruned_loss=0.1885, over 3715450.73 frames. ], batch size: 98, lr: 4.53e-02, grad_scale: 4.0
2022-12-22 11:44:27,624 WARNING [train.py:1060] (2/4) Exclude cut with ID 7859-102521-0017-7548-0_sp1.1 from training. Duration: 22.2954375
2022-12-22 11:44:37,382 INFO [train.py:894] (2/4) Epoch 2, batch 0, loss[loss=0.3293, simple_loss=0.373, pruned_loss=0.1428, over 18450.00 frames. ], tot_loss[loss=0.3293, simple_loss=0.373, pruned_loss=0.1428, over 18450.00 frames. ], batch size: 43, lr: 4.44e-02, grad_scale: 8.0
2022-12-22 11:44:37,383 INFO [train.py:919] (2/4) Computing validation loss
2022-12-22 11:44:49,009 INFO [train.py:928] (2/4) Epoch 2, validation: loss=0.3008, simple_loss=0.3844, pruned_loss=0.1086, over 944034.00 frames. 
2022-12-22 11:44:49,009 INFO [train.py:929] (2/4) Maximum memory allocated so far is 22490MB
2022-12-22 11:44:51,669 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 4.103e+02 7.617e+02 9.580e+02 1.359e+03 2.916e+03, threshold=1.916e+03, percent-clipped=9.0
2022-12-22 11:45:20,035 INFO [zipformer.py:660] (2/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=3528.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 11:45:36,148 WARNING [train.py:1060] (2/4) Exclude cut with ID 298-126791-0067-24026-0_sp0.9 from training. Duration: 21.438875
2022-12-22 11:45:40,217 WARNING [train.py:1060] (2/4) Exclude cut with ID 5652-39938-0025-23684-0_sp0.9 from training. Duration: 22.2055625
2022-12-22 11:45:44,550 INFO [zipformer.py:660] (2/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=3546.0, num_to_drop=1, layers_to_drop={1}
2022-12-22 11:45:59,794 INFO [train.py:894] (2/4) Epoch 2, batch 50, loss[loss=0.3338, simple_loss=0.3886, pruned_loss=0.1395, over 18691.00 frames. ], tot_loss[loss=0.3576, simple_loss=0.4025, pruned_loss=0.1563, over 837744.96 frames. ], batch size: 60, lr: 4.43e-02, grad_scale: 8.0
2022-12-22 11:46:52,004 INFO [zipformer.py:660] (2/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=3593.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 11:47:11,766 INFO [train.py:894] (2/4) Epoch 2, batch 100, loss[loss=0.3793, simple_loss=0.4234, pruned_loss=0.1676, over 18537.00 frames. ], tot_loss[loss=0.3491, simple_loss=0.3963, pruned_loss=0.1509, over 1476111.10 frames. ], batch size: 55, lr: 4.42e-02, grad_scale: 8.0
2022-12-22 11:47:12,852 INFO [zipformer.py:660] (2/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=3607.0, num_to_drop=2, layers_to_drop={1, 3}
2022-12-22 11:47:15,577 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 4.976e+02 8.232e+02 1.042e+03 1.328e+03 2.766e+03, threshold=2.084e+03, percent-clipped=5.0
2022-12-22 11:48:01,633 INFO [zipformer.py:660] (2/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=3641.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 11:48:05,719 INFO [zipformer.py:660] (2/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=3644.0, num_to_drop=1, layers_to_drop={2}
2022-12-22 11:48:24,881 INFO [train.py:894] (2/4) Epoch 2, batch 150, loss[loss=0.3116, simple_loss=0.3798, pruned_loss=0.1217, over 18698.00 frames. ], tot_loss[loss=0.3388, simple_loss=0.3889, pruned_loss=0.1444, over 1971606.47 frames. ], batch size: 62, lr: 4.40e-02, grad_scale: 8.0
2022-12-22 11:48:33,365 WARNING [train.py:1060] (2/4) Exclude cut with ID 7859-102521-0017-7548-0 from training. Duration: 24.525
2022-12-22 11:48:48,237 INFO [scaling.py:679] (2/4) Whitening: num_groups=8, num_channels=192, metric=2.28 vs. limit=2.0
2022-12-22 11:48:52,276 INFO [zipformer.py:660] (2/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=3675.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 11:48:53,690 INFO [zipformer.py:660] (2/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=3676.0, num_to_drop=1, layers_to_drop={1}
2022-12-22 11:49:07,521 WARNING [train.py:1060] (2/4) Exclude cut with ID 3699-47246-0007-3408-0_sp0.9 from training. Duration: 20.26675
2022-12-22 11:49:20,100 WARNING [train.py:1060] (2/4) Exclude cut with ID 7859-102521-0017-7548-0_sp0.9 from training. Duration: 27.25
2022-12-22 11:49:38,187 INFO [train.py:894] (2/4) Epoch 2, batch 200, loss[loss=0.3219, simple_loss=0.3833, pruned_loss=0.1302, over 18542.00 frames. ], tot_loss[loss=0.3326, simple_loss=0.3838, pruned_loss=0.1407, over 2358559.28 frames. ], batch size: 69, lr: 4.39e-02, grad_scale: 8.0
2022-12-22 11:49:41,123 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 3.513e+02 7.302e+02 9.000e+02 1.037e+03 2.044e+03, threshold=1.800e+03, percent-clipped=0.0
2022-12-22 11:50:02,797 INFO [zipformer.py:660] (2/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=3723.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 11:50:14,951 INFO [scaling.py:679] (2/4) Whitening: num_groups=8, num_channels=192, metric=2.02 vs. limit=2.0
2022-12-22 11:50:23,537 INFO [zipformer.py:660] (2/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=3737.0, num_to_drop=2, layers_to_drop={2, 3}
2022-12-22 11:50:34,698 WARNING [train.py:1060] (2/4) Exclude cut with ID 6426-64292-0017-15984-0 from training. Duration: 21.68
2022-12-22 11:50:46,776 WARNING [train.py:1060] (2/4) Exclude cut with ID 4278-13270-0007-59342-0 from training. Duration: 21.6300625
2022-12-22 11:50:52,891 INFO [train.py:894] (2/4) Epoch 2, batch 250, loss[loss=0.2971, simple_loss=0.3418, pruned_loss=0.1262, over 18404.00 frames. ], tot_loss[loss=0.3286, simple_loss=0.3805, pruned_loss=0.1383, over 2658088.35 frames. ], batch size: 46, lr: 4.38e-02, grad_scale: 8.0
2022-12-22 11:51:03,193 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.4505, 1.7239, 1.0686, 1.4290, 1.4148, 1.3227, 1.4229, 2.1986],
       device='cuda:2'), covar=tensor([0.2721, 0.1775, 0.3286, 0.2179, 0.3209, 0.2573, 0.2488, 0.1684],
       device='cuda:2'), in_proj_covar=tensor([0.0081, 0.0089, 0.0129, 0.0088, 0.0096, 0.0103, 0.0082, 0.0089],
       device='cuda:2'), out_proj_covar=tensor([0.0001, 0.0001, 0.0002, 0.0001, 0.0001, 0.0001, 0.0001, 0.0001],
       device='cuda:2')
2022-12-22 11:51:11,828 WARNING [train.py:1060] (2/4) Exclude cut with ID 4278-13270-0007-59342-0_sp0.9 from training. Duration: 24.033375
2022-12-22 11:51:17,447 INFO [scaling.py:679] (2/4) Whitening: num_groups=8, num_channels=96, metric=1.27 vs. limit=2.0
2022-12-22 11:52:06,374 INFO [train.py:894] (2/4) Epoch 2, batch 300, loss[loss=0.337, simple_loss=0.3957, pruned_loss=0.1392, over 18557.00 frames. ], tot_loss[loss=0.3296, simple_loss=0.3811, pruned_loss=0.139, over 2892510.18 frames. ], batch size: 57, lr: 4.37e-02, grad_scale: 8.0
2022-12-22 11:52:06,440 WARNING [train.py:1060] (2/4) Exclude cut with ID 4278-13270-0009-59344-0 from training. Duration: 22.905
2022-12-22 11:52:07,923 WARNING [train.py:1060] (2/4) Exclude cut with ID 5622-44585-0006-90525-0_sp1.1 from training. Duration: 23.4318125
2022-12-22 11:52:09,150 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 4.194e+02 6.939e+02 8.430e+02 1.162e+03 2.242e+03, threshold=1.686e+03, percent-clipped=6.0
2022-12-22 11:53:19,700 INFO [train.py:894] (2/4) Epoch 2, batch 350, loss[loss=0.3612, simple_loss=0.4068, pruned_loss=0.1578, over 18522.00 frames. ], tot_loss[loss=0.3327, simple_loss=0.3838, pruned_loss=0.1409, over 3074856.69 frames. ], batch size: 55, lr: 4.36e-02, grad_scale: 8.0
2022-12-22 11:54:05,324 WARNING [train.py:1060] (2/4) Exclude cut with ID 4278-13270-0009-59344-0_sp1.1 from training. Duration: 20.82275
2022-12-22 11:54:06,787 WARNING [train.py:1060] (2/4) Exclude cut with ID 4278-13270-0009-59344-0_sp0.9 from training. Duration: 25.45
2022-12-22 11:54:29,248 INFO [zipformer.py:660] (2/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=3902.0, num_to_drop=1, layers_to_drop={2}
2022-12-22 11:54:35,945 INFO [train.py:894] (2/4) Epoch 2, batch 400, loss[loss=0.3041, simple_loss=0.3599, pruned_loss=0.1241, over 18460.00 frames. ], tot_loss[loss=0.3338, simple_loss=0.3841, pruned_loss=0.1418, over 3215795.26 frames. ], batch size: 43, lr: 4.35e-02, grad_scale: 8.0
2022-12-22 11:54:38,882 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 4.591e+02 7.650e+02 9.752e+02 1.196e+03 2.094e+03, threshold=1.950e+03, percent-clipped=3.0
2022-12-22 11:55:05,437 INFO [zipformer.py:660] (2/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=3926.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 11:55:06,469 WARNING [train.py:1060] (2/4) Exclude cut with ID 5622-44585-0006-90525-0 from training. Duration: 25.775
2022-12-22 11:55:28,485 WARNING [train.py:1060] (2/4) Exclude cut with ID 2929-85685-0071-62375-0_sp0.9 from training. Duration: 22.25
2022-12-22 11:55:31,576 INFO [zipformer.py:660] (2/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=3944.0, num_to_drop=1, layers_to_drop={1}
2022-12-22 11:55:50,780 INFO [train.py:894] (2/4) Epoch 2, batch 450, loss[loss=0.3919, simple_loss=0.4321, pruned_loss=0.1759, over 18590.00 frames. ], tot_loss[loss=0.3363, simple_loss=0.3866, pruned_loss=0.143, over 3327053.48 frames. ], batch size: 175, lr: 4.34e-02, grad_scale: 8.0
2022-12-22 11:55:55,182 WARNING [train.py:1060] (2/4) Exclude cut with ID 3972-170212-0014-23379-0 from training. Duration: 26.205
2022-12-22 11:56:11,128 WARNING [train.py:1060] (2/4) Exclude cut with ID 5239-32139-0047-9341-0_sp0.9 from training. Duration: 30.1555625
2022-12-22 11:56:14,511 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.2802, 0.9569, 0.4661, 1.2594, 1.1620, 2.2821, 0.9724, 1.2808],
       device='cuda:2'), covar=tensor([0.1507, 0.2083, 0.2615, 0.1628, 0.1799, 0.0592, 0.1771, 0.2257],
       device='cuda:2'), in_proj_covar=tensor([0.0089, 0.0083, 0.0100, 0.0093, 0.0096, 0.0078, 0.0088, 0.0090],
       device='cuda:2'), out_proj_covar=tensor([0.0001, 0.0001, 0.0002, 0.0002, 0.0002, 0.0002, 0.0001, 0.0001],
       device='cuda:2')
2022-12-22 11:56:17,019 WARNING [train.py:1060] (2/4) Exclude cut with ID 1265-135635-0050-6781-0_sp0.9 from training. Duration: 21.8333125
2022-12-22 11:56:25,390 WARNING [train.py:1060] (2/4) Exclude cut with ID 1914-133440-0024-94914-0_sp1.1 from training. Duration: 20.6545625
2022-12-22 11:56:36,333 INFO [zipformer.py:660] (2/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=3987.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 11:56:38,121 INFO [scaling.py:679] (2/4) Whitening: num_groups=1, num_channels=384, metric=3.67 vs. limit=5.0
2022-12-22 11:56:43,464 INFO [zipformer.py:660] (2/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=3992.0, num_to_drop=1, layers_to_drop={1}
2022-12-22 11:57:09,988 INFO [train.py:894] (2/4) Epoch 2, batch 500, loss[loss=0.3053, simple_loss=0.368, pruned_loss=0.1212, over 18584.00 frames. ], tot_loss[loss=0.3373, simple_loss=0.3876, pruned_loss=0.1434, over 3412219.42 frames. ], batch size: 49, lr: 4.33e-02, grad_scale: 8.0
2022-12-22 11:57:12,286 WARNING [train.py:1060] (2/4) Exclude cut with ID 7395-89880-0045-39920-0_sp0.9 from training. Duration: 20.52225
2022-12-22 11:57:13,497 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 4.346e+02 6.819e+02 8.199e+02 9.839e+02 2.413e+03, threshold=1.640e+03, percent-clipped=2.0
2022-12-22 11:57:33,783 WARNING [train.py:1060] (2/4) Exclude cut with ID 3972-170212-0014-23379-0_sp0.9 from training. Duration: 29.1166875
2022-12-22 11:57:48,771 INFO [zipformer.py:660] (2/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=4032.0, num_to_drop=1, layers_to_drop={1}
2022-12-22 11:58:28,034 INFO [train.py:894] (2/4) Epoch 2, batch 550, loss[loss=0.3376, simple_loss=0.383, pruned_loss=0.146, over 18450.00 frames. ], tot_loss[loss=0.3366, simple_loss=0.387, pruned_loss=0.1431, over 3478895.54 frames. ], batch size: 50, lr: 4.32e-02, grad_scale: 8.0
2022-12-22 11:58:34,148 WARNING [train.py:1060] (2/4) Exclude cut with ID 543-133211-0007-59831-0_sp0.9 from training. Duration: 21.388875
2022-12-22 11:59:09,578 WARNING [train.py:1060] (2/4) Exclude cut with ID 1914-133440-0024-94914-0 from training. Duration: 22.72
2022-12-22 11:59:11,307 WARNING [train.py:1060] (2/4) Exclude cut with ID 1914-133440-0031-94921-0_sp0.9 from training. Duration: 22.7444375
2022-12-22 11:59:44,277 INFO [train.py:894] (2/4) Epoch 2, batch 600, loss[loss=0.349, simple_loss=0.4057, pruned_loss=0.1462, over 18690.00 frames. ], tot_loss[loss=0.3379, simple_loss=0.3881, pruned_loss=0.1438, over 3530635.51 frames. ], batch size: 96, lr: 4.31e-02, grad_scale: 8.0
2022-12-22 11:59:47,288 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 4.183e+02 7.677e+02 9.029e+02 1.105e+03 3.557e+03, threshold=1.806e+03, percent-clipped=7.0
2022-12-22 11:59:56,447 WARNING [train.py:1060] (2/4) Exclude cut with ID 4133-6541-0027-40495-0_sp1.1 from training. Duration: 0.9681875
2022-12-22 12:00:01,108 WARNING [train.py:1060] (2/4) Exclude cut with ID 6330-62851-0022-91297-0_sp0.9 from training. Duration: 22.3166875
2022-12-22 12:00:05,388 WARNING [train.py:1060] (2/4) Exclude cut with ID 543-133212-0015-59917-0_sp0.9 from training. Duration: 21.8166875
2022-12-22 12:00:35,781 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([2.3291, 1.7104, 1.6402, 1.0093, 1.6954, 2.1170, 1.2197, 1.6425],
       device='cuda:2'), covar=tensor([0.1008, 0.1270, 0.2226, 0.3086, 0.1747, 0.0929, 0.2667, 0.2375],
       device='cuda:2'), in_proj_covar=tensor([0.0090, 0.0125, 0.0152, 0.0160, 0.0123, 0.0106, 0.0136, 0.0151],
       device='cuda:2'), out_proj_covar=tensor([0.0001, 0.0001, 0.0002, 0.0002, 0.0001, 0.0001, 0.0001, 0.0002],
       device='cuda:2')
2022-12-22 12:00:35,784 INFO [zipformer.py:660] (2/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=4141.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 12:00:59,843 INFO [train.py:894] (2/4) Epoch 2, batch 650, loss[loss=0.304, simple_loss=0.3492, pruned_loss=0.1295, over 18616.00 frames. ], tot_loss[loss=0.3351, simple_loss=0.3859, pruned_loss=0.1421, over 3571871.66 frames. ], batch size: 45, lr: 4.30e-02, grad_scale: 8.0
2022-12-22 12:01:18,577 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.1778, 1.3300, 1.4386, 1.2969, 1.1815, 2.0625, 0.6130, 1.7538],
       device='cuda:2'), covar=tensor([0.4735, 0.2864, 0.1873, 0.3124, 0.2403, 0.0478, 0.2701, 0.1183],
       device='cuda:2'), in_proj_covar=tensor([0.0142, 0.0123, 0.0123, 0.0122, 0.0133, 0.0074, 0.0125, 0.0109],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0001, 0.0002, 0.0002],
       device='cuda:2')
2022-12-22 12:01:47,898 WARNING [train.py:1060] (2/4) Exclude cut with ID 4957-30119-0041-23990-0_sp0.9 from training. Duration: 20.22775
2022-12-22 12:02:10,288 INFO [zipformer.py:660] (2/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=4202.0, num_to_drop=1, layers_to_drop={0}
2022-12-22 12:02:10,410 INFO [zipformer.py:660] (2/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=4202.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 12:02:17,456 INFO [train.py:894] (2/4) Epoch 2, batch 700, loss[loss=0.3475, simple_loss=0.3939, pruned_loss=0.1505, over 18710.00 frames. ], tot_loss[loss=0.3348, simple_loss=0.3854, pruned_loss=0.1421, over 3602329.09 frames. ], batch size: 46, lr: 4.29e-02, grad_scale: 8.0
2022-12-22 12:02:20,236 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 4.497e+02 7.136e+02 8.417e+02 1.055e+03 2.568e+03, threshold=1.683e+03, percent-clipped=3.0
2022-12-22 12:02:34,047 WARNING [train.py:1060] (2/4) Exclude cut with ID 5239-32139-0047-9341-0_sp1.1 from training. Duration: 24.67275
2022-12-22 12:02:58,948 WARNING [train.py:1060] (2/4) Exclude cut with ID 3082-165428-0081-50734-0_sp0.9 from training. Duration: 21.8055625
2022-12-22 12:03:24,563 INFO [zipformer.py:660] (2/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=4250.0, num_to_drop=1, layers_to_drop={0}
2022-12-22 12:03:34,914 INFO [train.py:894] (2/4) Epoch 2, batch 750, loss[loss=0.3386, simple_loss=0.3819, pruned_loss=0.1477, over 18540.00 frames. ], tot_loss[loss=0.3349, simple_loss=0.3858, pruned_loss=0.142, over 3627209.74 frames. ], batch size: 47, lr: 4.28e-02, grad_scale: 8.0
2022-12-22 12:03:39,605 WARNING [train.py:1060] (2/4) Exclude cut with ID 3340-169293-0054-76830-0_sp0.9 from training. Duration: 22.6666875
2022-12-22 12:04:03,672 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.3996, 1.5936, 1.0640, 1.8261, 2.0270, 1.2332, 1.3417, 0.9105],
       device='cuda:2'), covar=tensor([0.1398, 0.1066, 0.1134, 0.0756, 0.0612, 0.1009, 0.0960, 0.1360],
       device='cuda:2'), in_proj_covar=tensor([0.0120, 0.0093, 0.0098, 0.0091, 0.0098, 0.0098, 0.0087, 0.0100],
       device='cuda:2'), out_proj_covar=tensor([1.1867e-04, 9.6182e-05, 9.5857e-05, 9.1261e-05, 9.7257e-05, 9.5707e-05,
        9.0361e-05, 9.5828e-05], device='cuda:2')
2022-12-22 12:04:12,159 INFO [zipformer.py:660] (2/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=4282.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 12:04:22,645 INFO [zipformer.py:660] (2/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=4288.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 12:04:43,620 WARNING [train.py:1060] (2/4) Exclude cut with ID 2411-132532-0017-82279-0_sp1.1 from training. Duration: 0.9681875
2022-12-22 12:04:51,496 INFO [train.py:894] (2/4) Epoch 2, batch 800, loss[loss=0.3054, simple_loss=0.3746, pruned_loss=0.1181, over 18690.00 frames. ], tot_loss[loss=0.3333, simple_loss=0.3844, pruned_loss=0.1411, over 3646967.91 frames. ], batch size: 60, lr: 4.27e-02, grad_scale: 8.0
2022-12-22 12:04:54,411 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 3.666e+02 6.822e+02 8.702e+02 1.097e+03 2.358e+03, threshold=1.740e+03, percent-clipped=4.0
2022-12-22 12:05:09,502 WARNING [train.py:1060] (2/4) Exclude cut with ID 6330-62850-0007-91323-0 from training. Duration: 22.485
2022-12-22 12:05:14,438 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.6005, 1.9114, 1.0415, 1.7373, 1.7033, 1.2475, 2.8027, 1.7713],
       device='cuda:2'), covar=tensor([0.0794, 0.1026, 0.1481, 0.1215, 0.1685, 0.0875, 0.0226, 0.0776],
       device='cuda:2'), in_proj_covar=tensor([0.0076, 0.0073, 0.0083, 0.0096, 0.0106, 0.0076, 0.0044, 0.0075],
       device='cuda:2'), out_proj_covar=tensor([7.1245e-05, 7.2676e-05, 7.7973e-05, 8.6921e-05, 1.0207e-04, 6.7826e-05,
        4.0473e-05, 6.9757e-05], device='cuda:2')
2022-12-22 12:05:28,807 INFO [zipformer.py:660] (2/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=4332.0, num_to_drop=1, layers_to_drop={2}
2022-12-22 12:05:50,021 WARNING [train.py:1060] (2/4) Exclude cut with ID 3972-170212-0014-23379-0_sp1.1 from training. Duration: 23.82275
2022-12-22 12:05:55,891 INFO [zipformer.py:660] (2/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=4349.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 12:06:03,199 WARNING [train.py:1060] (2/4) Exclude cut with ID 4860-13185-0032-76709-0 from training. Duration: 20.77
2022-12-22 12:06:07,418 INFO [train.py:894] (2/4) Epoch 2, batch 850, loss[loss=0.3794, simple_loss=0.4204, pruned_loss=0.1692, over 18629.00 frames. ], tot_loss[loss=0.3323, simple_loss=0.3838, pruned_loss=0.1405, over 3662258.20 frames. ], batch size: 69, lr: 4.26e-02, grad_scale: 8.0
2022-12-22 12:06:10,542 WARNING [train.py:1060] (2/4) Exclude cut with ID 6426-64292-0017-15984-0_sp0.9 from training. Duration: 24.088875
2022-12-22 12:06:28,878 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.4213, 0.4746, 1.4432, 0.9633, 1.5300, 1.5357, 1.4263, 1.3004],
       device='cuda:2'), covar=tensor([0.1360, 0.1722, 0.0647, 0.1519, 0.0872, 0.0588, 0.0720, 0.0882],
       device='cuda:2'), in_proj_covar=tensor([0.0061, 0.0098, 0.0069, 0.0108, 0.0091, 0.0062, 0.0063, 0.0088],
       device='cuda:2'), out_proj_covar=tensor([6.7046e-05, 9.5345e-05, 7.3548e-05, 1.0631e-04, 9.3365e-05, 6.9509e-05,
        6.7948e-05, 8.7630e-05], device='cuda:2')
2022-12-22 12:06:41,371 WARNING [train.py:1060] (2/4) Exclude cut with ID 6330-62850-0007-91323-0_sp1.1 from training. Duration: 20.4409375
2022-12-22 12:06:41,517 INFO [zipformer.py:660] (2/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=4380.0, num_to_drop=1, layers_to_drop={1}
2022-12-22 12:07:23,238 INFO [train.py:894] (2/4) Epoch 2, batch 900, loss[loss=0.3301, simple_loss=0.3917, pruned_loss=0.1343, over 18544.00 frames. ], tot_loss[loss=0.332, simple_loss=0.3832, pruned_loss=0.1404, over 3673605.88 frames. ], batch size: 55, lr: 4.25e-02, grad_scale: 8.0
2022-12-22 12:07:26,211 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 3.602e+02 7.350e+02 9.194e+02 1.218e+03 2.503e+03, threshold=1.839e+03, percent-clipped=6.0
2022-12-22 12:08:00,131 WARNING [train.py:1060] (2/4) Exclude cut with ID 2929-85685-0062-62366-0_sp0.9 from training. Duration: 22.511125
2022-12-22 12:08:00,154 WARNING [train.py:1060] (2/4) Exclude cut with ID 7395-89880-0031-39906-0 from training. Duration: 20.675
2022-12-22 12:08:39,521 INFO [train.py:894] (2/4) Epoch 2, batch 950, loss[loss=0.2732, simple_loss=0.3421, pruned_loss=0.1021, over 18579.00 frames. ], tot_loss[loss=0.3313, simple_loss=0.383, pruned_loss=0.1398, over 3683073.56 frames. ], batch size: 49, lr: 4.24e-02, grad_scale: 8.0
2022-12-22 12:08:42,974 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.4139, 0.7526, 1.0540, 1.4822, 1.2694, 1.3186, 1.2498, 1.0924],
       device='cuda:2'), covar=tensor([0.0894, 0.1049, 0.2400, 0.0891, 0.0738, 0.0881, 0.0782, 0.0930],
       device='cuda:2'), in_proj_covar=tensor([0.0045, 0.0049, 0.0063, 0.0046, 0.0043, 0.0049, 0.0041, 0.0046],
       device='cuda:2'), out_proj_covar=tensor([4.1770e-05, 3.7996e-05, 5.6020e-05, 4.2563e-05, 3.6965e-05, 4.1575e-05,
        3.5717e-05, 3.9869e-05], device='cuda:2')
2022-12-22 12:09:40,236 WARNING [train.py:1060] (2/4) Exclude cut with ID 6330-62850-0007-91323-0_sp0.9 from training. Duration: 24.9833125
2022-12-22 12:09:41,870 INFO [zipformer.py:660] (2/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=4497.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 12:09:56,186 INFO [train.py:894] (2/4) Epoch 2, batch 1000, loss[loss=0.3105, simple_loss=0.361, pruned_loss=0.13, over 18575.00 frames. ], tot_loss[loss=0.3295, simple_loss=0.3817, pruned_loss=0.1386, over 3690574.29 frames. ], batch size: 49, lr: 4.23e-02, grad_scale: 8.0
2022-12-22 12:09:59,097 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 4.331e+02 6.849e+02 8.394e+02 1.021e+03 1.875e+03, threshold=1.679e+03, percent-clipped=1.0
2022-12-22 12:10:11,261 WARNING [train.py:1060] (2/4) Exclude cut with ID 5239-32139-0047-9341-0 from training. Duration: 27.14
2022-12-22 12:10:27,471 WARNING [train.py:1060] (2/4) Exclude cut with ID 2929-85685-0044-62348-0 from training. Duration: 22.44
2022-12-22 12:10:45,873 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([2.7083, 2.0515, 1.9565, 1.3610, 1.7074, 2.1847, 1.5535, 1.7240],
       device='cuda:2'), covar=tensor([0.0904, 0.1184, 0.2279, 0.2685, 0.1644, 0.0840, 0.1949, 0.2145],
       device='cuda:2'), in_proj_covar=tensor([0.0094, 0.0131, 0.0156, 0.0160, 0.0128, 0.0109, 0.0136, 0.0148],
       device='cuda:2'), out_proj_covar=tensor([0.0001, 0.0001, 0.0002, 0.0002, 0.0001, 0.0001, 0.0001, 0.0002],
       device='cuda:2')
2022-12-22 12:11:12,767 INFO [train.py:894] (2/4) Epoch 2, batch 1050, loss[loss=0.3568, simple_loss=0.4104, pruned_loss=0.1516, over 18725.00 frames. ], tot_loss[loss=0.3293, simple_loss=0.382, pruned_loss=0.1383, over 3695858.43 frames. ], batch size: 54, lr: 4.22e-02, grad_scale: 8.0
2022-12-22 12:11:22,942 INFO [scaling.py:679] (2/4) Whitening: num_groups=8, num_channels=96, metric=1.28 vs. limit=2.0
2022-12-22 12:11:45,506 WARNING [train.py:1060] (2/4) Exclude cut with ID 2929-85685-0060-62364-0_sp0.9 from training. Duration: 21.361125
2022-12-22 12:11:50,998 INFO [zipformer.py:660] (2/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=4582.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 12:11:52,104 WARNING [train.py:1060] (2/4) Exclude cut with ID 2929-85685-0079-62383-0_sp1.1 from training. Duration: 27.0318125
2022-12-22 12:12:00,633 WARNING [train.py:1060] (2/4) Exclude cut with ID 5622-44585-0006-90525-0_sp0.9 from training. Duration: 28.638875
2022-12-22 12:12:15,328 WARNING [train.py:1060] (2/4) Exclude cut with ID 3340-169293-0054-76830-0 from training. Duration: 20.4
2022-12-22 12:12:30,560 INFO [train.py:894] (2/4) Epoch 2, batch 1100, loss[loss=0.3646, simple_loss=0.4031, pruned_loss=0.1631, over 18670.00 frames. ], tot_loss[loss=0.3291, simple_loss=0.3815, pruned_loss=0.1384, over 3700645.95 frames. ], batch size: 175, lr: 4.20e-02, grad_scale: 8.0
2022-12-22 12:12:33,501 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 3.704e+02 6.945e+02 8.446e+02 1.102e+03 2.369e+03, threshold=1.689e+03, percent-clipped=6.0
2022-12-22 12:12:47,136 WARNING [train.py:1060] (2/4) Exclude cut with ID 2929-85685-0071-62375-0 from training. Duration: 20.025
2022-12-22 12:12:47,144 WARNING [train.py:1060] (2/4) Exclude cut with ID 2364-131735-0112-64612-0_sp0.9 from training. Duration: 20.488875
2022-12-22 12:12:53,672 WARNING [train.py:1060] (2/4) Exclude cut with ID 2929-85685-0079-62383-0 from training. Duration: 29.735
2022-12-22 12:13:06,187 INFO [zipformer.py:660] (2/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=4630.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 12:13:27,777 INFO [zipformer.py:660] (2/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=4644.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 12:13:47,611 INFO [train.py:894] (2/4) Epoch 2, batch 1150, loss[loss=0.3264, simple_loss=0.3788, pruned_loss=0.137, over 18667.00 frames. ], tot_loss[loss=0.3264, simple_loss=0.3796, pruned_loss=0.1366, over 3703777.40 frames. ], batch size: 48, lr: 4.19e-02, grad_scale: 8.0
2022-12-22 12:14:19,106 WARNING [train.py:1060] (2/4) Exclude cut with ID 7276-92427-0014-12983-0_sp0.9 from training. Duration: 21.3055625
2022-12-22 12:14:19,150 WARNING [train.py:1060] (2/4) Exclude cut with ID 1025-75365-0008-79168-0_sp0.9 from training. Duration: 22.0666875
2022-12-22 12:14:35,601 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.6034, 0.3770, 1.3752, 0.9608, 1.5789, 1.6159, 1.5029, 1.3108],
       device='cuda:2'), covar=tensor([0.0746, 0.1623, 0.0716, 0.1279, 0.0716, 0.0572, 0.0640, 0.0773],
       device='cuda:2'), in_proj_covar=tensor([0.0060, 0.0099, 0.0077, 0.0112, 0.0093, 0.0063, 0.0063, 0.0090],
       device='cuda:2'), out_proj_covar=tensor([6.6280e-05, 9.8405e-05, 8.2919e-05, 1.1028e-04, 9.6534e-05, 7.2144e-05,
        6.8178e-05, 8.9422e-05], device='cuda:2')
2022-12-22 12:15:00,610 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([2.5349, 2.6080, 2.7583, 1.7322, 2.5315, 1.8456, 0.8728, 1.7506],
       device='cuda:2'), covar=tensor([0.1663, 0.0876, 0.2002, 0.2850, 0.1268, 0.1867, 0.5463, 0.2625],
       device='cuda:2'), in_proj_covar=tensor([0.0095, 0.0060, 0.0125, 0.0089, 0.0071, 0.0082, 0.0129, 0.0096],
       device='cuda:2'), out_proj_covar=tensor([1.2518e-04, 7.9892e-05, 1.6731e-04, 1.0801e-04, 1.0147e-04, 1.0836e-04,
        1.4087e-04, 1.2466e-04], device='cuda:2')
2022-12-22 12:15:04,744 INFO [train.py:894] (2/4) Epoch 2, batch 1200, loss[loss=0.3438, simple_loss=0.3707, pruned_loss=0.1585, over 18588.00 frames. ], tot_loss[loss=0.3248, simple_loss=0.3777, pruned_loss=0.1359, over 3705891.12 frames. ], batch size: 41, lr: 4.18e-02, grad_scale: 8.0
2022-12-22 12:15:07,490 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 3.237e+02 6.724e+02 8.402e+02 1.025e+03 3.359e+03, threshold=1.680e+03, percent-clipped=3.0
2022-12-22 12:16:10,605 WARNING [train.py:1060] (2/4) Exclude cut with ID 2929-85685-0062-62366-0 from training. Duration: 20.26
2022-12-22 12:16:21,794 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([4.5207, 3.9140, 3.6769, 4.0645, 4.0009, 3.9555, 4.6223, 1.4295],
       device='cuda:2'), covar=tensor([0.0544, 0.0497, 0.0717, 0.0324, 0.1277, 0.0643, 0.0476, 0.4202],
       device='cuda:2'), in_proj_covar=tensor([0.0120, 0.0108, 0.0103, 0.0077, 0.0147, 0.0110, 0.0101, 0.0164],
       device='cuda:2'), out_proj_covar=tensor([1.3697e-04, 1.1871e-04, 1.2182e-04, 8.9092e-05, 1.4783e-04, 1.2250e-04,
        1.0973e-04, 1.5231e-04], device='cuda:2')
2022-12-22 12:16:22,828 INFO [train.py:894] (2/4) Epoch 2, batch 1250, loss[loss=0.281, simple_loss=0.3286, pruned_loss=0.1167, over 18527.00 frames. ], tot_loss[loss=0.3235, simple_loss=0.3765, pruned_loss=0.1353, over 3707139.37 frames. ], batch size: 44, lr: 4.17e-02, grad_scale: 8.0
2022-12-22 12:16:24,341 WARNING [train.py:1060] (2/4) Exclude cut with ID 5239-32139-0030-9324-0_sp0.9 from training. Duration: 21.3444375
2022-12-22 12:16:34,032 INFO [scaling.py:679] (2/4) Whitening: num_groups=8, num_channels=96, metric=1.36 vs. limit=2.0
2022-12-22 12:16:51,203 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.2815, 1.2823, 1.1431, 0.9297, 1.3030, 1.3273, 0.5690, 1.2552],
       device='cuda:2'), covar=tensor([0.2364, 0.1872, 0.1869, 0.1163, 0.1813, 0.1760, 0.2606, 0.1385],
       device='cuda:2'), in_proj_covar=tensor([0.0086, 0.0089, 0.0105, 0.0073, 0.0085, 0.0104, 0.0119, 0.0080],
       device='cuda:2'), out_proj_covar=tensor([0.0001, 0.0001, 0.0002, 0.0001, 0.0001, 0.0002, 0.0002, 0.0001],
       device='cuda:2')
2022-12-22 12:17:22,513 WARNING [train.py:1060] (2/4) Exclude cut with ID 497-129325-0061-62254-0_sp1.1 from training. Duration: 0.97725
2022-12-22 12:17:24,339 INFO [zipformer.py:660] (2/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=4797.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 12:17:39,046 INFO [train.py:894] (2/4) Epoch 2, batch 1300, loss[loss=0.3413, simple_loss=0.3975, pruned_loss=0.1426, over 18687.00 frames. ], tot_loss[loss=0.3245, simple_loss=0.3772, pruned_loss=0.1359, over 3708579.55 frames. ], batch size: 60, lr: 4.16e-02, grad_scale: 8.0
2022-12-22 12:17:42,185 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 3.728e+02 7.714e+02 8.750e+02 1.105e+03 2.805e+03, threshold=1.750e+03, percent-clipped=4.0
2022-12-22 12:17:59,955 INFO [scaling.py:679] (2/4) Whitening: num_groups=8, num_channels=192, metric=1.92 vs. limit=2.0
2022-12-22 12:18:03,986 WARNING [train.py:1060] (2/4) Exclude cut with ID 7395-89880-0031-39906-0_sp0.9 from training. Duration: 22.97225
2022-12-22 12:18:14,737 INFO [scaling.py:679] (2/4) Whitening: num_groups=8, num_channels=96, metric=1.52 vs. limit=2.0
2022-12-22 12:18:35,143 WARNING [train.py:1060] (2/4) Exclude cut with ID 7395-89880-0047-39922-0_sp0.9 from training. Duration: 21.97775
2022-12-22 12:18:38,242 INFO [zipformer.py:660] (2/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=4845.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 12:18:49,878 WARNING [train.py:1060] (2/4) Exclude cut with ID 1112-1043-0006-89194-0_sp0.9 from training. Duration: 21.8333125
2022-12-22 12:18:55,618 INFO [train.py:894] (2/4) Epoch 2, batch 1350, loss[loss=0.3242, simple_loss=0.3749, pruned_loss=0.1367, over 18515.00 frames. ], tot_loss[loss=0.3236, simple_loss=0.3768, pruned_loss=0.1352, over 3709852.18 frames. ], batch size: 52, lr: 4.15e-02, grad_scale: 8.0
2022-12-22 12:19:00,303 WARNING [train.py:1060] (2/4) Exclude cut with ID 1914-133440-0031-94921-0 from training. Duration: 20.47
2022-12-22 12:19:02,994 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.2842, 1.3008, 0.9686, 1.5159, 1.6854, 1.2864, 2.9726, 1.3407],
       device='cuda:2'), covar=tensor([0.1044, 0.0918, 0.1089, 0.0694, 0.1075, 0.0887, 0.0098, 0.0864],
       device='cuda:2'), in_proj_covar=tensor([0.0087, 0.0083, 0.0093, 0.0110, 0.0119, 0.0084, 0.0046, 0.0083],
       device='cuda:2'), out_proj_covar=tensor([8.5209e-05, 8.5524e-05, 8.9663e-05, 1.0453e-04, 1.1583e-04, 7.9554e-05,
        4.6330e-05, 7.9660e-05], device='cuda:2')
2022-12-22 12:20:01,739 INFO [scaling.py:679] (2/4) Whitening: num_groups=8, num_channels=192, metric=1.92 vs. limit=2.0
2022-12-22 12:20:08,121 WARNING [train.py:1060] (2/4) Exclude cut with ID 7395-89880-0037-39912-0_sp0.9 from training. Duration: 20.67225
2022-12-22 12:20:12,447 INFO [train.py:894] (2/4) Epoch 2, batch 1400, loss[loss=0.3246, simple_loss=0.377, pruned_loss=0.1361, over 18561.00 frames. ], tot_loss[loss=0.3204, simple_loss=0.3743, pruned_loss=0.1333, over 3710365.26 frames. ], batch size: 78, lr: 4.14e-02, grad_scale: 8.0
2022-12-22 12:20:16,046 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 3.445e+02 6.272e+02 7.556e+02 9.448e+02 1.744e+03, threshold=1.511e+03, percent-clipped=0.0
2022-12-22 12:20:26,276 WARNING [train.py:1060] (2/4) Exclude cut with ID 1914-133440-0024-94914-0_sp0.9 from training. Duration: 25.2444375
2022-12-22 12:20:52,484 WARNING [train.py:1060] (2/4) Exclude cut with ID 3340-169293-0021-76797-0_sp0.9 from training. Duration: 21.1445
2022-12-22 12:21:08,895 INFO [zipformer.py:660] (2/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=4944.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 12:21:28,042 INFO [train.py:894] (2/4) Epoch 2, batch 1450, loss[loss=0.3228, simple_loss=0.3763, pruned_loss=0.1346, over 18726.00 frames. ], tot_loss[loss=0.3205, simple_loss=0.3744, pruned_loss=0.1333, over 3712433.73 frames. ], batch size: 52, lr: 4.13e-02, grad_scale: 8.0
2022-12-22 12:22:06,285 WARNING [train.py:1060] (2/4) Exclude cut with ID 2929-85685-0079-62383-0_sp0.9 from training. Duration: 33.038875
2022-12-22 12:22:21,189 INFO [zipformer.py:660] (2/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=4992.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 12:22:44,246 INFO [train.py:894] (2/4) Epoch 2, batch 1500, loss[loss=0.3447, simple_loss=0.3908, pruned_loss=0.1494, over 18453.00 frames. ], tot_loss[loss=0.3183, simple_loss=0.3728, pruned_loss=0.1318, over 3712628.85 frames. ], batch size: 48, lr: 4.12e-02, grad_scale: 8.0
2022-12-22 12:22:44,289 WARNING [train.py:1060] (2/4) Exclude cut with ID 6426-64291-0000-16059-0_sp0.9 from training. Duration: 20.0944375
2022-12-22 12:22:47,216 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 3.674e+02 6.252e+02 7.708e+02 1.003e+03 2.600e+03, threshold=1.542e+03, percent-clipped=7.0
2022-12-22 12:22:57,590 WARNING [train.py:1060] (2/4) Exclude cut with ID 2929-85685-0044-62348-0_sp1.1 from training. Duration: 20.4
2022-12-22 12:23:04,767 WARNING [train.py:1060] (2/4) Exclude cut with ID 6330-62851-0022-91297-0 from training. Duration: 20.085
2022-12-22 12:23:17,208 WARNING [train.py:1060] (2/4) Exclude cut with ID 4860-13185-0032-76709-0_sp0.9 from training. Duration: 23.07775
2022-12-22 12:23:59,413 INFO [train.py:894] (2/4) Epoch 2, batch 1550, loss[loss=0.3381, simple_loss=0.3748, pruned_loss=0.1507, over 18466.00 frames. ], tot_loss[loss=0.318, simple_loss=0.3729, pruned_loss=0.1316, over 3711866.38 frames. ], batch size: 50, lr: 4.11e-02, grad_scale: 8.0
2022-12-22 12:24:02,436 WARNING [train.py:1060] (2/4) Exclude cut with ID 2929-85685-0044-62348-0_sp0.9 from training. Duration: 24.9333125
2022-12-22 12:24:45,936 WARNING [train.py:1060] (2/4) Exclude cut with ID 5118-111612-0016-124680-0_sp0.9 from training. Duration: 20.388875
2022-12-22 12:24:51,415 INFO [scaling.py:679] (2/4) Whitening: num_groups=8, num_channels=96, metric=1.24 vs. limit=2.0
2022-12-22 12:24:53,294 WARNING [train.py:1060] (2/4) Exclude cut with ID 432-122774-0017-62487-0_sp1.1 from training. Duration: 20.3590625
2022-12-22 12:25:15,875 INFO [train.py:894] (2/4) Epoch 2, batch 1600, loss[loss=0.3148, simple_loss=0.3773, pruned_loss=0.1261, over 18642.00 frames. ], tot_loss[loss=0.3204, simple_loss=0.3743, pruned_loss=0.1332, over 3712378.76 frames. ], batch size: 53, lr: 4.10e-02, grad_scale: 8.0
2022-12-22 12:25:20,224 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 3.493e+02 7.158e+02 8.150e+02 1.051e+03 2.578e+03, threshold=1.630e+03, percent-clipped=2.0
2022-12-22 12:26:02,082 WARNING [train.py:1060] (2/4) Exclude cut with ID 3557-8342-0013-54691-0_sp1.1 from training. Duration: 0.836375
2022-12-22 12:26:31,554 INFO [train.py:894] (2/4) Epoch 2, batch 1650, loss[loss=0.4052, simple_loss=0.421, pruned_loss=0.1947, over 18588.00 frames. ], tot_loss[loss=0.3237, simple_loss=0.3761, pruned_loss=0.1357, over 3713712.65 frames. ], batch size: 51, lr: 4.09e-02, grad_scale: 8.0
2022-12-22 12:26:48,246 WARNING [train.py:1060] (2/4) Exclude cut with ID 8565-290391-0049-67394-0_sp0.9 from training. Duration: 21.3166875
2022-12-22 12:27:19,640 WARNING [train.py:1060] (2/4) Exclude cut with ID 6533-399-0029-104863-0_sp0.9 from training. Duration: 22.1055625
2022-12-22 12:27:28,619 WARNING [train.py:1060] (2/4) Exclude cut with ID 7699-105389-0094-26379-0_sp1.1 from training. Duration: 21.77725
2022-12-22 12:27:49,268 INFO [train.py:894] (2/4) Epoch 2, batch 1700, loss[loss=0.352, simple_loss=0.3708, pruned_loss=0.1666, over 18422.00 frames. ], tot_loss[loss=0.3282, simple_loss=0.378, pruned_loss=0.1392, over 3713062.41 frames. ], batch size: 42, lr: 4.08e-02, grad_scale: 8.0
2022-12-22 12:27:49,350 WARNING [train.py:1060] (2/4) Exclude cut with ID 7255-291500-0005-134304-0_sp0.9 from training. Duration: 27.8166875
2022-12-22 12:27:53,992 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 4.221e+02 7.752e+02 9.535e+02 1.253e+03 3.396e+03, threshold=1.907e+03, percent-clipped=8.0
2022-12-22 12:28:16,168 WARNING [train.py:1060] (2/4) Exclude cut with ID 7357-94126-0021-15852-0_sp1.1 from training. Duration: 22.5090625
2022-12-22 12:28:21,963 WARNING [train.py:1060] (2/4) Exclude cut with ID 7255-291500-0005-134304-0 from training. Duration: 25.035
2022-12-22 12:28:38,848 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.5970, 1.9192, 2.0014, 2.1363, 2.2358, 4.3778, 1.9381, 2.8684],
       device='cuda:2'), covar=tensor([0.4656, 0.2666, 0.2015, 0.2792, 0.2084, 0.0156, 0.2412, 0.1415],
       device='cuda:2'), in_proj_covar=tensor([0.0145, 0.0123, 0.0128, 0.0124, 0.0134, 0.0079, 0.0119, 0.0114],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0001, 0.0002, 0.0002],
       device='cuda:2')
2022-12-22 12:28:41,503 WARNING [train.py:1060] (2/4) Exclude cut with ID 774-127930-0014-10412-0_sp1.1 from training. Duration: 0.95
2022-12-22 12:28:59,844 WARNING [train.py:1060] (2/4) Exclude cut with ID 3033-130750-0096-55598-0_sp0.9 from training. Duration: 0.92225
2022-12-22 12:29:05,313 INFO [train.py:894] (2/4) Epoch 2, batch 1750, loss[loss=0.3951, simple_loss=0.416, pruned_loss=0.1871, over 18700.00 frames. ], tot_loss[loss=0.3359, simple_loss=0.3825, pruned_loss=0.1447, over 3714053.35 frames. ], batch size: 50, lr: 4.07e-02, grad_scale: 8.0
2022-12-22 12:29:26,101 WARNING [train.py:1060] (2/4) Exclude cut with ID 4511-76322-0006-80011-0 from training. Duration: 21.97
2022-12-22 12:29:40,102 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([0.3631, 1.1837, 1.2694, 0.3903, 1.2862, 1.3635, 1.1982, 1.2991],
       device='cuda:2'), covar=tensor([0.0925, 0.0639, 0.0672, 0.0988, 0.0607, 0.0412, 0.0692, 0.0604],
       device='cuda:2'), in_proj_covar=tensor([0.0080, 0.0081, 0.0061, 0.0093, 0.0075, 0.0052, 0.0089, 0.0068],
       device='cuda:2'), out_proj_covar=tensor([8.5350e-05, 9.0216e-05, 7.4741e-05, 1.0204e-04, 8.6115e-05, 6.2266e-05,
        9.9880e-05, 7.6020e-05], device='cuda:2')
2022-12-22 12:29:45,574 WARNING [train.py:1060] (2/4) Exclude cut with ID 7492-105653-0055-62765-0_sp0.9 from training. Duration: 21.97225
2022-12-22 12:29:46,879 WARNING [train.py:1060] (2/4) Exclude cut with ID 453-131332-0000-47844-0_sp0.9 from training. Duration: 25.3333125
2022-12-22 12:29:59,180 WARNING [train.py:1060] (2/4) Exclude cut with ID 5172-29468-0015-19128-0_sp0.9 from training. Duration: 21.5055625
2022-12-22 12:30:09,171 WARNING [train.py:1060] (2/4) Exclude cut with ID 453-131332-0000-47844-0_sp1.1 from training. Duration: 20.72725
2022-12-22 12:30:20,456 INFO [train.py:894] (2/4) Epoch 2, batch 1800, loss[loss=0.3843, simple_loss=0.4119, pruned_loss=0.1784, over 18594.00 frames. ], tot_loss[loss=0.3405, simple_loss=0.3843, pruned_loss=0.1484, over 3713734.71 frames. ], batch size: 51, lr: 4.06e-02, grad_scale: 8.0
2022-12-22 12:30:25,429 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 4.783e+02 7.590e+02 9.192e+02 1.099e+03 2.002e+03, threshold=1.838e+03, percent-clipped=3.0
2022-12-22 12:30:41,021 WARNING [train.py:1060] (2/4) Exclude cut with ID 8631-249866-0030-130156-0_sp0.9 from training. Duration: 26.32775
2022-12-22 12:31:15,485 WARNING [train.py:1060] (2/4) Exclude cut with ID 3867-173237-0077-144769-0 from training. Duration: 20.025
2022-12-22 12:31:21,379 WARNING [train.py:1060] (2/4) Exclude cut with ID 6709-74022-0004-86860-0_sp1.1 from training. Duration: 0.9409375
2022-12-22 12:31:21,392 WARNING [train.py:1060] (2/4) Exclude cut with ID 4757-1811-0023-62229-0_sp0.9 from training. Duration: 21.37775
2022-12-22 12:31:36,685 INFO [train.py:894] (2/4) Epoch 2, batch 1850, loss[loss=0.353, simple_loss=0.3769, pruned_loss=0.1645, over 18518.00 frames. ], tot_loss[loss=0.3443, simple_loss=0.3859, pruned_loss=0.1514, over 3713862.01 frames. ], batch size: 44, lr: 4.05e-02, grad_scale: 8.0
2022-12-22 12:31:40,284 WARNING [train.py:1060] (2/4) Exclude cut with ID 1250-135782-0004-25974-0_sp0.9 from training. Duration: 21.17225
2022-12-22 12:31:40,294 WARNING [train.py:1060] (2/4) Exclude cut with ID 7357-94126-0021-15852-0_sp0.9 from training. Duration: 27.511125
2022-12-22 12:32:11,963 WARNING [train.py:1060] (2/4) Exclude cut with ID 453-131332-0000-47844-0 from training. Duration: 22.8
2022-12-22 12:32:18,035 WARNING [train.py:1060] (2/4) Exclude cut with ID 4964-30587-0040-44509-0 from training. Duration: 22.585
2022-12-22 12:32:48,615 WARNING [train.py:1060] (2/4) Exclude cut with ID 6978-92210-0001-146967-0_sp0.9 from training. Duration: 22.0166875
2022-12-22 12:32:54,505 INFO [train.py:894] (2/4) Epoch 2, batch 1900, loss[loss=0.4196, simple_loss=0.4293, pruned_loss=0.205, over 18729.00 frames. ], tot_loss[loss=0.3482, simple_loss=0.388, pruned_loss=0.1543, over 3714272.77 frames. ], batch size: 54, lr: 4.04e-02, grad_scale: 8.0
2022-12-22 12:32:59,280 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 5.057e+02 7.619e+02 8.990e+02 1.158e+03 2.314e+03, threshold=1.798e+03, percent-clipped=4.0
2022-12-22 12:33:04,316 WARNING [train.py:1060] (2/4) Exclude cut with ID 7255-291500-0003-134302-0_sp1.1 from training. Duration: 24.395375
2022-12-22 12:33:10,208 WARNING [train.py:1060] (2/4) Exclude cut with ID 3488-85273-0017-41203-0_sp0.9 from training. Duration: 27.47775
2022-12-22 12:33:15,977 WARNING [train.py:1060] (2/4) Exclude cut with ID 432-122774-0017-62487-0_sp0.9 from training. Duration: 24.8833125
2022-12-22 12:33:19,031 WARNING [train.py:1060] (2/4) Exclude cut with ID 6758-72288-0033-108368-0 from training. Duration: 23.39
2022-12-22 12:33:24,844 WARNING [train.py:1060] (2/4) Exclude cut with ID 8291-282929-0007-12994-0_sp0.9 from training. Duration: 28.72225
2022-12-22 12:33:35,358 WARNING [train.py:1060] (2/4) Exclude cut with ID 585-294811-0110-133686-0_sp0.9 from training. Duration: 20.8944375
2022-12-22 12:33:50,023 WARNING [train.py:1060] (2/4) Exclude cut with ID 5796-66357-0007-116447-0_sp0.9 from training. Duration: 23.8444375
2022-12-22 12:34:11,298 INFO [train.py:894] (2/4) Epoch 2, batch 1950, loss[loss=0.365, simple_loss=0.3991, pruned_loss=0.1655, over 18646.00 frames. ], tot_loss[loss=0.3525, simple_loss=0.3903, pruned_loss=0.1574, over 3713957.31 frames. ], batch size: 53, lr: 4.03e-02, grad_scale: 8.0
2022-12-22 12:34:13,335 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.2913, 1.3971, 0.8766, 1.4001, 1.3738, 1.1757, 1.9588, 1.2944],
       device='cuda:2'), covar=tensor([0.0960, 0.1082, 0.1789, 0.1290, 0.1816, 0.0888, 0.0335, 0.0900],
       device='cuda:2'), in_proj_covar=tensor([0.0099, 0.0097, 0.0111, 0.0134, 0.0136, 0.0095, 0.0055, 0.0096],
       device='cuda:2'), out_proj_covar=tensor([1.0056e-04, 1.0207e-04, 1.1052e-04, 1.2986e-04, 1.3311e-04, 9.2272e-05,
        5.8917e-05, 9.6189e-05], device='cuda:2')
2022-12-22 12:34:17,307 WARNING [train.py:1060] (2/4) Exclude cut with ID 8291-282929-0007-12994-0 from training. Duration: 25.85
2022-12-22 12:34:17,317 WARNING [train.py:1060] (2/4) Exclude cut with ID 8291-282929-0023-13010-0 from training. Duration: 21.39
2022-12-22 12:34:27,547 WARNING [train.py:1060] (2/4) Exclude cut with ID 7357-94126-0014-15845-0 from training. Duration: 27.92
2022-12-22 12:34:50,685 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([2.2329, 1.8684, 1.6504, 2.7256, 2.2820, 3.7606, 2.2575, 2.2261],
       device='cuda:2'), covar=tensor([0.1600, 0.2265, 0.2239, 0.1608, 0.1862, 0.0418, 0.1487, 0.2106],
       device='cuda:2'), in_proj_covar=tensor([0.0095, 0.0088, 0.0101, 0.0093, 0.0105, 0.0081, 0.0095, 0.0092],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-22 12:34:56,272 WARNING [train.py:1060] (2/4) Exclude cut with ID 8631-249866-0039-130165-0_sp0.9 from training. Duration: 20.661125
2022-12-22 12:35:19,879 WARNING [train.py:1060] (2/4) Exclude cut with ID 7357-94126-0043-15874-0_sp0.9 from training. Duration: 20.07225
2022-12-22 12:35:27,421 INFO [train.py:894] (2/4) Epoch 2, batch 2000, loss[loss=0.3426, simple_loss=0.3758, pruned_loss=0.1547, over 18423.00 frames. ], tot_loss[loss=0.3534, simple_loss=0.39, pruned_loss=0.1584, over 3714298.82 frames. ], batch size: 48, lr: 4.02e-02, grad_scale: 8.0
2022-12-22 12:35:27,766 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.4744, 1.0829, 1.1723, 1.1430, 1.4667, 1.1980, 1.2859, 1.9119],
       device='cuda:2'), covar=tensor([0.2231, 0.2233, 0.2777, 0.2374, 0.3011, 0.2200, 0.2118, 0.1552],
       device='cuda:2'), in_proj_covar=tensor([0.0081, 0.0096, 0.0129, 0.0094, 0.0099, 0.0101, 0.0085, 0.0091],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-22 12:35:28,959 WARNING [train.py:1060] (2/4) Exclude cut with ID 1085-156170-0017-128270-0 from training. Duration: 21.01
2022-12-22 12:35:31,582 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 4.702e+02 7.346e+02 8.906e+02 1.153e+03 2.441e+03, threshold=1.781e+03, percent-clipped=4.0
2022-12-22 12:35:56,053 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([4.0906, 3.5595, 3.6334, 3.6940, 3.7229, 3.7227, 4.0695, 2.2088],
       device='cuda:2'), covar=tensor([0.0525, 0.0361, 0.0462, 0.0369, 0.1395, 0.0545, 0.0533, 0.2883],
       device='cuda:2'), in_proj_covar=tensor([0.0142, 0.0114, 0.0111, 0.0087, 0.0169, 0.0122, 0.0112, 0.0173],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0001, 0.0001, 0.0001, 0.0002, 0.0001, 0.0001, 0.0002],
       device='cuda:2')
2022-12-22 12:36:38,614 WARNING [train.py:1060] (2/4) Exclude cut with ID 2195-150901-0045-59933-0 from training. Duration: 20.65
2022-12-22 12:36:42,803 INFO [train.py:894] (2/4) Epoch 2, batch 2050, loss[loss=0.4018, simple_loss=0.4248, pruned_loss=0.1894, over 18388.00 frames. ], tot_loss[loss=0.3558, simple_loss=0.3912, pruned_loss=0.1601, over 3715119.85 frames. ], batch size: 53, lr: 4.01e-02, grad_scale: 8.0
2022-12-22 12:36:44,669 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.9011, 0.6255, 1.7186, 1.0560, 1.4307, 1.7659, 2.0711, 1.5836],
       device='cuda:2'), covar=tensor([0.0828, 0.1558, 0.0678, 0.1414, 0.0963, 0.0606, 0.0537, 0.0900],
       device='cuda:2'), in_proj_covar=tensor([0.0060, 0.0104, 0.0083, 0.0119, 0.0096, 0.0065, 0.0063, 0.0096],
       device='cuda:2'), out_proj_covar=tensor([6.8893e-05, 1.0565e-04, 9.0472e-05, 1.2164e-04, 1.0051e-04, 7.5959e-05,
        7.2171e-05, 9.8122e-05], device='cuda:2')
2022-12-22 12:36:45,630 WARNING [train.py:1060] (2/4) Exclude cut with ID 5796-66357-0007-116447-0 from training. Duration: 21.46
2022-12-22 12:37:22,489 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.8387, 1.9874, 1.0074, 2.2051, 2.1388, 1.7378, 3.2186, 1.8394],
       device='cuda:2'), covar=tensor([0.0834, 0.1206, 0.1716, 0.1403, 0.1459, 0.0741, 0.0164, 0.0910],
       device='cuda:2'), in_proj_covar=tensor([0.0099, 0.0097, 0.0112, 0.0139, 0.0136, 0.0096, 0.0055, 0.0097],
       device='cuda:2'), out_proj_covar=tensor([1.0084e-04, 1.0433e-04, 1.1213e-04, 1.3458e-04, 1.3429e-04, 9.4212e-05,
        5.9180e-05, 9.7661e-05], device='cuda:2')
2022-12-22 12:37:32,810 WARNING [train.py:1060] (2/4) Exclude cut with ID 3557-8342-0013-54691-0 from training. Duration: 0.92
2022-12-22 12:37:41,196 WARNING [train.py:1060] (2/4) Exclude cut with ID 8291-282929-0023-13010-0_sp0.9 from training. Duration: 23.7666875
2022-12-22 12:38:00,486 INFO [train.py:894] (2/4) Epoch 2, batch 2100, loss[loss=0.3703, simple_loss=0.4092, pruned_loss=0.1657, over 18568.00 frames. ], tot_loss[loss=0.3529, simple_loss=0.3891, pruned_loss=0.1584, over 3714484.92 frames. ], batch size: 57, lr: 4.00e-02, grad_scale: 8.0
2022-12-22 12:38:04,901 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 4.064e+02 6.944e+02 8.737e+02 1.108e+03 7.286e+03, threshold=1.747e+03, percent-clipped=11.0
2022-12-22 12:38:17,085 WARNING [train.py:1060] (2/4) Exclude cut with ID 8544-281189-0060-101339-0_sp0.9 from training. Duration: 20.861125
2022-12-22 12:38:27,924 WARNING [train.py:1060] (2/4) Exclude cut with ID 3488-65654-0031-41259-0_sp0.9 from training. Duration: 22.711125
2022-12-22 12:39:06,101 INFO [zipformer.py:660] (2/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=5650.0, num_to_drop=1, layers_to_drop={0}
2022-12-22 12:39:08,759 WARNING [train.py:1060] (2/4) Exclude cut with ID 6951-79737-0043-132310-0_sp1.1 from training. Duration: 22.986375
2022-12-22 12:39:15,886 INFO [train.py:894] (2/4) Epoch 2, batch 2150, loss[loss=0.2812, simple_loss=0.3274, pruned_loss=0.1175, over 18484.00 frames. ], tot_loss[loss=0.3532, simple_loss=0.3886, pruned_loss=0.159, over 3714269.93 frames. ], batch size: 43, lr: 3.99e-02, grad_scale: 8.0
2022-12-22 12:39:24,953 WARNING [train.py:1060] (2/4) Exclude cut with ID 8040-260924-0003-80960-0_sp0.9 from training. Duration: 22.07225
2022-12-22 12:39:29,347 WARNING [train.py:1060] (2/4) Exclude cut with ID 7699-105389-0045-26330-0_sp0.9 from training. Duration: 20.3055625
2022-12-22 12:39:32,971 WARNING [train.py:1060] (2/4) Exclude cut with ID 6356-271890-0060-94317-0_sp0.9 from training. Duration: 20.72225
2022-12-22 12:39:52,290 WARNING [train.py:1060] (2/4) Exclude cut with ID 3488-85273-0017-41203-0_sp1.1 from training. Duration: 22.4818125
2022-12-22 12:40:12,203 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.5939, 1.7552, 0.9920, 2.0188, 1.9725, 1.5564, 2.7397, 1.5861],
       device='cuda:2'), covar=tensor([0.0900, 0.1214, 0.1798, 0.1225, 0.1417, 0.0729, 0.0190, 0.1000],
       device='cuda:2'), in_proj_covar=tensor([0.0103, 0.0101, 0.0115, 0.0142, 0.0138, 0.0099, 0.0058, 0.0100],
       device='cuda:2'), out_proj_covar=tensor([1.0580e-04, 1.0899e-04, 1.1529e-04, 1.3843e-04, 1.3734e-04, 9.7677e-05,
        6.2464e-05, 1.0112e-04], device='cuda:2')
2022-12-22 12:40:15,301 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.6990, 0.7617, 1.2253, 1.6078, 1.3474, 1.3128, 1.4858, 1.0017],
       device='cuda:2'), covar=tensor([0.0735, 0.1362, 0.2228, 0.1023, 0.0632, 0.0725, 0.0917, 0.0866],
       device='cuda:2'), in_proj_covar=tensor([0.0068, 0.0082, 0.0100, 0.0080, 0.0067, 0.0072, 0.0069, 0.0072],
       device='cuda:2'), out_proj_covar=tensor([6.9005e-05, 7.0055e-05, 9.3035e-05, 8.2759e-05, 6.4025e-05, 6.9523e-05,
        6.7031e-05, 6.7194e-05], device='cuda:2')
2022-12-22 12:40:17,789 WARNING [train.py:1060] (2/4) Exclude cut with ID 4964-30587-0040-44509-0_sp0.9 from training. Duration: 25.0944375
2022-12-22 12:40:21,791 WARNING [train.py:1060] (2/4) Exclude cut with ID 6533-399-0047-104881-0 from training. Duration: 21.515
2022-12-22 12:40:28,915 WARNING [train.py:1060] (2/4) Exclude cut with ID 7357-94126-0009-15840-0_sp0.9 from training. Duration: 27.02225
2022-12-22 12:40:31,689 INFO [train.py:894] (2/4) Epoch 2, batch 2200, loss[loss=0.3439, simple_loss=0.3924, pruned_loss=0.1477, over 18554.00 frames. ], tot_loss[loss=0.3544, simple_loss=0.3897, pruned_loss=0.1595, over 3714401.15 frames. ], batch size: 55, lr: 3.98e-02, grad_scale: 8.0
2022-12-22 12:40:34,685 WARNING [train.py:1060] (2/4) Exclude cut with ID 432-122774-0010-62480-0_sp0.9 from training. Duration: 22.22225
2022-12-22 12:40:36,068 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 4.664e+02 7.690e+02 8.741e+02 1.076e+03 1.668e+03, threshold=1.748e+03, percent-clipped=0.0
2022-12-22 12:40:38,050 INFO [zipformer.py:660] (2/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=5711.0, num_to_drop=1, layers_to_drop={1}
2022-12-22 12:40:41,879 WARNING [train.py:1060] (2/4) Exclude cut with ID 4964-30587-0085-44554-0_sp0.9 from training. Duration: 20.85
2022-12-22 12:40:48,805 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.8699, 1.2484, 1.3518, 1.9055, 1.4803, 1.4586, 1.6058, 1.0739],
       device='cuda:2'), covar=tensor([0.0633, 0.1148, 0.1926, 0.0979, 0.0560, 0.0657, 0.0799, 0.0834],
       device='cuda:2'), in_proj_covar=tensor([0.0069, 0.0083, 0.0100, 0.0080, 0.0067, 0.0073, 0.0070, 0.0073],
       device='cuda:2'), out_proj_covar=tensor([6.9444e-05, 7.1143e-05, 9.3573e-05, 8.3632e-05, 6.4121e-05, 6.9906e-05,
        6.7781e-05, 6.8014e-05], device='cuda:2')
2022-12-22 12:41:01,948 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.8791, 0.8048, 1.7912, 1.2344, 1.6251, 1.8603, 2.0172, 1.6337],
       device='cuda:2'), covar=tensor([0.0860, 0.1825, 0.0913, 0.1464, 0.1026, 0.0602, 0.0676, 0.1039],
       device='cuda:2'), in_proj_covar=tensor([0.0060, 0.0106, 0.0088, 0.0119, 0.0099, 0.0068, 0.0064, 0.0100],
       device='cuda:2'), out_proj_covar=tensor([6.9809e-05, 1.0826e-04, 9.5093e-05, 1.2208e-04, 1.0477e-04, 7.9357e-05,
        7.3793e-05, 1.0299e-04], device='cuda:2')
2022-12-22 12:41:15,261 WARNING [train.py:1060] (2/4) Exclude cut with ID 4295-39940-0007-92567-0 from training. Duration: 21.54
2022-12-22 12:41:20,997 WARNING [train.py:1060] (2/4) Exclude cut with ID 4964-30587-0040-44509-0_sp1.1 from training. Duration: 20.5318125
2022-12-22 12:41:29,887 WARNING [train.py:1060] (2/4) Exclude cut with ID 7255-291500-0012-134311-0_sp0.9 from training. Duration: 21.9333125
2022-12-22 12:41:31,633 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.8701, 1.2678, 0.8394, 1.6198, 1.5682, 3.1375, 1.3370, 1.2693],
       device='cuda:2'), covar=tensor([0.1400, 0.2476, 0.2392, 0.1888, 0.1986, 0.0544, 0.1884, 0.2463],
       device='cuda:2'), in_proj_covar=tensor([0.0089, 0.0088, 0.0097, 0.0094, 0.0104, 0.0080, 0.0095, 0.0092],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-22 12:41:44,799 INFO [zipformer.py:660] (2/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=5755.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 12:41:47,251 INFO [train.py:894] (2/4) Epoch 2, batch 2250, loss[loss=0.3564, simple_loss=0.3999, pruned_loss=0.1564, over 18397.00 frames. ], tot_loss[loss=0.3556, simple_loss=0.39, pruned_loss=0.1606, over 3714187.22 frames. ], batch size: 53, lr: 3.97e-02, grad_scale: 8.0
2022-12-22 12:42:13,646 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.9582, 1.1190, 1.3423, 1.6620, 1.5298, 1.4060, 1.5659, 1.0075],
       device='cuda:2'), covar=tensor([0.0613, 0.1231, 0.2091, 0.0920, 0.0564, 0.0676, 0.0779, 0.0805],
       device='cuda:2'), in_proj_covar=tensor([0.0069, 0.0083, 0.0099, 0.0080, 0.0067, 0.0073, 0.0069, 0.0072],
       device='cuda:2'), out_proj_covar=tensor([7.0182e-05, 7.1658e-05, 9.2505e-05, 8.2593e-05, 6.4087e-05, 7.0140e-05,
        6.7129e-05, 6.7406e-05], device='cuda:2')
2022-12-22 12:42:17,592 WARNING [train.py:1060] (2/4) Exclude cut with ID 8631-249866-0025-130151-0_sp0.9 from training. Duration: 21.7944375
2022-12-22 12:42:25,408 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([2.2818, 1.7770, 2.5692, 3.0678, 2.2202, 1.9627, 1.1756, 2.1572],
       device='cuda:2'), covar=tensor([0.2745, 0.2409, 0.2197, 0.0682, 0.2176, 0.2027, 0.3054, 0.1523],
       device='cuda:2'), in_proj_covar=tensor([0.0092, 0.0096, 0.0107, 0.0075, 0.0088, 0.0108, 0.0126, 0.0083],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0001, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-22 12:42:31,806 WARNING [train.py:1060] (2/4) Exclude cut with ID 8291-282929-0002-12989-0_sp0.9 from training. Duration: 22.4666875
2022-12-22 12:42:37,492 WARNING [train.py:1060] (2/4) Exclude cut with ID 6121-9014-0076-24124-0 from training. Duration: 21.635
2022-12-22 12:42:43,708 WARNING [train.py:1060] (2/4) Exclude cut with ID 6121-9014-0076-24124-0_sp0.9 from training. Duration: 24.038875
2022-12-22 12:43:04,005 INFO [train.py:894] (2/4) Epoch 2, batch 2300, loss[loss=0.3862, simple_loss=0.4103, pruned_loss=0.181, over 18592.00 frames. ], tot_loss[loss=0.3545, simple_loss=0.3889, pruned_loss=0.16, over 3713800.13 frames. ], batch size: 56, lr: 3.96e-02, grad_scale: 8.0
2022-12-22 12:43:08,408 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 5.429e+02 8.168e+02 9.351e+02 1.113e+03 2.351e+03, threshold=1.870e+03, percent-clipped=5.0
2022-12-22 12:43:17,857 INFO [zipformer.py:660] (2/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=5816.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 12:43:26,147 WARNING [train.py:1060] (2/4) Exclude cut with ID 7255-291500-0009-134308-0_sp1.1 from training. Duration: 21.786375
2022-12-22 12:43:39,426 WARNING [train.py:1060] (2/4) Exclude cut with ID 8291-282929-0002-12989-0 from training. Duration: 20.22
2022-12-22 12:44:19,724 INFO [train.py:894] (2/4) Epoch 2, batch 2350, loss[loss=0.37, simple_loss=0.3984, pruned_loss=0.1708, over 18588.00 frames. ], tot_loss[loss=0.3542, simple_loss=0.3885, pruned_loss=0.1599, over 3714610.95 frames. ], batch size: 183, lr: 3.95e-02, grad_scale: 8.0
2022-12-22 12:44:21,665 INFO [zipformer.py:660] (2/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=5858.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 12:45:29,286 INFO [scaling.py:679] (2/4) Whitening: num_groups=8, num_channels=192, metric=1.85 vs. limit=2.0
2022-12-22 12:45:35,756 INFO [train.py:894] (2/4) Epoch 2, batch 2400, loss[loss=0.4379, simple_loss=0.4482, pruned_loss=0.2138, over 18518.00 frames. ], tot_loss[loss=0.3522, simple_loss=0.3872, pruned_loss=0.1586, over 3715459.04 frames. ], batch size: 58, lr: 3.94e-02, grad_scale: 8.0
2022-12-22 12:45:40,240 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 4.433e+02 7.534e+02 9.546e+02 1.123e+03 1.593e+03, threshold=1.909e+03, percent-clipped=0.0
2022-12-22 12:45:40,294 WARNING [train.py:1060] (2/4) Exclude cut with ID 6951-79737-0043-132310-0 from training. Duration: 25.285
2022-12-22 12:45:49,341 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.9565, 1.3854, 2.1138, 1.6151, 2.1799, 2.7291, 2.4833, 2.1343],
       device='cuda:2'), covar=tensor([0.1529, 0.1835, 0.0786, 0.1279, 0.1050, 0.0403, 0.0419, 0.0891],
       device='cuda:2'), in_proj_covar=tensor([0.0063, 0.0111, 0.0095, 0.0124, 0.0104, 0.0069, 0.0068, 0.0100],
       device='cuda:2'), out_proj_covar=tensor([7.2551e-05, 1.1395e-04, 1.0341e-04, 1.2686e-04, 1.1006e-04, 8.1334e-05,
        7.7367e-05, 1.0313e-04], device='cuda:2')
2022-12-22 12:45:53,632 INFO [zipformer.py:660] (2/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=5919.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 12:46:24,091 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([0.8820, 1.5030, 2.6184, 0.7977, 1.9054, 2.3133, 1.4823, 1.9972],
       device='cuda:2'), covar=tensor([0.1714, 0.0894, 0.0723, 0.1604, 0.0936, 0.0640, 0.1208, 0.1383],
       device='cuda:2'), in_proj_covar=tensor([0.0084, 0.0085, 0.0061, 0.0098, 0.0079, 0.0055, 0.0095, 0.0068],
       device='cuda:2'), out_proj_covar=tensor([9.0792e-05, 9.4513e-05, 7.5866e-05, 1.0819e-04, 9.1292e-05, 6.4824e-05,
        1.0825e-04, 7.7134e-05], device='cuda:2')
2022-12-22 12:46:44,686 WARNING [train.py:1060] (2/4) Exclude cut with ID 811-130148-0001-63453-0_sp0.9 from training. Duration: 20.861125
2022-12-22 12:46:52,383 INFO [train.py:894] (2/4) Epoch 2, batch 2450, loss[loss=0.4107, simple_loss=0.4357, pruned_loss=0.1929, over 18545.00 frames. ], tot_loss[loss=0.3526, simple_loss=0.3877, pruned_loss=0.1587, over 3714834.43 frames. ], batch size: 58, lr: 3.93e-02, grad_scale: 8.0
2022-12-22 12:47:07,530 WARNING [train.py:1060] (2/4) Exclude cut with ID 6010-56788-0055-90261-0 from training. Duration: 20.88
2022-12-22 12:47:41,289 WARNING [train.py:1060] (2/4) Exclude cut with ID 7357-94126-0045-15876-0_sp0.9 from training. Duration: 23.4166875
2022-12-22 12:47:43,117 INFO [zipformer.py:660] (2/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=5990.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 12:47:44,635 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.4711, 1.4984, 0.8030, 1.5004, 1.6972, 1.4380, 2.6678, 1.5869],
       device='cuda:2'), covar=tensor([0.1079, 0.1220, 0.1937, 0.1412, 0.1532, 0.0954, 0.0241, 0.1020],
       device='cuda:2'), in_proj_covar=tensor([0.0104, 0.0106, 0.0119, 0.0148, 0.0144, 0.0102, 0.0062, 0.0105],
       device='cuda:2'), out_proj_covar=tensor([1.0699e-04, 1.1440e-04, 1.2096e-04, 1.4477e-04, 1.4386e-04, 1.0126e-04,
        6.8393e-05, 1.0749e-04], device='cuda:2')
2022-12-22 12:48:09,207 INFO [zipformer.py:660] (2/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=6006.0, num_to_drop=1, layers_to_drop={0}
2022-12-22 12:48:10,618 INFO [train.py:894] (2/4) Epoch 2, batch 2500, loss[loss=0.3544, simple_loss=0.383, pruned_loss=0.1629, over 18711.00 frames. ], tot_loss[loss=0.3521, simple_loss=0.3875, pruned_loss=0.1583, over 3714789.35 frames. ], batch size: 50, lr: 3.92e-02, grad_scale: 8.0
2022-12-22 12:48:14,727 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 4.130e+02 7.982e+02 9.439e+02 1.211e+03 2.213e+03, threshold=1.888e+03, percent-clipped=1.0
2022-12-22 12:48:59,390 WARNING [train.py:1060] (2/4) Exclude cut with ID 7255-291500-0006-134305-0 from training. Duration: 21.24
2022-12-22 12:48:59,404 WARNING [train.py:1060] (2/4) Exclude cut with ID 6533-399-0047-104881-0_sp0.9 from training. Duration: 23.9055625
2022-12-22 12:49:05,948 INFO [zipformer.py:660] (2/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=6043.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 12:49:11,672 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.8157, 1.7491, 1.3101, 2.0304, 1.6302, 1.5329, 1.6540, 1.8178],
       device='cuda:2'), covar=tensor([0.1463, 0.1147, 0.1202, 0.1200, 0.1143, 0.0785, 0.1415, 0.0609],
       device='cuda:2'), in_proj_covar=tensor([0.0131, 0.0085, 0.0098, 0.0129, 0.0089, 0.0094, 0.0110, 0.0081],
       device='cuda:2'), out_proj_covar=tensor([1.4172e-04, 9.8203e-05, 9.6221e-05, 1.2978e-04, 1.0538e-04, 9.7122e-05,
        1.2186e-04, 8.9423e-05], device='cuda:2')
2022-12-22 12:49:17,271 INFO [zipformer.py:660] (2/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=6051.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 12:49:19,269 INFO [scaling.py:679] (2/4) Whitening: num_groups=1, num_channels=384, metric=3.59 vs. limit=5.0
2022-12-22 12:49:26,038 INFO [train.py:894] (2/4) Epoch 2, batch 2550, loss[loss=0.39, simple_loss=0.4151, pruned_loss=0.1824, over 18446.00 frames. ], tot_loss[loss=0.3528, simple_loss=0.3877, pruned_loss=0.1589, over 3714708.23 frames. ], batch size: 68, lr: 3.91e-02, grad_scale: 8.0
2022-12-22 12:49:32,419 WARNING [train.py:1060] (2/4) Exclude cut with ID 6758-72288-0033-108368-0_sp0.9 from training. Duration: 25.988875
2022-12-22 12:49:40,788 WARNING [train.py:1060] (2/4) Exclude cut with ID 7255-291500-0001-134300-0_sp0.9 from training. Duration: 20.67225
2022-12-22 12:50:14,076 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.7674, 2.1372, 1.4902, 2.4170, 2.7764, 1.3932, 2.1072, 1.2523],
       device='cuda:2'), covar=tensor([0.1557, 0.1721, 0.1348, 0.0898, 0.1063, 0.1397, 0.1416, 0.1568],
       device='cuda:2'), in_proj_covar=tensor([0.0146, 0.0125, 0.0126, 0.0113, 0.0142, 0.0123, 0.0115, 0.0126],
       device='cuda:2'), out_proj_covar=tensor([0.0001, 0.0001, 0.0001, 0.0001, 0.0002, 0.0001, 0.0001, 0.0001],
       device='cuda:2')
2022-12-22 12:50:21,181 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.8337, 1.8175, 0.8912, 1.8341, 1.9134, 1.4634, 3.1447, 1.7661],
       device='cuda:2'), covar=tensor([0.0961, 0.1388, 0.2137, 0.1891, 0.2071, 0.0902, 0.0244, 0.1065],
       device='cuda:2'), in_proj_covar=tensor([0.0107, 0.0105, 0.0123, 0.0152, 0.0149, 0.0103, 0.0064, 0.0109],
       device='cuda:2'), out_proj_covar=tensor([1.1085e-04, 1.1354e-04, 1.2572e-04, 1.4896e-04, 1.4824e-04, 1.0329e-04,
        7.0715e-05, 1.1228e-04], device='cuda:2')
2022-12-22 12:50:29,562 WARNING [train.py:1060] (2/4) Exclude cut with ID 3488-85273-0038-41224-0 from training. Duration: 20.34
2022-12-22 12:50:38,726 INFO [zipformer.py:660] (2/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=6104.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 12:50:40,444 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.3253, 1.0473, 1.3597, 1.3171, 1.1932, 1.7677, 0.8528, 1.3040],
       device='cuda:2'), covar=tensor([0.2476, 0.2182, 0.1794, 0.0993, 0.2169, 0.1286, 0.2407, 0.1464],
       device='cuda:2'), in_proj_covar=tensor([0.0093, 0.0095, 0.0106, 0.0072, 0.0089, 0.0105, 0.0124, 0.0085],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0001, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-22 12:50:43,045 INFO [train.py:894] (2/4) Epoch 2, batch 2600, loss[loss=0.3278, simple_loss=0.3617, pruned_loss=0.1469, over 18704.00 frames. ], tot_loss[loss=0.3512, simple_loss=0.3861, pruned_loss=0.1581, over 3714720.43 frames. ], batch size: 50, lr: 3.90e-02, grad_scale: 8.0
2022-12-22 12:50:47,638 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 3.975e+02 7.215e+02 8.875e+02 1.232e+03 1.892e+03, threshold=1.775e+03, percent-clipped=1.0
2022-12-22 12:50:49,975 INFO [zipformer.py:660] (2/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=6111.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 12:50:57,447 INFO [scaling.py:679] (2/4) Whitening: num_groups=1, num_channels=384, metric=3.20 vs. limit=5.0
2022-12-22 12:51:08,049 INFO [scaling.py:679] (2/4) Whitening: num_groups=8, num_channels=96, metric=1.52 vs. limit=2.0
2022-12-22 12:51:26,022 INFO [zipformer.py:660] (2/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=6135.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 12:51:42,642 WARNING [train.py:1060] (2/4) Exclude cut with ID 7357-94126-0026-15857-0_sp0.9 from training. Duration: 25.061125
2022-12-22 12:51:53,977 WARNING [train.py:1060] (2/4) Exclude cut with ID 3033-130750-0096-55598-0 from training. Duration: 0.83
2022-12-22 12:52:01,065 INFO [train.py:894] (2/4) Epoch 2, batch 2650, loss[loss=0.3393, simple_loss=0.3806, pruned_loss=0.149, over 18713.00 frames. ], tot_loss[loss=0.3505, simple_loss=0.3858, pruned_loss=0.1576, over 3715334.60 frames. ], batch size: 50, lr: 3.89e-02, grad_scale: 8.0
2022-12-22 12:52:18,853 WARNING [train.py:1060] (2/4) Exclude cut with ID 3488-85273-0017-41203-0 from training. Duration: 24.73
2022-12-22 12:52:20,494 INFO [zipformer.py:660] (2/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=6170.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 12:52:32,337 WARNING [train.py:1060] (2/4) Exclude cut with ID 7255-291500-0009-134308-0 from training. Duration: 23.965
2022-12-22 12:52:39,653 WARNING [train.py:1060] (2/4) Exclude cut with ID 6978-92210-0030-146996-0_sp0.9 from training. Duration: 22.088875
2022-12-22 12:52:44,959 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([3.8354, 3.1816, 3.1877, 3.3988, 3.3065, 3.3164, 3.9586, 1.4795],
       device='cuda:2'), covar=tensor([0.0903, 0.0701, 0.0917, 0.0529, 0.1894, 0.0868, 0.0522, 0.4501],
       device='cuda:2'), in_proj_covar=tensor([0.0151, 0.0122, 0.0112, 0.0091, 0.0169, 0.0125, 0.0118, 0.0179],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0001, 0.0001, 0.0001, 0.0002, 0.0001, 0.0001, 0.0002],
       device='cuda:2')
2022-12-22 12:52:56,993 WARNING [train.py:1060] (2/4) Exclude cut with ID 7255-291500-0006-134305-0_sp0.9 from training. Duration: 23.6
2022-12-22 12:53:00,815 INFO [zipformer.py:660] (2/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=6196.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 12:53:11,399 INFO [zipformer.py:660] (2/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=6203.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 12:53:17,461 INFO [train.py:894] (2/4) Epoch 2, batch 2700, loss[loss=0.3, simple_loss=0.3409, pruned_loss=0.1295, over 18602.00 frames. ], tot_loss[loss=0.3512, simple_loss=0.3861, pruned_loss=0.1582, over 3715567.67 frames. ], batch size: 45, lr: 3.88e-02, grad_scale: 8.0
2022-12-22 12:53:22,119 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 4.865e+02 7.324e+02 9.004e+02 1.200e+03 2.759e+03, threshold=1.801e+03, percent-clipped=6.0
2022-12-22 12:53:23,845 INFO [zipformer.py:660] (2/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=6211.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 12:53:28,220 INFO [zipformer.py:660] (2/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=6214.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 12:53:28,842 INFO [scaling.py:679] (2/4) Whitening: num_groups=8, num_channels=192, metric=1.95 vs. limit=2.0
2022-12-22 12:53:53,607 INFO [zipformer.py:660] (2/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=6231.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 12:54:33,135 INFO [scaling.py:679] (2/4) Whitening: num_groups=1, num_channels=384, metric=3.64 vs. limit=5.0
2022-12-22 12:54:33,802 INFO [train.py:894] (2/4) Epoch 2, batch 2750, loss[loss=0.3624, simple_loss=0.4064, pruned_loss=0.1592, over 18547.00 frames. ], tot_loss[loss=0.3481, simple_loss=0.3843, pruned_loss=0.156, over 3715312.50 frames. ], batch size: 55, lr: 3.87e-02, grad_scale: 8.0
2022-12-22 12:54:39,440 WARNING [train.py:1060] (2/4) Exclude cut with ID 8291-282929-0024-13011-0 from training. Duration: 23.795
2022-12-22 12:54:44,676 INFO [zipformer.py:660] (2/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=6264.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 12:54:47,519 INFO [zipformer.py:660] (2/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=6266.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 12:54:54,421 WARNING [train.py:1060] (2/4) Exclude cut with ID 8631-249866-0030-130156-0_sp1.1 from training. Duration: 21.5409375
2022-12-22 12:54:56,111 WARNING [train.py:1060] (2/4) Exclude cut with ID 6978-92210-0019-146985-0_sp0.9 from training. Duration: 24.97775
2022-12-22 12:54:56,511 INFO [zipformer.py:660] (2/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=6272.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 12:55:06,554 WARNING [train.py:1060] (2/4) Exclude cut with ID 1085-156170-0017-128270-0_sp0.9 from training. Duration: 23.3444375
2022-12-22 12:55:30,059 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.6806, 1.5290, 1.1969, 0.4020, 1.4086, 1.5313, 1.2125, 1.2073],
       device='cuda:2'), covar=tensor([0.0709, 0.0602, 0.1304, 0.1664, 0.0977, 0.1192, 0.1454, 0.1134],
       device='cuda:2'), in_proj_covar=tensor([0.0105, 0.0140, 0.0180, 0.0173, 0.0153, 0.0142, 0.0146, 0.0167],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-22 12:55:32,639 WARNING [train.py:1060] (2/4) Exclude cut with ID 6010-56788-0055-90261-0_sp0.9 from training. Duration: 23.2
2022-12-22 12:55:40,595 WARNING [train.py:1060] (2/4) Exclude cut with ID 5653-46179-0060-117930-0_sp0.9 from training. Duration: 21.17225
2022-12-22 12:55:48,323 INFO [zipformer.py:660] (2/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=6306.0, num_to_drop=1, layers_to_drop={0}
2022-12-22 12:55:49,450 INFO [train.py:894] (2/4) Epoch 2, batch 2800, loss[loss=0.2902, simple_loss=0.3355, pruned_loss=0.1225, over 18669.00 frames. ], tot_loss[loss=0.351, simple_loss=0.3862, pruned_loss=0.1579, over 3715625.39 frames. ], batch size: 48, lr: 3.86e-02, grad_scale: 8.0
2022-12-22 12:55:53,787 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 4.276e+02 7.690e+02 9.238e+02 1.166e+03 2.440e+03, threshold=1.848e+03, percent-clipped=6.0
2022-12-22 12:55:59,639 WARNING [train.py:1060] (2/4) Exclude cut with ID 7255-291500-0008-134307-0_sp0.9 from training. Duration: 24.6555625
2022-12-22 12:56:18,845 INFO [zipformer.py:660] (2/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=6327.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 12:56:48,184 INFO [zipformer.py:660] (2/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=6346.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 12:56:57,061 WARNING [train.py:1060] (2/4) Exclude cut with ID 3488-65654-0031-41259-0 from training. Duration: 20.44
2022-12-22 12:57:00,656 INFO [zipformer.py:660] (2/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=6354.0, num_to_drop=1, layers_to_drop={0}
2022-12-22 12:57:04,353 INFO [train.py:894] (2/4) Epoch 2, batch 2850, loss[loss=0.3359, simple_loss=0.3673, pruned_loss=0.1523, over 18407.00 frames. ], tot_loss[loss=0.3521, simple_loss=0.3873, pruned_loss=0.1585, over 3715427.56 frames. ], batch size: 46, lr: 3.85e-02, grad_scale: 8.0
2022-12-22 12:57:11,450 WARNING [train.py:1060] (2/4) Exclude cut with ID 6951-79737-0018-132285-0_sp0.9 from training. Duration: 23.45
2022-12-22 12:57:26,082 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([2.4046, 2.2755, 1.5990, 0.7456, 1.8254, 1.8593, 1.3067, 1.4844],
       device='cuda:2'), covar=tensor([0.0838, 0.0480, 0.1731, 0.2132, 0.1275, 0.1285, 0.1600, 0.1456],
       device='cuda:2'), in_proj_covar=tensor([0.0103, 0.0139, 0.0182, 0.0174, 0.0154, 0.0140, 0.0143, 0.0167],
       device='cuda:2'), out_proj_covar=tensor([0.0001, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-22 12:57:43,971 WARNING [train.py:1060] (2/4) Exclude cut with ID 6945-60535-0076-12784-0_sp0.9 from training. Duration: 20.52225
2022-12-22 12:57:49,730 WARNING [train.py:1060] (2/4) Exclude cut with ID 7255-291500-0008-134307-0 from training. Duration: 22.19
2022-12-22 12:57:55,991 INFO [scaling.py:679] (2/4) Whitening: num_groups=8, num_channels=96, metric=1.57 vs. limit=2.0
2022-12-22 12:58:00,018 WARNING [train.py:1060] (2/4) Exclude cut with ID 7357-94126-0014-15845-0_sp1.1 from training. Duration: 25.3818125
2022-12-22 12:58:08,925 INFO [zipformer.py:660] (2/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=6399.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 12:58:16,777 WARNING [train.py:1060] (2/4) Exclude cut with ID 6951-79737-0043-132310-0_sp0.9 from training. Duration: 28.0944375
2022-12-22 12:58:21,112 INFO [train.py:894] (2/4) Epoch 2, batch 2900, loss[loss=0.3737, simple_loss=0.4014, pruned_loss=0.173, over 18668.00 frames. ], tot_loss[loss=0.3513, simple_loss=0.3873, pruned_loss=0.1577, over 3715253.12 frames. ], batch size: 60, lr: 3.85e-02, grad_scale: 8.0
2022-12-22 12:58:24,162 WARNING [train.py:1060] (2/4) Exclude cut with ID 2195-150901-0045-59933-0_sp0.9 from training. Duration: 22.9444375
2022-12-22 12:58:25,477 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 3.875e+02 7.430e+02 9.420e+02 1.139e+03 3.139e+03, threshold=1.884e+03, percent-clipped=2.0
2022-12-22 12:58:27,325 INFO [zipformer.py:660] (2/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=6411.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 12:58:33,093 WARNING [train.py:1060] (2/4) Exclude cut with ID 8291-282929-0024-13011-0_sp1.1 from training. Duration: 21.6318125
2022-12-22 12:58:37,766 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([3.2257, 5.0355, 5.1826, 3.1162, 5.3315, 3.9373, 1.3571, 3.5287],
       device='cuda:2'), covar=tensor([0.1126, 0.0511, 0.0865, 0.2444, 0.0502, 0.0919, 0.4706, 0.1600],
       device='cuda:2'), in_proj_covar=tensor([0.0104, 0.0074, 0.0133, 0.0103, 0.0085, 0.0088, 0.0136, 0.0104],
       device='cuda:2'), out_proj_covar=tensor([0.0001, 0.0001, 0.0002, 0.0001, 0.0001, 0.0001, 0.0002, 0.0001],
       device='cuda:2')
2022-12-22 12:58:42,627 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.7901, 0.7828, 1.3135, 1.4419, 1.4208, 1.3634, 1.3933, 0.9599],
       device='cuda:2'), covar=tensor([0.0679, 0.1408, 0.2253, 0.1147, 0.0642, 0.0719, 0.0926, 0.0897],
       device='cuda:2'), in_proj_covar=tensor([0.0078, 0.0103, 0.0118, 0.0100, 0.0079, 0.0086, 0.0086, 0.0086],
       device='cuda:2'), out_proj_covar=tensor([8.4504e-05, 9.2659e-05, 1.1210e-04, 1.0636e-04, 7.8295e-05, 8.4151e-05,
        8.6340e-05, 8.1618e-05], device='cuda:2')
2022-12-22 12:58:48,138 WARNING [train.py:1060] (2/4) Exclude cut with ID 8631-249866-0030-130156-0 from training. Duration: 23.695
2022-12-22 12:58:55,183 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.9613, 1.3979, 2.2402, 1.4774, 2.0438, 2.6944, 2.2664, 2.5163],
       device='cuda:2'), covar=tensor([0.1539, 0.1678, 0.0773, 0.1633, 0.0960, 0.0613, 0.0972, 0.1022],
       device='cuda:2'), in_proj_covar=tensor([0.0060, 0.0100, 0.0091, 0.0118, 0.0097, 0.0067, 0.0067, 0.0093],
       device='cuda:2'), out_proj_covar=tensor([6.9624e-05, 1.0350e-04, 9.9562e-05, 1.2323e-04, 1.0372e-04, 7.9067e-05,
        7.7837e-05, 9.7244e-05], device='cuda:2')
2022-12-22 12:58:55,841 INFO [scaling.py:679] (2/4) Whitening: num_groups=1, num_channels=384, metric=4.67 vs. limit=5.0
2022-12-22 12:59:16,852 WARNING [train.py:1060] (2/4) Exclude cut with ID 7699-105389-0094-26379-0 from training. Duration: 23.955
2022-12-22 12:59:37,903 INFO [train.py:894] (2/4) Epoch 2, batch 2950, loss[loss=0.3117, simple_loss=0.3637, pruned_loss=0.1298, over 18440.00 frames. ], tot_loss[loss=0.3484, simple_loss=0.3852, pruned_loss=0.1558, over 3715000.91 frames. ], batch size: 48, lr: 3.84e-02, grad_scale: 8.0
2022-12-22 12:59:40,992 INFO [zipformer.py:660] (2/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=6459.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 12:59:41,187 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.2928, 1.0762, 1.2791, 1.3773, 1.1535, 1.6559, 0.7309, 1.2844],
       device='cuda:2'), covar=tensor([0.2235, 0.2006, 0.1727, 0.0847, 0.1864, 0.1299, 0.2426, 0.1443],
       device='cuda:2'), in_proj_covar=tensor([0.0094, 0.0099, 0.0112, 0.0073, 0.0091, 0.0106, 0.0128, 0.0087],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0001, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-22 12:59:45,900 INFO [zipformer.py:660] (2/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=6462.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 12:59:51,245 WARNING [train.py:1060] (2/4) Exclude cut with ID 8291-282929-0024-13011-0_sp0.9 from training. Duration: 26.438875
2022-12-22 12:59:53,017 INFO [zipformer.py:660] (2/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=6467.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 13:00:30,415 INFO [zipformer.py:660] (2/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=6491.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 13:00:36,293 WARNING [train.py:1060] (2/4) Exclude cut with ID 7699-105389-0021-26306-0_sp0.9 from training. Duration: 21.2444375
2022-12-22 13:00:37,693 WARNING [train.py:1060] (2/4) Exclude cut with ID 7357-94126-0014-15845-0_sp0.9 from training. Duration: 31.02225
2022-12-22 13:00:48,238 WARNING [train.py:1060] (2/4) Exclude cut with ID 432-122774-0017-62487-0 from training. Duration: 22.395
2022-12-22 13:00:54,416 INFO [train.py:894] (2/4) Epoch 2, batch 3000, loss[loss=0.3624, simple_loss=0.3853, pruned_loss=0.1698, over 18693.00 frames. ], tot_loss[loss=0.347, simple_loss=0.3842, pruned_loss=0.1549, over 3714994.62 frames. ], batch size: 48, lr: 3.83e-02, grad_scale: 8.0
2022-12-22 13:00:54,416 INFO [train.py:919] (2/4) Computing validation loss
2022-12-22 13:01:05,452 INFO [train.py:928] (2/4) Epoch 2, validation: loss=0.2458, simple_loss=0.336, pruned_loss=0.07777, over 944034.00 frames. 
2022-12-22 13:01:05,452 INFO [train.py:929] (2/4) Maximum memory allocated so far is 23986MB
2022-12-22 13:01:05,987 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.7615, 1.6834, 0.8885, 1.5781, 1.8828, 1.7207, 2.8601, 1.7489],
       device='cuda:2'), covar=tensor([0.1022, 0.1346, 0.2227, 0.1938, 0.1958, 0.0884, 0.0298, 0.1094],
       device='cuda:2'), in_proj_covar=tensor([0.0111, 0.0113, 0.0128, 0.0163, 0.0154, 0.0109, 0.0068, 0.0113],
       device='cuda:2'), out_proj_covar=tensor([1.1618e-04, 1.2390e-04, 1.3180e-04, 1.6040e-04, 1.5593e-04, 1.1010e-04,
        7.7618e-05, 1.1706e-04], device='cuda:2')
2022-12-22 13:01:10,819 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 4.073e+02 7.196e+02 8.571e+02 1.049e+03 2.506e+03, threshold=1.714e+03, percent-clipped=2.0
2022-12-22 13:01:13,888 WARNING [train.py:1060] (2/4) Exclude cut with ID 7357-94126-0045-15876-0 from training. Duration: 21.075
2022-12-22 13:01:17,212 INFO [zipformer.py:660] (2/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=6514.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 13:01:18,338 WARNING [train.py:1060] (2/4) Exclude cut with ID 6482-98857-0025-147532-0_sp0.9 from training. Duration: 20.0055625
2022-12-22 13:01:18,349 WARNING [train.py:1060] (2/4) Exclude cut with ID 6951-79737-0037-132304-0_sp0.9 from training. Duration: 22.05
2022-12-22 13:01:18,360 WARNING [train.py:1060] (2/4) Exclude cut with ID 7255-291500-0003-134302-0 from training. Duration: 26.8349375
2022-12-22 13:01:23,537 WARNING [train.py:1060] (2/4) Exclude cut with ID 7357-94126-0009-15840-0_sp1.1 from training. Duration: 22.1090625
2022-12-22 13:01:30,793 WARNING [train.py:1060] (2/4) Exclude cut with ID 7699-105389-0094-26379-0_sp0.9 from training. Duration: 26.6166875
2022-12-22 13:01:31,138 INFO [zipformer.py:660] (2/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=6523.0, num_to_drop=1, layers_to_drop={0}
2022-12-22 13:01:35,686 INFO [zipformer.py:660] (2/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=6526.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 13:01:38,858 INFO [zipformer.py:660] (2/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=6528.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 13:01:50,215 WARNING [train.py:1060] (2/4) Exclude cut with ID 2046-178027-0000-53705-0_sp0.9 from training. Duration: 20.3055625
2022-12-22 13:02:11,588 WARNING [train.py:1060] (2/4) Exclude cut with ID 7205-50138-0008-5373-0_sp0.9 from training. Duration: 20.7
2022-12-22 13:02:22,258 INFO [train.py:894] (2/4) Epoch 2, batch 3050, loss[loss=0.3428, simple_loss=0.3834, pruned_loss=0.1511, over 18664.00 frames. ], tot_loss[loss=0.3459, simple_loss=0.3839, pruned_loss=0.1539, over 3714724.32 frames. ], batch size: 48, lr: 3.82e-02, grad_scale: 8.0
2022-12-22 13:02:25,502 INFO [zipformer.py:660] (2/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=6559.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 13:02:30,752 INFO [zipformer.py:660] (2/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=6562.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 13:02:38,372 INFO [zipformer.py:660] (2/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=6567.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 13:02:55,450 WARNING [train.py:1060] (2/4) Exclude cut with ID 6978-92210-0019-146985-0 from training. Duration: 22.48
2022-12-22 13:03:10,284 WARNING [train.py:1060] (2/4) Exclude cut with ID 7255-291500-0003-134302-0_sp0.9 from training. Duration: 29.816625
2022-12-22 13:03:29,477 WARNING [train.py:1060] (2/4) Exclude cut with ID 7255-291500-0005-134304-0_sp1.1 from training. Duration: 22.7590625
2022-12-22 13:03:35,688 WARNING [train.py:1060] (2/4) Exclude cut with ID 7357-94126-0026-15857-0 from training. Duration: 22.555
2022-12-22 13:03:40,719 INFO [train.py:894] (2/4) Epoch 2, batch 3100, loss[loss=0.3565, simple_loss=0.3892, pruned_loss=0.162, over 18418.00 frames. ], tot_loss[loss=0.3438, simple_loss=0.3826, pruned_loss=0.1525, over 3713888.66 frames. ], batch size: 48, lr: 3.81e-02, grad_scale: 8.0
2022-12-22 13:03:44,969 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 4.125e+02 7.487e+02 9.188e+02 1.082e+03 2.139e+03, threshold=1.838e+03, percent-clipped=7.0
2022-12-22 13:03:56,553 WARNING [train.py:1060] (2/4) Exclude cut with ID 1250-135782-0005-25975-0_sp0.9 from training. Duration: 21.688875
2022-12-22 13:04:03,029 INFO [zipformer.py:660] (2/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=6622.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 13:04:13,373 INFO [zipformer.py:660] (2/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=6629.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 13:04:30,211 WARNING [train.py:1060] (2/4) Exclude cut with ID 3488-85273-0038-41224-0_sp0.9 from training. Duration: 22.6
2022-12-22 13:04:39,996 INFO [zipformer.py:660] (2/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=6646.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 13:04:55,896 INFO [train.py:894] (2/4) Epoch 2, batch 3150, loss[loss=0.3461, simple_loss=0.3906, pruned_loss=0.1508, over 18478.00 frames. ], tot_loss[loss=0.3429, simple_loss=0.382, pruned_loss=0.1519, over 3714029.23 frames. ], batch size: 64, lr: 3.80e-02, grad_scale: 8.0
2022-12-22 13:05:05,878 WARNING [train.py:1060] (2/4) Exclude cut with ID 7357-94126-0009-15840-0 from training. Duration: 24.32
2022-12-22 13:05:45,521 INFO [zipformer.py:660] (2/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=6690.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 13:05:52,603 INFO [zipformer.py:660] (2/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=6694.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 13:06:00,323 INFO [zipformer.py:660] (2/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=6699.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 13:06:06,684 WARNING [train.py:1060] (2/4) Exclude cut with ID 8291-276745-0093-13116-0_sp0.9 from training. Duration: 21.061125
2022-12-22 13:06:11,501 INFO [scaling.py:679] (2/4) Whitening: num_groups=8, num_channels=96, metric=1.37 vs. limit=2.0
2022-12-22 13:06:11,979 INFO [train.py:894] (2/4) Epoch 2, batch 3200, loss[loss=0.39, simple_loss=0.4158, pruned_loss=0.1821, over 18721.00 frames. ], tot_loss[loss=0.3424, simple_loss=0.3812, pruned_loss=0.1518, over 3713018.36 frames. ], batch size: 52, lr: 3.79e-02, grad_scale: 8.0
2022-12-22 13:06:16,407 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 4.683e+02 7.806e+02 9.589e+02 1.263e+03 2.859e+03, threshold=1.918e+03, percent-clipped=8.0
2022-12-22 13:06:21,374 WARNING [train.py:1060] (2/4) Exclude cut with ID 7357-94126-0024-15855-0_sp0.9 from training. Duration: 20.32225
2022-12-22 13:06:33,581 WARNING [train.py:1060] (2/4) Exclude cut with ID 3033-130750-0096-55598-0_sp1.1 from training. Duration: 0.7545625
2022-12-22 13:06:35,459 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.2700, 1.3704, 1.6650, 1.7115, 1.8199, 2.9629, 1.4427, 2.1404],
       device='cuda:2'), covar=tensor([0.5145, 0.3172, 0.2125, 0.2754, 0.2039, 0.0351, 0.2252, 0.1470],
       device='cuda:2'), in_proj_covar=tensor([0.0155, 0.0132, 0.0139, 0.0129, 0.0137, 0.0085, 0.0123, 0.0121],
       device='cuda:2'), out_proj_covar=tensor([0.0003, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-22 13:06:49,484 WARNING [train.py:1060] (2/4) Exclude cut with ID 4295-39940-0007-92567-0_sp0.9 from training. Duration: 23.9333125
2022-12-22 13:07:12,633 INFO [zipformer.py:660] (2/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=6747.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 13:07:16,290 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.7810, 1.0438, 0.9354, 1.1614, 1.2657, 1.1141, 1.4305, 1.8488],
       device='cuda:2'), covar=tensor([0.2484, 0.2984, 0.3568, 0.2668, 0.3740, 0.2719, 0.2460, 0.2324],
       device='cuda:2'), in_proj_covar=tensor([0.0080, 0.0098, 0.0125, 0.0094, 0.0099, 0.0097, 0.0085, 0.0093],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-22 13:07:22,992 WARNING [train.py:1060] (2/4) Exclude cut with ID 7255-291500-0008-134307-0_sp1.1 from training. Duration: 20.17275
2022-12-22 13:07:27,091 INFO [train.py:894] (2/4) Epoch 2, batch 3250, loss[loss=0.3227, simple_loss=0.3542, pruned_loss=0.1456, over 18576.00 frames. ], tot_loss[loss=0.3433, simple_loss=0.3821, pruned_loss=0.1522, over 3712793.10 frames. ], batch size: 45, lr: 3.78e-02, grad_scale: 8.0
2022-12-22 13:07:27,739 WARNING [train.py:1060] (2/4) Exclude cut with ID 6978-92210-0019-146985-0_sp1.1 from training. Duration: 20.436375
2022-12-22 13:08:19,734 INFO [zipformer.py:660] (2/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=6791.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 13:08:25,513 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.7335, 1.1241, 2.0257, 1.1275, 1.9181, 2.2363, 2.4522, 1.9345],
       device='cuda:2'), covar=tensor([0.2126, 0.1971, 0.0881, 0.1741, 0.0981, 0.0655, 0.0584, 0.1458],
       device='cuda:2'), in_proj_covar=tensor([0.0063, 0.0106, 0.0099, 0.0122, 0.0101, 0.0070, 0.0070, 0.0098],
       device='cuda:2'), out_proj_covar=tensor([7.3973e-05, 1.0941e-04, 1.0835e-04, 1.2717e-04, 1.0868e-04, 8.2839e-05,
        8.0635e-05, 1.0232e-04], device='cuda:2')
2022-12-22 13:08:44,169 INFO [train.py:894] (2/4) Epoch 2, batch 3300, loss[loss=0.2684, simple_loss=0.3119, pruned_loss=0.1125, over 18408.00 frames. ], tot_loss[loss=0.3417, simple_loss=0.3806, pruned_loss=0.1514, over 3713412.68 frames. ], batch size: 42, lr: 3.77e-02, grad_scale: 8.0
2022-12-22 13:08:48,834 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 4.425e+02 7.176e+02 8.285e+02 1.070e+03 1.922e+03, threshold=1.657e+03, percent-clipped=1.0
2022-12-22 13:08:51,717 WARNING [train.py:1060] (2/4) Exclude cut with ID 4234-40345-0022-142709-0_sp0.9 from training. Duration: 23.1055625
2022-12-22 13:08:53,153 WARNING [train.py:1060] (2/4) Exclude cut with ID 8291-282929-0007-12994-0_sp1.1 from training. Duration: 23.5
2022-12-22 13:09:01,082 INFO [zipformer.py:660] (2/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=6818.0, num_to_drop=1, layers_to_drop={0}
2022-12-22 13:09:03,857 WARNING [train.py:1060] (2/4) Exclude cut with ID 7255-291500-0009-134308-0_sp0.9 from training. Duration: 26.62775
2022-12-22 13:09:08,374 INFO [zipformer.py:660] (2/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=6823.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 13:09:12,969 INFO [zipformer.py:660] (2/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=6826.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 13:09:18,390 WARNING [train.py:1060] (2/4) Exclude cut with ID 6951-79737-0018-132285-0 from training. Duration: 21.105
2022-12-22 13:09:23,075 WARNING [train.py:1060] (2/4) Exclude cut with ID 4511-76322-0006-80011-0_sp0.9 from training. Duration: 24.411125
2022-12-22 13:09:28,547 INFO [scaling.py:679] (2/4) Whitening: num_groups=8, num_channels=96, metric=1.55 vs. limit=2.0
2022-12-22 13:09:32,604 INFO [zipformer.py:660] (2/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=6839.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 13:09:50,854 WARNING [train.py:1060] (2/4) Exclude cut with ID 6758-72288-0033-108368-0_sp1.1 from training. Duration: 21.263625
2022-12-22 13:10:00,172 INFO [train.py:894] (2/4) Epoch 2, batch 3350, loss[loss=0.3862, simple_loss=0.41, pruned_loss=0.1812, over 18612.00 frames. ], tot_loss[loss=0.3421, simple_loss=0.381, pruned_loss=0.1516, over 3713340.22 frames. ], batch size: 69, lr: 3.76e-02, grad_scale: 8.0
2022-12-22 13:10:03,376 INFO [zipformer.py:660] (2/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=6859.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 13:10:15,292 INFO [zipformer.py:660] (2/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=6867.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 13:10:22,492 WARNING [train.py:1060] (2/4) Exclude cut with ID 4234-40345-0022-142709-0 from training. Duration: 20.795
2022-12-22 13:10:25,742 INFO [zipformer.py:660] (2/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=6874.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 13:10:32,002 INFO [zipformer.py:660] (2/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=6878.0, num_to_drop=1, layers_to_drop={0}
2022-12-22 13:10:33,159 WARNING [train.py:1060] (2/4) Exclude cut with ID 7357-94126-0021-15852-0 from training. Duration: 24.76
2022-12-22 13:10:33,176 WARNING [train.py:1060] (2/4) Exclude cut with ID 3867-173237-0077-144769-0_sp0.9 from training. Duration: 22.25
2022-12-22 13:11:00,065 WARNING [train.py:1060] (2/4) Exclude cut with ID 7357-94126-0026-15857-0_sp1.1 from training. Duration: 20.5045625
2022-12-22 13:11:17,337 INFO [train.py:894] (2/4) Epoch 2, batch 3400, loss[loss=0.3909, simple_loss=0.4174, pruned_loss=0.1822, over 18567.00 frames. ], tot_loss[loss=0.3419, simple_loss=0.3811, pruned_loss=0.1513, over 3712941.13 frames. ], batch size: 98, lr: 3.75e-02, grad_scale: 8.0
2022-12-22 13:11:17,514 INFO [zipformer.py:660] (2/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=6907.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 13:11:20,766 INFO [zipformer.py:660] (2/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=6909.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 13:11:22,033 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 5.317e+02 7.342e+02 9.217e+02 1.220e+03 2.617e+03, threshold=1.843e+03, percent-clipped=11.0
2022-12-22 13:11:29,550 INFO [zipformer.py:660] (2/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=6915.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 13:11:39,478 INFO [zipformer.py:660] (2/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=6922.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 13:11:53,901 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([3.0097, 3.9399, 4.0009, 2.0452, 3.7954, 2.8446, 0.9938, 2.7625],
       device='cuda:2'), covar=tensor([0.1366, 0.0627, 0.1364, 0.2978, 0.1109, 0.1413, 0.5073, 0.1829],
       device='cuda:2'), in_proj_covar=tensor([0.0108, 0.0074, 0.0133, 0.0101, 0.0086, 0.0089, 0.0135, 0.0102],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0001, 0.0002, 0.0001, 0.0001, 0.0001, 0.0002, 0.0001],
       device='cuda:2')
2022-12-22 13:12:04,121 INFO [zipformer.py:660] (2/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=6939.0, num_to_drop=1, layers_to_drop={0}
2022-12-22 13:12:04,486 INFO [scaling.py:679] (2/4) Whitening: num_groups=8, num_channels=96, metric=1.98 vs. limit=2.0
2022-12-22 13:12:32,076 INFO [train.py:894] (2/4) Epoch 2, batch 3450, loss[loss=0.3278, simple_loss=0.3697, pruned_loss=0.143, over 18426.00 frames. ], tot_loss[loss=0.3422, simple_loss=0.3809, pruned_loss=0.1517, over 3713341.08 frames. ], batch size: 48, lr: 3.75e-02, grad_scale: 8.0
2022-12-22 13:12:50,992 INFO [zipformer.py:660] (2/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=6970.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 13:12:51,240 INFO [zipformer.py:660] (2/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=6970.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 13:13:13,380 INFO [zipformer.py:660] (2/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=6985.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 13:13:46,393 INFO [train.py:894] (2/4) Epoch 2, batch 3500, loss[loss=0.3878, simple_loss=0.4059, pruned_loss=0.1849, over 18620.00 frames. ], tot_loss[loss=0.3416, simple_loss=0.3804, pruned_loss=0.1514, over 3714292.35 frames. ], batch size: 176, lr: 3.74e-02, grad_scale: 8.0
2022-12-22 13:13:51,974 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 4.008e+02 6.904e+02 8.316e+02 1.072e+03 2.058e+03, threshold=1.663e+03, percent-clipped=2.0
2022-12-22 13:14:08,155 WARNING [train.py:1060] (2/4) Exclude cut with ID 7859-102521-0017-7548-0_sp1.1 from training. Duration: 22.2954375
2022-12-22 13:14:17,621 INFO [train.py:894] (2/4) Epoch 3, batch 0, loss[loss=0.3776, simple_loss=0.4174, pruned_loss=0.169, over 18688.00 frames. ], tot_loss[loss=0.3776, simple_loss=0.4174, pruned_loss=0.169, over 18688.00 frames. ], batch size: 97, lr: 3.55e-02, grad_scale: 8.0
2022-12-22 13:14:17,621 INFO [train.py:919] (2/4) Computing validation loss
2022-12-22 13:14:28,602 INFO [train.py:928] (2/4) Epoch 3, validation: loss=0.2448, simple_loss=0.3366, pruned_loss=0.07644, over 944034.00 frames. 
2022-12-22 13:14:28,602 INFO [train.py:929] (2/4) Maximum memory allocated so far is 23995MB
2022-12-22 13:15:21,026 INFO [zipformer.py:660] (2/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=7048.0, num_to_drop=1, layers_to_drop={0}
2022-12-22 13:15:23,490 WARNING [train.py:1060] (2/4) Exclude cut with ID 298-126791-0067-24026-0_sp0.9 from training. Duration: 21.438875
2022-12-22 13:15:27,952 WARNING [train.py:1060] (2/4) Exclude cut with ID 5652-39938-0025-23684-0_sp0.9 from training. Duration: 22.2055625
2022-12-22 13:15:43,246 INFO [train.py:894] (2/4) Epoch 3, batch 50, loss[loss=0.2495, simple_loss=0.3173, pruned_loss=0.09089, over 18590.00 frames. ], tot_loss[loss=0.2964, simple_loss=0.3546, pruned_loss=0.1191, over 837508.16 frames. ], batch size: 45, lr: 3.54e-02, grad_scale: 16.0
2022-12-22 13:16:55,261 INFO [zipformer.py:660] (2/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=7109.0, num_to_drop=1, layers_to_drop={2}
2022-12-22 13:16:58,255 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 3.649e+02 5.613e+02 6.780e+02 8.484e+02 1.673e+03, threshold=1.356e+03, percent-clipped=1.0
2022-12-22 13:17:01,179 INFO [train.py:894] (2/4) Epoch 3, batch 100, loss[loss=0.2711, simple_loss=0.3225, pruned_loss=0.1098, over 18695.00 frames. ], tot_loss[loss=0.2933, simple_loss=0.3532, pruned_loss=0.1167, over 1474930.70 frames. ], batch size: 46, lr: 3.53e-02, grad_scale: 8.0
2022-12-22 13:17:10,071 INFO [zipformer.py:660] (2/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=7118.0, num_to_drop=1, layers_to_drop={1}
2022-12-22 13:17:17,543 INFO [zipformer.py:660] (2/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=7123.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 13:17:20,602 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([3.5434, 2.7022, 2.1054, 1.3083, 2.3147, 2.9785, 1.9300, 2.0606],
       device='cuda:2'), covar=tensor([0.0663, 0.0785, 0.2078, 0.2636, 0.1747, 0.0984, 0.1412, 0.1771],
       device='cuda:2'), in_proj_covar=tensor([0.0108, 0.0145, 0.0186, 0.0177, 0.0162, 0.0150, 0.0147, 0.0167],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-22 13:18:12,337 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.4680, 1.6791, 1.8091, 2.1713, 1.8945, 3.7314, 1.7747, 2.4592],
       device='cuda:2'), covar=tensor([0.4775, 0.2946, 0.2304, 0.2495, 0.2043, 0.0220, 0.1991, 0.1435],
       device='cuda:2'), in_proj_covar=tensor([0.0159, 0.0135, 0.0146, 0.0134, 0.0138, 0.0089, 0.0125, 0.0123],
       device='cuda:2'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0002, 0.0003, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-22 13:18:16,839 INFO [train.py:894] (2/4) Epoch 3, batch 150, loss[loss=0.2569, simple_loss=0.331, pruned_loss=0.09143, over 18582.00 frames. ], tot_loss[loss=0.2863, simple_loss=0.3476, pruned_loss=0.1125, over 1970849.90 frames. ], batch size: 51, lr: 3.52e-02, grad_scale: 8.0
2022-12-22 13:18:17,335 INFO [zipformer.py:660] (2/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=7163.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 13:18:22,195 INFO [zipformer.py:660] (2/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=7166.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 13:18:26,207 WARNING [train.py:1060] (2/4) Exclude cut with ID 7859-102521-0017-7548-0 from training. Duration: 24.525
2022-12-22 13:18:30,298 INFO [zipformer.py:660] (2/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=7171.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 13:18:55,519 INFO [scaling.py:679] (2/4) Whitening: num_groups=8, num_channels=192, metric=1.91 vs. limit=2.0
2022-12-22 13:19:00,640 WARNING [train.py:1060] (2/4) Exclude cut with ID 3699-47246-0007-3408-0_sp0.9 from training. Duration: 20.26675
2022-12-22 13:19:15,739 WARNING [train.py:1060] (2/4) Exclude cut with ID 7859-102521-0017-7548-0_sp0.9 from training. Duration: 27.25
2022-12-22 13:19:28,829 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 3.213e+02 5.590e+02 6.850e+02 8.940e+02 2.169e+03, threshold=1.370e+03, percent-clipped=7.0
2022-12-22 13:19:32,463 INFO [train.py:894] (2/4) Epoch 3, batch 200, loss[loss=0.2574, simple_loss=0.3309, pruned_loss=0.09188, over 18727.00 frames. ], tot_loss[loss=0.2837, simple_loss=0.3462, pruned_loss=0.1106, over 2357069.14 frames. ], batch size: 52, lr: 3.52e-02, grad_scale: 8.0
2022-12-22 13:19:38,402 INFO [scaling.py:679] (2/4) Whitening: num_groups=1, num_channels=384, metric=3.85 vs. limit=5.0
2022-12-22 13:19:50,930 INFO [zipformer.py:660] (2/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=7224.0, num_to_drop=1, layers_to_drop={0}
2022-12-22 13:20:04,790 INFO [zipformer.py:660] (2/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=7234.0, num_to_drop=1, layers_to_drop={1}
2022-12-22 13:20:29,651 WARNING [train.py:1060] (2/4) Exclude cut with ID 6426-64292-0017-15984-0 from training. Duration: 21.68
2022-12-22 13:20:41,964 WARNING [train.py:1060] (2/4) Exclude cut with ID 4278-13270-0007-59342-0 from training. Duration: 21.6300625
2022-12-22 13:20:47,788 INFO [train.py:894] (2/4) Epoch 3, batch 250, loss[loss=0.2632, simple_loss=0.3277, pruned_loss=0.0993, over 18680.00 frames. ], tot_loss[loss=0.2845, simple_loss=0.3471, pruned_loss=0.1109, over 2656913.73 frames. ], batch size: 46, lr: 3.51e-02, grad_scale: 8.0
2022-12-22 13:20:51,274 INFO [zipformer.py:660] (2/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=7265.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 13:21:06,515 WARNING [train.py:1060] (2/4) Exclude cut with ID 4278-13270-0007-59342-0_sp0.9 from training. Duration: 24.033375
2022-12-22 13:21:06,870 INFO [zipformer.py:660] (2/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=7275.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 13:21:08,837 INFO [scaling.py:679] (2/4) Whitening: num_groups=8, num_channels=96, metric=1.22 vs. limit=2.0
2022-12-22 13:21:21,790 INFO [zipformer.py:660] (2/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=7285.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 13:21:58,026 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.3067, 0.9191, 1.5288, 1.6675, 1.6525, 1.7960, 0.6538, 1.4193],
       device='cuda:2'), covar=tensor([0.2506, 0.2371, 0.1813, 0.0862, 0.1728, 0.1539, 0.3268, 0.1535],
       device='cuda:2'), in_proj_covar=tensor([0.0094, 0.0101, 0.0109, 0.0075, 0.0090, 0.0108, 0.0128, 0.0087],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-22 13:22:00,480 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 3.728e+02 5.754e+02 6.972e+02 8.878e+02 1.547e+03, threshold=1.394e+03, percent-clipped=2.0
2022-12-22 13:22:03,712 INFO [train.py:894] (2/4) Epoch 3, batch 300, loss[loss=0.2685, simple_loss=0.343, pruned_loss=0.09699, over 18725.00 frames. ], tot_loss[loss=0.282, simple_loss=0.3448, pruned_loss=0.1096, over 2891628.67 frames. ], batch size: 52, lr: 3.50e-02, grad_scale: 8.0
2022-12-22 13:22:06,409 WARNING [train.py:1060] (2/4) Exclude cut with ID 4278-13270-0009-59344-0 from training. Duration: 22.905
2022-12-22 13:22:08,013 WARNING [train.py:1060] (2/4) Exclude cut with ID 5622-44585-0006-90525-0_sp1.1 from training. Duration: 23.4318125
2022-12-22 13:22:34,151 INFO [zipformer.py:660] (2/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=7333.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 13:22:39,142 INFO [zipformer.py:660] (2/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=7336.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 13:23:19,970 INFO [train.py:894] (2/4) Epoch 3, batch 350, loss[loss=0.2926, simple_loss=0.3473, pruned_loss=0.1189, over 18405.00 frames. ], tot_loss[loss=0.2851, simple_loss=0.3469, pruned_loss=0.1116, over 3074199.62 frames. ], batch size: 46, lr: 3.49e-02, grad_scale: 8.0
2022-12-22 13:24:06,703 WARNING [train.py:1060] (2/4) Exclude cut with ID 4278-13270-0009-59344-0_sp1.1 from training. Duration: 20.82275
2022-12-22 13:24:08,057 WARNING [train.py:1060] (2/4) Exclude cut with ID 4278-13270-0009-59344-0_sp0.9 from training. Duration: 25.45
2022-12-22 13:24:22,096 INFO [zipformer.py:660] (2/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=7404.0, num_to_drop=1, layers_to_drop={3}
2022-12-22 13:24:27,656 INFO [zipformer.py:660] (2/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=7407.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 13:24:33,298 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 4.000e+02 6.102e+02 7.979e+02 1.069e+03 3.939e+03, threshold=1.596e+03, percent-clipped=10.0
2022-12-22 13:24:36,870 INFO [train.py:894] (2/4) Epoch 3, batch 400, loss[loss=0.2549, simple_loss=0.3197, pruned_loss=0.09507, over 18522.00 frames. ], tot_loss[loss=0.2881, simple_loss=0.3498, pruned_loss=0.1132, over 3215895.04 frames. ], batch size: 47, lr: 3.48e-02, grad_scale: 8.0
2022-12-22 13:25:05,698 WARNING [train.py:1060] (2/4) Exclude cut with ID 5622-44585-0006-90525-0 from training. Duration: 25.775
2022-12-22 13:25:21,548 INFO [scaling.py:679] (2/4) Whitening: num_groups=8, num_channels=96, metric=1.35 vs. limit=2.0
2022-12-22 13:25:26,367 WARNING [train.py:1060] (2/4) Exclude cut with ID 2929-85685-0071-62375-0_sp0.9 from training. Duration: 22.25
2022-12-22 13:25:52,870 INFO [train.py:894] (2/4) Epoch 3, batch 450, loss[loss=0.2755, simple_loss=0.3544, pruned_loss=0.09823, over 18600.00 frames. ], tot_loss[loss=0.2889, simple_loss=0.3507, pruned_loss=0.1135, over 3327582.22 frames. ], batch size: 56, lr: 3.48e-02, grad_scale: 8.0
2022-12-22 13:25:55,884 WARNING [train.py:1060] (2/4) Exclude cut with ID 3972-170212-0014-23379-0 from training. Duration: 26.205
2022-12-22 13:26:01,213 INFO [zipformer.py:660] (2/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=7468.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 13:26:13,407 WARNING [train.py:1060] (2/4) Exclude cut with ID 5239-32139-0047-9341-0_sp0.9 from training. Duration: 30.1555625
2022-12-22 13:26:17,636 WARNING [train.py:1060] (2/4) Exclude cut with ID 1265-135635-0050-6781-0_sp0.9 from training. Duration: 21.8333125
2022-12-22 13:26:24,796 WARNING [train.py:1060] (2/4) Exclude cut with ID 1914-133440-0024-94914-0_sp1.1 from training. Duration: 20.6545625
2022-12-22 13:27:04,842 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 3.631e+02 6.334e+02 7.684e+02 9.294e+02 1.756e+03, threshold=1.537e+03, percent-clipped=3.0
2022-12-22 13:27:07,849 INFO [train.py:894] (2/4) Epoch 3, batch 500, loss[loss=0.3073, simple_loss=0.3687, pruned_loss=0.123, over 18390.00 frames. ], tot_loss[loss=0.2912, simple_loss=0.3524, pruned_loss=0.115, over 3413305.36 frames. ], batch size: 53, lr: 3.47e-02, grad_scale: 8.0
2022-12-22 13:27:09,220 WARNING [train.py:1060] (2/4) Exclude cut with ID 7395-89880-0045-39920-0_sp0.9 from training. Duration: 20.52225
2022-12-22 13:27:17,090 INFO [zipformer.py:660] (2/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=7519.0, num_to_drop=1, layers_to_drop={0}
2022-12-22 13:27:28,607 WARNING [train.py:1060] (2/4) Exclude cut with ID 3972-170212-0014-23379-0_sp0.9 from training. Duration: 29.1166875
2022-12-22 13:27:38,605 INFO [zipformer.py:660] (2/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=7534.0, num_to_drop=1, layers_to_drop={1}
2022-12-22 13:28:23,479 INFO [train.py:894] (2/4) Epoch 3, batch 550, loss[loss=0.2473, simple_loss=0.3098, pruned_loss=0.09243, over 18535.00 frames. ], tot_loss[loss=0.2944, simple_loss=0.3545, pruned_loss=0.1171, over 3479231.15 frames. ], batch size: 44, lr: 3.46e-02, grad_scale: 8.0
2022-12-22 13:28:26,643 INFO [zipformer.py:660] (2/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=7565.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 13:28:27,676 WARNING [train.py:1060] (2/4) Exclude cut with ID 543-133211-0007-59831-0_sp0.9 from training. Duration: 21.388875
2022-12-22 13:28:50,606 INFO [zipformer.py:660] (2/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=7582.0, num_to_drop=1, layers_to_drop={1}
2022-12-22 13:28:58,827 WARNING [train.py:1060] (2/4) Exclude cut with ID 1914-133440-0024-94914-0 from training. Duration: 22.72
2022-12-22 13:29:00,298 WARNING [train.py:1060] (2/4) Exclude cut with ID 1914-133440-0031-94921-0_sp0.9 from training. Duration: 22.7444375
2022-12-22 13:29:14,480 INFO [scaling.py:679] (2/4) Whitening: num_groups=8, num_channels=96, metric=1.30 vs. limit=2.0
2022-12-22 13:29:22,727 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.8849, 1.2424, 1.6169, 1.6289, 1.6526, 1.4817, 1.6430, 1.2401],
       device='cuda:2'), covar=tensor([0.0633, 0.1379, 0.1536, 0.0996, 0.0553, 0.0565, 0.0997, 0.0708],
       device='cuda:2'), in_proj_covar=tensor([0.0099, 0.0131, 0.0136, 0.0122, 0.0099, 0.0102, 0.0112, 0.0106],
       device='cuda:2'), out_proj_covar=tensor([0.0001, 0.0001, 0.0001, 0.0001, 0.0001, 0.0001, 0.0001, 0.0001],
       device='cuda:2')
2022-12-22 13:29:34,402 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 4.376e+02 6.274e+02 7.834e+02 9.587e+02 2.337e+03, threshold=1.567e+03, percent-clipped=2.0
2022-12-22 13:29:37,660 INFO [train.py:894] (2/4) Epoch 3, batch 600, loss[loss=0.2522, simple_loss=0.3119, pruned_loss=0.09627, over 18635.00 frames. ], tot_loss[loss=0.294, simple_loss=0.3544, pruned_loss=0.1168, over 3532051.85 frames. ], batch size: 41, lr: 3.45e-02, grad_scale: 8.0
2022-12-22 13:29:37,831 INFO [zipformer.py:660] (2/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=7613.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 13:29:44,934 WARNING [train.py:1060] (2/4) Exclude cut with ID 4133-6541-0027-40495-0_sp1.1 from training. Duration: 0.9681875
2022-12-22 13:29:47,006 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.7374, 1.0498, 1.4063, 1.7723, 1.5295, 1.3789, 1.6242, 1.0532],
       device='cuda:2'), covar=tensor([0.0766, 0.1526, 0.1714, 0.1072, 0.0603, 0.0626, 0.1036, 0.0833],
       device='cuda:2'), in_proj_covar=tensor([0.0100, 0.0131, 0.0136, 0.0122, 0.0099, 0.0102, 0.0111, 0.0106],
       device='cuda:2'), out_proj_covar=tensor([0.0001, 0.0001, 0.0001, 0.0001, 0.0001, 0.0001, 0.0001, 0.0001],
       device='cuda:2')
2022-12-22 13:29:48,067 WARNING [train.py:1060] (2/4) Exclude cut with ID 6330-62851-0022-91297-0_sp0.9 from training. Duration: 22.3166875
2022-12-22 13:29:54,180 WARNING [train.py:1060] (2/4) Exclude cut with ID 543-133212-0015-59917-0_sp0.9 from training. Duration: 21.8166875
2022-12-22 13:29:54,494 INFO [zipformer.py:660] (2/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=7624.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 13:29:59,038 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.6540, 1.3565, 1.4338, 1.2288, 1.7458, 1.4288, 1.6076, 2.1494],
       device='cuda:2'), covar=tensor([0.1904, 0.2093, 0.2458, 0.2166, 0.2445, 0.1823, 0.1892, 0.1608],
       device='cuda:2'), in_proj_covar=tensor([0.0082, 0.0100, 0.0128, 0.0098, 0.0102, 0.0099, 0.0088, 0.0097],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-22 13:30:04,468 INFO [zipformer.py:660] (2/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=7631.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 13:30:53,723 INFO [train.py:894] (2/4) Epoch 3, batch 650, loss[loss=0.2889, simple_loss=0.3496, pruned_loss=0.1141, over 18581.00 frames. ], tot_loss[loss=0.2928, simple_loss=0.3529, pruned_loss=0.1164, over 3572755.16 frames. ], batch size: 51, lr: 3.44e-02, grad_scale: 8.0
2022-12-22 13:31:25,922 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.4417, 1.5723, 1.1119, 1.7191, 1.4581, 1.4057, 1.4310, 1.5783],
       device='cuda:2'), covar=tensor([0.1607, 0.1349, 0.1289, 0.1273, 0.1236, 0.0747, 0.1377, 0.0655],
       device='cuda:2'), in_proj_covar=tensor([0.0155, 0.0120, 0.0120, 0.0175, 0.0117, 0.0117, 0.0139, 0.0100],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0001, 0.0001, 0.0002, 0.0001, 0.0001, 0.0002, 0.0001],
       device='cuda:2')
2022-12-22 13:31:27,274 INFO [zipformer.py:660] (2/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=7685.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 13:31:37,056 WARNING [train.py:1060] (2/4) Exclude cut with ID 4957-30119-0041-23990-0_sp0.9 from training. Duration: 20.22775
2022-12-22 13:31:56,612 INFO [zipformer.py:660] (2/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=7704.0, num_to_drop=1, layers_to_drop={2}
2022-12-22 13:32:06,836 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 3.551e+02 6.142e+02 7.581e+02 9.638e+02 3.494e+03, threshold=1.516e+03, percent-clipped=4.0
2022-12-22 13:32:09,742 INFO [train.py:894] (2/4) Epoch 3, batch 700, loss[loss=0.2973, simple_loss=0.3645, pruned_loss=0.1151, over 18538.00 frames. ], tot_loss[loss=0.2909, simple_loss=0.3515, pruned_loss=0.1151, over 3603766.26 frames. ], batch size: 55, lr: 3.44e-02, grad_scale: 8.0
2022-12-22 13:32:23,202 WARNING [train.py:1060] (2/4) Exclude cut with ID 5239-32139-0047-9341-0_sp1.1 from training. Duration: 24.67275
2022-12-22 13:32:45,411 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([2.5661, 2.4428, 2.0545, 1.0528, 2.2046, 2.0246, 1.9775, 2.3185],
       device='cuda:2'), covar=tensor([0.0748, 0.0613, 0.1841, 0.2397, 0.1447, 0.1439, 0.1336, 0.1060],
       device='cuda:2'), in_proj_covar=tensor([0.0106, 0.0140, 0.0178, 0.0176, 0.0160, 0.0148, 0.0148, 0.0163],
       device='cuda:2'), out_proj_covar=tensor([0.0001, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-22 13:32:46,011 INFO [scaling.py:679] (2/4) Whitening: num_groups=8, num_channels=192, metric=2.06 vs. limit=2.0
2022-12-22 13:32:51,373 WARNING [train.py:1060] (2/4) Exclude cut with ID 3082-165428-0081-50734-0_sp0.9 from training. Duration: 21.8055625
2022-12-22 13:32:54,693 INFO [zipformer.py:660] (2/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=7743.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 13:33:08,090 INFO [zipformer.py:660] (2/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=7752.0, num_to_drop=1, layers_to_drop={1}
2022-12-22 13:33:24,433 INFO [train.py:894] (2/4) Epoch 3, batch 750, loss[loss=0.3543, simple_loss=0.4127, pruned_loss=0.1479, over 18726.00 frames. ], tot_loss[loss=0.2918, simple_loss=0.3522, pruned_loss=0.1157, over 3628373.60 frames. ], batch size: 54, lr: 3.43e-02, grad_scale: 8.0
2022-12-22 13:33:24,589 INFO [zipformer.py:660] (2/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=7763.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 13:33:28,778 WARNING [train.py:1060] (2/4) Exclude cut with ID 3340-169293-0054-76830-0_sp0.9 from training. Duration: 22.6666875
2022-12-22 13:34:23,921 INFO [zipformer.py:660] (2/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=7802.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 13:34:27,415 INFO [zipformer.py:660] (2/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=7804.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 13:34:29,949 WARNING [train.py:1060] (2/4) Exclude cut with ID 2411-132532-0017-82279-0_sp1.1 from training. Duration: 0.9681875
2022-12-22 13:34:37,381 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 4.051e+02 6.250e+02 7.704e+02 9.761e+02 2.933e+03, threshold=1.541e+03, percent-clipped=4.0
2022-12-22 13:34:40,173 INFO [train.py:894] (2/4) Epoch 3, batch 800, loss[loss=0.2779, simple_loss=0.3271, pruned_loss=0.1144, over 18621.00 frames. ], tot_loss[loss=0.293, simple_loss=0.3535, pruned_loss=0.1163, over 3647406.26 frames. ], batch size: 45, lr: 3.42e-02, grad_scale: 8.0
2022-12-22 13:34:49,471 INFO [zipformer.py:660] (2/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=7819.0, num_to_drop=1, layers_to_drop={2}
2022-12-22 13:34:54,609 WARNING [train.py:1060] (2/4) Exclude cut with ID 6330-62850-0007-91323-0 from training. Duration: 22.485
2022-12-22 13:34:54,866 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([4.5887, 3.7829, 3.8656, 4.1855, 4.1095, 4.0572, 4.6790, 1.3607],
       device='cuda:2'), covar=tensor([0.0503, 0.0548, 0.0489, 0.0331, 0.1085, 0.0580, 0.0342, 0.4430],
       device='cuda:2'), in_proj_covar=tensor([0.0164, 0.0137, 0.0117, 0.0097, 0.0183, 0.0134, 0.0137, 0.0187],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0001, 0.0001, 0.0001, 0.0002, 0.0001, 0.0001, 0.0002],
       device='cuda:2')
2022-12-22 13:35:32,123 WARNING [train.py:1060] (2/4) Exclude cut with ID 3972-170212-0014-23379-0_sp1.1 from training. Duration: 23.82275
2022-12-22 13:35:47,989 WARNING [train.py:1060] (2/4) Exclude cut with ID 4860-13185-0032-76709-0 from training. Duration: 20.77
2022-12-22 13:35:55,071 INFO [train.py:894] (2/4) Epoch 3, batch 850, loss[loss=0.2695, simple_loss=0.334, pruned_loss=0.1025, over 18559.00 frames. ], tot_loss[loss=0.2912, simple_loss=0.352, pruned_loss=0.1152, over 3661764.63 frames. ], batch size: 49, lr: 3.41e-02, grad_scale: 8.0
2022-12-22 13:35:55,132 WARNING [train.py:1060] (2/4) Exclude cut with ID 6426-64292-0017-15984-0_sp0.9 from training. Duration: 24.088875
2022-12-22 13:35:55,427 INFO [zipformer.py:660] (2/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=7863.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 13:36:00,805 INFO [zipformer.py:660] (2/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=7867.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 13:36:25,296 WARNING [train.py:1060] (2/4) Exclude cut with ID 6330-62850-0007-91323-0_sp1.1 from training. Duration: 20.4409375
2022-12-22 13:37:07,071 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 4.016e+02 5.777e+02 7.379e+02 9.226e+02 3.443e+03, threshold=1.476e+03, percent-clipped=5.0
2022-12-22 13:37:09,972 INFO [train.py:894] (2/4) Epoch 3, batch 900, loss[loss=0.2708, simple_loss=0.3276, pruned_loss=0.107, over 18506.00 frames. ], tot_loss[loss=0.2907, simple_loss=0.3512, pruned_loss=0.1151, over 3672171.52 frames. ], batch size: 44, lr: 3.41e-02, grad_scale: 8.0
2022-12-22 13:37:36,423 INFO [zipformer.py:660] (2/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=7931.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 13:37:41,857 WARNING [train.py:1060] (2/4) Exclude cut with ID 2929-85685-0062-62366-0_sp0.9 from training. Duration: 22.511125
2022-12-22 13:37:41,878 WARNING [train.py:1060] (2/4) Exclude cut with ID 7395-89880-0031-39906-0 from training. Duration: 20.675
2022-12-22 13:38:24,396 INFO [train.py:894] (2/4) Epoch 3, batch 950, loss[loss=0.2437, simple_loss=0.3096, pruned_loss=0.08894, over 18551.00 frames. ], tot_loss[loss=0.2911, simple_loss=0.3519, pruned_loss=0.1151, over 3680780.42 frames. ], batch size: 44, lr: 3.40e-02, grad_scale: 8.0
2022-12-22 13:38:32,734 INFO [scaling.py:679] (2/4) Whitening: num_groups=8, num_channels=192, metric=1.87 vs. limit=2.0
2022-12-22 13:38:48,063 INFO [zipformer.py:660] (2/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=7979.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 13:38:49,743 INFO [zipformer.py:660] (2/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=7980.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 13:39:25,674 WARNING [train.py:1060] (2/4) Exclude cut with ID 6330-62850-0007-91323-0_sp0.9 from training. Duration: 24.9833125
2022-12-22 13:39:28,852 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([5.9488, 5.0006, 5.0793, 5.5552, 5.4282, 4.9854, 5.4665, 1.5367],
       device='cuda:2'), covar=tensor([0.0411, 0.0429, 0.0392, 0.0251, 0.1026, 0.0719, 0.0443, 0.4130],
       device='cuda:2'), in_proj_covar=tensor([0.0158, 0.0132, 0.0115, 0.0095, 0.0175, 0.0132, 0.0132, 0.0180],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0001, 0.0001, 0.0001, 0.0002, 0.0001, 0.0001, 0.0002],
       device='cuda:2')
2022-12-22 13:39:40,172 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 2.748e+02 5.844e+02 7.550e+02 9.469e+02 1.849e+03, threshold=1.510e+03, percent-clipped=2.0
2022-12-22 13:39:43,199 INFO [train.py:894] (2/4) Epoch 3, batch 1000, loss[loss=0.2772, simple_loss=0.3304, pruned_loss=0.112, over 18478.00 frames. ], tot_loss[loss=0.2922, simple_loss=0.3533, pruned_loss=0.1156, over 3688053.59 frames. ], batch size: 44, lr: 3.39e-02, grad_scale: 8.0
2022-12-22 13:39:56,342 WARNING [train.py:1060] (2/4) Exclude cut with ID 5239-32139-0047-9341-0 from training. Duration: 27.14
2022-12-22 13:40:10,711 WARNING [train.py:1060] (2/4) Exclude cut with ID 2929-85685-0044-62348-0 from training. Duration: 22.44
2022-12-22 13:40:57,480 INFO [train.py:894] (2/4) Epoch 3, batch 1050, loss[loss=0.2552, simple_loss=0.3176, pruned_loss=0.09642, over 18430.00 frames. ], tot_loss[loss=0.2913, simple_loss=0.3528, pruned_loss=0.1149, over 3693094.27 frames. ], batch size: 42, lr: 3.38e-02, grad_scale: 8.0
2022-12-22 13:40:57,883 INFO [zipformer.py:660] (2/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=8063.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 13:41:30,998 WARNING [train.py:1060] (2/4) Exclude cut with ID 2929-85685-0060-62364-0_sp0.9 from training. Duration: 21.361125
2022-12-22 13:41:39,284 WARNING [train.py:1060] (2/4) Exclude cut with ID 2929-85685-0079-62383-0_sp1.1 from training. Duration: 27.0318125
2022-12-22 13:41:47,139 WARNING [train.py:1060] (2/4) Exclude cut with ID 5622-44585-0006-90525-0_sp0.9 from training. Duration: 28.638875
2022-12-22 13:41:52,772 INFO [zipformer.py:660] (2/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=8099.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 13:42:04,427 WARNING [train.py:1060] (2/4) Exclude cut with ID 3340-169293-0054-76830-0 from training. Duration: 20.4
2022-12-22 13:42:10,267 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 3.547e+02 6.153e+02 7.411e+02 9.093e+02 1.546e+03, threshold=1.482e+03, percent-clipped=1.0
2022-12-22 13:42:10,472 INFO [zipformer.py:660] (2/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=8111.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 13:42:13,252 INFO [train.py:894] (2/4) Epoch 3, batch 1100, loss[loss=0.2838, simple_loss=0.341, pruned_loss=0.1133, over 18532.00 frames. ], tot_loss[loss=0.2904, simple_loss=0.3524, pruned_loss=0.1142, over 3697509.26 frames. ], batch size: 47, lr: 3.37e-02, grad_scale: 8.0
2022-12-22 13:42:25,464 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.2997, 1.1630, 1.1357, 1.2635, 1.5424, 1.4803, 1.6153, 1.9326],
       device='cuda:2'), covar=tensor([0.2057, 0.2425, 0.2661, 0.1952, 0.2584, 0.1735, 0.1750, 0.1649],
       device='cuda:2'), in_proj_covar=tensor([0.0084, 0.0102, 0.0126, 0.0097, 0.0102, 0.0097, 0.0090, 0.0095],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-22 13:42:34,099 WARNING [train.py:1060] (2/4) Exclude cut with ID 2929-85685-0071-62375-0 from training. Duration: 20.025
2022-12-22 13:42:35,617 WARNING [train.py:1060] (2/4) Exclude cut with ID 2364-131735-0112-64612-0_sp0.9 from training. Duration: 20.488875
2022-12-22 13:42:41,024 WARNING [train.py:1060] (2/4) Exclude cut with ID 2929-85685-0079-62383-0 from training. Duration: 29.735
2022-12-22 13:43:22,526 INFO [zipformer.py:660] (2/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=8158.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 13:43:29,556 INFO [train.py:894] (2/4) Epoch 3, batch 1150, loss[loss=0.3374, simple_loss=0.3822, pruned_loss=0.1463, over 18720.00 frames. ], tot_loss[loss=0.2902, simple_loss=0.3522, pruned_loss=0.1141, over 3700519.12 frames. ], batch size: 54, lr: 3.37e-02, grad_scale: 8.0
2022-12-22 13:44:02,570 WARNING [train.py:1060] (2/4) Exclude cut with ID 7276-92427-0014-12983-0_sp0.9 from training. Duration: 21.3055625
2022-12-22 13:44:03,832 WARNING [train.py:1060] (2/4) Exclude cut with ID 1025-75365-0008-79168-0_sp0.9 from training. Duration: 22.0666875
2022-12-22 13:44:31,030 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([2.7338, 1.5770, 1.7997, 2.5529, 2.0706, 1.8290, 2.1461, 1.1899],
       device='cuda:2'), covar=tensor([0.0571, 0.1506, 0.1485, 0.0995, 0.0533, 0.0537, 0.1037, 0.0797],
       device='cuda:2'), in_proj_covar=tensor([0.0108, 0.0140, 0.0143, 0.0135, 0.0109, 0.0113, 0.0120, 0.0114],
       device='cuda:2'), out_proj_covar=tensor([0.0001, 0.0001, 0.0001, 0.0001, 0.0001, 0.0001, 0.0001, 0.0001],
       device='cuda:2')
2022-12-22 13:44:42,386 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 3.950e+02 6.195e+02 7.771e+02 9.937e+02 1.486e+03, threshold=1.554e+03, percent-clipped=1.0
2022-12-22 13:44:45,406 INFO [train.py:894] (2/4) Epoch 3, batch 1200, loss[loss=0.3483, simple_loss=0.3959, pruned_loss=0.1504, over 18656.00 frames. ], tot_loss[loss=0.2913, simple_loss=0.3534, pruned_loss=0.1146, over 3703270.83 frames. ], batch size: 62, lr: 3.36e-02, grad_scale: 8.0
2022-12-22 13:45:31,087 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([5.8611, 4.8018, 4.7487, 5.3898, 5.2813, 5.0459, 5.5323, 1.5846],
       device='cuda:2'), covar=tensor([0.0357, 0.0524, 0.0509, 0.0288, 0.0937, 0.0515, 0.0336, 0.3856],
       device='cuda:2'), in_proj_covar=tensor([0.0164, 0.0138, 0.0121, 0.0100, 0.0183, 0.0135, 0.0137, 0.0188],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0001, 0.0001, 0.0001, 0.0002, 0.0001, 0.0001, 0.0002],
       device='cuda:2')
2022-12-22 13:45:50,216 WARNING [train.py:1060] (2/4) Exclude cut with ID 2929-85685-0062-62366-0 from training. Duration: 20.26
2022-12-22 13:45:58,966 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.5713, 1.8016, 1.8193, 1.8762, 2.2377, 3.4770, 1.7296, 2.4615],
       device='cuda:2'), covar=tensor([0.4266, 0.2476, 0.1997, 0.2516, 0.1611, 0.0288, 0.2007, 0.1396],
       device='cuda:2'), in_proj_covar=tensor([0.0154, 0.0133, 0.0142, 0.0131, 0.0139, 0.0090, 0.0121, 0.0124],
       device='cuda:2'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0002, 0.0003, 0.0003],
       device='cuda:2')
2022-12-22 13:46:00,042 INFO [train.py:894] (2/4) Epoch 3, batch 1250, loss[loss=0.2514, simple_loss=0.3225, pruned_loss=0.09014, over 18448.00 frames. ], tot_loss[loss=0.2904, simple_loss=0.352, pruned_loss=0.1144, over 3705627.54 frames. ], batch size: 50, lr: 3.35e-02, grad_scale: 8.0
2022-12-22 13:46:03,399 INFO [zipformer.py:660] (2/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=8265.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 13:46:04,562 WARNING [train.py:1060] (2/4) Exclude cut with ID 5239-32139-0030-9324-0_sp0.9 from training. Duration: 21.3444375
2022-12-22 13:46:25,983 INFO [zipformer.py:660] (2/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=8280.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 13:46:30,500 INFO [scaling.py:679] (2/4) Whitening: num_groups=8, num_channels=96, metric=1.30 vs. limit=2.0
2022-12-22 13:47:00,048 WARNING [train.py:1060] (2/4) Exclude cut with ID 497-129325-0061-62254-0_sp1.1 from training. Duration: 0.97725
2022-12-22 13:47:11,730 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 3.563e+02 6.682e+02 7.893e+02 1.062e+03 4.114e+03, threshold=1.579e+03, percent-clipped=7.0
2022-12-22 13:47:14,618 INFO [train.py:894] (2/4) Epoch 3, batch 1300, loss[loss=0.2642, simple_loss=0.3351, pruned_loss=0.09668, over 18589.00 frames. ], tot_loss[loss=0.2896, simple_loss=0.3516, pruned_loss=0.1138, over 3707539.69 frames. ], batch size: 51, lr: 3.34e-02, grad_scale: 8.0
2022-12-22 13:47:35,095 INFO [zipformer.py:660] (2/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=8326.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 13:47:37,581 INFO [zipformer.py:660] (2/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=8328.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 13:47:43,561 WARNING [train.py:1060] (2/4) Exclude cut with ID 7395-89880-0031-39906-0_sp0.9 from training. Duration: 22.97225
2022-12-22 13:48:15,428 WARNING [train.py:1060] (2/4) Exclude cut with ID 7395-89880-0047-39922-0_sp0.9 from training. Duration: 21.97775
2022-12-22 13:48:22,176 INFO [scaling.py:679] (2/4) Whitening: num_groups=8, num_channels=96, metric=1.70 vs. limit=2.0
2022-12-22 13:48:27,765 WARNING [train.py:1060] (2/4) Exclude cut with ID 1112-1043-0006-89194-0_sp0.9 from training. Duration: 21.8333125
2022-12-22 13:48:30,489 INFO [train.py:894] (2/4) Epoch 3, batch 1350, loss[loss=0.2882, simple_loss=0.3456, pruned_loss=0.1154, over 18706.00 frames. ], tot_loss[loss=0.287, simple_loss=0.3498, pruned_loss=0.1121, over 3708480.50 frames. ], batch size: 50, lr: 3.34e-02, grad_scale: 8.0
2022-12-22 13:48:37,815 WARNING [train.py:1060] (2/4) Exclude cut with ID 1914-133440-0031-94921-0 from training. Duration: 20.47
2022-12-22 13:49:26,198 INFO [zipformer.py:660] (2/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=8399.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 13:49:32,515 INFO [scaling.py:679] (2/4) Whitening: num_groups=8, num_channels=192, metric=2.03 vs. limit=2.0
2022-12-22 13:49:37,902 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.8574, 2.0610, 0.9239, 2.0140, 2.0321, 1.6458, 3.2174, 1.9698],
       device='cuda:2'), covar=tensor([0.0876, 0.1346, 0.2226, 0.1779, 0.1642, 0.0869, 0.0331, 0.0986],
       device='cuda:2'), in_proj_covar=tensor([0.0124, 0.0130, 0.0153, 0.0188, 0.0165, 0.0123, 0.0093, 0.0132],
       device='cuda:2'), out_proj_covar=tensor([0.0001, 0.0001, 0.0002, 0.0002, 0.0002, 0.0001, 0.0001, 0.0001],
       device='cuda:2')
2022-12-22 13:49:42,261 WARNING [train.py:1060] (2/4) Exclude cut with ID 7395-89880-0037-39912-0_sp0.9 from training. Duration: 20.67225
2022-12-22 13:49:43,613 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 3.311e+02 6.013e+02 7.179e+02 8.761e+02 1.780e+03, threshold=1.436e+03, percent-clipped=1.0
2022-12-22 13:49:46,608 INFO [train.py:894] (2/4) Epoch 3, batch 1400, loss[loss=0.2413, simple_loss=0.3076, pruned_loss=0.08744, over 18386.00 frames. ], tot_loss[loss=0.2871, simple_loss=0.3495, pruned_loss=0.1123, over 3710652.81 frames. ], batch size: 46, lr: 3.33e-02, grad_scale: 8.0
2022-12-22 13:50:00,315 WARNING [train.py:1060] (2/4) Exclude cut with ID 1914-133440-0024-94914-0_sp0.9 from training. Duration: 25.2444375
2022-12-22 13:50:05,901 INFO [scaling.py:679] (2/4) Whitening: num_groups=8, num_channels=96, metric=1.29 vs. limit=2.0
2022-12-22 13:50:06,719 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([3.5623, 2.9362, 2.6063, 1.3254, 2.2391, 2.5178, 1.8138, 1.9428],
       device='cuda:2'), covar=tensor([0.0812, 0.0718, 0.1890, 0.2728, 0.2280, 0.1429, 0.1953, 0.1790],
       device='cuda:2'), in_proj_covar=tensor([0.0106, 0.0141, 0.0175, 0.0174, 0.0159, 0.0143, 0.0151, 0.0162],
       device='cuda:2'), out_proj_covar=tensor([0.0001, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-22 13:50:24,682 WARNING [train.py:1060] (2/4) Exclude cut with ID 3340-169293-0021-76797-0_sp0.9 from training. Duration: 21.1445
2022-12-22 13:50:37,661 INFO [zipformer.py:660] (2/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=8447.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 13:50:54,218 INFO [zipformer.py:660] (2/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=8458.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 13:51:01,642 INFO [train.py:894] (2/4) Epoch 3, batch 1450, loss[loss=0.2357, simple_loss=0.2985, pruned_loss=0.08642, over 18555.00 frames. ], tot_loss[loss=0.2902, simple_loss=0.3518, pruned_loss=0.1143, over 3711977.86 frames. ], batch size: 44, lr: 3.32e-02, grad_scale: 8.0
2022-12-22 13:51:38,509 WARNING [train.py:1060] (2/4) Exclude cut with ID 2929-85685-0079-62383-0_sp0.9 from training. Duration: 33.038875
2022-12-22 13:52:05,963 INFO [zipformer.py:660] (2/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=8506.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 13:52:13,221 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 3.559e+02 6.368e+02 8.032e+02 1.005e+03 2.835e+03, threshold=1.606e+03, percent-clipped=10.0
2022-12-22 13:52:15,351 WARNING [train.py:1060] (2/4) Exclude cut with ID 6426-64291-0000-16059-0_sp0.9 from training. Duration: 20.0944375
2022-12-22 13:52:16,674 INFO [train.py:894] (2/4) Epoch 3, batch 1500, loss[loss=0.2947, simple_loss=0.3651, pruned_loss=0.1122, over 18523.00 frames. ], tot_loss[loss=0.2878, simple_loss=0.3494, pruned_loss=0.1131, over 3711546.34 frames. ], batch size: 64, lr: 3.32e-02, grad_scale: 8.0
2022-12-22 13:52:30,738 WARNING [train.py:1060] (2/4) Exclude cut with ID 2929-85685-0044-62348-0_sp1.1 from training. Duration: 20.4
2022-12-22 13:52:40,033 WARNING [train.py:1060] (2/4) Exclude cut with ID 6330-62851-0022-91297-0 from training. Duration: 20.085
2022-12-22 13:52:49,628 WARNING [train.py:1060] (2/4) Exclude cut with ID 4860-13185-0032-76709-0_sp0.9 from training. Duration: 23.07775
2022-12-22 13:52:52,655 INFO [scaling.py:679] (2/4) Whitening: num_groups=1, num_channels=384, metric=4.76 vs. limit=5.0
2022-12-22 13:53:30,916 INFO [train.py:894] (2/4) Epoch 3, batch 1550, loss[loss=0.2545, simple_loss=0.3115, pruned_loss=0.09881, over 18439.00 frames. ], tot_loss[loss=0.2879, simple_loss=0.3495, pruned_loss=0.1132, over 3711981.50 frames. ], batch size: 42, lr: 3.31e-02, grad_scale: 8.0
2022-12-22 13:53:37,295 WARNING [train.py:1060] (2/4) Exclude cut with ID 2929-85685-0044-62348-0_sp0.9 from training. Duration: 24.9333125
2022-12-22 13:54:19,646 INFO [scaling.py:679] (2/4) Whitening: num_groups=8, num_channels=192, metric=1.91 vs. limit=2.0
2022-12-22 13:54:20,337 WARNING [train.py:1060] (2/4) Exclude cut with ID 5118-111612-0016-124680-0_sp0.9 from training. Duration: 20.388875
2022-12-22 13:54:26,231 WARNING [train.py:1060] (2/4) Exclude cut with ID 432-122774-0017-62487-0_sp1.1 from training. Duration: 20.3590625
2022-12-22 13:54:43,758 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 4.316e+02 6.096e+02 7.223e+02 9.116e+02 1.940e+03, threshold=1.445e+03, percent-clipped=4.0
2022-12-22 13:54:47,241 INFO [train.py:894] (2/4) Epoch 3, batch 1600, loss[loss=0.3138, simple_loss=0.3683, pruned_loss=0.1296, over 18645.00 frames. ], tot_loss[loss=0.2873, simple_loss=0.3496, pruned_loss=0.1125, over 3712367.30 frames. ], batch size: 182, lr: 3.30e-02, grad_scale: 8.0
2022-12-22 13:54:59,166 INFO [zipformer.py:660] (2/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=8621.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 13:55:33,363 WARNING [train.py:1060] (2/4) Exclude cut with ID 3557-8342-0013-54691-0_sp1.1 from training. Duration: 0.836375
2022-12-22 13:56:01,307 INFO [train.py:894] (2/4) Epoch 3, batch 1650, loss[loss=0.2946, simple_loss=0.3467, pruned_loss=0.1213, over 18474.00 frames. ], tot_loss[loss=0.2924, simple_loss=0.353, pruned_loss=0.1159, over 3712955.96 frames. ], batch size: 50, lr: 3.29e-02, grad_scale: 8.0
2022-12-22 13:56:04,518 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([3.6672, 3.1941, 3.1552, 3.4250, 2.8959, 3.2111, 3.9869, 1.2666],
       device='cuda:2'), covar=tensor([0.1395, 0.1339, 0.1189, 0.0923, 0.2994, 0.1500, 0.0872, 0.5960],
       device='cuda:2'), in_proj_covar=tensor([0.0174, 0.0140, 0.0129, 0.0105, 0.0193, 0.0146, 0.0142, 0.0195],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0001, 0.0001, 0.0001, 0.0002, 0.0001, 0.0001, 0.0002],
       device='cuda:2')
2022-12-22 13:56:19,134 WARNING [train.py:1060] (2/4) Exclude cut with ID 8565-290391-0049-67394-0_sp0.9 from training. Duration: 21.3166875
2022-12-22 13:56:49,610 WARNING [train.py:1060] (2/4) Exclude cut with ID 6533-399-0029-104863-0_sp0.9 from training. Duration: 22.1055625
2022-12-22 13:56:58,458 WARNING [train.py:1060] (2/4) Exclude cut with ID 7699-105389-0094-26379-0_sp1.1 from training. Duration: 21.77725
2022-12-22 13:57:12,830 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 4.680e+02 6.678e+02 8.781e+02 1.140e+03 4.926e+03, threshold=1.756e+03, percent-clipped=10.0
2022-12-22 13:57:15,870 INFO [train.py:894] (2/4) Epoch 3, batch 1700, loss[loss=0.3585, simple_loss=0.39, pruned_loss=0.1634, over 18374.00 frames. ], tot_loss[loss=0.2981, simple_loss=0.356, pruned_loss=0.1202, over 3713106.58 frames. ], batch size: 51, lr: 3.29e-02, grad_scale: 8.0
2022-12-22 13:57:16,313 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.8046, 1.7948, 0.8814, 1.5257, 1.7461, 1.6117, 2.6366, 1.9279],
       device='cuda:2'), covar=tensor([0.0740, 0.1077, 0.2068, 0.1683, 0.1665, 0.0770, 0.0403, 0.0820],
       device='cuda:2'), in_proj_covar=tensor([0.0126, 0.0133, 0.0161, 0.0198, 0.0173, 0.0128, 0.0104, 0.0135],
       device='cuda:2'), out_proj_covar=tensor([0.0001, 0.0001, 0.0002, 0.0002, 0.0002, 0.0001, 0.0001, 0.0001],
       device='cuda:2')
2022-12-22 13:57:18,752 WARNING [train.py:1060] (2/4) Exclude cut with ID 7255-291500-0005-134304-0_sp0.9 from training. Duration: 27.8166875
2022-12-22 13:57:43,426 WARNING [train.py:1060] (2/4) Exclude cut with ID 7357-94126-0021-15852-0_sp1.1 from training. Duration: 22.5090625
2022-12-22 13:57:49,865 WARNING [train.py:1060] (2/4) Exclude cut with ID 7255-291500-0005-134304-0 from training. Duration: 25.035
2022-12-22 13:58:08,304 WARNING [train.py:1060] (2/4) Exclude cut with ID 774-127930-0014-10412-0_sp1.1 from training. Duration: 0.95
2022-12-22 13:58:13,315 INFO [scaling.py:679] (2/4) Whitening: num_groups=8, num_channels=96, metric=1.91 vs. limit=2.0
2022-12-22 13:58:19,855 INFO [scaling.py:679] (2/4) Whitening: num_groups=1, num_channels=384, metric=4.66 vs. limit=5.0
2022-12-22 13:58:26,767 WARNING [train.py:1060] (2/4) Exclude cut with ID 3033-130750-0096-55598-0_sp0.9 from training. Duration: 0.92225
2022-12-22 13:58:27,110 INFO [zipformer.py:660] (2/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=8760.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 13:58:31,064 INFO [train.py:894] (2/4) Epoch 3, batch 1750, loss[loss=0.3471, simple_loss=0.3929, pruned_loss=0.1507, over 18486.00 frames. ], tot_loss[loss=0.3032, simple_loss=0.3589, pruned_loss=0.1238, over 3713321.87 frames. ], batch size: 54, lr: 3.28e-02, grad_scale: 8.0
2022-12-22 13:58:55,294 WARNING [train.py:1060] (2/4) Exclude cut with ID 4511-76322-0006-80011-0 from training. Duration: 21.97
2022-12-22 13:59:12,766 WARNING [train.py:1060] (2/4) Exclude cut with ID 7492-105653-0055-62765-0_sp0.9 from training. Duration: 21.97225
2022-12-22 13:59:14,181 WARNING [train.py:1060] (2/4) Exclude cut with ID 453-131332-0000-47844-0_sp0.9 from training. Duration: 25.3333125
2022-12-22 13:59:22,908 WARNING [train.py:1060] (2/4) Exclude cut with ID 5172-29468-0015-19128-0_sp0.9 from training. Duration: 21.5055625
2022-12-22 13:59:34,733 WARNING [train.py:1060] (2/4) Exclude cut with ID 453-131332-0000-47844-0_sp1.1 from training. Duration: 20.72725
2022-12-22 13:59:43,097 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 4.523e+02 7.062e+02 8.790e+02 1.163e+03 3.166e+03, threshold=1.758e+03, percent-clipped=4.0
2022-12-22 13:59:46,003 INFO [train.py:894] (2/4) Epoch 3, batch 1800, loss[loss=0.3326, simple_loss=0.3621, pruned_loss=0.1516, over 18540.00 frames. ], tot_loss[loss=0.3089, simple_loss=0.3614, pruned_loss=0.1282, over 3713627.15 frames. ], batch size: 47, lr: 3.27e-02, grad_scale: 8.0
2022-12-22 13:59:52,477 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([2.4888, 3.2612, 3.4995, 1.7643, 3.2451, 2.2566, 0.8106, 2.3908],
       device='cuda:2'), covar=tensor([0.1920, 0.0926, 0.1418, 0.3388, 0.1034, 0.1823, 0.6263, 0.2212],
       device='cuda:2'), in_proj_covar=tensor([0.0112, 0.0083, 0.0141, 0.0112, 0.0089, 0.0093, 0.0142, 0.0109],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0001, 0.0002, 0.0002, 0.0001, 0.0001, 0.0002, 0.0002],
       device='cuda:2')
2022-12-22 13:59:59,144 INFO [zipformer.py:660] (2/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=8821.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 14:00:03,563 WARNING [train.py:1060] (2/4) Exclude cut with ID 8631-249866-0030-130156-0_sp0.9 from training. Duration: 26.32775
2022-12-22 14:00:36,948 WARNING [train.py:1060] (2/4) Exclude cut with ID 3867-173237-0077-144769-0 from training. Duration: 20.025
2022-12-22 14:00:40,917 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([2.3809, 2.0815, 1.7104, 2.9047, 2.0422, 4.1593, 2.3388, 1.8135],
       device='cuda:2'), covar=tensor([0.1020, 0.1872, 0.1670, 0.1195, 0.1667, 0.0293, 0.1372, 0.1906],
       device='cuda:2'), in_proj_covar=tensor([0.0091, 0.0091, 0.0095, 0.0092, 0.0110, 0.0079, 0.0098, 0.0090],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0003, 0.0002, 0.0003, 0.0002],
       device='cuda:2')
2022-12-22 14:00:43,603 WARNING [train.py:1060] (2/4) Exclude cut with ID 6709-74022-0004-86860-0_sp1.1 from training. Duration: 0.9409375
2022-12-22 14:00:43,613 WARNING [train.py:1060] (2/4) Exclude cut with ID 4757-1811-0023-62229-0_sp0.9 from training. Duration: 21.37775
2022-12-22 14:01:02,095 INFO [train.py:894] (2/4) Epoch 3, batch 1850, loss[loss=0.288, simple_loss=0.3378, pruned_loss=0.1191, over 18611.00 frames. ], tot_loss[loss=0.3156, simple_loss=0.3649, pruned_loss=0.1332, over 3713212.54 frames. ], batch size: 45, lr: 3.27e-02, grad_scale: 8.0
2022-12-22 14:01:04,235 WARNING [train.py:1060] (2/4) Exclude cut with ID 1250-135782-0004-25974-0_sp0.9 from training. Duration: 21.17225
2022-12-22 14:01:04,247 WARNING [train.py:1060] (2/4) Exclude cut with ID 7357-94126-0021-15852-0_sp0.9 from training. Duration: 27.511125
2022-12-22 14:01:35,272 WARNING [train.py:1060] (2/4) Exclude cut with ID 453-131332-0000-47844-0 from training. Duration: 22.8
2022-12-22 14:01:39,591 WARNING [train.py:1060] (2/4) Exclude cut with ID 4964-30587-0040-44509-0 from training. Duration: 22.585
2022-12-22 14:02:09,070 WARNING [train.py:1060] (2/4) Exclude cut with ID 6978-92210-0001-146967-0_sp0.9 from training. Duration: 22.0166875
2022-12-22 14:02:15,217 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 4.126e+02 6.920e+02 8.514e+02 1.073e+03 2.311e+03, threshold=1.703e+03, percent-clipped=5.0
2022-12-22 14:02:18,063 INFO [train.py:894] (2/4) Epoch 3, batch 1900, loss[loss=0.3145, simple_loss=0.3682, pruned_loss=0.1304, over 18593.00 frames. ], tot_loss[loss=0.3196, simple_loss=0.3672, pruned_loss=0.136, over 3713043.56 frames. ], batch size: 56, lr: 3.26e-02, grad_scale: 8.0
2022-12-22 14:02:25,477 WARNING [train.py:1060] (2/4) Exclude cut with ID 7255-291500-0003-134302-0_sp1.1 from training. Duration: 24.395375
2022-12-22 14:02:29,910 INFO [zipformer.py:660] (2/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=8921.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 14:02:32,768 WARNING [train.py:1060] (2/4) Exclude cut with ID 3488-85273-0017-41203-0_sp0.9 from training. Duration: 27.47775
2022-12-22 14:02:37,760 WARNING [train.py:1060] (2/4) Exclude cut with ID 432-122774-0017-62487-0_sp0.9 from training. Duration: 24.8833125
2022-12-22 14:02:39,385 INFO [zipformer.py:660] (2/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=8927.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 14:02:40,284 WARNING [train.py:1060] (2/4) Exclude cut with ID 6758-72288-0033-108368-0 from training. Duration: 23.39
2022-12-22 14:02:45,078 WARNING [train.py:1060] (2/4) Exclude cut with ID 8291-282929-0007-12994-0_sp0.9 from training. Duration: 28.72225
2022-12-22 14:02:54,421 WARNING [train.py:1060] (2/4) Exclude cut with ID 585-294811-0110-133686-0_sp0.9 from training. Duration: 20.8944375
2022-12-22 14:02:54,713 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([2.2719, 1.6625, 2.7959, 3.7194, 2.6452, 2.1516, 0.6977, 2.2306],
       device='cuda:2'), covar=tensor([0.2015, 0.2242, 0.1550, 0.0516, 0.1691, 0.1683, 0.3269, 0.1512],
       device='cuda:2'), in_proj_covar=tensor([0.0096, 0.0106, 0.0115, 0.0078, 0.0096, 0.0110, 0.0132, 0.0093],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-22 14:03:09,355 WARNING [train.py:1060] (2/4) Exclude cut with ID 5796-66357-0007-116447-0_sp0.9 from training. Duration: 23.8444375
2022-12-22 14:03:32,686 INFO [train.py:894] (2/4) Epoch 3, batch 1950, loss[loss=0.3272, simple_loss=0.3772, pruned_loss=0.1386, over 18586.00 frames. ], tot_loss[loss=0.3191, simple_loss=0.3667, pruned_loss=0.1358, over 3714020.81 frames. ], batch size: 51, lr: 3.25e-02, grad_scale: 8.0
2022-12-22 14:03:34,190 WARNING [train.py:1060] (2/4) Exclude cut with ID 8291-282929-0007-12994-0 from training. Duration: 25.85
2022-12-22 14:03:34,200 WARNING [train.py:1060] (2/4) Exclude cut with ID 8291-282929-0023-13010-0 from training. Duration: 21.39
2022-12-22 14:03:41,379 INFO [zipformer.py:660] (2/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=8969.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 14:03:47,608 WARNING [train.py:1060] (2/4) Exclude cut with ID 7357-94126-0014-15845-0 from training. Duration: 27.92
2022-12-22 14:04:06,825 INFO [zipformer.py:660] (2/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=8986.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 14:04:10,660 INFO [zipformer.py:660] (2/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=8988.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 14:04:16,415 WARNING [train.py:1060] (2/4) Exclude cut with ID 8631-249866-0039-130165-0_sp0.9 from training. Duration: 20.661125
2022-12-22 14:04:40,597 WARNING [train.py:1060] (2/4) Exclude cut with ID 7357-94126-0043-15874-0_sp0.9 from training. Duration: 20.07225
2022-12-22 14:04:44,891 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 3.540e+02 6.702e+02 8.505e+02 1.065e+03 2.063e+03, threshold=1.701e+03, percent-clipped=2.0
2022-12-22 14:04:47,916 INFO [train.py:894] (2/4) Epoch 3, batch 2000, loss[loss=0.3011, simple_loss=0.3353, pruned_loss=0.1334, over 18612.00 frames. ], tot_loss[loss=0.319, simple_loss=0.3659, pruned_loss=0.136, over 3714151.52 frames. ], batch size: 41, lr: 3.24e-02, grad_scale: 8.0
2022-12-22 14:04:47,954 WARNING [train.py:1060] (2/4) Exclude cut with ID 1085-156170-0017-128270-0 from training. Duration: 21.01
2022-12-22 14:04:48,262 INFO [zipformer.py:660] (2/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=9013.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 14:04:59,937 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([2.2180, 2.2425, 0.9375, 2.3183, 2.5721, 2.1077, 3.8048, 2.0236],
       device='cuda:2'), covar=tensor([0.0742, 0.1457, 0.2326, 0.2012, 0.1713, 0.0810, 0.0368, 0.1063],
       device='cuda:2'), in_proj_covar=tensor([0.0129, 0.0137, 0.0167, 0.0203, 0.0179, 0.0132, 0.0114, 0.0140],
       device='cuda:2'), out_proj_covar=tensor([0.0001, 0.0001, 0.0002, 0.0002, 0.0002, 0.0001, 0.0001, 0.0001],
       device='cuda:2')
2022-12-22 14:05:39,004 INFO [zipformer.py:660] (2/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=9047.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 14:05:44,681 INFO [scaling.py:679] (2/4) Whitening: num_groups=8, num_channels=96, metric=1.38 vs. limit=2.0
2022-12-22 14:05:57,256 WARNING [train.py:1060] (2/4) Exclude cut with ID 2195-150901-0045-59933-0 from training. Duration: 20.65
2022-12-22 14:06:03,147 INFO [train.py:894] (2/4) Epoch 3, batch 2050, loss[loss=0.2879, simple_loss=0.3418, pruned_loss=0.117, over 18573.00 frames. ], tot_loss[loss=0.3214, simple_loss=0.3672, pruned_loss=0.1378, over 3714960.29 frames. ], batch size: 49, lr: 3.24e-02, grad_scale: 8.0
2022-12-22 14:06:03,202 WARNING [train.py:1060] (2/4) Exclude cut with ID 5796-66357-0007-116447-0 from training. Duration: 21.46
2022-12-22 14:06:20,874 INFO [zipformer.py:660] (2/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=9074.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 14:06:47,895 WARNING [train.py:1060] (2/4) Exclude cut with ID 3557-8342-0013-54691-0 from training. Duration: 0.92
2022-12-22 14:06:55,381 WARNING [train.py:1060] (2/4) Exclude cut with ID 8291-282929-0023-13010-0_sp0.9 from training. Duration: 23.7666875
2022-12-22 14:07:16,204 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 4.185e+02 7.079e+02 8.812e+02 1.071e+03 1.633e+03, threshold=1.762e+03, percent-clipped=0.0
2022-12-22 14:07:19,196 INFO [train.py:894] (2/4) Epoch 3, batch 2100, loss[loss=0.3399, simple_loss=0.3747, pruned_loss=0.1525, over 18514.00 frames. ], tot_loss[loss=0.3213, simple_loss=0.3667, pruned_loss=0.1379, over 3714657.15 frames. ], batch size: 52, lr: 3.23e-02, grad_scale: 16.0
2022-12-22 14:07:24,386 INFO [zipformer.py:660] (2/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=9116.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 14:07:31,870 WARNING [train.py:1060] (2/4) Exclude cut with ID 8544-281189-0060-101339-0_sp0.9 from training. Duration: 20.861125
2022-12-22 14:07:42,334 WARNING [train.py:1060] (2/4) Exclude cut with ID 3488-65654-0031-41259-0_sp0.9 from training. Duration: 22.711125
2022-12-22 14:08:09,708 INFO [scaling.py:679] (2/4) Whitening: num_groups=1, num_channels=384, metric=5.61 vs. limit=5.0
2022-12-22 14:08:21,772 WARNING [train.py:1060] (2/4) Exclude cut with ID 6951-79737-0043-132310-0_sp1.1 from training. Duration: 22.986375
2022-12-22 14:08:35,393 INFO [train.py:894] (2/4) Epoch 3, batch 2150, loss[loss=0.2785, simple_loss=0.3242, pruned_loss=0.1164, over 18597.00 frames. ], tot_loss[loss=0.3212, simple_loss=0.3659, pruned_loss=0.1382, over 3714750.80 frames. ], batch size: 45, lr: 3.22e-02, grad_scale: 8.0
2022-12-22 14:08:39,772 WARNING [train.py:1060] (2/4) Exclude cut with ID 8040-260924-0003-80960-0_sp0.9 from training. Duration: 22.07225
2022-12-22 14:08:43,566 WARNING [train.py:1060] (2/4) Exclude cut with ID 7699-105389-0045-26330-0_sp0.9 from training. Duration: 20.3055625
2022-12-22 14:08:45,327 WARNING [train.py:1060] (2/4) Exclude cut with ID 6356-271890-0060-94317-0_sp0.9 from training. Duration: 20.72225
2022-12-22 14:09:01,115 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.5655, 0.9248, 1.8331, 2.2550, 1.7013, 1.9581, 0.3639, 1.6798],
       device='cuda:2'), covar=tensor([0.2703, 0.3229, 0.2124, 0.0967, 0.2197, 0.1667, 0.4198, 0.1935],
       device='cuda:2'), in_proj_covar=tensor([0.0097, 0.0106, 0.0115, 0.0080, 0.0097, 0.0110, 0.0135, 0.0093],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-22 14:09:03,801 WARNING [train.py:1060] (2/4) Exclude cut with ID 3488-85273-0017-41203-0_sp1.1 from training. Duration: 22.4818125
2022-12-22 14:09:30,363 WARNING [train.py:1060] (2/4) Exclude cut with ID 4964-30587-0040-44509-0_sp0.9 from training. Duration: 25.0944375
2022-12-22 14:09:34,922 WARNING [train.py:1060] (2/4) Exclude cut with ID 6533-399-0047-104881-0 from training. Duration: 21.515
2022-12-22 14:09:40,667 WARNING [train.py:1060] (2/4) Exclude cut with ID 7357-94126-0009-15840-0_sp0.9 from training. Duration: 27.02225
2022-12-22 14:09:46,709 WARNING [train.py:1060] (2/4) Exclude cut with ID 432-122774-0010-62480-0_sp0.9 from training. Duration: 22.22225
2022-12-22 14:09:48,381 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 4.146e+02 7.205e+02 8.913e+02 1.057e+03 2.694e+03, threshold=1.783e+03, percent-clipped=4.0
2022-12-22 14:09:49,870 INFO [train.py:894] (2/4) Epoch 3, batch 2200, loss[loss=0.2731, simple_loss=0.3229, pruned_loss=0.1116, over 18609.00 frames. ], tot_loss[loss=0.3222, simple_loss=0.3665, pruned_loss=0.139, over 3714601.58 frames. ], batch size: 45, lr: 3.22e-02, grad_scale: 8.0
2022-12-22 14:09:52,886 WARNING [train.py:1060] (2/4) Exclude cut with ID 4964-30587-0085-44554-0_sp0.9 from training. Duration: 20.85
2022-12-22 14:09:53,365 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.3195, 1.5191, 1.0769, 1.8940, 2.1007, 1.2360, 1.7644, 0.9684],
       device='cuda:2'), covar=tensor([0.1875, 0.1693, 0.1495, 0.0938, 0.1383, 0.1335, 0.1383, 0.1742],
       device='cuda:2'), in_proj_covar=tensor([0.0183, 0.0151, 0.0149, 0.0134, 0.0183, 0.0143, 0.0150, 0.0149],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0001, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-22 14:10:29,106 WARNING [train.py:1060] (2/4) Exclude cut with ID 4295-39940-0007-92567-0 from training. Duration: 21.54
2022-12-22 14:10:33,660 WARNING [train.py:1060] (2/4) Exclude cut with ID 4964-30587-0040-44509-0_sp1.1 from training. Duration: 20.5318125
2022-12-22 14:10:43,911 WARNING [train.py:1060] (2/4) Exclude cut with ID 7255-291500-0012-134311-0_sp0.9 from training. Duration: 21.9333125
2022-12-22 14:11:04,623 INFO [train.py:894] (2/4) Epoch 3, batch 2250, loss[loss=0.3041, simple_loss=0.3572, pruned_loss=0.1255, over 18704.00 frames. ], tot_loss[loss=0.3223, simple_loss=0.3664, pruned_loss=0.1391, over 3713856.59 frames. ], batch size: 50, lr: 3.21e-02, grad_scale: 8.0
2022-12-22 14:11:32,443 WARNING [train.py:1060] (2/4) Exclude cut with ID 8631-249866-0025-130151-0_sp0.9 from training. Duration: 21.7944375
2022-12-22 14:11:35,393 INFO [zipformer.py:660] (2/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=9283.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 14:11:46,396 WARNING [train.py:1060] (2/4) Exclude cut with ID 8291-282929-0002-12989-0_sp0.9 from training. Duration: 22.4666875
2022-12-22 14:11:52,244 WARNING [train.py:1060] (2/4) Exclude cut with ID 6121-9014-0076-24124-0 from training. Duration: 21.635
2022-12-22 14:11:58,125 WARNING [train.py:1060] (2/4) Exclude cut with ID 6121-9014-0076-24124-0_sp0.9 from training. Duration: 24.038875
2022-12-22 14:12:18,740 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 4.113e+02 7.064e+02 8.441e+02 1.095e+03 1.855e+03, threshold=1.688e+03, percent-clipped=2.0
2022-12-22 14:12:20,127 INFO [train.py:894] (2/4) Epoch 3, batch 2300, loss[loss=0.3416, simple_loss=0.3886, pruned_loss=0.1473, over 18623.00 frames. ], tot_loss[loss=0.3222, simple_loss=0.3666, pruned_loss=0.1389, over 3714118.82 frames. ], batch size: 69, lr: 3.20e-02, grad_scale: 8.0
2022-12-22 14:12:41,350 WARNING [train.py:1060] (2/4) Exclude cut with ID 7255-291500-0009-134308-0_sp1.1 from training. Duration: 21.786375
2022-12-22 14:12:48,311 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([0.5519, 1.5019, 1.9423, 0.3636, 1.3845, 2.2650, 1.5160, 1.4002],
       device='cuda:2'), covar=tensor([0.1311, 0.0667, 0.0495, 0.1219, 0.0573, 0.0305, 0.0666, 0.0918],
       device='cuda:2'), in_proj_covar=tensor([0.0097, 0.0104, 0.0073, 0.0110, 0.0095, 0.0067, 0.0108, 0.0082],
       device='cuda:2'), out_proj_covar=tensor([1.0732e-04, 1.1802e-04, 8.7429e-05, 1.2412e-04, 1.0695e-04, 8.0245e-05,
        1.2496e-04, 9.5119e-05], device='cuda:2')
2022-12-22 14:12:52,354 WARNING [train.py:1060] (2/4) Exclude cut with ID 8291-282929-0002-12989-0 from training. Duration: 20.22
2022-12-22 14:13:06,043 INFO [zipformer.py:660] (2/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=9342.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 14:13:37,570 INFO [train.py:894] (2/4) Epoch 3, batch 2350, loss[loss=0.2463, simple_loss=0.2989, pruned_loss=0.09686, over 18699.00 frames. ], tot_loss[loss=0.3217, simple_loss=0.3665, pruned_loss=0.1384, over 3715301.36 frames. ], batch size: 41, lr: 3.20e-02, grad_scale: 8.0
2022-12-22 14:13:48,228 INFO [zipformer.py:660] (2/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=9369.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 14:14:50,982 WARNING [train.py:1060] (2/4) Exclude cut with ID 6951-79737-0043-132310-0 from training. Duration: 25.285
2022-12-22 14:14:53,183 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 3.774e+02 6.974e+02 8.877e+02 1.045e+03 2.873e+03, threshold=1.775e+03, percent-clipped=4.0
2022-12-22 14:14:54,865 INFO [train.py:894] (2/4) Epoch 3, batch 2400, loss[loss=0.3137, simple_loss=0.3681, pruned_loss=0.1296, over 18671.00 frames. ], tot_loss[loss=0.3206, simple_loss=0.3652, pruned_loss=0.138, over 3714577.37 frames. ], batch size: 62, lr: 3.19e-02, grad_scale: 8.0
2022-12-22 14:14:56,738 INFO [zipformer.py:660] (2/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=9414.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 14:14:59,982 INFO [zipformer.py:660] (2/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=9416.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 14:15:57,195 WARNING [train.py:1060] (2/4) Exclude cut with ID 811-130148-0001-63453-0_sp0.9 from training. Duration: 20.861125
2022-12-22 14:16:09,908 INFO [train.py:894] (2/4) Epoch 3, batch 2450, loss[loss=0.2852, simple_loss=0.3479, pruned_loss=0.1113, over 18512.00 frames. ], tot_loss[loss=0.3183, simple_loss=0.3642, pruned_loss=0.1363, over 3714624.61 frames. ], batch size: 52, lr: 3.18e-02, grad_scale: 8.0
2022-12-22 14:16:11,468 INFO [zipformer.py:660] (2/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=9464.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 14:16:20,953 WARNING [train.py:1060] (2/4) Exclude cut with ID 6010-56788-0055-90261-0 from training. Duration: 20.88
2022-12-22 14:16:27,980 INFO [zipformer.py:660] (2/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=9475.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 14:16:52,520 WARNING [train.py:1060] (2/4) Exclude cut with ID 7357-94126-0045-15876-0_sp0.9 from training. Duration: 23.4166875
2022-12-22 14:17:24,905 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 4.333e+02 7.255e+02 8.783e+02 1.029e+03 1.927e+03, threshold=1.757e+03, percent-clipped=1.0
2022-12-22 14:17:25,381 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.8585, 2.0032, 1.1809, 0.9293, 2.3721, 2.0855, 1.3663, 1.1668],
       device='cuda:2'), covar=tensor([0.0586, 0.0493, 0.1110, 0.1292, 0.0214, 0.0504, 0.1115, 0.1887],
       device='cuda:2'), in_proj_covar=tensor([0.0118, 0.0110, 0.0136, 0.0125, 0.0077, 0.0114, 0.0137, 0.0149],
       device='cuda:2'), out_proj_covar=tensor([1.4690e-04, 1.3881e-04, 1.6429e-04, 1.5235e-04, 9.5723e-05, 1.3981e-04,
        1.6709e-04, 1.7629e-04], device='cuda:2')
2022-12-22 14:17:26,938 INFO [train.py:894] (2/4) Epoch 3, batch 2500, loss[loss=0.3147, simple_loss=0.3466, pruned_loss=0.1414, over 18615.00 frames. ], tot_loss[loss=0.3177, simple_loss=0.3638, pruned_loss=0.1358, over 3715793.82 frames. ], batch size: 41, lr: 3.18e-02, grad_scale: 8.0
2022-12-22 14:17:28,862 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([2.2849, 2.8901, 2.8622, 1.7530, 2.6016, 2.6637, 2.1421, 3.2556],
       device='cuda:2'), covar=tensor([0.1455, 0.0848, 0.1734, 0.2459, 0.1081, 0.1517, 0.2240, 0.0709],
       device='cuda:2'), in_proj_covar=tensor([0.0194, 0.0151, 0.0193, 0.0186, 0.0178, 0.0200, 0.0191, 0.0168],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-22 14:18:07,765 WARNING [train.py:1060] (2/4) Exclude cut with ID 7255-291500-0006-134305-0 from training. Duration: 21.24
2022-12-22 14:18:09,195 WARNING [train.py:1060] (2/4) Exclude cut with ID 6533-399-0047-104881-0_sp0.9 from training. Duration: 23.9055625
2022-12-22 14:18:42,848 INFO [train.py:894] (2/4) Epoch 3, batch 2550, loss[loss=0.2825, simple_loss=0.3241, pruned_loss=0.1204, over 18400.00 frames. ], tot_loss[loss=0.3164, simple_loss=0.3625, pruned_loss=0.1352, over 3715103.97 frames. ], batch size: 42, lr: 3.17e-02, grad_scale: 8.0
2022-12-22 14:18:42,901 WARNING [train.py:1060] (2/4) Exclude cut with ID 6758-72288-0033-108368-0_sp0.9 from training. Duration: 25.988875
2022-12-22 14:18:51,371 WARNING [train.py:1060] (2/4) Exclude cut with ID 7255-291500-0001-134300-0_sp0.9 from training. Duration: 20.67225
2022-12-22 14:19:12,631 INFO [zipformer.py:660] (2/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=9583.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 14:19:15,108 INFO [zipformer.py:660] (2/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=9584.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 14:19:17,870 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.1831, 0.9544, 1.3647, 1.6174, 1.1416, 1.7166, 0.8621, 1.2445],
       device='cuda:2'), covar=tensor([0.1915, 0.1886, 0.1338, 0.0708, 0.1762, 0.1232, 0.2062, 0.1467],
       device='cuda:2'), in_proj_covar=tensor([0.0100, 0.0108, 0.0119, 0.0083, 0.0100, 0.0113, 0.0135, 0.0093],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-22 14:19:38,105 WARNING [train.py:1060] (2/4) Exclude cut with ID 3488-85273-0038-41224-0 from training. Duration: 20.34
2022-12-22 14:19:58,220 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 3.647e+02 7.260e+02 8.987e+02 1.119e+03 2.066e+03, threshold=1.797e+03, percent-clipped=2.0
2022-12-22 14:19:59,699 INFO [train.py:894] (2/4) Epoch 3, batch 2600, loss[loss=0.3256, simple_loss=0.3735, pruned_loss=0.1388, over 18726.00 frames. ], tot_loss[loss=0.3175, simple_loss=0.3635, pruned_loss=0.1357, over 3716162.18 frames. ], batch size: 54, lr: 3.16e-02, grad_scale: 8.0
2022-12-22 14:20:26,422 INFO [zipformer.py:660] (2/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=9631.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 14:20:43,655 INFO [zipformer.py:660] (2/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=9642.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 14:20:48,817 INFO [zipformer.py:660] (2/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=9645.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 14:20:51,252 WARNING [train.py:1060] (2/4) Exclude cut with ID 7357-94126-0026-15857-0_sp0.9 from training. Duration: 25.061125
2022-12-22 14:21:02,745 WARNING [train.py:1060] (2/4) Exclude cut with ID 3033-130750-0096-55598-0 from training. Duration: 0.83
2022-12-22 14:21:07,116 INFO [zipformer.py:660] (2/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=9658.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 14:21:07,215 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([2.2167, 2.3832, 1.0444, 2.2796, 2.4914, 2.1218, 3.8791, 2.5066],
       device='cuda:2'), covar=tensor([0.0787, 0.1258, 0.2221, 0.2220, 0.1622, 0.0707, 0.0330, 0.0833],
       device='cuda:2'), in_proj_covar=tensor([0.0137, 0.0142, 0.0176, 0.0213, 0.0187, 0.0140, 0.0122, 0.0149],
       device='cuda:2'), out_proj_covar=tensor([0.0001, 0.0002, 0.0002, 0.0002, 0.0002, 0.0001, 0.0001, 0.0002],
       device='cuda:2')
2022-12-22 14:21:14,218 INFO [train.py:894] (2/4) Epoch 3, batch 2650, loss[loss=0.3005, simple_loss=0.3373, pruned_loss=0.1319, over 18619.00 frames. ], tot_loss[loss=0.3174, simple_loss=0.3636, pruned_loss=0.1356, over 3715369.80 frames. ], batch size: 45, lr: 3.16e-02, grad_scale: 8.0
2022-12-22 14:21:23,325 INFO [zipformer.py:660] (2/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=9669.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 14:21:28,843 WARNING [train.py:1060] (2/4) Exclude cut with ID 3488-85273-0017-41203-0 from training. Duration: 24.73
2022-12-22 14:21:43,015 WARNING [train.py:1060] (2/4) Exclude cut with ID 7255-291500-0009-134308-0 from training. Duration: 23.965
2022-12-22 14:21:46,601 INFO [scaling.py:679] (2/4) Whitening: num_groups=8, num_channels=96, metric=1.57 vs. limit=2.0
2022-12-22 14:21:52,773 WARNING [train.py:1060] (2/4) Exclude cut with ID 6978-92210-0030-146996-0_sp0.9 from training. Duration: 22.088875
2022-12-22 14:21:55,802 INFO [zipformer.py:660] (2/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=9690.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 14:21:58,949 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.7683, 1.9507, 1.1228, 0.8669, 2.4098, 2.1645, 1.3687, 1.1778],
       device='cuda:2'), covar=tensor([0.0670, 0.0621, 0.1167, 0.1317, 0.0186, 0.0494, 0.1104, 0.1814],
       device='cuda:2'), in_proj_covar=tensor([0.0116, 0.0111, 0.0135, 0.0125, 0.0076, 0.0112, 0.0133, 0.0147],
       device='cuda:2'), out_proj_covar=tensor([1.4491e-04, 1.3992e-04, 1.6317e-04, 1.5145e-04, 9.4589e-05, 1.3797e-04,
        1.6276e-04, 1.7344e-04], device='cuda:2')
2022-12-22 14:22:09,983 WARNING [train.py:1060] (2/4) Exclude cut with ID 7255-291500-0006-134305-0_sp0.9 from training. Duration: 23.6
2022-12-22 14:22:23,808 INFO [scaling.py:679] (2/4) Whitening: num_groups=8, num_channels=96, metric=2.11 vs. limit=2.0
2022-12-22 14:22:28,385 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 4.461e+02 7.252e+02 8.983e+02 1.112e+03 2.476e+03, threshold=1.797e+03, percent-clipped=3.0
2022-12-22 14:22:29,673 INFO [train.py:894] (2/4) Epoch 3, batch 2700, loss[loss=0.3555, simple_loss=0.3958, pruned_loss=0.1576, over 18661.00 frames. ], tot_loss[loss=0.3164, simple_loss=0.3625, pruned_loss=0.1352, over 3715750.78 frames. ], batch size: 98, lr: 3.15e-02, grad_scale: 8.0
2022-12-22 14:22:35,624 INFO [zipformer.py:660] (2/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=9717.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 14:22:38,845 INFO [zipformer.py:660] (2/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=9719.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 14:23:12,474 INFO [scaling.py:679] (2/4) Whitening: num_groups=8, num_channels=96, metric=1.40 vs. limit=2.0
2022-12-22 14:23:16,393 INFO [zipformer.py:660] (2/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=9743.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 14:23:45,409 INFO [train.py:894] (2/4) Epoch 3, batch 2750, loss[loss=0.3547, simple_loss=0.3893, pruned_loss=0.1601, over 18387.00 frames. ], tot_loss[loss=0.3159, simple_loss=0.362, pruned_loss=0.1349, over 3715265.38 frames. ], batch size: 53, lr: 3.15e-02, grad_scale: 8.0
2022-12-22 14:23:47,611 INFO [scaling.py:679] (2/4) Whitening: num_groups=8, num_channels=192, metric=1.80 vs. limit=2.0
2022-12-22 14:23:49,542 WARNING [train.py:1060] (2/4) Exclude cut with ID 8291-282929-0024-13011-0 from training. Duration: 23.795
2022-12-22 14:23:53,149 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.9934, 1.9033, 1.2043, 2.1498, 1.7913, 1.6151, 1.7031, 2.3313],
       device='cuda:2'), covar=tensor([0.1410, 0.1433, 0.1161, 0.1598, 0.1508, 0.0710, 0.1655, 0.0487],
       device='cuda:2'), in_proj_covar=tensor([0.0186, 0.0149, 0.0146, 0.0221, 0.0149, 0.0140, 0.0168, 0.0123],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0002, 0.0001, 0.0002, 0.0002, 0.0001, 0.0002, 0.0001],
       device='cuda:2')
2022-12-22 14:23:55,984 INFO [zipformer.py:660] (2/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=9770.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 14:24:06,602 WARNING [train.py:1060] (2/4) Exclude cut with ID 8631-249866-0030-130156-0_sp1.1 from training. Duration: 21.5409375
2022-12-22 14:24:07,041 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.6935, 1.6547, 0.9056, 1.6570, 1.7536, 1.4996, 2.6486, 1.7267],
       device='cuda:2'), covar=tensor([0.1001, 0.1369, 0.2585, 0.1807, 0.1898, 0.0980, 0.0575, 0.1102],
       device='cuda:2'), in_proj_covar=tensor([0.0137, 0.0143, 0.0177, 0.0216, 0.0187, 0.0141, 0.0124, 0.0148],
       device='cuda:2'), out_proj_covar=tensor([0.0001, 0.0002, 0.0002, 0.0002, 0.0002, 0.0001, 0.0001, 0.0001],
       device='cuda:2')
2022-12-22 14:24:09,260 WARNING [train.py:1060] (2/4) Exclude cut with ID 6978-92210-0019-146985-0_sp0.9 from training. Duration: 24.97775
2022-12-22 14:24:19,259 WARNING [train.py:1060] (2/4) Exclude cut with ID 1085-156170-0017-128270-0_sp0.9 from training. Duration: 23.3444375
2022-12-22 14:24:45,245 WARNING [train.py:1060] (2/4) Exclude cut with ID 6010-56788-0055-90261-0_sp0.9 from training. Duration: 23.2
2022-12-22 14:24:47,074 INFO [zipformer.py:660] (2/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=9804.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 14:24:50,147 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.1494, 1.1058, 1.5233, 0.8971, 1.3191, 1.3340, 0.9076, 1.5495],
       device='cuda:2'), covar=tensor([0.1131, 0.1333, 0.1220, 0.1696, 0.0949, 0.1097, 0.2537, 0.0660],
       device='cuda:2'), in_proj_covar=tensor([0.0198, 0.0158, 0.0197, 0.0191, 0.0190, 0.0201, 0.0195, 0.0172],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-22 14:24:51,216 WARNING [train.py:1060] (2/4) Exclude cut with ID 5653-46179-0060-117930-0_sp0.9 from training. Duration: 21.17225
2022-12-22 14:24:57,994 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 4.440e+02 7.867e+02 9.020e+02 1.115e+03 2.564e+03, threshold=1.804e+03, percent-clipped=3.0
2022-12-22 14:24:59,518 INFO [train.py:894] (2/4) Epoch 3, batch 2800, loss[loss=0.3879, simple_loss=0.4131, pruned_loss=0.1814, over 18672.00 frames. ], tot_loss[loss=0.3177, simple_loss=0.3632, pruned_loss=0.1362, over 3714248.22 frames. ], batch size: 60, lr: 3.14e-02, grad_scale: 8.0
2022-12-22 14:25:11,461 WARNING [train.py:1060] (2/4) Exclude cut with ID 7255-291500-0008-134307-0_sp0.9 from training. Duration: 24.6555625
2022-12-22 14:25:33,206 INFO [scaling.py:679] (2/4) Whitening: num_groups=1, num_channels=384, metric=3.06 vs. limit=5.0
2022-12-22 14:26:06,578 WARNING [train.py:1060] (2/4) Exclude cut with ID 3488-65654-0031-41259-0 from training. Duration: 20.44
2022-12-22 14:26:15,230 INFO [train.py:894] (2/4) Epoch 3, batch 2850, loss[loss=0.3305, simple_loss=0.3763, pruned_loss=0.1424, over 18591.00 frames. ], tot_loss[loss=0.317, simple_loss=0.3628, pruned_loss=0.1356, over 3714626.70 frames. ], batch size: 51, lr: 3.13e-02, grad_scale: 8.0
2022-12-22 14:26:19,910 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([2.4833, 3.4255, 3.6211, 1.6124, 3.3677, 2.4339, 0.8647, 2.3784],
       device='cuda:2'), covar=tensor([0.2101, 0.1002, 0.1632, 0.4144, 0.1226, 0.1697, 0.6652, 0.2377],
       device='cuda:2'), in_proj_covar=tensor([0.0120, 0.0088, 0.0145, 0.0115, 0.0100, 0.0096, 0.0147, 0.0113],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0001, 0.0002, 0.0002, 0.0002, 0.0001, 0.0002, 0.0002],
       device='cuda:2')
2022-12-22 14:26:21,837 WARNING [train.py:1060] (2/4) Exclude cut with ID 6951-79737-0018-132285-0_sp0.9 from training. Duration: 23.45
2022-12-22 14:26:50,247 WARNING [train.py:1060] (2/4) Exclude cut with ID 6945-60535-0076-12784-0_sp0.9 from training. Duration: 20.52225
2022-12-22 14:26:57,795 WARNING [train.py:1060] (2/4) Exclude cut with ID 7255-291500-0008-134307-0 from training. Duration: 22.19
2022-12-22 14:27:08,352 WARNING [train.py:1060] (2/4) Exclude cut with ID 7357-94126-0014-15845-0_sp1.1 from training. Duration: 25.3818125
2022-12-22 14:27:21,929 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.7279, 1.9845, 1.1049, 2.1650, 2.1278, 1.8156, 2.9911, 1.8653],
       device='cuda:2'), covar=tensor([0.0969, 0.1337, 0.2273, 0.1640, 0.1576, 0.0768, 0.0455, 0.1066],
       device='cuda:2'), in_proj_covar=tensor([0.0140, 0.0145, 0.0175, 0.0214, 0.0186, 0.0142, 0.0125, 0.0147],
       device='cuda:2'), out_proj_covar=tensor([0.0001, 0.0002, 0.0002, 0.0002, 0.0002, 0.0001, 0.0001, 0.0001],
       device='cuda:2')
2022-12-22 14:27:25,633 WARNING [train.py:1060] (2/4) Exclude cut with ID 6951-79737-0043-132310-0_sp0.9 from training. Duration: 28.0944375
2022-12-22 14:27:29,791 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 3.837e+02 6.577e+02 8.557e+02 1.075e+03 2.821e+03, threshold=1.711e+03, percent-clipped=5.0
2022-12-22 14:27:31,269 INFO [train.py:894] (2/4) Epoch 3, batch 2900, loss[loss=0.3801, simple_loss=0.4127, pruned_loss=0.1738, over 18530.00 frames. ], tot_loss[loss=0.3164, simple_loss=0.3624, pruned_loss=0.1352, over 3714889.64 frames. ], batch size: 55, lr: 3.13e-02, grad_scale: 8.0
2022-12-22 14:27:33,437 WARNING [train.py:1060] (2/4) Exclude cut with ID 2195-150901-0045-59933-0_sp0.9 from training. Duration: 22.9444375
2022-12-22 14:27:41,102 WARNING [train.py:1060] (2/4) Exclude cut with ID 8291-282929-0024-13011-0_sp1.1 from training. Duration: 21.6318125
2022-12-22 14:27:59,983 WARNING [train.py:1060] (2/4) Exclude cut with ID 8631-249866-0030-130156-0 from training. Duration: 23.695
2022-12-22 14:28:13,480 INFO [zipformer.py:660] (2/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=9940.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 14:28:24,890 WARNING [train.py:1060] (2/4) Exclude cut with ID 7699-105389-0094-26379-0 from training. Duration: 23.955
2022-12-22 14:28:47,832 INFO [train.py:894] (2/4) Epoch 3, batch 2950, loss[loss=0.3174, simple_loss=0.3546, pruned_loss=0.1401, over 18532.00 frames. ], tot_loss[loss=0.3155, simple_loss=0.3618, pruned_loss=0.1346, over 3714298.16 frames. ], batch size: 47, lr: 3.12e-02, grad_scale: 8.0
2022-12-22 14:28:57,554 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([2.8729, 3.6070, 3.9493, 2.1830, 3.7546, 2.6375, 0.8197, 2.4933],
       device='cuda:2'), covar=tensor([0.1527, 0.0805, 0.1246, 0.3043, 0.0916, 0.1364, 0.5351, 0.2111],
       device='cuda:2'), in_proj_covar=tensor([0.0120, 0.0087, 0.0145, 0.0114, 0.0099, 0.0095, 0.0144, 0.0110],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0001, 0.0002, 0.0002, 0.0002, 0.0001, 0.0002, 0.0002],
       device='cuda:2')
2022-12-22 14:28:58,833 WARNING [train.py:1060] (2/4) Exclude cut with ID 8291-282929-0024-13011-0_sp0.9 from training. Duration: 26.438875
2022-12-22 14:29:40,516 WARNING [train.py:1060] (2/4) Exclude cut with ID 7699-105389-0021-26306-0_sp0.9 from training. Duration: 21.2444375
2022-12-22 14:29:40,545 WARNING [train.py:1060] (2/4) Exclude cut with ID 7357-94126-0014-15845-0_sp0.9 from training. Duration: 31.02225
2022-12-22 14:29:40,825 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([5.9742, 5.0109, 5.0423, 5.3838, 5.3416, 5.1604, 5.7310, 1.6446],
       device='cuda:2'), covar=tensor([0.0409, 0.0372, 0.0460, 0.0306, 0.1140, 0.0825, 0.0333, 0.4053],
       device='cuda:2'), in_proj_covar=tensor([0.0186, 0.0148, 0.0136, 0.0109, 0.0203, 0.0154, 0.0155, 0.0197],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0001, 0.0001, 0.0001, 0.0002, 0.0001, 0.0002, 0.0002],
       device='cuda:2')
2022-12-22 14:29:54,245 WARNING [train.py:1060] (2/4) Exclude cut with ID 432-122774-0017-62487-0 from training. Duration: 22.395
2022-12-22 14:30:05,697 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 3.928e+02 7.043e+02 9.169e+02 1.169e+03 2.051e+03, threshold=1.834e+03, percent-clipped=1.0
2022-12-22 14:30:07,879 INFO [train.py:894] (2/4) Epoch 3, batch 3000, loss[loss=0.2751, simple_loss=0.3423, pruned_loss=0.1039, over 18496.00 frames. ], tot_loss[loss=0.3164, simple_loss=0.3624, pruned_loss=0.1353, over 3714406.57 frames. ], batch size: 52, lr: 3.11e-02, grad_scale: 8.0
2022-12-22 14:30:07,879 INFO [train.py:919] (2/4) Computing validation loss
2022-12-22 14:30:18,924 INFO [train.py:928] (2/4) Epoch 3, validation: loss=0.2259, simple_loss=0.3201, pruned_loss=0.06583, over 944034.00 frames. 
2022-12-22 14:30:18,925 INFO [train.py:929] (2/4) Maximum memory allocated so far is 24257MB
2022-12-22 14:30:20,716 INFO [zipformer.py:660] (2/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=10014.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 14:30:23,429 WARNING [train.py:1060] (2/4) Exclude cut with ID 7357-94126-0045-15876-0 from training. Duration: 21.075
2022-12-22 14:30:29,599 WARNING [train.py:1060] (2/4) Exclude cut with ID 6482-98857-0025-147532-0_sp0.9 from training. Duration: 20.0055625
2022-12-22 14:30:29,607 WARNING [train.py:1060] (2/4) Exclude cut with ID 6951-79737-0037-132304-0_sp0.9 from training. Duration: 22.05
2022-12-22 14:30:29,619 WARNING [train.py:1060] (2/4) Exclude cut with ID 7255-291500-0003-134302-0 from training. Duration: 26.8349375
2022-12-22 14:30:32,500 WARNING [train.py:1060] (2/4) Exclude cut with ID 7357-94126-0009-15840-0_sp1.1 from training. Duration: 22.1090625
2022-12-22 14:30:38,457 WARNING [train.py:1060] (2/4) Exclude cut with ID 7699-105389-0094-26379-0_sp0.9 from training. Duration: 26.6166875
2022-12-22 14:30:58,380 WARNING [train.py:1060] (2/4) Exclude cut with ID 2046-178027-0000-53705-0_sp0.9 from training. Duration: 20.3055625
2022-12-22 14:30:58,682 INFO [zipformer.py:660] (2/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=10039.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 14:31:19,871 WARNING [train.py:1060] (2/4) Exclude cut with ID 7205-50138-0008-5373-0_sp0.9 from training. Duration: 20.7
2022-12-22 14:31:33,791 INFO [zipformer.py:660] (2/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=10062.0, num_to_drop=1, layers_to_drop={0}
2022-12-22 14:31:34,710 INFO [train.py:894] (2/4) Epoch 3, batch 3050, loss[loss=0.2518, simple_loss=0.3062, pruned_loss=0.09868, over 18476.00 frames. ], tot_loss[loss=0.3156, simple_loss=0.362, pruned_loss=0.1346, over 3713935.11 frames. ], batch size: 43, lr: 3.11e-02, grad_scale: 8.0
2022-12-22 14:31:45,000 INFO [zipformer.py:660] (2/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=10070.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 14:32:02,327 WARNING [train.py:1060] (2/4) Exclude cut with ID 6978-92210-0019-146985-0 from training. Duration: 22.48
2022-12-22 14:32:18,999 WARNING [train.py:1060] (2/4) Exclude cut with ID 7255-291500-0003-134302-0_sp0.9 from training. Duration: 29.816625
2022-12-22 14:32:30,175 INFO [zipformer.py:660] (2/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=10099.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 14:32:31,832 INFO [zipformer.py:660] (2/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=10100.0, num_to_drop=1, layers_to_drop={0}
2022-12-22 14:32:39,048 WARNING [train.py:1060] (2/4) Exclude cut with ID 7255-291500-0005-134304-0_sp1.1 from training. Duration: 22.7590625
2022-12-22 14:32:44,701 WARNING [train.py:1060] (2/4) Exclude cut with ID 7357-94126-0026-15857-0 from training. Duration: 22.555
2022-12-22 14:32:48,877 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 4.989e+02 6.948e+02 8.486e+02 9.943e+02 2.176e+03, threshold=1.697e+03, percent-clipped=1.0
2022-12-22 14:32:50,368 INFO [train.py:894] (2/4) Epoch 3, batch 3100, loss[loss=0.3042, simple_loss=0.3427, pruned_loss=0.1328, over 18458.00 frames. ], tot_loss[loss=0.3158, simple_loss=0.3622, pruned_loss=0.1347, over 3713775.01 frames. ], batch size: 43, lr: 3.10e-02, grad_scale: 8.0
2022-12-22 14:32:58,180 INFO [zipformer.py:660] (2/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=10118.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 14:33:02,308 WARNING [train.py:1060] (2/4) Exclude cut with ID 1250-135782-0005-25975-0_sp0.9 from training. Duration: 21.688875
2022-12-22 14:33:05,299 INFO [zipformer.py:660] (2/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=10123.0, num_to_drop=1, layers_to_drop={3}
2022-12-22 14:33:37,122 WARNING [train.py:1060] (2/4) Exclude cut with ID 3488-85273-0038-41224-0_sp0.9 from training. Duration: 22.6
2022-12-22 14:34:04,925 INFO [train.py:894] (2/4) Epoch 3, batch 3150, loss[loss=0.3896, simple_loss=0.4151, pruned_loss=0.182, over 18373.00 frames. ], tot_loss[loss=0.3143, simple_loss=0.3612, pruned_loss=0.1337, over 3714286.56 frames. ], batch size: 51, lr: 3.10e-02, grad_scale: 4.0
2022-12-22 14:34:11,709 WARNING [train.py:1060] (2/4) Exclude cut with ID 7357-94126-0009-15840-0 from training. Duration: 24.32
2022-12-22 14:35:11,110 WARNING [train.py:1060] (2/4) Exclude cut with ID 8291-276745-0093-13116-0_sp0.9 from training. Duration: 21.061125
2022-12-22 14:35:21,377 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 3.845e+02 7.390e+02 8.631e+02 1.171e+03 2.300e+03, threshold=1.726e+03, percent-clipped=5.0
2022-12-22 14:35:21,392 INFO [train.py:894] (2/4) Epoch 3, batch 3200, loss[loss=0.3013, simple_loss=0.3508, pruned_loss=0.126, over 18412.00 frames. ], tot_loss[loss=0.312, simple_loss=0.3594, pruned_loss=0.1323, over 3713111.08 frames. ], batch size: 46, lr: 3.09e-02, grad_scale: 8.0
2022-12-22 14:35:23,158 WARNING [train.py:1060] (2/4) Exclude cut with ID 7357-94126-0024-15855-0_sp0.9 from training. Duration: 20.32225
2022-12-22 14:35:34,688 WARNING [train.py:1060] (2/4) Exclude cut with ID 3033-130750-0096-55598-0_sp1.1 from training. Duration: 0.7545625
2022-12-22 14:35:54,145 WARNING [train.py:1060] (2/4) Exclude cut with ID 4295-39940-0007-92567-0_sp0.9 from training. Duration: 23.9333125
2022-12-22 14:36:04,669 INFO [zipformer.py:660] (2/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=10240.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 14:36:23,260 INFO [scaling.py:679] (2/4) Whitening: num_groups=8, num_channels=192, metric=1.84 vs. limit=2.0
2022-12-22 14:36:25,350 WARNING [train.py:1060] (2/4) Exclude cut with ID 7255-291500-0008-134307-0_sp1.1 from training. Duration: 20.17275
2022-12-22 14:36:32,481 WARNING [train.py:1060] (2/4) Exclude cut with ID 6978-92210-0019-146985-0_sp1.1 from training. Duration: 20.436375
2022-12-22 14:36:32,982 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([2.5442, 1.5227, 1.9297, 2.2487, 2.2009, 1.6883, 2.1458, 1.3561],
       device='cuda:2'), covar=tensor([0.0779, 0.1677, 0.1431, 0.1254, 0.0560, 0.0519, 0.1205, 0.0658],
       device='cuda:2'), in_proj_covar=tensor([0.0149, 0.0182, 0.0176, 0.0169, 0.0145, 0.0144, 0.0156, 0.0143],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0001, 0.0001, 0.0002, 0.0001],
       device='cuda:2')
2022-12-22 14:36:38,065 INFO [train.py:894] (2/4) Epoch 3, batch 3250, loss[loss=0.2936, simple_loss=0.3585, pruned_loss=0.1143, over 18461.00 frames. ], tot_loss[loss=0.3097, simple_loss=0.3578, pruned_loss=0.1308, over 3713391.76 frames. ], batch size: 54, lr: 3.08e-02, grad_scale: 8.0
2022-12-22 14:37:17,017 INFO [zipformer.py:660] (2/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=10288.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 14:37:53,205 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 4.532e+02 6.935e+02 8.611e+02 1.028e+03 2.623e+03, threshold=1.722e+03, percent-clipped=3.0
2022-12-22 14:37:53,221 INFO [train.py:894] (2/4) Epoch 3, batch 3300, loss[loss=0.2828, simple_loss=0.3456, pruned_loss=0.1101, over 18552.00 frames. ], tot_loss[loss=0.3073, simple_loss=0.3556, pruned_loss=0.1296, over 3713175.41 frames. ], batch size: 55, lr: 3.08e-02, grad_scale: 8.0
2022-12-22 14:37:54,867 WARNING [train.py:1060] (2/4) Exclude cut with ID 4234-40345-0022-142709-0_sp0.9 from training. Duration: 23.1055625
2022-12-22 14:37:55,044 INFO [zipformer.py:660] (2/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=10314.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 14:37:56,479 WARNING [train.py:1060] (2/4) Exclude cut with ID 8291-282929-0007-12994-0_sp1.1 from training. Duration: 23.5
2022-12-22 14:38:08,103 WARNING [train.py:1060] (2/4) Exclude cut with ID 7255-291500-0009-134308-0_sp0.9 from training. Duration: 26.62775
2022-12-22 14:38:22,264 WARNING [train.py:1060] (2/4) Exclude cut with ID 6951-79737-0018-132285-0 from training. Duration: 21.105
2022-12-22 14:38:26,870 WARNING [train.py:1060] (2/4) Exclude cut with ID 4511-76322-0006-80011-0_sp0.9 from training. Duration: 24.411125
2022-12-22 14:38:39,600 INFO [scaling.py:679] (2/4) Whitening: num_groups=8, num_channels=192, metric=2.03 vs. limit=2.0
2022-12-22 14:38:54,541 WARNING [train.py:1060] (2/4) Exclude cut with ID 6758-72288-0033-108368-0_sp1.1 from training. Duration: 21.263625
2022-12-22 14:39:07,889 INFO [zipformer.py:660] (2/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=10362.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 14:39:09,165 INFO [train.py:894] (2/4) Epoch 3, batch 3350, loss[loss=0.2943, simple_loss=0.3565, pruned_loss=0.116, over 18547.00 frames. ], tot_loss[loss=0.307, simple_loss=0.3555, pruned_loss=0.1292, over 3713765.28 frames. ], batch size: 55, lr: 3.07e-02, grad_scale: 8.0
2022-12-22 14:39:26,261 WARNING [train.py:1060] (2/4) Exclude cut with ID 4234-40345-0022-142709-0 from training. Duration: 20.795
2022-12-22 14:39:37,120 WARNING [train.py:1060] (2/4) Exclude cut with ID 7357-94126-0021-15852-0 from training. Duration: 24.76
2022-12-22 14:39:37,658 INFO [scaling.py:679] (2/4) Whitening: num_groups=1, num_channels=384, metric=2.89 vs. limit=5.0
2022-12-22 14:39:38,415 WARNING [train.py:1060] (2/4) Exclude cut with ID 3867-173237-0077-144769-0_sp0.9 from training. Duration: 22.25
2022-12-22 14:39:59,654 INFO [zipformer.py:660] (2/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=10395.0, num_to_drop=1, layers_to_drop={2}
2022-12-22 14:40:02,560 WARNING [train.py:1060] (2/4) Exclude cut with ID 7357-94126-0026-15857-0_sp1.1 from training. Duration: 20.5045625
2022-12-22 14:40:05,667 INFO [zipformer.py:660] (2/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=10399.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 14:40:21,231 INFO [zipformer.py:660] (2/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=10409.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 14:40:27,019 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 4.373e+02 7.368e+02 9.134e+02 1.174e+03 1.740e+03, threshold=1.827e+03, percent-clipped=1.0
2022-12-22 14:40:27,034 INFO [train.py:894] (2/4) Epoch 3, batch 3400, loss[loss=0.3083, simple_loss=0.3693, pruned_loss=0.1237, over 18693.00 frames. ], tot_loss[loss=0.3061, simple_loss=0.3553, pruned_loss=0.1284, over 3713942.22 frames. ], batch size: 62, lr: 3.07e-02, grad_scale: 8.0
2022-12-22 14:40:34,846 INFO [zipformer.py:660] (2/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=10418.0, num_to_drop=1, layers_to_drop={2}
2022-12-22 14:41:03,209 INFO [scaling.py:679] (2/4) Whitening: num_groups=1, num_channels=384, metric=3.90 vs. limit=5.0
2022-12-22 14:41:16,461 INFO [zipformer.py:660] (2/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=10447.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 14:41:26,345 INFO [zipformer.py:660] (2/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=10454.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 14:41:38,779 INFO [train.py:894] (2/4) Epoch 3, batch 3450, loss[loss=0.2377, simple_loss=0.2929, pruned_loss=0.09122, over 18548.00 frames. ], tot_loss[loss=0.3053, simple_loss=0.3551, pruned_loss=0.1278, over 3714215.06 frames. ], batch size: 41, lr: 3.06e-02, grad_scale: 8.0
2022-12-22 14:41:49,338 INFO [zipformer.py:660] (2/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=10470.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 14:42:01,706 INFO [scaling.py:679] (2/4) Whitening: num_groups=8, num_channels=192, metric=1.99 vs. limit=2.0
2022-12-22 14:42:07,116 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.5737, 1.9223, 1.5893, 2.5235, 2.6340, 1.3486, 1.7633, 1.1448],
       device='cuda:2'), covar=tensor([0.1860, 0.1830, 0.1351, 0.0814, 0.1186, 0.1336, 0.1537, 0.1700],
       device='cuda:2'), in_proj_covar=tensor([0.0191, 0.0160, 0.0157, 0.0144, 0.0193, 0.0148, 0.0159, 0.0154],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-22 14:42:12,718 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([2.1367, 1.8130, 1.1859, 0.4531, 1.4106, 1.7335, 1.2262, 1.7086],
       device='cuda:2'), covar=tensor([0.0720, 0.0564, 0.1514, 0.2132, 0.1129, 0.1548, 0.1775, 0.0877],
       device='cuda:2'), in_proj_covar=tensor([0.0124, 0.0158, 0.0199, 0.0192, 0.0180, 0.0162, 0.0172, 0.0172],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-22 14:42:22,422 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([2.6867, 2.4533, 1.7402, 0.7338, 1.9787, 2.1434, 1.4992, 2.0804],
       device='cuda:2'), covar=tensor([0.0730, 0.0578, 0.1674, 0.2390, 0.1378, 0.1330, 0.1669, 0.1060],
       device='cuda:2'), in_proj_covar=tensor([0.0123, 0.0157, 0.0197, 0.0190, 0.0178, 0.0161, 0.0172, 0.0171],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-22 14:42:51,940 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 3.608e+02 7.309e+02 8.677e+02 1.147e+03 1.668e+03, threshold=1.735e+03, percent-clipped=0.0
2022-12-22 14:42:51,962 INFO [train.py:894] (2/4) Epoch 3, batch 3500, loss[loss=0.3847, simple_loss=0.4071, pruned_loss=0.1812, over 18673.00 frames. ], tot_loss[loss=0.3068, simple_loss=0.3562, pruned_loss=0.1287, over 3714632.30 frames. ], batch size: 171, lr: 3.05e-02, grad_scale: 8.0
2022-12-22 14:42:55,770 INFO [zipformer.py:660] (2/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=10515.0, num_to_drop=1, layers_to_drop={2}
2022-12-22 14:43:14,510 WARNING [train.py:1060] (2/4) Exclude cut with ID 7859-102521-0017-7548-0_sp1.1 from training. Duration: 22.2954375
2022-12-22 14:43:26,380 INFO [train.py:894] (2/4) Epoch 4, batch 0, loss[loss=0.2633, simple_loss=0.3298, pruned_loss=0.09837, over 18458.00 frames. ], tot_loss[loss=0.2633, simple_loss=0.3298, pruned_loss=0.09837, over 18458.00 frames. ], batch size: 50, lr: 2.85e-02, grad_scale: 8.0
2022-12-22 14:43:26,380 INFO [train.py:919] (2/4) Computing validation loss
2022-12-22 14:43:37,536 INFO [train.py:928] (2/4) Epoch 4, validation: loss=0.2246, simple_loss=0.3191, pruned_loss=0.06499, over 944034.00 frames. 
2022-12-22 14:43:37,537 INFO [train.py:929] (2/4) Maximum memory allocated so far is 24257MB
2022-12-22 14:44:12,706 INFO [scaling.py:679] (2/4) Whitening: num_groups=8, num_channels=192, metric=1.91 vs. limit=2.0
2022-12-22 14:44:18,433 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([2.6905, 2.5505, 1.8589, 0.8872, 2.2858, 2.3152, 1.8631, 2.1350],
       device='cuda:2'), covar=tensor([0.0754, 0.0582, 0.1725, 0.2207, 0.1364, 0.1100, 0.1315, 0.1028],
       device='cuda:2'), in_proj_covar=tensor([0.0123, 0.0156, 0.0194, 0.0187, 0.0178, 0.0158, 0.0169, 0.0170],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-22 14:44:28,748 WARNING [train.py:1060] (2/4) Exclude cut with ID 298-126791-0067-24026-0_sp0.9 from training. Duration: 21.438875
2022-12-22 14:44:34,661 WARNING [train.py:1060] (2/4) Exclude cut with ID 5652-39938-0025-23684-0_sp0.9 from training. Duration: 22.2055625
2022-12-22 14:44:53,490 INFO [train.py:894] (2/4) Epoch 4, batch 50, loss[loss=0.2593, simple_loss=0.3285, pruned_loss=0.09503, over 18579.00 frames. ], tot_loss[loss=0.2766, simple_loss=0.3411, pruned_loss=0.1061, over 837192.58 frames. ], batch size: 57, lr: 2.85e-02, grad_scale: 8.0
2022-12-22 14:45:59,174 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 3.242e+02 5.273e+02 6.714e+02 8.694e+02 2.081e+03, threshold=1.343e+03, percent-clipped=1.0
2022-12-22 14:46:10,523 INFO [train.py:894] (2/4) Epoch 4, batch 100, loss[loss=0.2377, simple_loss=0.2985, pruned_loss=0.08844, over 18499.00 frames. ], tot_loss[loss=0.267, simple_loss=0.333, pruned_loss=0.1005, over 1474549.58 frames. ], batch size: 43, lr: 2.84e-02, grad_scale: 8.0
2022-12-22 14:47:00,829 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([3.7790, 2.9970, 2.3602, 1.1938, 2.3324, 2.7145, 2.0025, 1.9660],
       device='cuda:2'), covar=tensor([0.0553, 0.0711, 0.1758, 0.2516, 0.1915, 0.1151, 0.1611, 0.1535],
       device='cuda:2'), in_proj_covar=tensor([0.0121, 0.0157, 0.0194, 0.0184, 0.0176, 0.0158, 0.0169, 0.0169],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-22 14:47:25,415 INFO [train.py:894] (2/4) Epoch 4, batch 150, loss[loss=0.2634, simple_loss=0.3382, pruned_loss=0.09425, over 18569.00 frames. ], tot_loss[loss=0.2632, simple_loss=0.3314, pruned_loss=0.09754, over 1972059.02 frames. ], batch size: 57, lr: 2.84e-02, grad_scale: 8.0
2022-12-22 14:47:35,128 WARNING [train.py:1060] (2/4) Exclude cut with ID 7859-102521-0017-7548-0 from training. Duration: 24.525
2022-12-22 14:48:01,990 INFO [zipformer.py:660] (2/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=10695.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 14:48:08,925 WARNING [train.py:1060] (2/4) Exclude cut with ID 3699-47246-0007-3408-0_sp0.9 from training. Duration: 20.26675
2022-12-22 14:48:22,546 WARNING [train.py:1060] (2/4) Exclude cut with ID 7859-102521-0017-7548-0_sp0.9 from training. Duration: 27.25
2022-12-22 14:48:30,169 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 3.370e+02 5.528e+02 6.608e+02 7.818e+02 1.640e+03, threshold=1.322e+03, percent-clipped=2.0
2022-12-22 14:48:32,026 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.4207, 1.6460, 1.6142, 1.6197, 1.6392, 2.7308, 1.4536, 1.9387],
       device='cuda:2'), covar=tensor([0.4022, 0.2271, 0.2006, 0.2219, 0.1616, 0.0360, 0.1731, 0.1283],
       device='cuda:2'), in_proj_covar=tensor([0.0153, 0.0130, 0.0143, 0.0130, 0.0134, 0.0095, 0.0117, 0.0116],
       device='cuda:2'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0002, 0.0003, 0.0003],
       device='cuda:2')
2022-12-22 14:48:38,185 INFO [zipformer.py:660] (2/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=10718.0, num_to_drop=1, layers_to_drop={0}
2022-12-22 14:48:40,681 INFO [train.py:894] (2/4) Epoch 4, batch 200, loss[loss=0.2483, simple_loss=0.3029, pruned_loss=0.09684, over 18551.00 frames. ], tot_loss[loss=0.2606, simple_loss=0.329, pruned_loss=0.09605, over 2357242.86 frames. ], batch size: 44, lr: 2.83e-02, grad_scale: 8.0
2022-12-22 14:49:14,302 INFO [zipformer.py:660] (2/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=10743.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 14:49:16,050 INFO [zipformer.py:660] (2/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=10744.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 14:49:22,496 INFO [scaling.py:679] (2/4) Whitening: num_groups=8, num_channels=96, metric=1.51 vs. limit=2.0
2022-12-22 14:49:42,106 WARNING [train.py:1060] (2/4) Exclude cut with ID 6426-64292-0017-15984-0 from training. Duration: 21.68
2022-12-22 14:49:47,954 INFO [zipformer.py:660] (2/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=10765.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 14:49:49,373 INFO [zipformer.py:660] (2/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=10766.0, num_to_drop=1, layers_to_drop={0}
2022-12-22 14:49:52,830 WARNING [train.py:1060] (2/4) Exclude cut with ID 4278-13270-0007-59342-0 from training. Duration: 21.6300625
2022-12-22 14:49:55,755 INFO [train.py:894] (2/4) Epoch 4, batch 250, loss[loss=0.2208, simple_loss=0.2957, pruned_loss=0.07292, over 18548.00 frames. ], tot_loss[loss=0.2595, simple_loss=0.3288, pruned_loss=0.09509, over 2658223.19 frames. ], batch size: 47, lr: 2.83e-02, grad_scale: 8.0
2022-12-22 14:50:01,984 INFO [zipformer.py:660] (2/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=10774.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 14:50:15,110 WARNING [train.py:1060] (2/4) Exclude cut with ID 4278-13270-0007-59342-0_sp0.9 from training. Duration: 24.033375
2022-12-22 14:50:18,407 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.2682, 1.2024, 1.3179, 1.3420, 1.3523, 2.2168, 0.8611, 1.7830],
       device='cuda:2'), covar=tensor([0.5376, 0.3238, 0.2448, 0.2844, 0.1734, 0.0462, 0.1981, 0.1176],
       device='cuda:2'), in_proj_covar=tensor([0.0153, 0.0130, 0.0143, 0.0130, 0.0134, 0.0094, 0.0118, 0.0116],
       device='cuda:2'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0002, 0.0003, 0.0003],
       device='cuda:2')
2022-12-22 14:50:38,607 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.6740, 1.5835, 1.2235, 1.8856, 1.5857, 1.4837, 1.4305, 1.5956],
       device='cuda:2'), covar=tensor([0.1469, 0.1728, 0.1188, 0.1597, 0.1642, 0.0767, 0.1612, 0.0588],
       device='cuda:2'), in_proj_covar=tensor([0.0196, 0.0175, 0.0161, 0.0247, 0.0167, 0.0156, 0.0184, 0.0135],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0001],
       device='cuda:2')
2022-12-22 14:50:50,269 INFO [zipformer.py:660] (2/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=10805.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 14:50:57,728 INFO [zipformer.py:660] (2/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=10810.0, num_to_drop=1, layers_to_drop={2}
2022-12-22 14:51:02,533 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 3.039e+02 5.468e+02 6.455e+02 8.141e+02 3.038e+03, threshold=1.291e+03, percent-clipped=5.0
2022-12-22 14:51:10,925 WARNING [train.py:1060] (2/4) Exclude cut with ID 4278-13270-0009-59344-0 from training. Duration: 22.905
2022-12-22 14:51:12,324 INFO [train.py:894] (2/4) Epoch 4, batch 300, loss[loss=0.2856, simple_loss=0.3556, pruned_loss=0.1078, over 18715.00 frames. ], tot_loss[loss=0.2583, simple_loss=0.328, pruned_loss=0.09433, over 2891699.28 frames. ], batch size: 62, lr: 2.82e-02, grad_scale: 8.0
2022-12-22 14:51:12,344 WARNING [train.py:1060] (2/4) Exclude cut with ID 5622-44585-0006-90525-0_sp1.1 from training. Duration: 23.4318125
2022-12-22 14:51:33,954 INFO [zipformer.py:660] (2/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=10835.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 14:52:26,552 INFO [train.py:894] (2/4) Epoch 4, batch 350, loss[loss=0.3017, simple_loss=0.3677, pruned_loss=0.1179, over 18464.00 frames. ], tot_loss[loss=0.2616, simple_loss=0.3297, pruned_loss=0.0967, over 3073993.59 frames. ], batch size: 68, lr: 2.82e-02, grad_scale: 8.0
2022-12-22 14:52:37,291 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([2.4271, 1.5908, 2.0881, 2.2754, 1.9798, 1.7758, 2.0678, 1.2370],
       device='cuda:2'), covar=tensor([0.0812, 0.1682, 0.1223, 0.1185, 0.0640, 0.0557, 0.1280, 0.0725],
       device='cuda:2'), in_proj_covar=tensor([0.0154, 0.0186, 0.0171, 0.0175, 0.0150, 0.0145, 0.0163, 0.0145],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0001, 0.0002, 0.0001],
       device='cuda:2')
2022-12-22 14:52:48,894 INFO [zipformer.py:660] (2/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=10885.0, num_to_drop=1, layers_to_drop={0}
2022-12-22 14:53:06,402 WARNING [train.py:1060] (2/4) Exclude cut with ID 4278-13270-0009-59344-0_sp1.1 from training. Duration: 20.82275
2022-12-22 14:53:08,010 WARNING [train.py:1060] (2/4) Exclude cut with ID 4278-13270-0009-59344-0_sp0.9 from training. Duration: 25.45
2022-12-22 14:53:33,183 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 3.068e+02 5.924e+02 7.243e+02 8.972e+02 2.148e+03, threshold=1.449e+03, percent-clipped=9.0
2022-12-22 14:53:43,091 INFO [train.py:894] (2/4) Epoch 4, batch 400, loss[loss=0.2118, simple_loss=0.2853, pruned_loss=0.06913, over 18586.00 frames. ], tot_loss[loss=0.2642, simple_loss=0.332, pruned_loss=0.09814, over 3215404.56 frames. ], batch size: 45, lr: 2.81e-02, grad_scale: 8.0
2022-12-22 14:53:53,308 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([2.8588, 5.1658, 4.9413, 2.2009, 4.8225, 3.8086, 0.6317, 3.1548],
       device='cuda:2'), covar=tensor([0.1647, 0.0484, 0.1042, 0.3332, 0.0743, 0.0976, 0.6052, 0.1859],
       device='cuda:2'), in_proj_covar=tensor([0.0120, 0.0089, 0.0143, 0.0116, 0.0099, 0.0094, 0.0143, 0.0112],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0001, 0.0002, 0.0002, 0.0002, 0.0001, 0.0002, 0.0002],
       device='cuda:2')
2022-12-22 14:54:06,237 WARNING [train.py:1060] (2/4) Exclude cut with ID 5622-44585-0006-90525-0 from training. Duration: 25.775
2022-12-22 14:54:16,596 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.4775, 2.0805, 2.2060, 2.2110, 2.3846, 4.7640, 2.2845, 3.0879],
       device='cuda:2'), covar=tensor([0.4032, 0.2126, 0.1695, 0.1940, 0.1413, 0.0093, 0.1496, 0.1028],
       device='cuda:2'), in_proj_covar=tensor([0.0152, 0.0129, 0.0141, 0.0131, 0.0131, 0.0095, 0.0116, 0.0116],
       device='cuda:2'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0002, 0.0003, 0.0003],
       device='cuda:2')
2022-12-22 14:54:22,641 INFO [zipformer.py:660] (2/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=10946.0, num_to_drop=1, layers_to_drop={0}
2022-12-22 14:54:27,107 WARNING [train.py:1060] (2/4) Exclude cut with ID 2929-85685-0071-62375-0_sp0.9 from training. Duration: 22.25
2022-12-22 14:54:54,961 WARNING [train.py:1060] (2/4) Exclude cut with ID 3972-170212-0014-23379-0 from training. Duration: 26.205
2022-12-22 14:54:57,936 INFO [train.py:894] (2/4) Epoch 4, batch 450, loss[loss=0.2813, simple_loss=0.3481, pruned_loss=0.1073, over 18644.00 frames. ], tot_loss[loss=0.2674, simple_loss=0.3349, pruned_loss=0.09993, over 3325779.54 frames. ], batch size: 53, lr: 2.81e-02, grad_scale: 8.0
2022-12-22 14:55:12,645 WARNING [train.py:1060] (2/4) Exclude cut with ID 5239-32139-0047-9341-0_sp0.9 from training. Duration: 30.1555625
2022-12-22 14:55:18,427 WARNING [train.py:1060] (2/4) Exclude cut with ID 1265-135635-0050-6781-0_sp0.9 from training. Duration: 21.8333125
2022-12-22 14:55:25,446 WARNING [train.py:1060] (2/4) Exclude cut with ID 1914-133440-0024-94914-0_sp1.1 from training. Duration: 20.6545625
2022-12-22 14:56:03,704 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 3.947e+02 6.290e+02 7.836e+02 1.078e+03 2.986e+03, threshold=1.567e+03, percent-clipped=7.0
2022-12-22 14:56:08,091 WARNING [train.py:1060] (2/4) Exclude cut with ID 7395-89880-0045-39920-0_sp0.9 from training. Duration: 20.52225
2022-12-22 14:56:13,914 INFO [train.py:894] (2/4) Epoch 4, batch 500, loss[loss=0.2468, simple_loss=0.3086, pruned_loss=0.09254, over 18442.00 frames. ], tot_loss[loss=0.2693, simple_loss=0.3366, pruned_loss=0.101, over 3412027.18 frames. ], batch size: 42, lr: 2.80e-02, grad_scale: 8.0
2022-12-22 14:56:14,147 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([2.6944, 5.0978, 4.7571, 2.2721, 5.1925, 3.7405, 1.2722, 3.3741],
       device='cuda:2'), covar=tensor([0.1650, 0.0381, 0.0995, 0.2900, 0.0561, 0.0993, 0.4455, 0.1498],
       device='cuda:2'), in_proj_covar=tensor([0.0119, 0.0088, 0.0144, 0.0113, 0.0098, 0.0096, 0.0141, 0.0110],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0001, 0.0002, 0.0002, 0.0002, 0.0001, 0.0002, 0.0002],
       device='cuda:2')
2022-12-22 14:56:29,931 WARNING [train.py:1060] (2/4) Exclude cut with ID 3972-170212-0014-23379-0_sp0.9 from training. Duration: 29.1166875
2022-12-22 14:57:21,949 INFO [zipformer.py:660] (2/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=11065.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 14:57:28,741 INFO [train.py:894] (2/4) Epoch 4, batch 550, loss[loss=0.2636, simple_loss=0.3337, pruned_loss=0.09678, over 18522.00 frames. ], tot_loss[loss=0.2695, simple_loss=0.3372, pruned_loss=0.1009, over 3479472.90 frames. ], batch size: 58, lr: 2.80e-02, grad_scale: 8.0
2022-12-22 14:57:32,926 WARNING [train.py:1060] (2/4) Exclude cut with ID 543-133211-0007-59831-0_sp0.9 from training. Duration: 21.388875
2022-12-22 14:58:08,781 WARNING [train.py:1060] (2/4) Exclude cut with ID 1914-133440-0024-94914-0 from training. Duration: 22.72
2022-12-22 14:58:10,293 WARNING [train.py:1060] (2/4) Exclude cut with ID 1914-133440-0031-94921-0_sp0.9 from training. Duration: 22.7444375
2022-12-22 14:58:13,751 INFO [zipformer.py:660] (2/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=11100.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 14:58:27,401 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.3947, 1.4006, 1.3655, 1.3460, 1.3101, 1.9505, 1.0697, 1.7314],
       device='cuda:2'), covar=tensor([0.3883, 0.2357, 0.2110, 0.2518, 0.1586, 0.0493, 0.1684, 0.1108],
       device='cuda:2'), in_proj_covar=tensor([0.0154, 0.0129, 0.0141, 0.0132, 0.0130, 0.0093, 0.0118, 0.0116],
       device='cuda:2'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0002, 0.0003, 0.0003],
       device='cuda:2')
2022-12-22 14:58:28,861 INFO [zipformer.py:660] (2/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=11110.0, num_to_drop=1, layers_to_drop={0}
2022-12-22 14:58:33,135 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 2.935e+02 5.294e+02 6.541e+02 8.578e+02 1.518e+03, threshold=1.308e+03, percent-clipped=0.0
2022-12-22 14:58:33,309 INFO [zipformer.py:660] (2/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=11113.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 14:58:43,491 INFO [train.py:894] (2/4) Epoch 4, batch 600, loss[loss=0.2625, simple_loss=0.3397, pruned_loss=0.09267, over 18644.00 frames. ], tot_loss[loss=0.2711, simple_loss=0.3388, pruned_loss=0.1016, over 3532686.09 frames. ], batch size: 53, lr: 2.79e-02, grad_scale: 8.0
2022-12-22 14:58:52,568 WARNING [train.py:1060] (2/4) Exclude cut with ID 4133-6541-0027-40495-0_sp1.1 from training. Duration: 0.9681875
2022-12-22 14:58:55,456 WARNING [train.py:1060] (2/4) Exclude cut with ID 6330-62851-0022-91297-0_sp0.9 from training. Duration: 22.3166875
2022-12-22 14:58:58,531 INFO [zipformer.py:660] (2/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=11130.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 14:59:01,130 WARNING [train.py:1060] (2/4) Exclude cut with ID 543-133212-0015-59917-0_sp0.9 from training. Duration: 21.8166875
2022-12-22 14:59:42,951 INFO [zipformer.py:660] (2/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=11158.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 14:59:55,608 INFO [scaling.py:679] (2/4) Whitening: num_groups=1, num_channels=384, metric=4.71 vs. limit=5.0
2022-12-22 15:00:00,322 INFO [train.py:894] (2/4) Epoch 4, batch 650, loss[loss=0.2889, simple_loss=0.3554, pruned_loss=0.1112, over 18685.00 frames. ], tot_loss[loss=0.2703, simple_loss=0.338, pruned_loss=0.1013, over 3572967.05 frames. ], batch size: 60, lr: 2.78e-02, grad_scale: 8.0
2022-12-22 15:00:06,665 INFO [zipformer.py:660] (2/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=11174.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 15:00:17,033 INFO [scaling.py:679] (2/4) Whitening: num_groups=8, num_channels=96, metric=1.54 vs. limit=2.0
2022-12-22 15:00:44,715 WARNING [train.py:1060] (2/4) Exclude cut with ID 4957-30119-0041-23990-0_sp0.9 from training. Duration: 20.22775
2022-12-22 15:01:06,121 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 3.895e+02 6.331e+02 7.829e+02 9.597e+02 3.856e+03, threshold=1.566e+03, percent-clipped=9.0
2022-12-22 15:01:16,830 INFO [train.py:894] (2/4) Epoch 4, batch 700, loss[loss=0.2572, simple_loss=0.3314, pruned_loss=0.09149, over 18393.00 frames. ], tot_loss[loss=0.2693, simple_loss=0.3369, pruned_loss=0.1008, over 3604556.75 frames. ], batch size: 53, lr: 2.78e-02, grad_scale: 8.0
2022-12-22 15:01:27,145 WARNING [train.py:1060] (2/4) Exclude cut with ID 5239-32139-0047-9341-0_sp1.1 from training. Duration: 24.67275
2022-12-22 15:01:41,164 INFO [zipformer.py:660] (2/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=11235.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 15:01:50,548 INFO [zipformer.py:660] (2/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=11241.0, num_to_drop=1, layers_to_drop={0}
2022-12-22 15:01:59,310 WARNING [train.py:1060] (2/4) Exclude cut with ID 3082-165428-0081-50734-0_sp0.9 from training. Duration: 21.8055625
2022-12-22 15:02:05,321 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.5078, 1.2257, 0.7365, 1.5127, 1.5058, 3.0158, 1.2953, 1.4057],
       device='cuda:2'), covar=tensor([0.1500, 0.2966, 0.2286, 0.1709, 0.2149, 0.0411, 0.2287, 0.2843],
       device='cuda:2'), in_proj_covar=tensor([0.0087, 0.0092, 0.0092, 0.0090, 0.0110, 0.0080, 0.0098, 0.0091],
       device='cuda:2'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0002, 0.0003, 0.0003],
       device='cuda:2')
2022-12-22 15:02:33,317 INFO [train.py:894] (2/4) Epoch 4, batch 750, loss[loss=0.2214, simple_loss=0.2988, pruned_loss=0.07194, over 18437.00 frames. ], tot_loss[loss=0.2686, simple_loss=0.3362, pruned_loss=0.1005, over 3627936.59 frames. ], batch size: 48, lr: 2.77e-02, grad_scale: 8.0
2022-12-22 15:02:36,269 WARNING [train.py:1060] (2/4) Exclude cut with ID 3340-169293-0054-76830-0_sp0.9 from training. Duration: 22.6666875
2022-12-22 15:03:35,345 INFO [scaling.py:679] (2/4) Whitening: num_groups=8, num_channels=96, metric=1.59 vs. limit=2.0
2022-12-22 15:03:38,711 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 3.263e+02 5.618e+02 6.637e+02 8.413e+02 1.667e+03, threshold=1.327e+03, percent-clipped=1.0
2022-12-22 15:03:40,050 WARNING [train.py:1060] (2/4) Exclude cut with ID 2411-132532-0017-82279-0_sp1.1 from training. Duration: 0.9681875
2022-12-22 15:03:48,445 INFO [train.py:894] (2/4) Epoch 4, batch 800, loss[loss=0.2863, simple_loss=0.3579, pruned_loss=0.1073, over 18693.00 frames. ], tot_loss[loss=0.2678, simple_loss=0.3357, pruned_loss=0.09993, over 3647986.80 frames. ], batch size: 60, lr: 2.77e-02, grad_scale: 8.0
2022-12-22 15:04:07,182 WARNING [train.py:1060] (2/4) Exclude cut with ID 6330-62850-0007-91323-0 from training. Duration: 22.485
2022-12-22 15:04:46,180 WARNING [train.py:1060] (2/4) Exclude cut with ID 3972-170212-0014-23379-0_sp1.1 from training. Duration: 23.82275
2022-12-22 15:04:55,530 INFO [scaling.py:679] (2/4) Whitening: num_groups=1, num_channels=384, metric=2.96 vs. limit=5.0
2022-12-22 15:04:59,241 WARNING [train.py:1060] (2/4) Exclude cut with ID 4860-13185-0032-76709-0 from training. Duration: 20.77
2022-12-22 15:05:05,213 INFO [train.py:894] (2/4) Epoch 4, batch 850, loss[loss=0.2862, simple_loss=0.3558, pruned_loss=0.1083, over 18713.00 frames. ], tot_loss[loss=0.2675, simple_loss=0.3351, pruned_loss=0.09997, over 3661957.37 frames. ], batch size: 78, lr: 2.76e-02, grad_scale: 8.0
2022-12-22 15:05:06,700 WARNING [train.py:1060] (2/4) Exclude cut with ID 6426-64292-0017-15984-0_sp0.9 from training. Duration: 24.088875
2022-12-22 15:05:25,520 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.4712, 1.2095, 0.7394, 1.6914, 1.2172, 2.8998, 1.4088, 1.2162],
       device='cuda:2'), covar=tensor([0.1307, 0.2244, 0.1936, 0.1209, 0.2024, 0.0371, 0.1577, 0.2054],
       device='cuda:2'), in_proj_covar=tensor([0.0088, 0.0093, 0.0093, 0.0089, 0.0110, 0.0080, 0.0099, 0.0091],
       device='cuda:2'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0002, 0.0003, 0.0003],
       device='cuda:2')
2022-12-22 15:05:36,636 WARNING [train.py:1060] (2/4) Exclude cut with ID 6330-62850-0007-91323-0_sp1.1 from training. Duration: 20.4409375
2022-12-22 15:05:42,049 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.8729, 2.1351, 2.0578, 2.2940, 2.6261, 4.9014, 2.6835, 3.3313],
       device='cuda:2'), covar=tensor([0.3920, 0.2017, 0.1860, 0.1981, 0.1371, 0.0086, 0.1357, 0.0944],
       device='cuda:2'), in_proj_covar=tensor([0.0155, 0.0131, 0.0143, 0.0131, 0.0129, 0.0095, 0.0118, 0.0115],
       device='cuda:2'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0002, 0.0003, 0.0003],
       device='cuda:2')
2022-12-22 15:05:52,204 INFO [zipformer.py:660] (2/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=11400.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 15:06:10,635 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 3.650e+02 5.684e+02 6.986e+02 8.807e+02 1.585e+03, threshold=1.397e+03, percent-clipped=2.0
2022-12-22 15:06:21,136 INFO [train.py:894] (2/4) Epoch 4, batch 900, loss[loss=0.288, simple_loss=0.3437, pruned_loss=0.1162, over 18579.00 frames. ], tot_loss[loss=0.2666, simple_loss=0.3343, pruned_loss=0.0995, over 3673304.63 frames. ], batch size: 49, lr: 2.76e-02, grad_scale: 8.0
2022-12-22 15:06:37,057 INFO [zipformer.py:660] (2/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=11430.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 15:06:54,332 WARNING [train.py:1060] (2/4) Exclude cut with ID 2929-85685-0062-62366-0_sp0.9 from training. Duration: 22.511125
2022-12-22 15:06:54,352 WARNING [train.py:1060] (2/4) Exclude cut with ID 7395-89880-0031-39906-0 from training. Duration: 20.675
2022-12-22 15:07:05,476 INFO [zipformer.py:660] (2/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=11448.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 15:07:37,404 INFO [train.py:894] (2/4) Epoch 4, batch 950, loss[loss=0.2519, simple_loss=0.3269, pruned_loss=0.08845, over 18553.00 frames. ], tot_loss[loss=0.2663, simple_loss=0.3342, pruned_loss=0.09926, over 3681886.80 frames. ], batch size: 49, lr: 2.75e-02, grad_scale: 8.0
2022-12-22 15:07:50,376 INFO [zipformer.py:660] (2/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=11478.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 15:08:15,650 INFO [scaling.py:679] (2/4) Whitening: num_groups=1, num_channels=384, metric=4.61 vs. limit=5.0
2022-12-22 15:08:34,230 WARNING [train.py:1060] (2/4) Exclude cut with ID 6330-62850-0007-91323-0_sp0.9 from training. Duration: 24.9833125
2022-12-22 15:08:34,617 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([2.5048, 2.2704, 1.7247, 0.6693, 1.5352, 1.9238, 1.5075, 1.7683],
       device='cuda:2'), covar=tensor([0.0611, 0.0428, 0.1320, 0.1910, 0.1356, 0.1433, 0.1681, 0.0921],
       device='cuda:2'), in_proj_covar=tensor([0.0124, 0.0156, 0.0191, 0.0184, 0.0178, 0.0155, 0.0171, 0.0164],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-22 15:08:42,485 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 3.740e+02 6.104e+02 6.909e+02 8.530e+02 1.884e+03, threshold=1.382e+03, percent-clipped=3.0
2022-12-22 15:08:52,606 INFO [train.py:894] (2/4) Epoch 4, batch 1000, loss[loss=0.2773, simple_loss=0.3504, pruned_loss=0.1021, over 18507.00 frames. ], tot_loss[loss=0.2654, simple_loss=0.3339, pruned_loss=0.09846, over 3688688.38 frames. ], batch size: 77, lr: 2.75e-02, grad_scale: 8.0
2022-12-22 15:09:05,599 WARNING [train.py:1060] (2/4) Exclude cut with ID 5239-32139-0047-9341-0 from training. Duration: 27.14
2022-12-22 15:09:08,548 INFO [zipformer.py:660] (2/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=11530.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 15:09:11,564 INFO [zipformer.py:660] (2/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=11532.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 15:09:20,961 WARNING [train.py:1060] (2/4) Exclude cut with ID 2929-85685-0044-62348-0 from training. Duration: 22.44
2022-12-22 15:09:25,342 INFO [zipformer.py:660] (2/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=11541.0, num_to_drop=1, layers_to_drop={1}
2022-12-22 15:10:07,534 INFO [train.py:894] (2/4) Epoch 4, batch 1050, loss[loss=0.2578, simple_loss=0.3116, pruned_loss=0.102, over 18487.00 frames. ], tot_loss[loss=0.2648, simple_loss=0.3335, pruned_loss=0.09807, over 3694861.84 frames. ], batch size: 43, lr: 2.74e-02, grad_scale: 8.0
2022-12-22 15:10:37,230 INFO [zipformer.py:660] (2/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=11589.0, num_to_drop=1, layers_to_drop={0}
2022-12-22 15:10:40,271 WARNING [train.py:1060] (2/4) Exclude cut with ID 2929-85685-0060-62364-0_sp0.9 from training. Duration: 21.361125
2022-12-22 15:10:43,418 INFO [zipformer.py:660] (2/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=11593.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 15:10:45,653 WARNING [train.py:1060] (2/4) Exclude cut with ID 2929-85685-0079-62383-0_sp1.1 from training. Duration: 27.0318125
2022-12-22 15:10:57,866 WARNING [train.py:1060] (2/4) Exclude cut with ID 5622-44585-0006-90525-0_sp0.9 from training. Duration: 28.638875
2022-12-22 15:11:12,460 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 4.335e+02 6.019e+02 7.505e+02 8.612e+02 1.443e+03, threshold=1.501e+03, percent-clipped=1.0
2022-12-22 15:11:13,956 WARNING [train.py:1060] (2/4) Exclude cut with ID 3340-169293-0054-76830-0 from training. Duration: 20.4
2022-12-22 15:11:22,525 INFO [train.py:894] (2/4) Epoch 4, batch 1100, loss[loss=0.184, simple_loss=0.2606, pruned_loss=0.05367, over 18526.00 frames. ], tot_loss[loss=0.2631, simple_loss=0.3318, pruned_loss=0.09717, over 3698117.30 frames. ], batch size: 47, lr: 2.74e-02, grad_scale: 8.0
2022-12-22 15:11:39,387 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.1614, 0.4788, 0.9045, 0.9086, 1.1102, 1.1316, 1.1089, 0.9078],
       device='cuda:2'), covar=tensor([0.0461, 0.0713, 0.0933, 0.0528, 0.0387, 0.0459, 0.0405, 0.0587],
       device='cuda:2'), in_proj_covar=tensor([0.0069, 0.0102, 0.0117, 0.0119, 0.0097, 0.0077, 0.0071, 0.0107],
       device='cuda:2'), out_proj_covar=tensor([7.6741e-05, 1.0370e-04, 1.2396e-04, 1.2207e-04, 1.0457e-04, 8.1536e-05,
        7.6578e-05, 1.1012e-04], device='cuda:2')
2022-12-22 15:11:46,716 WARNING [train.py:1060] (2/4) Exclude cut with ID 2929-85685-0071-62375-0 from training. Duration: 20.025
2022-12-22 15:11:46,729 WARNING [train.py:1060] (2/4) Exclude cut with ID 2364-131735-0112-64612-0_sp0.9 from training. Duration: 20.488875
2022-12-22 15:11:47,093 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.5358, 1.6259, 1.5567, 1.6464, 1.4122, 2.8313, 1.2096, 2.1000],
       device='cuda:2'), covar=tensor([0.3670, 0.2173, 0.1974, 0.2054, 0.1657, 0.0310, 0.1866, 0.1145],
       device='cuda:2'), in_proj_covar=tensor([0.0153, 0.0132, 0.0143, 0.0131, 0.0129, 0.0095, 0.0120, 0.0117],
       device='cuda:2'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0002, 0.0003, 0.0003],
       device='cuda:2')
2022-12-22 15:11:47,113 INFO [zipformer.py:660] (2/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=11635.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 15:11:53,192 WARNING [train.py:1060] (2/4) Exclude cut with ID 2929-85685-0079-62383-0 from training. Duration: 29.735
2022-12-22 15:12:38,692 INFO [train.py:894] (2/4) Epoch 4, batch 1150, loss[loss=0.2382, simple_loss=0.3198, pruned_loss=0.07829, over 18581.00 frames. ], tot_loss[loss=0.2625, simple_loss=0.3315, pruned_loss=0.09677, over 3701258.22 frames. ], batch size: 51, lr: 2.73e-02, grad_scale: 8.0
2022-12-22 15:13:15,997 WARNING [train.py:1060] (2/4) Exclude cut with ID 7276-92427-0014-12983-0_sp0.9 from training. Duration: 21.3055625
2022-12-22 15:13:17,644 WARNING [train.py:1060] (2/4) Exclude cut with ID 1025-75365-0008-79168-0_sp0.9 from training. Duration: 22.0666875
2022-12-22 15:13:19,355 INFO [zipformer.py:660] (2/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=11696.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 15:13:21,315 INFO [scaling.py:679] (2/4) Whitening: num_groups=8, num_channels=192, metric=2.03 vs. limit=2.0
2022-12-22 15:13:23,686 INFO [zipformer.py:660] (2/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=11699.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 15:13:43,177 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 3.231e+02 5.643e+02 6.547e+02 8.776e+02 2.408e+03, threshold=1.309e+03, percent-clipped=2.0
2022-12-22 15:13:53,669 INFO [train.py:894] (2/4) Epoch 4, batch 1200, loss[loss=0.3419, simple_loss=0.3805, pruned_loss=0.1516, over 18574.00 frames. ], tot_loss[loss=0.2632, simple_loss=0.3319, pruned_loss=0.09719, over 3704859.15 frames. ], batch size: 57, lr: 2.73e-02, grad_scale: 8.0
2022-12-22 15:14:08,239 INFO [scaling.py:679] (2/4) Whitening: num_groups=8, num_channels=96, metric=1.29 vs. limit=2.0
2022-12-22 15:14:28,895 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([2.7587, 3.8147, 4.0381, 2.0380, 3.7740, 2.9358, 0.9128, 2.6552],
       device='cuda:2'), covar=tensor([0.1712, 0.0656, 0.1103, 0.2922, 0.0793, 0.1087, 0.5068, 0.1719],
       device='cuda:2'), in_proj_covar=tensor([0.0121, 0.0089, 0.0142, 0.0111, 0.0097, 0.0095, 0.0140, 0.0110],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0001, 0.0002, 0.0002, 0.0002, 0.0001, 0.0002, 0.0002],
       device='cuda:2')
2022-12-22 15:14:29,011 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.6155, 0.8416, 1.8679, 2.6881, 1.8310, 2.1073, 0.5695, 1.6453],
       device='cuda:2'), covar=tensor([0.2139, 0.2321, 0.1748, 0.0590, 0.1606, 0.1562, 0.3119, 0.1575],
       device='cuda:2'), in_proj_covar=tensor([0.0103, 0.0114, 0.0126, 0.0086, 0.0102, 0.0121, 0.0139, 0.0102],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0002, 0.0003, 0.0002, 0.0002, 0.0003, 0.0003, 0.0002],
       device='cuda:2')
2022-12-22 15:14:46,560 INFO [scaling.py:679] (2/4) Whitening: num_groups=8, num_channels=96, metric=1.33 vs. limit=2.0
2022-12-22 15:14:55,597 INFO [zipformer.py:660] (2/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=11760.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 15:14:58,464 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.6729, 1.6060, 1.0745, 2.0412, 1.5902, 3.2168, 1.8070, 1.3419],
       device='cuda:2'), covar=tensor([0.1103, 0.1726, 0.1739, 0.1107, 0.1706, 0.0316, 0.1261, 0.1850],
       device='cuda:2'), in_proj_covar=tensor([0.0087, 0.0092, 0.0092, 0.0090, 0.0108, 0.0078, 0.0098, 0.0089],
       device='cuda:2'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0002, 0.0003, 0.0003],
       device='cuda:2')
2022-12-22 15:15:08,695 WARNING [train.py:1060] (2/4) Exclude cut with ID 2929-85685-0062-62366-0 from training. Duration: 20.26
2022-12-22 15:15:10,219 INFO [train.py:894] (2/4) Epoch 4, batch 1250, loss[loss=0.2838, simple_loss=0.3452, pruned_loss=0.1112, over 18469.00 frames. ], tot_loss[loss=0.2632, simple_loss=0.3323, pruned_loss=0.09704, over 3706976.93 frames. ], batch size: 50, lr: 2.72e-02, grad_scale: 8.0
2022-12-22 15:15:21,044 WARNING [train.py:1060] (2/4) Exclude cut with ID 5239-32139-0030-9324-0_sp0.9 from training. Duration: 21.3444375
2022-12-22 15:16:16,647 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 3.456e+02 5.375e+02 6.633e+02 8.647e+02 2.282e+03, threshold=1.327e+03, percent-clipped=7.0
2022-12-22 15:16:18,018 WARNING [train.py:1060] (2/4) Exclude cut with ID 497-129325-0061-62254-0_sp1.1 from training. Duration: 0.97725
2022-12-22 15:16:21,560 INFO [zipformer.py:660] (2/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=11816.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 15:16:27,456 INFO [train.py:894] (2/4) Epoch 4, batch 1300, loss[loss=0.2604, simple_loss=0.3319, pruned_loss=0.09445, over 18699.00 frames. ], tot_loss[loss=0.2639, simple_loss=0.3329, pruned_loss=0.09751, over 3709056.16 frames. ], batch size: 50, lr: 2.72e-02, grad_scale: 8.0
2022-12-22 15:16:42,968 INFO [zipformer.py:660] (2/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=11830.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 15:16:58,132 WARNING [train.py:1060] (2/4) Exclude cut with ID 7395-89880-0031-39906-0_sp0.9 from training. Duration: 22.97225
2022-12-22 15:17:07,587 INFO [zipformer.py:660] (2/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=11846.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 15:17:30,556 WARNING [train.py:1060] (2/4) Exclude cut with ID 7395-89880-0047-39922-0_sp0.9 from training. Duration: 21.97775
2022-12-22 15:17:43,187 INFO [scaling.py:679] (2/4) Whitening: num_groups=8, num_channels=192, metric=1.93 vs. limit=2.0
2022-12-22 15:17:43,823 INFO [train.py:894] (2/4) Epoch 4, batch 1350, loss[loss=0.2408, simple_loss=0.3063, pruned_loss=0.08771, over 18686.00 frames. ], tot_loss[loss=0.2637, simple_loss=0.3325, pruned_loss=0.09745, over 3709494.86 frames. ], batch size: 46, lr: 2.71e-02, grad_scale: 8.0
2022-12-22 15:17:43,859 WARNING [train.py:1060] (2/4) Exclude cut with ID 1112-1043-0006-89194-0_sp0.9 from training. Duration: 21.8333125
2022-12-22 15:17:55,072 WARNING [train.py:1060] (2/4) Exclude cut with ID 1914-133440-0031-94921-0 from training. Duration: 20.47
2022-12-22 15:17:55,385 INFO [zipformer.py:660] (2/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=11877.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 15:17:56,631 INFO [zipformer.py:660] (2/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=11878.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 15:18:11,799 INFO [zipformer.py:660] (2/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=11888.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 15:18:39,499 INFO [zipformer.py:660] (2/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=11907.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 15:18:48,466 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 3.033e+02 5.388e+02 6.547e+02 8.025e+02 2.001e+03, threshold=1.309e+03, percent-clipped=2.0
2022-12-22 15:18:50,351 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([3.7225, 3.0679, 2.9636, 3.4367, 3.2056, 3.2283, 3.7935, 1.0228],
       device='cuda:2'), covar=tensor([0.0594, 0.0680, 0.0708, 0.0428, 0.1347, 0.0840, 0.0459, 0.4187],
       device='cuda:2'), in_proj_covar=tensor([0.0189, 0.0152, 0.0139, 0.0122, 0.0206, 0.0159, 0.0153, 0.0201],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0001, 0.0001, 0.0001, 0.0002, 0.0001, 0.0001, 0.0002],
       device='cuda:2')
2022-12-22 15:18:59,685 INFO [train.py:894] (2/4) Epoch 4, batch 1400, loss[loss=0.244, simple_loss=0.3042, pruned_loss=0.09189, over 18408.00 frames. ], tot_loss[loss=0.2621, simple_loss=0.331, pruned_loss=0.09663, over 3709827.54 frames. ], batch size: 42, lr: 2.71e-02, grad_scale: 8.0
2022-12-22 15:19:00,702 INFO [zipformer.py:660] (2/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=11920.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 15:19:01,851 WARNING [train.py:1060] (2/4) Exclude cut with ID 7395-89880-0037-39912-0_sp0.9 from training. Duration: 20.67225
2022-12-22 15:19:21,518 WARNING [train.py:1060] (2/4) Exclude cut with ID 1914-133440-0024-94914-0_sp0.9 from training. Duration: 25.2444375
2022-12-22 15:19:42,332 INFO [zipformer.py:660] (2/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=11948.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 15:19:43,537 WARNING [train.py:1060] (2/4) Exclude cut with ID 3340-169293-0021-76797-0_sp0.9 from training. Duration: 21.1445
2022-12-22 15:19:54,568 INFO [scaling.py:679] (2/4) Whitening: num_groups=8, num_channels=96, metric=1.30 vs. limit=2.0
2022-12-22 15:20:01,739 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.4913, 1.5848, 1.4044, 1.4852, 1.4369, 2.8170, 1.3814, 1.9391],
       device='cuda:2'), covar=tensor([0.3632, 0.2204, 0.1999, 0.2159, 0.1473, 0.0292, 0.1616, 0.1089],
       device='cuda:2'), in_proj_covar=tensor([0.0150, 0.0129, 0.0141, 0.0128, 0.0125, 0.0094, 0.0115, 0.0114],
       device='cuda:2'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0002, 0.0003, 0.0003],
       device='cuda:2')
2022-12-22 15:20:15,658 INFO [train.py:894] (2/4) Epoch 4, batch 1450, loss[loss=0.2607, simple_loss=0.3327, pruned_loss=0.09432, over 18629.00 frames. ], tot_loss[loss=0.2628, simple_loss=0.3315, pruned_loss=0.09703, over 3710092.83 frames. ], batch size: 53, lr: 2.70e-02, grad_scale: 8.0
2022-12-22 15:20:32,915 INFO [zipformer.py:660] (2/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=11981.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 15:20:37,211 INFO [zipformer.py:660] (2/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=11984.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 15:20:47,005 INFO [zipformer.py:660] (2/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=11991.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 15:20:48,719 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.7363, 1.0283, 1.5817, 1.1751, 1.7611, 1.9702, 2.0339, 1.2368],
       device='cuda:2'), covar=tensor([0.0545, 0.0551, 0.0667, 0.0544, 0.0360, 0.0330, 0.0288, 0.0576],
       device='cuda:2'), in_proj_covar=tensor([0.0069, 0.0101, 0.0118, 0.0115, 0.0097, 0.0077, 0.0071, 0.0108],
       device='cuda:2'), out_proj_covar=tensor([7.5919e-05, 1.0189e-04, 1.2384e-04, 1.1747e-04, 1.0442e-04, 7.9379e-05,
        7.5268e-05, 1.1045e-04], device='cuda:2')
2022-12-22 15:20:58,201 WARNING [train.py:1060] (2/4) Exclude cut with ID 2929-85685-0079-62383-0_sp0.9 from training. Duration: 33.038875
2022-12-22 15:21:16,121 INFO [scaling.py:679] (2/4) Whitening: num_groups=8, num_channels=192, metric=2.08 vs. limit=2.0
2022-12-22 15:21:17,759 INFO [zipformer.py:660] (2/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=12009.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 15:21:24,162 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 3.421e+02 5.689e+02 8.122e+02 1.002e+03 1.607e+03, threshold=1.624e+03, percent-clipped=6.0
2022-12-22 15:21:35,445 INFO [train.py:894] (2/4) Epoch 4, batch 1500, loss[loss=0.2517, simple_loss=0.3258, pruned_loss=0.0888, over 18424.00 frames. ], tot_loss[loss=0.2612, simple_loss=0.3303, pruned_loss=0.09602, over 3710703.52 frames. ], batch size: 48, lr: 2.70e-02, grad_scale: 8.0
2022-12-22 15:21:41,170 WARNING [train.py:1060] (2/4) Exclude cut with ID 6426-64291-0000-16059-0_sp0.9 from training. Duration: 20.0944375
2022-12-22 15:21:56,434 WARNING [train.py:1060] (2/4) Exclude cut with ID 2929-85685-0044-62348-0_sp1.1 from training. Duration: 20.4
2022-12-22 15:22:03,704 WARNING [train.py:1060] (2/4) Exclude cut with ID 6330-62851-0022-91297-0 from training. Duration: 20.085
2022-12-22 15:22:12,762 INFO [zipformer.py:660] (2/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=12045.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 15:22:13,800 WARNING [train.py:1060] (2/4) Exclude cut with ID 4860-13185-0032-76709-0_sp0.9 from training. Duration: 23.07775
2022-12-22 15:22:27,309 INFO [zipformer.py:660] (2/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=12055.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 15:22:50,192 INFO [train.py:894] (2/4) Epoch 4, batch 1550, loss[loss=0.3413, simple_loss=0.3872, pruned_loss=0.1477, over 18577.00 frames. ], tot_loss[loss=0.2604, simple_loss=0.3299, pruned_loss=0.09546, over 3711756.96 frames. ], batch size: 56, lr: 2.70e-02, grad_scale: 8.0
2022-12-22 15:23:00,300 INFO [zipformer.py:660] (2/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=12076.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 15:23:01,332 WARNING [train.py:1060] (2/4) Exclude cut with ID 2929-85685-0044-62348-0_sp0.9 from training. Duration: 24.9333125
2022-12-22 15:23:30,054 INFO [scaling.py:679] (2/4) Whitening: num_groups=8, num_channels=96, metric=1.26 vs. limit=2.0
2022-12-22 15:23:44,140 WARNING [train.py:1060] (2/4) Exclude cut with ID 5118-111612-0016-124680-0_sp0.9 from training. Duration: 20.388875
2022-12-22 15:23:52,585 WARNING [train.py:1060] (2/4) Exclude cut with ID 432-122774-0017-62487-0_sp1.1 from training. Duration: 20.3590625
2022-12-22 15:23:55,520 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 3.456e+02 6.070e+02 7.487e+02 9.437e+02 1.926e+03, threshold=1.497e+03, percent-clipped=1.0
2022-12-22 15:24:06,565 INFO [train.py:894] (2/4) Epoch 4, batch 1600, loss[loss=0.2222, simple_loss=0.3099, pruned_loss=0.06721, over 18569.00 frames. ], tot_loss[loss=0.2614, simple_loss=0.3303, pruned_loss=0.09629, over 3712563.74 frames. ], batch size: 51, lr: 2.69e-02, grad_scale: 8.0
2022-12-22 15:24:32,385 INFO [zipformer.py:660] (2/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=12137.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 15:24:39,176 INFO [scaling.py:679] (2/4) Whitening: num_groups=8, num_channels=96, metric=1.67 vs. limit=2.0
2022-12-22 15:25:00,613 WARNING [train.py:1060] (2/4) Exclude cut with ID 3557-8342-0013-54691-0_sp1.1 from training. Duration: 0.836375
2022-12-22 15:25:22,154 INFO [train.py:894] (2/4) Epoch 4, batch 1650, loss[loss=0.2314, simple_loss=0.2969, pruned_loss=0.08292, over 18526.00 frames. ], tot_loss[loss=0.2656, simple_loss=0.3328, pruned_loss=0.09918, over 3713475.34 frames. ], batch size: 44, lr: 2.69e-02, grad_scale: 16.0
2022-12-22 15:25:25,118 INFO [zipformer.py:660] (2/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=12172.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 15:25:41,401 WARNING [train.py:1060] (2/4) Exclude cut with ID 8565-290391-0049-67394-0_sp0.9 from training. Duration: 21.3166875
2022-12-22 15:25:47,074 INFO [scaling.py:679] (2/4) Whitening: num_groups=1, num_channels=384, metric=4.26 vs. limit=5.0
2022-12-22 15:25:49,390 INFO [zipformer.py:660] (2/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=12188.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 15:25:57,716 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([2.2279, 1.8098, 1.1908, 2.0196, 2.4453, 1.9658, 2.6429, 2.7628],
       device='cuda:2'), covar=tensor([0.1611, 0.1808, 0.2314, 0.1566, 0.1643, 0.1355, 0.1257, 0.1309],
       device='cuda:2'), in_proj_covar=tensor([0.0092, 0.0105, 0.0132, 0.0103, 0.0107, 0.0097, 0.0098, 0.0101],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0002, 0.0003, 0.0003, 0.0003, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-22 15:26:09,625 INFO [zipformer.py:660] (2/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=12202.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 15:26:12,974 WARNING [train.py:1060] (2/4) Exclude cut with ID 6533-399-0029-104863-0_sp0.9 from training. Duration: 22.1055625
2022-12-22 15:26:23,738 WARNING [train.py:1060] (2/4) Exclude cut with ID 7699-105389-0094-26379-0_sp1.1 from training. Duration: 21.77725
2022-12-22 15:26:27,786 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 4.397e+02 7.185e+02 8.672e+02 1.039e+03 2.165e+03, threshold=1.734e+03, percent-clipped=6.0
2022-12-22 15:26:37,340 INFO [train.py:894] (2/4) Epoch 4, batch 1700, loss[loss=0.2867, simple_loss=0.3513, pruned_loss=0.1111, over 18508.00 frames. ], tot_loss[loss=0.2715, simple_loss=0.3359, pruned_loss=0.1036, over 3712722.32 frames. ], batch size: 52, lr: 2.68e-02, grad_scale: 8.0
2022-12-22 15:26:44,740 WARNING [train.py:1060] (2/4) Exclude cut with ID 7255-291500-0005-134304-0_sp0.9 from training. Duration: 27.8166875
2022-12-22 15:27:01,074 INFO [zipformer.py:660] (2/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=12236.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 15:27:09,569 WARNING [train.py:1060] (2/4) Exclude cut with ID 7357-94126-0021-15852-0_sp1.1 from training. Duration: 22.5090625
2022-12-22 15:27:17,054 WARNING [train.py:1060] (2/4) Exclude cut with ID 7255-291500-0005-134304-0 from training. Duration: 25.035
2022-12-22 15:27:34,427 WARNING [train.py:1060] (2/4) Exclude cut with ID 774-127930-0014-10412-0_sp1.1 from training. Duration: 0.95
2022-12-22 15:27:54,444 INFO [train.py:894] (2/4) Epoch 4, batch 1750, loss[loss=0.2516, simple_loss=0.3068, pruned_loss=0.09824, over 18373.00 frames. ], tot_loss[loss=0.2761, simple_loss=0.338, pruned_loss=0.1071, over 3712927.28 frames. ], batch size: 42, lr: 2.68e-02, grad_scale: 4.0
2022-12-22 15:27:54,498 WARNING [train.py:1060] (2/4) Exclude cut with ID 3033-130750-0096-55598-0_sp0.9 from training. Duration: 0.92225
2022-12-22 15:28:03,289 INFO [zipformer.py:660] (2/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=12276.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 15:28:20,549 WARNING [train.py:1060] (2/4) Exclude cut with ID 4511-76322-0006-80011-0 from training. Duration: 21.97
2022-12-22 15:28:25,155 INFO [zipformer.py:660] (2/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=12291.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 15:28:40,100 WARNING [train.py:1060] (2/4) Exclude cut with ID 7492-105653-0055-62765-0_sp0.9 from training. Duration: 21.97225
2022-12-22 15:28:41,534 WARNING [train.py:1060] (2/4) Exclude cut with ID 453-131332-0000-47844-0_sp0.9 from training. Duration: 25.3333125
2022-12-22 15:28:44,636 INFO [zipformer.py:660] (2/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=12304.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 15:28:54,622 WARNING [train.py:1060] (2/4) Exclude cut with ID 5172-29468-0015-19128-0_sp0.9 from training. Duration: 21.5055625
2022-12-22 15:29:02,124 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 3.995e+02 6.529e+02 8.437e+02 1.023e+03 2.036e+03, threshold=1.687e+03, percent-clipped=2.0
2022-12-22 15:29:04,838 WARNING [train.py:1060] (2/4) Exclude cut with ID 453-131332-0000-47844-0_sp1.1 from training. Duration: 20.72725
2022-12-22 15:29:08,971 INFO [train.py:894] (2/4) Epoch 4, batch 1800, loss[loss=0.3312, simple_loss=0.3795, pruned_loss=0.1414, over 18689.00 frames. ], tot_loss[loss=0.2822, simple_loss=0.342, pruned_loss=0.1112, over 3712950.46 frames. ], batch size: 60, lr: 2.67e-02, grad_scale: 4.0
2022-12-22 15:29:33,351 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([2.2955, 1.1822, 1.8197, 2.0395, 1.9747, 1.6973, 1.9709, 1.1564],
       device='cuda:2'), covar=tensor([0.0611, 0.1182, 0.1049, 0.0778, 0.0520, 0.0502, 0.0782, 0.0634],
       device='cuda:2'), in_proj_covar=tensor([0.0160, 0.0194, 0.0184, 0.0185, 0.0163, 0.0154, 0.0172, 0.0152],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0001, 0.0002, 0.0001],
       device='cuda:2')
2022-12-22 15:29:34,291 WARNING [train.py:1060] (2/4) Exclude cut with ID 8631-249866-0030-130156-0_sp0.9 from training. Duration: 26.32775
2022-12-22 15:29:37,397 INFO [zipformer.py:660] (2/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=12339.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 15:29:38,841 INFO [zipformer.py:660] (2/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=12340.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 15:30:01,945 INFO [zipformer.py:660] (2/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=12355.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 15:30:08,349 WARNING [train.py:1060] (2/4) Exclude cut with ID 3867-173237-0077-144769-0 from training. Duration: 20.025
2022-12-22 15:30:11,401 WARNING [train.py:1060] (2/4) Exclude cut with ID 6709-74022-0004-86860-0_sp1.1 from training. Duration: 0.9409375
2022-12-22 15:30:11,414 WARNING [train.py:1060] (2/4) Exclude cut with ID 4757-1811-0023-62229-0_sp0.9 from training. Duration: 21.37775
2022-12-22 15:30:24,822 INFO [train.py:894] (2/4) Epoch 4, batch 1850, loss[loss=0.3484, simple_loss=0.3843, pruned_loss=0.1563, over 18466.00 frames. ], tot_loss[loss=0.2883, simple_loss=0.3453, pruned_loss=0.1157, over 3712901.52 frames. ], batch size: 50, lr: 2.67e-02, grad_scale: 4.0
2022-12-22 15:30:33,708 WARNING [train.py:1060] (2/4) Exclude cut with ID 1250-135782-0004-25974-0_sp0.9 from training. Duration: 21.17225
2022-12-22 15:30:33,722 WARNING [train.py:1060] (2/4) Exclude cut with ID 7357-94126-0021-15852-0_sp0.9 from training. Duration: 27.511125
2022-12-22 15:31:04,789 WARNING [train.py:1060] (2/4) Exclude cut with ID 453-131332-0000-47844-0 from training. Duration: 22.8
2022-12-22 15:31:09,264 WARNING [train.py:1060] (2/4) Exclude cut with ID 4964-30587-0040-44509-0 from training. Duration: 22.585
2022-12-22 15:31:14,793 INFO [zipformer.py:660] (2/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=12403.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 15:31:33,079 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 3.783e+02 7.220e+02 8.554e+02 1.149e+03 2.507e+03, threshold=1.711e+03, percent-clipped=5.0
2022-12-22 15:31:40,395 INFO [train.py:894] (2/4) Epoch 4, batch 1900, loss[loss=0.3919, simple_loss=0.4157, pruned_loss=0.184, over 18452.00 frames. ], tot_loss[loss=0.2925, simple_loss=0.3477, pruned_loss=0.1187, over 3713602.32 frames. ], batch size: 54, lr: 2.66e-02, grad_scale: 4.0
2022-12-22 15:31:40,397 WARNING [train.py:1060] (2/4) Exclude cut with ID 6978-92210-0001-146967-0_sp0.9 from training. Duration: 22.0166875
2022-12-22 15:31:55,406 WARNING [train.py:1060] (2/4) Exclude cut with ID 7255-291500-0003-134302-0_sp1.1 from training. Duration: 24.395375
2022-12-22 15:31:58,787 INFO [zipformer.py:660] (2/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=12432.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 15:32:03,239 WARNING [train.py:1060] (2/4) Exclude cut with ID 3488-85273-0017-41203-0_sp0.9 from training. Duration: 27.47775
2022-12-22 15:32:07,310 WARNING [train.py:1060] (2/4) Exclude cut with ID 432-122774-0017-62487-0_sp0.9 from training. Duration: 24.8833125
2022-12-22 15:32:10,968 WARNING [train.py:1060] (2/4) Exclude cut with ID 6758-72288-0033-108368-0 from training. Duration: 23.39
2022-12-22 15:32:16,785 WARNING [train.py:1060] (2/4) Exclude cut with ID 8291-282929-0007-12994-0_sp0.9 from training. Duration: 28.72225
2022-12-22 15:32:19,932 INFO [zipformer.py:660] (2/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=12446.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 15:32:26,323 WARNING [train.py:1060] (2/4) Exclude cut with ID 585-294811-0110-133686-0_sp0.9 from training. Duration: 20.8944375
2022-12-22 15:32:39,524 WARNING [train.py:1060] (2/4) Exclude cut with ID 5796-66357-0007-116447-0_sp0.9 from training. Duration: 23.8444375
2022-12-22 15:32:55,864 INFO [train.py:894] (2/4) Epoch 4, batch 1950, loss[loss=0.3423, simple_loss=0.382, pruned_loss=0.1513, over 18649.00 frames. ], tot_loss[loss=0.2955, simple_loss=0.349, pruned_loss=0.121, over 3713573.05 frames. ], batch size: 60, lr: 2.66e-02, grad_scale: 4.0
2022-12-22 15:32:59,283 INFO [zipformer.py:660] (2/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=12472.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 15:33:02,046 WARNING [train.py:1060] (2/4) Exclude cut with ID 8291-282929-0007-12994-0 from training. Duration: 25.85
2022-12-22 15:33:02,058 WARNING [train.py:1060] (2/4) Exclude cut with ID 8291-282929-0023-13010-0 from training. Duration: 21.39
2022-12-22 15:33:12,698 WARNING [train.py:1060] (2/4) Exclude cut with ID 7357-94126-0014-15845-0 from training. Duration: 27.92
2022-12-22 15:33:40,809 WARNING [train.py:1060] (2/4) Exclude cut with ID 8631-249866-0039-130165-0_sp0.9 from training. Duration: 20.661125
2022-12-22 15:33:45,444 INFO [zipformer.py:660] (2/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=12502.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 15:33:52,955 INFO [zipformer.py:660] (2/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=12507.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 15:34:04,115 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 5.166e+02 7.819e+02 9.278e+02 1.167e+03 2.388e+03, threshold=1.856e+03, percent-clipped=8.0
2022-12-22 15:34:04,181 WARNING [train.py:1060] (2/4) Exclude cut with ID 7357-94126-0043-15874-0_sp0.9 from training. Duration: 20.07225
2022-12-22 15:34:06,218 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.5464, 1.0430, 0.9100, 1.2463, 1.7845, 0.9069, 1.2641, 1.6331],
       device='cuda:2'), covar=tensor([0.2008, 0.2299, 0.2487, 0.1699, 0.2042, 0.1719, 0.1532, 0.1597],
       device='cuda:2'), in_proj_covar=tensor([0.0092, 0.0105, 0.0130, 0.0101, 0.0109, 0.0095, 0.0097, 0.0100],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0002, 0.0003, 0.0003, 0.0003, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-22 15:34:12,013 INFO [train.py:894] (2/4) Epoch 4, batch 2000, loss[loss=0.2617, simple_loss=0.3155, pruned_loss=0.104, over 18528.00 frames. ], tot_loss[loss=0.2952, simple_loss=0.3483, pruned_loss=0.121, over 3713603.07 frames. ], batch size: 47, lr: 2.65e-02, grad_scale: 8.0
2022-12-22 15:34:12,118 WARNING [train.py:1060] (2/4) Exclude cut with ID 1085-156170-0017-128270-0 from training. Duration: 21.01
2022-12-22 15:34:12,186 INFO [zipformer.py:660] (2/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=12520.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 15:34:32,291 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([4.2211, 3.6126, 3.6078, 3.7908, 3.7851, 3.8863, 4.3895, 1.5337],
       device='cuda:2'), covar=tensor([0.0785, 0.0648, 0.0696, 0.0692, 0.1613, 0.1105, 0.0568, 0.4361],
       device='cuda:2'), in_proj_covar=tensor([0.0197, 0.0156, 0.0146, 0.0133, 0.0212, 0.0165, 0.0164, 0.0205],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0001, 0.0001, 0.0001, 0.0002, 0.0001, 0.0002, 0.0002],
       device='cuda:2')
2022-12-22 15:34:58,657 INFO [zipformer.py:660] (2/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=12550.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 15:35:22,241 WARNING [train.py:1060] (2/4) Exclude cut with ID 2195-150901-0045-59933-0 from training. Duration: 20.65
2022-12-22 15:35:27,886 INFO [train.py:894] (2/4) Epoch 4, batch 2050, loss[loss=0.3021, simple_loss=0.3611, pruned_loss=0.1216, over 18545.00 frames. ], tot_loss[loss=0.2954, simple_loss=0.3483, pruned_loss=0.1212, over 3715578.26 frames. ], batch size: 96, lr: 2.65e-02, grad_scale: 8.0
2022-12-22 15:35:29,484 WARNING [train.py:1060] (2/4) Exclude cut with ID 5796-66357-0007-116447-0 from training. Duration: 21.46
2022-12-22 15:35:37,196 INFO [zipformer.py:660] (2/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=12576.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 15:35:37,977 INFO [scaling.py:679] (2/4) Whitening: num_groups=8, num_channels=96, metric=1.54 vs. limit=2.0
2022-12-22 15:36:12,248 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.6842, 1.1603, 0.9305, 1.1385, 1.8134, 0.8092, 1.2643, 1.6628],
       device='cuda:2'), covar=tensor([0.1871, 0.2313, 0.2563, 0.2038, 0.2379, 0.2042, 0.1773, 0.1760],
       device='cuda:2'), in_proj_covar=tensor([0.0093, 0.0107, 0.0133, 0.0104, 0.0113, 0.0099, 0.0099, 0.0103],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0003, 0.0003, 0.0003, 0.0003, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-22 15:36:16,503 WARNING [train.py:1060] (2/4) Exclude cut with ID 3557-8342-0013-54691-0 from training. Duration: 0.92
2022-12-22 15:36:21,167 INFO [zipformer.py:660] (2/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=12604.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 15:36:22,386 WARNING [train.py:1060] (2/4) Exclude cut with ID 8291-282929-0023-13010-0_sp0.9 from training. Duration: 23.7666875
2022-12-22 15:36:36,980 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 3.601e+02 6.890e+02 8.468e+02 1.080e+03 1.755e+03, threshold=1.694e+03, percent-clipped=0.0
2022-12-22 15:36:44,321 INFO [train.py:894] (2/4) Epoch 4, batch 2100, loss[loss=0.3081, simple_loss=0.3581, pruned_loss=0.129, over 18726.00 frames. ], tot_loss[loss=0.2951, simple_loss=0.3482, pruned_loss=0.121, over 3715329.85 frames. ], batch size: 52, lr: 2.64e-02, grad_scale: 8.0
2022-12-22 15:36:50,797 INFO [zipformer.py:660] (2/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=12624.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 15:37:00,631 WARNING [train.py:1060] (2/4) Exclude cut with ID 8544-281189-0060-101339-0_sp0.9 from training. Duration: 20.861125
2022-12-22 15:37:13,542 WARNING [train.py:1060] (2/4) Exclude cut with ID 3488-65654-0031-41259-0_sp0.9 from training. Duration: 22.711125
2022-12-22 15:37:16,591 INFO [zipformer.py:660] (2/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=12640.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 15:37:34,285 INFO [zipformer.py:660] (2/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=12652.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 15:37:52,281 WARNING [train.py:1060] (2/4) Exclude cut with ID 6951-79737-0043-132310-0_sp1.1 from training. Duration: 22.986375
2022-12-22 15:38:00,053 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.5441, 1.9465, 1.2577, 2.2878, 2.5674, 1.4964, 1.8841, 1.1048],
       device='cuda:2'), covar=tensor([0.1800, 0.1509, 0.1403, 0.0792, 0.1167, 0.1177, 0.1282, 0.1606],
       device='cuda:2'), in_proj_covar=tensor([0.0204, 0.0173, 0.0166, 0.0151, 0.0212, 0.0157, 0.0168, 0.0163],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-22 15:38:00,977 INFO [train.py:894] (2/4) Epoch 4, batch 2150, loss[loss=0.2954, simple_loss=0.3454, pruned_loss=0.1227, over 18572.00 frames. ], tot_loss[loss=0.296, simple_loss=0.3483, pruned_loss=0.1218, over 3715221.52 frames. ], batch size: 49, lr: 2.64e-02, grad_scale: 8.0
2022-12-22 15:38:09,499 WARNING [train.py:1060] (2/4) Exclude cut with ID 8040-260924-0003-80960-0_sp0.9 from training. Duration: 22.07225
2022-12-22 15:38:12,567 WARNING [train.py:1060] (2/4) Exclude cut with ID 7699-105389-0045-26330-0_sp0.9 from training. Duration: 20.3055625
2022-12-22 15:38:16,229 WARNING [train.py:1060] (2/4) Exclude cut with ID 6356-271890-0060-94317-0_sp0.9 from training. Duration: 20.72225
2022-12-22 15:38:21,144 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([2.6101, 3.6538, 3.7101, 1.6485, 3.6401, 2.7498, 0.7071, 2.5427],
       device='cuda:2'), covar=tensor([0.1714, 0.0770, 0.1477, 0.3446, 0.0851, 0.1223, 0.5735, 0.1808],
       device='cuda:2'), in_proj_covar=tensor([0.0122, 0.0097, 0.0148, 0.0118, 0.0104, 0.0103, 0.0146, 0.0111],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-22 15:38:30,026 INFO [zipformer.py:660] (2/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=12688.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 15:38:30,905 INFO [scaling.py:679] (2/4) Whitening: num_groups=8, num_channels=96, metric=1.59 vs. limit=2.0
2022-12-22 15:38:37,029 WARNING [train.py:1060] (2/4) Exclude cut with ID 3488-85273-0017-41203-0_sp1.1 from training. Duration: 22.4818125
2022-12-22 15:39:02,217 WARNING [train.py:1060] (2/4) Exclude cut with ID 4964-30587-0040-44509-0_sp0.9 from training. Duration: 25.0944375
2022-12-22 15:39:06,471 WARNING [train.py:1060] (2/4) Exclude cut with ID 6533-399-0047-104881-0 from training. Duration: 21.515
2022-12-22 15:39:09,168 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 3.874e+02 7.222e+02 8.655e+02 1.105e+03 2.731e+03, threshold=1.731e+03, percent-clipped=2.0
2022-12-22 15:39:14,381 WARNING [train.py:1060] (2/4) Exclude cut with ID 7357-94126-0009-15840-0_sp0.9 from training. Duration: 27.02225
2022-12-22 15:39:17,189 INFO [train.py:894] (2/4) Epoch 4, batch 2200, loss[loss=0.3239, simple_loss=0.3768, pruned_loss=0.1355, over 18480.00 frames. ], tot_loss[loss=0.2945, simple_loss=0.3476, pruned_loss=0.1207, over 3714163.44 frames. ], batch size: 64, lr: 2.64e-02, grad_scale: 8.0
2022-12-22 15:39:19,475 WARNING [train.py:1060] (2/4) Exclude cut with ID 432-122774-0010-62480-0_sp0.9 from training. Duration: 22.22225
2022-12-22 15:39:26,834 WARNING [train.py:1060] (2/4) Exclude cut with ID 4964-30587-0085-44554-0_sp0.9 from training. Duration: 20.85
2022-12-22 15:39:37,075 INFO [zipformer.py:660] (2/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=12732.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 15:39:57,102 WARNING [train.py:1060] (2/4) Exclude cut with ID 4295-39940-0007-92567-0 from training. Duration: 21.54
2022-12-22 15:40:01,900 WARNING [train.py:1060] (2/4) Exclude cut with ID 4964-30587-0040-44509-0_sp1.1 from training. Duration: 20.5318125
2022-12-22 15:40:07,016 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.3565, 1.7981, 1.1744, 2.0307, 2.2213, 1.3675, 1.6641, 1.0127],
       device='cuda:2'), covar=tensor([0.2037, 0.1515, 0.1587, 0.0846, 0.1135, 0.1251, 0.1300, 0.1752],
       device='cuda:2'), in_proj_covar=tensor([0.0211, 0.0179, 0.0173, 0.0157, 0.0220, 0.0163, 0.0173, 0.0171],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-22 15:40:09,895 WARNING [train.py:1060] (2/4) Exclude cut with ID 7255-291500-0012-134311-0_sp0.9 from training. Duration: 21.9333125
2022-12-22 15:40:25,790 INFO [zipformer.py:660] (2/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=12763.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 15:40:36,692 INFO [train.py:894] (2/4) Epoch 4, batch 2250, loss[loss=0.2904, simple_loss=0.354, pruned_loss=0.1134, over 18555.00 frames. ], tot_loss[loss=0.294, simple_loss=0.3466, pruned_loss=0.1207, over 3714755.67 frames. ], batch size: 57, lr: 2.63e-02, grad_scale: 8.0
2022-12-22 15:40:53,270 INFO [zipformer.py:660] (2/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=12780.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 15:40:58,888 WARNING [train.py:1060] (2/4) Exclude cut with ID 8631-249866-0025-130151-0_sp0.9 from training. Duration: 21.7944375
2022-12-22 15:41:10,551 WARNING [train.py:1060] (2/4) Exclude cut with ID 8291-282929-0002-12989-0_sp0.9 from training. Duration: 22.4666875
2022-12-22 15:41:16,474 WARNING [train.py:1060] (2/4) Exclude cut with ID 6121-9014-0076-24124-0 from training. Duration: 21.635
2022-12-22 15:41:22,619 WARNING [train.py:1060] (2/4) Exclude cut with ID 6121-9014-0076-24124-0_sp0.9 from training. Duration: 24.038875
2022-12-22 15:41:25,728 INFO [zipformer.py:660] (2/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=12802.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 15:41:47,262 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 4.686e+02 7.014e+02 8.865e+02 1.069e+03 2.354e+03, threshold=1.773e+03, percent-clipped=4.0
2022-12-22 15:41:55,103 INFO [train.py:894] (2/4) Epoch 4, batch 2300, loss[loss=0.3102, simple_loss=0.3614, pruned_loss=0.1295, over 18416.00 frames. ], tot_loss[loss=0.2957, simple_loss=0.3481, pruned_loss=0.1217, over 3714445.07 frames. ], batch size: 53, lr: 2.63e-02, grad_scale: 8.0
2022-12-22 15:42:01,630 INFO [zipformer.py:660] (2/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=12824.0, num_to_drop=1, layers_to_drop={0}
2022-12-22 15:42:05,538 WARNING [train.py:1060] (2/4) Exclude cut with ID 7255-291500-0009-134308-0_sp1.1 from training. Duration: 21.786375
2022-12-22 15:42:15,564 WARNING [train.py:1060] (2/4) Exclude cut with ID 8291-282929-0002-12989-0 from training. Duration: 20.22
2022-12-22 15:43:11,114 INFO [train.py:894] (2/4) Epoch 4, batch 2350, loss[loss=0.2646, simple_loss=0.3199, pruned_loss=0.1047, over 18672.00 frames. ], tot_loss[loss=0.2945, simple_loss=0.3474, pruned_loss=0.1208, over 3714272.79 frames. ], batch size: 46, lr: 2.62e-02, grad_scale: 8.0
2022-12-22 15:43:24,907 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([0.1706, 1.4495, 1.7402, 0.3406, 0.9954, 2.0863, 1.6617, 1.5813],
       device='cuda:2'), covar=tensor([0.0782, 0.0369, 0.0391, 0.0612, 0.0575, 0.0258, 0.0324, 0.0511],
       device='cuda:2'), in_proj_covar=tensor([0.0105, 0.0114, 0.0084, 0.0109, 0.0108, 0.0077, 0.0115, 0.0092],
       device='cuda:2'), out_proj_covar=tensor([1.1284e-04, 1.2284e-04, 9.2690e-05, 1.1671e-04, 1.1512e-04, 8.4218e-05,
        1.2692e-04, 1.0022e-04], device='cuda:2')
2022-12-22 15:44:16,077 INFO [scaling.py:679] (2/4) Whitening: num_groups=8, num_channels=192, metric=1.95 vs. limit=2.0
2022-12-22 15:44:20,056 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 4.811e+02 6.843e+02 8.099e+02 1.087e+03 2.395e+03, threshold=1.620e+03, percent-clipped=1.0
2022-12-22 15:44:21,450 WARNING [train.py:1060] (2/4) Exclude cut with ID 6951-79737-0043-132310-0 from training. Duration: 25.285
2022-12-22 15:44:27,503 INFO [train.py:894] (2/4) Epoch 4, batch 2400, loss[loss=0.3258, simple_loss=0.3676, pruned_loss=0.142, over 18640.00 frames. ], tot_loss[loss=0.2927, simple_loss=0.3463, pruned_loss=0.1196, over 3715085.95 frames. ], batch size: 69, lr: 2.62e-02, grad_scale: 8.0
2022-12-22 15:44:29,292 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([3.8343, 3.3511, 3.4023, 3.5707, 3.4605, 3.5329, 3.9103, 1.8429],
       device='cuda:2'), covar=tensor([0.0605, 0.0480, 0.0565, 0.0572, 0.1236, 0.0750, 0.0535, 0.3162],
       device='cuda:2'), in_proj_covar=tensor([0.0201, 0.0156, 0.0149, 0.0135, 0.0219, 0.0166, 0.0164, 0.0206],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0001, 0.0001, 0.0001, 0.0002, 0.0001, 0.0002, 0.0002],
       device='cuda:2')
2022-12-22 15:44:51,842 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([2.0176, 2.2414, 1.5411, 1.1860, 2.9276, 2.5271, 1.9517, 0.9872],
       device='cuda:2'), covar=tensor([0.0597, 0.0556, 0.1025, 0.1097, 0.0119, 0.0386, 0.0807, 0.1958],
       device='cuda:2'), in_proj_covar=tensor([0.0118, 0.0113, 0.0135, 0.0124, 0.0074, 0.0113, 0.0134, 0.0147],
       device='cuda:2'), out_proj_covar=tensor([1.4816e-04, 1.4347e-04, 1.6441e-04, 1.5182e-04, 9.4459e-05, 1.3759e-04,
        1.6431e-04, 1.7988e-04], device='cuda:2')
2022-12-22 15:45:24,194 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([3.1757, 2.6982, 2.7812, 3.0160, 2.7852, 2.8118, 3.2691, 1.0353],
       device='cuda:2'), covar=tensor([0.0747, 0.0703, 0.0697, 0.0519, 0.1536, 0.1020, 0.0654, 0.3783],
       device='cuda:2'), in_proj_covar=tensor([0.0204, 0.0158, 0.0151, 0.0137, 0.0218, 0.0168, 0.0168, 0.0209],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0001, 0.0001, 0.0001, 0.0002, 0.0001, 0.0002, 0.0002],
       device='cuda:2')
2022-12-22 15:45:26,829 WARNING [train.py:1060] (2/4) Exclude cut with ID 811-130148-0001-63453-0_sp0.9 from training. Duration: 20.861125
2022-12-22 15:45:43,625 INFO [train.py:894] (2/4) Epoch 4, batch 2450, loss[loss=0.2455, simple_loss=0.3169, pruned_loss=0.0871, over 18451.00 frames. ], tot_loss[loss=0.2934, simple_loss=0.3466, pruned_loss=0.1201, over 3715660.94 frames. ], batch size: 50, lr: 2.61e-02, grad_scale: 8.0
2022-12-22 15:45:49,679 WARNING [train.py:1060] (2/4) Exclude cut with ID 6010-56788-0055-90261-0 from training. Duration: 20.88
2022-12-22 15:45:57,466 INFO [scaling.py:679] (2/4) Whitening: num_groups=1, num_channels=384, metric=3.48 vs. limit=5.0
2022-12-22 15:46:20,303 WARNING [train.py:1060] (2/4) Exclude cut with ID 7357-94126-0045-15876-0_sp0.9 from training. Duration: 23.4166875
2022-12-22 15:46:51,982 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 4.109e+02 7.565e+02 9.195e+02 1.170e+03 3.094e+03, threshold=1.839e+03, percent-clipped=13.0
2022-12-22 15:46:59,550 INFO [train.py:894] (2/4) Epoch 4, batch 2500, loss[loss=0.2975, simple_loss=0.3545, pruned_loss=0.1202, over 18629.00 frames. ], tot_loss[loss=0.2936, simple_loss=0.3467, pruned_loss=0.1203, over 3715643.69 frames. ], batch size: 53, lr: 2.61e-02, grad_scale: 8.0
2022-12-22 15:47:13,086 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.5894, 1.4795, 0.9122, 1.6509, 1.6391, 1.4526, 2.0999, 1.6316],
       device='cuda:2'), covar=tensor([0.0878, 0.1283, 0.2363, 0.1401, 0.1667, 0.0821, 0.0742, 0.0995],
       device='cuda:2'), in_proj_covar=tensor([0.0158, 0.0166, 0.0205, 0.0246, 0.0208, 0.0165, 0.0159, 0.0171],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-22 15:47:26,738 INFO [scaling.py:679] (2/4) Whitening: num_groups=8, num_channels=192, metric=1.98 vs. limit=2.0
2022-12-22 15:47:38,526 WARNING [train.py:1060] (2/4) Exclude cut with ID 7255-291500-0006-134305-0 from training. Duration: 21.24
2022-12-22 15:47:40,192 WARNING [train.py:1060] (2/4) Exclude cut with ID 6533-399-0047-104881-0_sp0.9 from training. Duration: 23.9055625
2022-12-22 15:47:44,071 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([2.5787, 2.1217, 1.7167, 0.6073, 1.7160, 2.0058, 1.3491, 1.7215],
       device='cuda:2'), covar=tensor([0.0601, 0.0550, 0.1186, 0.1842, 0.1189, 0.1392, 0.1675, 0.0863],
       device='cuda:2'), in_proj_covar=tensor([0.0133, 0.0164, 0.0196, 0.0190, 0.0183, 0.0170, 0.0183, 0.0174],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-22 15:47:45,961 INFO [scaling.py:679] (2/4) Whitening: num_groups=8, num_channels=192, metric=1.90 vs. limit=2.0
2022-12-22 15:48:13,946 WARNING [train.py:1060] (2/4) Exclude cut with ID 6758-72288-0033-108368-0_sp0.9 from training. Duration: 25.988875
2022-12-22 15:48:15,368 INFO [train.py:894] (2/4) Epoch 4, batch 2550, loss[loss=0.2223, simple_loss=0.2894, pruned_loss=0.07762, over 18576.00 frames. ], tot_loss[loss=0.2928, simple_loss=0.3455, pruned_loss=0.1201, over 3715264.35 frames. ], batch size: 45, lr: 2.60e-02, grad_scale: 8.0
2022-12-22 15:48:22,603 WARNING [train.py:1060] (2/4) Exclude cut with ID 7255-291500-0001-134300-0_sp0.9 from training. Duration: 20.67225
2022-12-22 15:49:05,464 INFO [zipformer.py:660] (2/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=13102.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 15:49:09,824 WARNING [train.py:1060] (2/4) Exclude cut with ID 3488-85273-0038-41224-0 from training. Duration: 20.34
2022-12-22 15:49:19,083 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.6549, 1.4009, 1.2922, 2.1380, 1.6413, 3.3334, 1.5314, 1.6067],
       device='cuda:2'), covar=tensor([0.1183, 0.2179, 0.1619, 0.1161, 0.1836, 0.0310, 0.1609, 0.1863],
       device='cuda:2'), in_proj_covar=tensor([0.0085, 0.0091, 0.0088, 0.0085, 0.0106, 0.0077, 0.0096, 0.0086],
       device='cuda:2'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003],
       device='cuda:2')
2022-12-22 15:49:24,508 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 4.669e+02 7.177e+02 8.673e+02 1.022e+03 2.536e+03, threshold=1.735e+03, percent-clipped=2.0
2022-12-22 15:49:30,804 INFO [zipformer.py:660] (2/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=13119.0, num_to_drop=1, layers_to_drop={0}
2022-12-22 15:49:31,958 INFO [train.py:894] (2/4) Epoch 4, batch 2600, loss[loss=0.3508, simple_loss=0.3861, pruned_loss=0.1578, over 18597.00 frames. ], tot_loss[loss=0.2928, simple_loss=0.3453, pruned_loss=0.1201, over 3715009.42 frames. ], batch size: 98, lr: 2.60e-02, grad_scale: 8.0
2022-12-22 15:50:18,442 INFO [zipformer.py:660] (2/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=13150.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 15:50:22,575 WARNING [train.py:1060] (2/4) Exclude cut with ID 7357-94126-0026-15857-0_sp0.9 from training. Duration: 25.061125
2022-12-22 15:50:29,040 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([0.3458, 1.6176, 2.0662, 0.5820, 1.3445, 2.4848, 1.8944, 1.6309],
       device='cuda:2'), covar=tensor([0.0941, 0.0450, 0.0403, 0.0605, 0.0466, 0.0215, 0.0335, 0.0617],
       device='cuda:2'), in_proj_covar=tensor([0.0107, 0.0118, 0.0086, 0.0112, 0.0109, 0.0080, 0.0114, 0.0095],
       device='cuda:2'), out_proj_covar=tensor([1.1472e-04, 1.2711e-04, 9.3626e-05, 1.1922e-04, 1.1553e-04, 8.6853e-05,
        1.2543e-04, 1.0385e-04], device='cuda:2')
2022-12-22 15:50:34,826 WARNING [train.py:1060] (2/4) Exclude cut with ID 3033-130750-0096-55598-0 from training. Duration: 0.83
2022-12-22 15:50:48,033 INFO [train.py:894] (2/4) Epoch 4, batch 2650, loss[loss=0.3614, simple_loss=0.3902, pruned_loss=0.1663, over 18604.00 frames. ], tot_loss[loss=0.294, simple_loss=0.3461, pruned_loss=0.121, over 3714575.16 frames. ], batch size: 175, lr: 2.60e-02, grad_scale: 8.0
2022-12-22 15:50:58,521 WARNING [train.py:1060] (2/4) Exclude cut with ID 3488-85273-0017-41203-0 from training. Duration: 24.73
2022-12-22 15:51:10,930 WARNING [train.py:1060] (2/4) Exclude cut with ID 7255-291500-0009-134308-0 from training. Duration: 23.965
2022-12-22 15:51:21,675 WARNING [train.py:1060] (2/4) Exclude cut with ID 6978-92210-0030-146996-0_sp0.9 from training. Duration: 22.088875
2022-12-22 15:51:37,276 WARNING [train.py:1060] (2/4) Exclude cut with ID 7255-291500-0006-134305-0_sp0.9 from training. Duration: 23.6
2022-12-22 15:51:58,019 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 4.369e+02 7.191e+02 8.861e+02 1.051e+03 2.028e+03, threshold=1.772e+03, percent-clipped=5.0
2022-12-22 15:52:06,063 INFO [train.py:894] (2/4) Epoch 4, batch 2700, loss[loss=0.2508, simple_loss=0.3171, pruned_loss=0.09223, over 18386.00 frames. ], tot_loss[loss=0.293, simple_loss=0.3457, pruned_loss=0.1202, over 3714873.89 frames. ], batch size: 46, lr: 2.59e-02, grad_scale: 8.0
2022-12-22 15:53:19,647 WARNING [train.py:1060] (2/4) Exclude cut with ID 8291-282929-0024-13011-0 from training. Duration: 23.795
2022-12-22 15:53:22,785 INFO [train.py:894] (2/4) Epoch 4, batch 2750, loss[loss=0.3223, simple_loss=0.3727, pruned_loss=0.136, over 18708.00 frames. ], tot_loss[loss=0.2926, simple_loss=0.3458, pruned_loss=0.1197, over 3714614.66 frames. ], batch size: 65, lr: 2.59e-02, grad_scale: 8.0
2022-12-22 15:53:38,045 WARNING [train.py:1060] (2/4) Exclude cut with ID 8631-249866-0030-130156-0_sp1.1 from training. Duration: 21.5409375
2022-12-22 15:53:41,018 WARNING [train.py:1060] (2/4) Exclude cut with ID 6978-92210-0019-146985-0_sp0.9 from training. Duration: 24.97775
2022-12-22 15:53:53,109 WARNING [train.py:1060] (2/4) Exclude cut with ID 1085-156170-0017-128270-0_sp0.9 from training. Duration: 23.3444375
2022-12-22 15:54:04,655 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.5976, 1.6243, 1.1004, 2.0317, 1.4761, 3.3149, 1.4077, 1.4399],
       device='cuda:2'), covar=tensor([0.1222, 0.1943, 0.1591, 0.1062, 0.1890, 0.0336, 0.1585, 0.1846],
       device='cuda:2'), in_proj_covar=tensor([0.0086, 0.0093, 0.0089, 0.0087, 0.0108, 0.0078, 0.0097, 0.0088],
       device='cuda:2'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003],
       device='cuda:2')
2022-12-22 15:54:19,748 WARNING [train.py:1060] (2/4) Exclude cut with ID 6010-56788-0055-90261-0_sp0.9 from training. Duration: 23.2
2022-12-22 15:54:25,810 WARNING [train.py:1060] (2/4) Exclude cut with ID 5653-46179-0060-117930-0_sp0.9 from training. Duration: 21.17225
2022-12-22 15:54:27,680 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([3.8363, 3.0540, 2.4271, 1.1900, 2.2173, 2.8118, 2.3265, 2.1460],
       device='cuda:2'), covar=tensor([0.0542, 0.0562, 0.1485, 0.2379, 0.1885, 0.1132, 0.1288, 0.1405],
       device='cuda:2'), in_proj_covar=tensor([0.0134, 0.0163, 0.0194, 0.0192, 0.0183, 0.0169, 0.0184, 0.0178],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-22 15:54:31,513 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 4.064e+02 6.793e+02 8.291e+02 1.011e+03 1.777e+03, threshold=1.658e+03, percent-clipped=1.0
2022-12-22 15:54:39,275 INFO [train.py:894] (2/4) Epoch 4, batch 2800, loss[loss=0.3085, simple_loss=0.3629, pruned_loss=0.127, over 18614.00 frames. ], tot_loss[loss=0.2913, simple_loss=0.3448, pruned_loss=0.1189, over 3715043.74 frames. ], batch size: 53, lr: 2.58e-02, grad_scale: 8.0
2022-12-22 15:54:45,983 WARNING [train.py:1060] (2/4) Exclude cut with ID 7255-291500-0008-134307-0_sp0.9 from training. Duration: 24.6555625
2022-12-22 15:55:40,838 WARNING [train.py:1060] (2/4) Exclude cut with ID 3488-65654-0031-41259-0 from training. Duration: 20.44
2022-12-22 15:55:56,109 INFO [train.py:894] (2/4) Epoch 4, batch 2850, loss[loss=0.3093, simple_loss=0.3556, pruned_loss=0.1315, over 18460.00 frames. ], tot_loss[loss=0.2905, simple_loss=0.3442, pruned_loss=0.1184, over 3714837.82 frames. ], batch size: 50, lr: 2.58e-02, grad_scale: 8.0
2022-12-22 15:55:56,156 WARNING [train.py:1060] (2/4) Exclude cut with ID 6951-79737-0018-132285-0_sp0.9 from training. Duration: 23.45
2022-12-22 15:56:00,224 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.7324, 1.7704, 1.7365, 1.8037, 1.6061, 3.6636, 1.7927, 2.5616],
       device='cuda:2'), covar=tensor([0.3902, 0.2222, 0.1992, 0.2125, 0.1614, 0.0204, 0.1729, 0.0992],
       device='cuda:2'), in_proj_covar=tensor([0.0155, 0.0132, 0.0145, 0.0131, 0.0125, 0.0099, 0.0116, 0.0113],
       device='cuda:2'), out_proj_covar=tensor([0.0004, 0.0003, 0.0004, 0.0003, 0.0003, 0.0002, 0.0003, 0.0003],
       device='cuda:2')
2022-12-22 15:56:25,730 WARNING [train.py:1060] (2/4) Exclude cut with ID 6945-60535-0076-12784-0_sp0.9 from training. Duration: 20.52225
2022-12-22 15:56:32,904 WARNING [train.py:1060] (2/4) Exclude cut with ID 7255-291500-0008-134307-0 from training. Duration: 22.19
2022-12-22 15:56:39,294 INFO [scaling.py:679] (2/4) Whitening: num_groups=1, num_channels=384, metric=3.10 vs. limit=5.0
2022-12-22 15:56:43,406 WARNING [train.py:1060] (2/4) Exclude cut with ID 7357-94126-0014-15845-0_sp1.1 from training. Duration: 25.3818125
2022-12-22 15:56:52,354 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([4.2973, 3.5139, 3.6075, 4.0172, 3.8611, 3.8331, 4.4610, 1.2290],
       device='cuda:2'), covar=tensor([0.0685, 0.0627, 0.0636, 0.0538, 0.1415, 0.0970, 0.0422, 0.4562],
       device='cuda:2'), in_proj_covar=tensor([0.0215, 0.0165, 0.0150, 0.0142, 0.0220, 0.0173, 0.0172, 0.0210],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0001, 0.0001, 0.0001, 0.0002, 0.0001, 0.0002, 0.0002],
       device='cuda:2')
2022-12-22 15:56:53,162 INFO [scaling.py:679] (2/4) Whitening: num_groups=8, num_channels=192, metric=2.00 vs. limit=2.0
2022-12-22 15:56:59,888 WARNING [train.py:1060] (2/4) Exclude cut with ID 6951-79737-0043-132310-0_sp0.9 from training. Duration: 28.0944375
2022-12-22 15:57:03,921 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 4.958e+02 6.920e+02 8.531e+02 1.147e+03 2.504e+03, threshold=1.706e+03, percent-clipped=8.0
2022-12-22 15:57:04,537 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([2.5731, 2.0063, 1.5159, 2.6786, 1.8717, 1.6883, 2.0127, 2.7730],
       device='cuda:2'), covar=tensor([0.1206, 0.1875, 0.1190, 0.1787, 0.1867, 0.0781, 0.1840, 0.0381],
       device='cuda:2'), in_proj_covar=tensor([0.0228, 0.0206, 0.0187, 0.0286, 0.0195, 0.0179, 0.0212, 0.0154],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0003, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-22 15:57:05,505 WARNING [train.py:1060] (2/4) Exclude cut with ID 2195-150901-0045-59933-0_sp0.9 from training. Duration: 22.9444375
2022-12-22 15:57:10,532 INFO [zipformer.py:660] (2/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=13419.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 15:57:11,739 INFO [train.py:894] (2/4) Epoch 4, batch 2900, loss[loss=0.3095, simple_loss=0.3668, pruned_loss=0.1261, over 18610.00 frames. ], tot_loss[loss=0.2895, simple_loss=0.3435, pruned_loss=0.1178, over 3715488.64 frames. ], batch size: 53, lr: 2.57e-02, grad_scale: 8.0
2022-12-22 15:57:13,343 WARNING [train.py:1060] (2/4) Exclude cut with ID 8291-282929-0024-13011-0_sp1.1 from training. Duration: 21.6318125
2022-12-22 15:57:19,252 INFO [scaling.py:679] (2/4) Whitening: num_groups=8, num_channels=96, metric=1.94 vs. limit=2.0
2022-12-22 15:57:32,288 WARNING [train.py:1060] (2/4) Exclude cut with ID 8631-249866-0030-130156-0 from training. Duration: 23.695
2022-12-22 15:57:56,944 WARNING [train.py:1060] (2/4) Exclude cut with ID 7699-105389-0094-26379-0 from training. Duration: 23.955
2022-12-22 15:58:23,219 INFO [zipformer.py:660] (2/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=13467.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 15:58:27,988 INFO [train.py:894] (2/4) Epoch 4, batch 2950, loss[loss=0.2554, simple_loss=0.3226, pruned_loss=0.0941, over 18574.00 frames. ], tot_loss[loss=0.2883, simple_loss=0.3427, pruned_loss=0.1169, over 3715539.71 frames. ], batch size: 49, lr: 2.57e-02, grad_scale: 8.0
2022-12-22 15:58:34,154 WARNING [train.py:1060] (2/4) Exclude cut with ID 8291-282929-0024-13011-0_sp0.9 from training. Duration: 26.438875
2022-12-22 15:59:19,136 WARNING [train.py:1060] (2/4) Exclude cut with ID 7699-105389-0021-26306-0_sp0.9 from training. Duration: 21.2444375
2022-12-22 15:59:19,161 WARNING [train.py:1060] (2/4) Exclude cut with ID 7357-94126-0014-15845-0_sp0.9 from training. Duration: 31.02225
2022-12-22 15:59:28,409 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([2.2038, 1.5681, 2.3403, 3.8018, 2.2597, 2.5416, 0.9940, 2.3346],
       device='cuda:2'), covar=tensor([0.1762, 0.1874, 0.1445, 0.0417, 0.1557, 0.1642, 0.2586, 0.1312],
       device='cuda:2'), in_proj_covar=tensor([0.0107, 0.0114, 0.0124, 0.0090, 0.0106, 0.0123, 0.0137, 0.0104],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0002, 0.0003, 0.0002, 0.0003, 0.0003, 0.0003, 0.0003],
       device='cuda:2')
2022-12-22 15:59:31,011 WARNING [train.py:1060] (2/4) Exclude cut with ID 432-122774-0017-62487-0 from training. Duration: 22.395
2022-12-22 15:59:36,276 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 4.113e+02 6.895e+02 8.488e+02 1.043e+03 2.580e+03, threshold=1.698e+03, percent-clipped=2.0
2022-12-22 15:59:44,464 INFO [train.py:894] (2/4) Epoch 4, batch 3000, loss[loss=0.2556, simple_loss=0.3129, pruned_loss=0.09913, over 18534.00 frames. ], tot_loss[loss=0.2876, simple_loss=0.3422, pruned_loss=0.1165, over 3714768.61 frames. ], batch size: 47, lr: 2.57e-02, grad_scale: 8.0
2022-12-22 15:59:44,464 INFO [train.py:919] (2/4) Computing validation loss
2022-12-22 15:59:53,033 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([4.1039, 3.6713, 3.6397, 3.8938, 3.5574, 3.6583, 4.2008, 1.5346],
       device='cuda:2'), covar=tensor([0.0593, 0.0509, 0.0568, 0.0481, 0.1202, 0.0656, 0.0311, 0.3787],
       device='cuda:2'), in_proj_covar=tensor([0.0213, 0.0164, 0.0153, 0.0143, 0.0224, 0.0174, 0.0175, 0.0211],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0001, 0.0001, 0.0001, 0.0002, 0.0001, 0.0002, 0.0002],
       device='cuda:2')
2022-12-22 15:59:55,854 INFO [train.py:928] (2/4) Epoch 4, validation: loss=0.2143, simple_loss=0.31, pruned_loss=0.05929, over 944034.00 frames. 
2022-12-22 15:59:55,855 INFO [train.py:929] (2/4) Maximum memory allocated so far is 24320MB
2022-12-22 15:59:58,841 WARNING [train.py:1060] (2/4) Exclude cut with ID 7357-94126-0045-15876-0 from training. Duration: 21.075
2022-12-22 16:00:04,623 WARNING [train.py:1060] (2/4) Exclude cut with ID 6482-98857-0025-147532-0_sp0.9 from training. Duration: 20.0055625
2022-12-22 16:00:04,635 WARNING [train.py:1060] (2/4) Exclude cut with ID 6951-79737-0037-132304-0_sp0.9 from training. Duration: 22.05
2022-12-22 16:00:04,646 WARNING [train.py:1060] (2/4) Exclude cut with ID 7255-291500-0003-134302-0 from training. Duration: 26.8349375
2022-12-22 16:00:09,221 WARNING [train.py:1060] (2/4) Exclude cut with ID 7357-94126-0009-15840-0_sp1.1 from training. Duration: 22.1090625
2022-12-22 16:00:15,101 WARNING [train.py:1060] (2/4) Exclude cut with ID 7699-105389-0094-26379-0_sp0.9 from training. Duration: 26.6166875
2022-12-22 16:00:33,644 WARNING [train.py:1060] (2/4) Exclude cut with ID 2046-178027-0000-53705-0_sp0.9 from training. Duration: 20.3055625
2022-12-22 16:00:40,676 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([2.2406, 2.3914, 1.5897, 1.0777, 3.1168, 2.7395, 2.0236, 1.3495],
       device='cuda:2'), covar=tensor([0.0489, 0.0379, 0.0872, 0.1048, 0.0108, 0.0320, 0.0684, 0.1293],
       device='cuda:2'), in_proj_covar=tensor([0.0119, 0.0113, 0.0135, 0.0124, 0.0078, 0.0115, 0.0137, 0.0152],
       device='cuda:2'), out_proj_covar=tensor([1.5091e-04, 1.4308e-04, 1.6503e-04, 1.5321e-04, 9.9220e-05, 1.4009e-04,
        1.6813e-04, 1.8673e-04], device='cuda:2')
2022-12-22 16:00:53,984 WARNING [train.py:1060] (2/4) Exclude cut with ID 7205-50138-0008-5373-0_sp0.9 from training. Duration: 20.7
2022-12-22 16:01:05,121 INFO [scaling.py:679] (2/4) Whitening: num_groups=8, num_channels=96, metric=1.22 vs. limit=2.0
2022-12-22 16:01:11,979 INFO [train.py:894] (2/4) Epoch 4, batch 3050, loss[loss=0.2242, simple_loss=0.289, pruned_loss=0.07967, over 18416.00 frames. ], tot_loss[loss=0.287, simple_loss=0.3417, pruned_loss=0.1162, over 3714227.50 frames. ], batch size: 42, lr: 2.56e-02, grad_scale: 8.0
2022-12-22 16:01:36,233 WARNING [train.py:1060] (2/4) Exclude cut with ID 6978-92210-0019-146985-0 from training. Duration: 22.48
2022-12-22 16:01:51,666 WARNING [train.py:1060] (2/4) Exclude cut with ID 7255-291500-0003-134302-0_sp0.9 from training. Duration: 29.816625
2022-12-22 16:02:12,309 WARNING [train.py:1060] (2/4) Exclude cut with ID 7255-291500-0005-134304-0_sp1.1 from training. Duration: 22.7590625
2022-12-22 16:02:17,178 WARNING [train.py:1060] (2/4) Exclude cut with ID 7357-94126-0026-15857-0 from training. Duration: 22.555
2022-12-22 16:02:20,487 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([4.5391, 3.7612, 3.7666, 4.1976, 3.9285, 3.9569, 4.6019, 1.2829],
       device='cuda:2'), covar=tensor([0.0575, 0.0549, 0.0592, 0.0551, 0.1392, 0.0805, 0.0421, 0.4303],
       device='cuda:2'), in_proj_covar=tensor([0.0212, 0.0164, 0.0153, 0.0143, 0.0221, 0.0175, 0.0176, 0.0212],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0001, 0.0001, 0.0001, 0.0002, 0.0001, 0.0002, 0.0002],
       device='cuda:2')
2022-12-22 16:02:21,446 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 4.251e+02 7.437e+02 9.296e+02 1.089e+03 1.791e+03, threshold=1.859e+03, percent-clipped=1.0
2022-12-22 16:02:30,301 INFO [train.py:894] (2/4) Epoch 4, batch 3100, loss[loss=0.2767, simple_loss=0.3412, pruned_loss=0.1061, over 18649.00 frames. ], tot_loss[loss=0.2885, simple_loss=0.3429, pruned_loss=0.117, over 3714880.58 frames. ], batch size: 60, lr: 2.56e-02, grad_scale: 8.0
2022-12-22 16:02:38,989 WARNING [train.py:1060] (2/4) Exclude cut with ID 1250-135782-0005-25975-0_sp0.9 from training. Duration: 21.688875
2022-12-22 16:03:08,720 INFO [scaling.py:679] (2/4) Whitening: num_groups=8, num_channels=96, metric=1.78 vs. limit=2.0
2022-12-22 16:03:16,163 WARNING [train.py:1060] (2/4) Exclude cut with ID 3488-85273-0038-41224-0_sp0.9 from training. Duration: 22.6
2022-12-22 16:03:44,453 INFO [train.py:894] (2/4) Epoch 4, batch 3150, loss[loss=0.3395, simple_loss=0.3744, pruned_loss=0.1523, over 18723.00 frames. ], tot_loss[loss=0.2868, simple_loss=0.3414, pruned_loss=0.1161, over 3713923.89 frames. ], batch size: 52, lr: 2.55e-02, grad_scale: 8.0
2022-12-22 16:03:53,720 WARNING [train.py:1060] (2/4) Exclude cut with ID 7357-94126-0009-15840-0 from training. Duration: 24.32
2022-12-22 16:03:57,810 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.6197, 0.5901, 1.3556, 1.4578, 1.5286, 1.4192, 1.3215, 1.0651],
       device='cuda:2'), covar=tensor([0.0651, 0.1128, 0.1014, 0.0709, 0.0556, 0.0432, 0.0753, 0.0555],
       device='cuda:2'), in_proj_covar=tensor([0.0179, 0.0213, 0.0197, 0.0203, 0.0183, 0.0169, 0.0195, 0.0167],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-22 16:04:00,579 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([2.5711, 2.3952, 1.7405, 0.8615, 1.6344, 1.9415, 1.4296, 1.8812],
       device='cuda:2'), covar=tensor([0.0601, 0.0417, 0.1185, 0.1512, 0.1189, 0.1217, 0.1528, 0.0890],
       device='cuda:2'), in_proj_covar=tensor([0.0135, 0.0165, 0.0195, 0.0189, 0.0185, 0.0173, 0.0182, 0.0180],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-22 16:04:53,194 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 3.606e+02 7.220e+02 8.309e+02 1.076e+03 2.757e+03, threshold=1.662e+03, percent-clipped=2.0
2022-12-22 16:04:53,222 WARNING [train.py:1060] (2/4) Exclude cut with ID 8291-276745-0093-13116-0_sp0.9 from training. Duration: 21.061125
2022-12-22 16:05:01,302 INFO [train.py:894] (2/4) Epoch 4, batch 3200, loss[loss=0.2911, simple_loss=0.3473, pruned_loss=0.1175, over 18573.00 frames. ], tot_loss[loss=0.2876, simple_loss=0.3426, pruned_loss=0.1163, over 3714347.64 frames. ], batch size: 57, lr: 2.55e-02, grad_scale: 8.0
2022-12-22 16:05:07,407 WARNING [train.py:1060] (2/4) Exclude cut with ID 7357-94126-0024-15855-0_sp0.9 from training. Duration: 20.32225
2022-12-22 16:05:11,015 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([2.7171, 2.4337, 1.9623, 0.7500, 1.8295, 2.0279, 1.5458, 1.9809],
       device='cuda:2'), covar=tensor([0.0569, 0.0382, 0.1216, 0.1733, 0.1152, 0.1219, 0.1442, 0.0858],
       device='cuda:2'), in_proj_covar=tensor([0.0136, 0.0166, 0.0197, 0.0194, 0.0187, 0.0173, 0.0185, 0.0182],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-22 16:05:21,109 WARNING [train.py:1060] (2/4) Exclude cut with ID 3033-130750-0096-55598-0_sp1.1 from training. Duration: 0.7545625
2022-12-22 16:05:28,823 INFO [zipformer.py:660] (2/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=13737.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 16:05:35,492 WARNING [train.py:1060] (2/4) Exclude cut with ID 4295-39940-0007-92567-0_sp0.9 from training. Duration: 23.9333125
2022-12-22 16:06:00,910 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.5095, 2.0046, 1.2907, 2.4095, 2.2978, 1.3907, 1.8776, 1.1515],
       device='cuda:2'), covar=tensor([0.1867, 0.1406, 0.1415, 0.0727, 0.1386, 0.1205, 0.1411, 0.1478],
       device='cuda:2'), in_proj_covar=tensor([0.0215, 0.0182, 0.0172, 0.0160, 0.0225, 0.0165, 0.0180, 0.0171],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-22 16:06:04,856 WARNING [train.py:1060] (2/4) Exclude cut with ID 7255-291500-0008-134307-0_sp1.1 from training. Duration: 20.17275
2022-12-22 16:06:12,740 WARNING [train.py:1060] (2/4) Exclude cut with ID 6978-92210-0019-146985-0_sp1.1 from training. Duration: 20.436375
2022-12-22 16:06:16,908 INFO [train.py:894] (2/4) Epoch 4, batch 3250, loss[loss=0.2442, simple_loss=0.3025, pruned_loss=0.09293, over 18458.00 frames. ], tot_loss[loss=0.288, simple_loss=0.3427, pruned_loss=0.1167, over 3713523.02 frames. ], batch size: 43, lr: 2.55e-02, grad_scale: 8.0
2022-12-22 16:06:37,170 INFO [scaling.py:679] (2/4) Whitening: num_groups=8, num_channels=96, metric=1.17 vs. limit=2.0
2022-12-22 16:06:47,659 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([2.9114, 5.0689, 4.9190, 2.7112, 4.8050, 3.5588, 1.1286, 3.5115],
       device='cuda:2'), covar=tensor([0.1745, 0.0573, 0.0914, 0.3165, 0.0796, 0.1050, 0.5079, 0.1570],
       device='cuda:2'), in_proj_covar=tensor([0.0124, 0.0098, 0.0150, 0.0117, 0.0107, 0.0099, 0.0145, 0.0110],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-22 16:06:52,546 INFO [scaling.py:679] (2/4) Whitening: num_groups=1, num_channels=384, metric=3.17 vs. limit=5.0
2022-12-22 16:06:59,644 INFO [zipformer.py:660] (2/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=13798.0, num_to_drop=1, layers_to_drop={0}
2022-12-22 16:07:25,858 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 5.468e+02 7.247e+02 9.411e+02 1.182e+03 2.246e+03, threshold=1.882e+03, percent-clipped=4.0
2022-12-22 16:07:33,609 INFO [train.py:894] (2/4) Epoch 4, batch 3300, loss[loss=0.3262, simple_loss=0.3718, pruned_loss=0.1403, over 18563.00 frames. ], tot_loss[loss=0.2871, simple_loss=0.342, pruned_loss=0.1161, over 3713524.90 frames. ], batch size: 177, lr: 2.54e-02, grad_scale: 8.0
2022-12-22 16:07:34,965 WARNING [train.py:1060] (2/4) Exclude cut with ID 4234-40345-0022-142709-0_sp0.9 from training. Duration: 23.1055625
2022-12-22 16:07:37,781 WARNING [train.py:1060] (2/4) Exclude cut with ID 8291-282929-0007-12994-0_sp1.1 from training. Duration: 23.5
2022-12-22 16:07:49,566 WARNING [train.py:1060] (2/4) Exclude cut with ID 7255-291500-0009-134308-0_sp0.9 from training. Duration: 26.62775
2022-12-22 16:07:57,479 INFO [scaling.py:679] (2/4) Whitening: num_groups=8, num_channels=96, metric=1.48 vs. limit=2.0
2022-12-22 16:08:02,627 WARNING [train.py:1060] (2/4) Exclude cut with ID 6951-79737-0018-132285-0 from training. Duration: 21.105
2022-12-22 16:08:07,054 WARNING [train.py:1060] (2/4) Exclude cut with ID 4511-76322-0006-80011-0_sp0.9 from training. Duration: 24.411125
2022-12-22 16:08:11,906 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([2.4481, 2.0651, 1.5347, 0.6481, 1.4019, 1.6886, 1.0477, 1.7579],
       device='cuda:2'), covar=tensor([0.0658, 0.0526, 0.1432, 0.1859, 0.1303, 0.1490, 0.1909, 0.0944],
       device='cuda:2'), in_proj_covar=tensor([0.0138, 0.0166, 0.0195, 0.0190, 0.0189, 0.0173, 0.0185, 0.0177],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-22 16:08:35,084 WARNING [train.py:1060] (2/4) Exclude cut with ID 6758-72288-0033-108368-0_sp1.1 from training. Duration: 21.263625
2022-12-22 16:08:48,538 INFO [train.py:894] (2/4) Epoch 4, batch 3350, loss[loss=0.2775, simple_loss=0.3275, pruned_loss=0.1137, over 18432.00 frames. ], tot_loss[loss=0.2856, simple_loss=0.3411, pruned_loss=0.115, over 3713120.97 frames. ], batch size: 48, lr: 2.54e-02, grad_scale: 8.0
2022-12-22 16:09:06,259 WARNING [train.py:1060] (2/4) Exclude cut with ID 4234-40345-0022-142709-0 from training. Duration: 20.795
2022-12-22 16:09:15,266 WARNING [train.py:1060] (2/4) Exclude cut with ID 7357-94126-0021-15852-0 from training. Duration: 24.76
2022-12-22 16:09:15,281 WARNING [train.py:1060] (2/4) Exclude cut with ID 3867-173237-0077-144769-0_sp0.9 from training. Duration: 22.25
2022-12-22 16:09:19,099 INFO [scaling.py:679] (2/4) Whitening: num_groups=8, num_channels=192, metric=1.93 vs. limit=2.0
2022-12-22 16:09:39,815 WARNING [train.py:1060] (2/4) Exclude cut with ID 7357-94126-0026-15857-0_sp1.1 from training. Duration: 20.5045625
2022-12-22 16:09:57,418 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 5.151e+02 8.018e+02 9.433e+02 1.110e+03 2.180e+03, threshold=1.887e+03, percent-clipped=3.0
2022-12-22 16:10:06,396 INFO [train.py:894] (2/4) Epoch 4, batch 3400, loss[loss=0.2802, simple_loss=0.3324, pruned_loss=0.114, over 18421.00 frames. ], tot_loss[loss=0.2879, simple_loss=0.3426, pruned_loss=0.1166, over 3713503.45 frames. ], batch size: 48, lr: 2.53e-02, grad_scale: 8.0
2022-12-22 16:10:39,961 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.4206, 1.5864, 1.5244, 1.7727, 1.4799, 3.1321, 1.5146, 2.0621],
       device='cuda:2'), covar=tensor([0.3871, 0.2213, 0.2215, 0.2043, 0.1588, 0.0295, 0.1758, 0.1152],
       device='cuda:2'), in_proj_covar=tensor([0.0153, 0.0129, 0.0146, 0.0129, 0.0122, 0.0099, 0.0115, 0.0111],
       device='cuda:2'), out_proj_covar=tensor([0.0004, 0.0003, 0.0004, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003],
       device='cuda:2')
2022-12-22 16:10:44,110 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([2.3616, 3.3405, 2.7405, 1.5273, 3.0158, 3.2695, 1.9805, 3.0924],
       device='cuda:2'), covar=tensor([0.1155, 0.0818, 0.1610, 0.2231, 0.0902, 0.1010, 0.2022, 0.0662],
       device='cuda:2'), in_proj_covar=tensor([0.0191, 0.0170, 0.0196, 0.0188, 0.0183, 0.0203, 0.0195, 0.0176],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-22 16:11:18,206 INFO [train.py:894] (2/4) Epoch 4, batch 3450, loss[loss=0.2935, simple_loss=0.3574, pruned_loss=0.1149, over 18686.00 frames. ], tot_loss[loss=0.2884, simple_loss=0.3429, pruned_loss=0.117, over 3714513.19 frames. ], batch size: 60, lr: 2.53e-02, grad_scale: 8.0
2022-12-22 16:11:21,847 INFO [scaling.py:679] (2/4) Whitening: num_groups=8, num_channels=192, metric=1.96 vs. limit=2.0
2022-12-22 16:12:27,570 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 4.799e+02 6.955e+02 8.936e+02 1.193e+03 2.861e+03, threshold=1.787e+03, percent-clipped=3.0
2022-12-22 16:12:35,995 INFO [train.py:894] (2/4) Epoch 4, batch 3500, loss[loss=0.3173, simple_loss=0.3667, pruned_loss=0.134, over 18541.00 frames. ], tot_loss[loss=0.2895, simple_loss=0.3439, pruned_loss=0.1175, over 3715075.85 frames. ], batch size: 99, lr: 2.53e-02, grad_scale: 8.0
2022-12-22 16:12:57,110 WARNING [train.py:1060] (2/4) Exclude cut with ID 7859-102521-0017-7548-0_sp1.1 from training. Duration: 22.2954375
2022-12-22 16:13:07,750 INFO [train.py:894] (2/4) Epoch 5, batch 0, loss[loss=0.2994, simple_loss=0.3654, pruned_loss=0.1167, over 18614.00 frames. ], tot_loss[loss=0.2994, simple_loss=0.3654, pruned_loss=0.1167, over 18614.00 frames. ], batch size: 69, lr: 2.35e-02, grad_scale: 8.0
2022-12-22 16:13:07,750 INFO [train.py:919] (2/4) Computing validation loss
2022-12-22 16:13:19,097 INFO [train.py:928] (2/4) Epoch 5, validation: loss=0.2114, simple_loss=0.3085, pruned_loss=0.05719, over 944034.00 frames. 
2022-12-22 16:13:19,098 INFO [train.py:929] (2/4) Maximum memory allocated so far is 24320MB
2022-12-22 16:14:09,327 WARNING [train.py:1060] (2/4) Exclude cut with ID 298-126791-0067-24026-0_sp0.9 from training. Duration: 21.438875
2022-12-22 16:14:14,236 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.6694, 1.5643, 1.1218, 1.8405, 1.4145, 1.3220, 1.4399, 1.7902],
       device='cuda:2'), covar=tensor([0.0955, 0.1362, 0.0897, 0.1255, 0.1185, 0.0565, 0.1255, 0.0367],
       device='cuda:2'), in_proj_covar=tensor([0.0236, 0.0211, 0.0192, 0.0287, 0.0201, 0.0184, 0.0215, 0.0156],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0003, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-22 16:14:16,488 WARNING [train.py:1060] (2/4) Exclude cut with ID 5652-39938-0025-23684-0_sp0.9 from training. Duration: 22.2055625
2022-12-22 16:14:29,301 INFO [zipformer.py:660] (2/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=14072.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 16:14:34,719 INFO [train.py:894] (2/4) Epoch 5, batch 50, loss[loss=0.2743, simple_loss=0.3509, pruned_loss=0.0989, over 18474.00 frames. ], tot_loss[loss=0.2552, simple_loss=0.3274, pruned_loss=0.09154, over 837453.38 frames. ], batch size: 64, lr: 2.35e-02, grad_scale: 8.0
2022-12-22 16:14:59,971 INFO [zipformer.py:660] (2/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=14093.0, num_to_drop=1, layers_to_drop={0}
2022-12-22 16:15:14,723 INFO [zipformer.py:660] (2/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=14103.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 16:15:33,496 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 2.796e+02 5.002e+02 6.137e+02 7.910e+02 1.797e+03, threshold=1.227e+03, percent-clipped=1.0
2022-12-22 16:15:41,885 INFO [zipformer.py:660] (2/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=14120.0, num_to_drop=1, layers_to_drop={1}
2022-12-22 16:15:50,068 INFO [train.py:894] (2/4) Epoch 5, batch 100, loss[loss=0.2577, simple_loss=0.3297, pruned_loss=0.09282, over 18663.00 frames. ], tot_loss[loss=0.2508, simple_loss=0.323, pruned_loss=0.08928, over 1475384.30 frames. ], batch size: 48, lr: 2.34e-02, grad_scale: 8.0
2022-12-22 16:16:01,100 INFO [zipformer.py:660] (2/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=14133.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 16:16:03,949 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([3.4903, 3.1792, 1.4716, 1.3781, 4.3257, 3.7088, 2.4839, 2.0469],
       device='cuda:2'), covar=tensor([0.0324, 0.0345, 0.1043, 0.0958, 0.0041, 0.0328, 0.0670, 0.0933],
       device='cuda:2'), in_proj_covar=tensor([0.0117, 0.0112, 0.0134, 0.0121, 0.0077, 0.0117, 0.0138, 0.0151],
       device='cuda:2'), out_proj_covar=tensor([1.4776e-04, 1.4204e-04, 1.6415e-04, 1.4968e-04, 9.7681e-05, 1.4228e-04,
        1.7008e-04, 1.8514e-04], device='cuda:2')
2022-12-22 16:16:46,749 INFO [zipformer.py:660] (2/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=14164.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 16:17:04,863 INFO [train.py:894] (2/4) Epoch 5, batch 150, loss[loss=0.2151, simple_loss=0.289, pruned_loss=0.07057, over 18436.00 frames. ], tot_loss[loss=0.2473, simple_loss=0.3193, pruned_loss=0.08761, over 1970811.30 frames. ], batch size: 48, lr: 2.34e-02, grad_scale: 8.0
2022-12-22 16:17:13,357 INFO [zipformer.py:660] (2/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=14181.0, num_to_drop=1, layers_to_drop={3}
2022-12-22 16:17:17,295 WARNING [train.py:1060] (2/4) Exclude cut with ID 7859-102521-0017-7548-0 from training. Duration: 24.525
2022-12-22 16:17:50,586 WARNING [train.py:1060] (2/4) Exclude cut with ID 3699-47246-0007-3408-0_sp0.9 from training. Duration: 20.26675
2022-12-22 16:18:04,150 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 2.653e+02 5.430e+02 6.354e+02 8.496e+02 1.598e+03, threshold=1.271e+03, percent-clipped=4.0
2022-12-22 16:18:04,235 WARNING [train.py:1060] (2/4) Exclude cut with ID 7859-102521-0017-7548-0_sp0.9 from training. Duration: 27.25
2022-12-22 16:18:21,215 INFO [train.py:894] (2/4) Epoch 5, batch 200, loss[loss=0.2437, simple_loss=0.3183, pruned_loss=0.08459, over 18613.00 frames. ], tot_loss[loss=0.2476, simple_loss=0.3198, pruned_loss=0.08767, over 2357527.92 frames. ], batch size: 77, lr: 2.34e-02, grad_scale: 8.0
2022-12-22 16:19:18,252 WARNING [train.py:1060] (2/4) Exclude cut with ID 6426-64292-0017-15984-0 from training. Duration: 21.68
2022-12-22 16:19:29,167 WARNING [train.py:1060] (2/4) Exclude cut with ID 4278-13270-0007-59342-0 from training. Duration: 21.6300625
2022-12-22 16:19:29,371 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([3.2116, 5.4309, 5.0946, 2.7456, 5.4461, 3.7025, 0.9635, 4.0229],
       device='cuda:2'), covar=tensor([0.1770, 0.0494, 0.0971, 0.3202, 0.0481, 0.1001, 0.5731, 0.1435],
       device='cuda:2'), in_proj_covar=tensor([0.0126, 0.0098, 0.0148, 0.0117, 0.0105, 0.0098, 0.0146, 0.0110],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-22 16:19:37,225 INFO [train.py:894] (2/4) Epoch 5, batch 250, loss[loss=0.2467, simple_loss=0.3052, pruned_loss=0.09415, over 18700.00 frames. ], tot_loss[loss=0.245, simple_loss=0.3176, pruned_loss=0.08615, over 2658108.28 frames. ], batch size: 46, lr: 2.33e-02, grad_scale: 16.0
2022-12-22 16:19:52,124 WARNING [train.py:1060] (2/4) Exclude cut with ID 4278-13270-0007-59342-0_sp0.9 from training. Duration: 24.033375
2022-12-22 16:20:11,535 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.7327, 1.5167, 1.2091, 2.0030, 1.6840, 3.2946, 1.1721, 1.4270],
       device='cuda:2'), covar=tensor([0.1052, 0.2002, 0.1521, 0.1100, 0.1649, 0.0290, 0.1643, 0.1877],
       device='cuda:2'), in_proj_covar=tensor([0.0085, 0.0092, 0.0088, 0.0087, 0.0107, 0.0078, 0.0094, 0.0088],
       device='cuda:2'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003],
       device='cuda:2')
2022-12-22 16:20:36,226 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 3.207e+02 5.412e+02 6.626e+02 8.496e+02 2.185e+03, threshold=1.325e+03, percent-clipped=4.0
2022-12-22 16:20:51,198 WARNING [train.py:1060] (2/4) Exclude cut with ID 4278-13270-0009-59344-0 from training. Duration: 22.905
2022-12-22 16:20:52,581 INFO [train.py:894] (2/4) Epoch 5, batch 300, loss[loss=0.233, simple_loss=0.3187, pruned_loss=0.07363, over 18464.00 frames. ], tot_loss[loss=0.2443, simple_loss=0.3165, pruned_loss=0.08608, over 2891923.08 frames. ], batch size: 54, lr: 2.33e-02, grad_scale: 16.0
2022-12-22 16:20:52,654 WARNING [train.py:1060] (2/4) Exclude cut with ID 5622-44585-0006-90525-0_sp1.1 from training. Duration: 23.4318125
2022-12-22 16:22:06,648 INFO [train.py:894] (2/4) Epoch 5, batch 350, loss[loss=0.2486, simple_loss=0.3334, pruned_loss=0.08186, over 18561.00 frames. ], tot_loss[loss=0.2457, simple_loss=0.3176, pruned_loss=0.08688, over 3074308.16 frames. ], batch size: 57, lr: 2.33e-02, grad_scale: 16.0
2022-12-22 16:22:31,894 INFO [zipformer.py:660] (2/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=14393.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 16:22:48,894 WARNING [train.py:1060] (2/4) Exclude cut with ID 4278-13270-0009-59344-0_sp1.1 from training. Duration: 20.82275
2022-12-22 16:22:50,572 WARNING [train.py:1060] (2/4) Exclude cut with ID 4278-13270-0009-59344-0_sp0.9 from training. Duration: 25.45
2022-12-22 16:23:04,208 INFO [zipformer.py:660] (2/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=14413.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 16:23:06,651 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 3.028e+02 5.505e+02 6.625e+02 8.750e+02 2.038e+03, threshold=1.325e+03, percent-clipped=6.0
2022-12-22 16:23:23,907 INFO [train.py:894] (2/4) Epoch 5, batch 400, loss[loss=0.2007, simple_loss=0.2779, pruned_loss=0.06181, over 18690.00 frames. ], tot_loss[loss=0.2466, simple_loss=0.3184, pruned_loss=0.08743, over 3216855.11 frames. ], batch size: 46, lr: 2.32e-02, grad_scale: 16.0
2022-12-22 16:23:27,381 INFO [zipformer.py:660] (2/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=14428.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 16:23:40,419 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.5011, 0.8944, 1.6626, 2.5342, 1.7431, 2.1507, 0.7884, 1.6569],
       device='cuda:2'), covar=tensor([0.2478, 0.2982, 0.1860, 0.0915, 0.1753, 0.1470, 0.3175, 0.1791],
       device='cuda:2'), in_proj_covar=tensor([0.0106, 0.0115, 0.0123, 0.0092, 0.0104, 0.0122, 0.0138, 0.0106],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0002, 0.0003, 0.0002, 0.0003, 0.0003, 0.0003, 0.0003],
       device='cuda:2')
2022-12-22 16:23:46,115 INFO [zipformer.py:660] (2/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=14441.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 16:23:53,049 WARNING [train.py:1060] (2/4) Exclude cut with ID 5622-44585-0006-90525-0 from training. Duration: 25.775
2022-12-22 16:24:13,347 INFO [zipformer.py:660] (2/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=14459.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 16:24:15,838 WARNING [train.py:1060] (2/4) Exclude cut with ID 2929-85685-0071-62375-0_sp0.9 from training. Duration: 22.25
2022-12-22 16:24:36,142 INFO [zipformer.py:660] (2/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=14474.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 16:24:38,462 INFO [train.py:894] (2/4) Epoch 5, batch 450, loss[loss=0.3071, simple_loss=0.3625, pruned_loss=0.1259, over 18644.00 frames. ], tot_loss[loss=0.2491, simple_loss=0.3205, pruned_loss=0.0889, over 3327146.76 frames. ], batch size: 176, lr: 2.32e-02, grad_scale: 16.0
2022-12-22 16:24:38,635 INFO [zipformer.py:660] (2/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=14476.0, num_to_drop=1, layers_to_drop={1}
2022-12-22 16:24:42,678 WARNING [train.py:1060] (2/4) Exclude cut with ID 3972-170212-0014-23379-0 from training. Duration: 26.205
2022-12-22 16:24:54,866 INFO [zipformer.py:660] (2/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=14487.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 16:25:00,439 WARNING [train.py:1060] (2/4) Exclude cut with ID 5239-32139-0047-9341-0_sp0.9 from training. Duration: 30.1555625
2022-12-22 16:25:06,090 WARNING [train.py:1060] (2/4) Exclude cut with ID 1265-135635-0050-6781-0_sp0.9 from training. Duration: 21.8333125
2022-12-22 16:25:08,551 INFO [scaling.py:679] (2/4) Whitening: num_groups=8, num_channels=192, metric=1.95 vs. limit=2.0
2022-12-22 16:25:16,950 WARNING [train.py:1060] (2/4) Exclude cut with ID 1914-133440-0024-94914-0_sp1.1 from training. Duration: 20.6545625
2022-12-22 16:25:35,595 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 3.434e+02 5.649e+02 6.496e+02 8.214e+02 2.350e+03, threshold=1.299e+03, percent-clipped=4.0
2022-12-22 16:25:51,854 INFO [train.py:894] (2/4) Epoch 5, batch 500, loss[loss=0.2475, simple_loss=0.3296, pruned_loss=0.08265, over 18514.00 frames. ], tot_loss[loss=0.25, simple_loss=0.3215, pruned_loss=0.08921, over 3413241.40 frames. ], batch size: 58, lr: 2.31e-02, grad_scale: 16.0
2022-12-22 16:25:59,369 WARNING [train.py:1060] (2/4) Exclude cut with ID 7395-89880-0045-39920-0_sp0.9 from training. Duration: 20.52225
2022-12-22 16:26:17,867 WARNING [train.py:1060] (2/4) Exclude cut with ID 3972-170212-0014-23379-0_sp0.9 from training. Duration: 29.1166875
2022-12-22 16:26:24,215 INFO [zipformer.py:660] (2/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=14548.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 16:27:01,232 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.3727, 1.6700, 1.4466, 1.5931, 1.4059, 3.1325, 1.3640, 2.1217],
       device='cuda:2'), covar=tensor([0.3635, 0.1893, 0.2035, 0.2032, 0.1456, 0.0226, 0.1824, 0.1003],
       device='cuda:2'), in_proj_covar=tensor([0.0149, 0.0129, 0.0144, 0.0128, 0.0119, 0.0097, 0.0115, 0.0109],
       device='cuda:2'), out_proj_covar=tensor([0.0004, 0.0004, 0.0004, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003],
       device='cuda:2')
2022-12-22 16:27:06,688 INFO [train.py:894] (2/4) Epoch 5, batch 550, loss[loss=0.2111, simple_loss=0.2803, pruned_loss=0.07098, over 18407.00 frames. ], tot_loss[loss=0.2508, simple_loss=0.3219, pruned_loss=0.08984, over 3479996.46 frames. ], batch size: 42, lr: 2.31e-02, grad_scale: 16.0
2022-12-22 16:27:20,107 WARNING [train.py:1060] (2/4) Exclude cut with ID 543-133211-0007-59831-0_sp0.9 from training. Duration: 21.388875
2022-12-22 16:27:56,671 WARNING [train.py:1060] (2/4) Exclude cut with ID 1914-133440-0024-94914-0 from training. Duration: 22.72
2022-12-22 16:27:57,905 WARNING [train.py:1060] (2/4) Exclude cut with ID 1914-133440-0031-94921-0_sp0.9 from training. Duration: 22.7444375
2022-12-22 16:28:07,474 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 3.606e+02 5.759e+02 6.795e+02 8.562e+02 2.141e+03, threshold=1.359e+03, percent-clipped=4.0
2022-12-22 16:28:23,072 INFO [train.py:894] (2/4) Epoch 5, batch 600, loss[loss=0.2678, simple_loss=0.3413, pruned_loss=0.0972, over 18698.00 frames. ], tot_loss[loss=0.2526, simple_loss=0.3235, pruned_loss=0.09085, over 3532712.84 frames. ], batch size: 69, lr: 2.31e-02, grad_scale: 8.0
2022-12-22 16:28:44,130 WARNING [train.py:1060] (2/4) Exclude cut with ID 4133-6541-0027-40495-0_sp1.1 from training. Duration: 0.9681875
2022-12-22 16:28:46,980 WARNING [train.py:1060] (2/4) Exclude cut with ID 6330-62851-0022-91297-0_sp0.9 from training. Duration: 22.3166875
2022-12-22 16:28:51,311 WARNING [train.py:1060] (2/4) Exclude cut with ID 543-133212-0015-59917-0_sp0.9 from training. Duration: 21.8166875
2022-12-22 16:29:38,228 INFO [train.py:894] (2/4) Epoch 5, batch 650, loss[loss=0.3049, simple_loss=0.3732, pruned_loss=0.1183, over 18570.00 frames. ], tot_loss[loss=0.2527, simple_loss=0.3238, pruned_loss=0.0908, over 3572424.09 frames. ], batch size: 57, lr: 2.30e-02, grad_scale: 8.0
2022-12-22 16:30:34,675 WARNING [train.py:1060] (2/4) Exclude cut with ID 4957-30119-0041-23990-0_sp0.9 from training. Duration: 20.22775
2022-12-22 16:30:38,993 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 3.231e+02 5.539e+02 6.191e+02 8.462e+02 2.551e+03, threshold=1.238e+03, percent-clipped=2.0
2022-12-22 16:30:54,116 INFO [train.py:894] (2/4) Epoch 5, batch 700, loss[loss=0.2292, simple_loss=0.3076, pruned_loss=0.0754, over 18686.00 frames. ], tot_loss[loss=0.253, simple_loss=0.3242, pruned_loss=0.0909, over 3603733.24 frames. ], batch size: 48, lr: 2.30e-02, grad_scale: 8.0
2022-12-22 16:30:57,215 INFO [zipformer.py:660] (2/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=14728.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 16:31:09,552 INFO [scaling.py:679] (2/4) Whitening: num_groups=1, num_channels=384, metric=3.29 vs. limit=5.0
2022-12-22 16:31:15,052 INFO [zipformer.py:660] (2/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=14740.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 16:31:19,722 WARNING [train.py:1060] (2/4) Exclude cut with ID 5239-32139-0047-9341-0_sp1.1 from training. Duration: 24.67275
2022-12-22 16:31:44,846 INFO [zipformer.py:660] (2/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=14759.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 16:31:46,360 WARNING [train.py:1060] (2/4) Exclude cut with ID 3082-165428-0081-50734-0_sp0.9 from training. Duration: 21.8055625
2022-12-22 16:31:50,340 INFO [scaling.py:679] (2/4) Whitening: num_groups=8, num_channels=192, metric=1.90 vs. limit=2.0
2022-12-22 16:32:00,593 INFO [zipformer.py:660] (2/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=14769.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 16:32:10,898 INFO [train.py:894] (2/4) Epoch 5, batch 750, loss[loss=0.2609, simple_loss=0.3398, pruned_loss=0.09102, over 18728.00 frames. ], tot_loss[loss=0.2533, simple_loss=0.3249, pruned_loss=0.09091, over 3629042.04 frames. ], batch size: 54, lr: 2.30e-02, grad_scale: 8.0
2022-12-22 16:32:11,100 INFO [zipformer.py:660] (2/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=14776.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 16:32:11,320 INFO [zipformer.py:660] (2/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=14776.0, num_to_drop=1, layers_to_drop={1}
2022-12-22 16:32:25,257 WARNING [train.py:1060] (2/4) Exclude cut with ID 3340-169293-0054-76830-0_sp0.9 from training. Duration: 22.6666875
2022-12-22 16:32:49,042 INFO [zipformer.py:660] (2/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=14801.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 16:32:57,985 INFO [zipformer.py:660] (2/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=14807.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 16:33:11,565 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 3.661e+02 5.575e+02 6.678e+02 8.427e+02 1.558e+03, threshold=1.336e+03, percent-clipped=4.0
2022-12-22 16:33:20,457 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([0.3451, 1.6544, 2.0778, 0.2622, 1.1529, 2.3616, 1.4198, 1.6415],
       device='cuda:2'), covar=tensor([0.0782, 0.0362, 0.0296, 0.0569, 0.0459, 0.0232, 0.0400, 0.0505],
       device='cuda:2'), in_proj_covar=tensor([0.0109, 0.0122, 0.0083, 0.0112, 0.0107, 0.0084, 0.0114, 0.0099],
       device='cuda:2'), out_proj_covar=tensor([1.1344e-04, 1.2849e-04, 8.8293e-05, 1.1537e-04, 1.0982e-04, 8.8515e-05,
        1.2140e-04, 1.0416e-04], device='cuda:2')
2022-12-22 16:33:22,943 INFO [zipformer.py:660] (2/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=14824.0, num_to_drop=1, layers_to_drop={0}
2022-12-22 16:33:25,647 INFO [train.py:894] (2/4) Epoch 5, batch 800, loss[loss=0.3045, simple_loss=0.3666, pruned_loss=0.1212, over 18629.00 frames. ], tot_loss[loss=0.2533, simple_loss=0.3243, pruned_loss=0.09117, over 3649005.07 frames. ], batch size: 69, lr: 2.29e-02, grad_scale: 8.0
2022-12-22 16:33:27,098 WARNING [train.py:1060] (2/4) Exclude cut with ID 2411-132532-0017-82279-0_sp1.1 from training. Duration: 0.9681875
2022-12-22 16:33:50,306 WARNING [train.py:1060] (2/4) Exclude cut with ID 6330-62850-0007-91323-0 from training. Duration: 22.485
2022-12-22 16:33:52,008 INFO [zipformer.py:660] (2/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=14843.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 16:34:17,070 INFO [scaling.py:679] (2/4) Whitening: num_groups=1, num_channels=384, metric=3.00 vs. limit=5.0
2022-12-22 16:34:28,559 WARNING [train.py:1060] (2/4) Exclude cut with ID 3972-170212-0014-23379-0_sp1.1 from training. Duration: 23.82275
2022-12-22 16:34:37,471 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.1575, 0.9084, 1.4086, 2.1256, 1.3447, 2.1035, 0.7237, 1.3188],
       device='cuda:2'), covar=tensor([0.1931, 0.1974, 0.1285, 0.0631, 0.1472, 0.1011, 0.2328, 0.1488],
       device='cuda:2'), in_proj_covar=tensor([0.0105, 0.0116, 0.0120, 0.0095, 0.0103, 0.0122, 0.0139, 0.0104],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0003, 0.0003, 0.0002, 0.0003, 0.0003, 0.0003, 0.0003],
       device='cuda:2')
2022-12-22 16:34:41,451 INFO [train.py:894] (2/4) Epoch 5, batch 850, loss[loss=0.2802, simple_loss=0.3489, pruned_loss=0.1058, over 18541.00 frames. ], tot_loss[loss=0.2545, simple_loss=0.3255, pruned_loss=0.09179, over 3664152.14 frames. ], batch size: 55, lr: 2.29e-02, grad_scale: 8.0
2022-12-22 16:34:41,479 WARNING [train.py:1060] (2/4) Exclude cut with ID 4860-13185-0032-76709-0 from training. Duration: 20.77
2022-12-22 16:34:48,839 WARNING [train.py:1060] (2/4) Exclude cut with ID 6426-64292-0017-15984-0_sp0.9 from training. Duration: 24.088875
2022-12-22 16:35:19,373 WARNING [train.py:1060] (2/4) Exclude cut with ID 6330-62850-0007-91323-0_sp1.1 from training. Duration: 20.4409375
2022-12-22 16:35:44,371 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 3.622e+02 5.143e+02 7.004e+02 8.582e+02 1.972e+03, threshold=1.401e+03, percent-clipped=5.0
2022-12-22 16:35:58,972 INFO [train.py:894] (2/4) Epoch 5, batch 900, loss[loss=0.2775, simple_loss=0.3425, pruned_loss=0.1063, over 18575.00 frames. ], tot_loss[loss=0.2538, simple_loss=0.325, pruned_loss=0.0913, over 3675448.20 frames. ], batch size: 57, lr: 2.29e-02, grad_scale: 8.0
2022-12-22 16:36:36,101 WARNING [train.py:1060] (2/4) Exclude cut with ID 2929-85685-0062-62366-0_sp0.9 from training. Duration: 22.511125
2022-12-22 16:36:36,120 WARNING [train.py:1060] (2/4) Exclude cut with ID 7395-89880-0031-39906-0 from training. Duration: 20.675
2022-12-22 16:37:14,679 INFO [train.py:894] (2/4) Epoch 5, batch 950, loss[loss=0.275, simple_loss=0.3478, pruned_loss=0.1011, over 18600.00 frames. ], tot_loss[loss=0.2537, simple_loss=0.3247, pruned_loss=0.09133, over 3683660.14 frames. ], batch size: 56, lr: 2.28e-02, grad_scale: 8.0
2022-12-22 16:37:39,870 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.4256, 1.5316, 1.1521, 1.9698, 1.4236, 3.2137, 1.5037, 1.5005],
       device='cuda:2'), covar=tensor([0.1043, 0.1688, 0.1417, 0.0913, 0.1544, 0.0269, 0.1281, 0.1620],
       device='cuda:2'), in_proj_covar=tensor([0.0083, 0.0088, 0.0087, 0.0083, 0.0103, 0.0075, 0.0092, 0.0084],
       device='cuda:2'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003],
       device='cuda:2')
2022-12-22 16:38:16,017 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 3.793e+02 5.959e+02 7.329e+02 8.646e+02 1.591e+03, threshold=1.466e+03, percent-clipped=3.0
2022-12-22 16:38:16,130 WARNING [train.py:1060] (2/4) Exclude cut with ID 6330-62850-0007-91323-0_sp0.9 from training. Duration: 24.9833125
2022-12-22 16:38:30,749 INFO [train.py:894] (2/4) Epoch 5, batch 1000, loss[loss=0.258, simple_loss=0.3364, pruned_loss=0.08984, over 18527.00 frames. ], tot_loss[loss=0.2524, simple_loss=0.324, pruned_loss=0.09041, over 3690330.75 frames. ], batch size: 52, lr: 2.28e-02, grad_scale: 8.0
2022-12-22 16:38:49,241 WARNING [train.py:1060] (2/4) Exclude cut with ID 5239-32139-0047-9341-0 from training. Duration: 27.14
2022-12-22 16:38:53,484 INFO [scaling.py:679] (2/4) Whitening: num_groups=8, num_channels=96, metric=1.30 vs. limit=2.0
2022-12-22 16:39:04,489 WARNING [train.py:1060] (2/4) Exclude cut with ID 2929-85685-0044-62348-0 from training. Duration: 22.44
2022-12-22 16:39:36,625 INFO [zipformer.py:660] (2/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=15069.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 16:39:36,868 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([2.2051, 2.5079, 1.2576, 2.7862, 2.7418, 2.3584, 3.8979, 2.5081],
       device='cuda:2'), covar=tensor([0.0775, 0.1310, 0.2275, 0.1809, 0.1291, 0.0712, 0.0488, 0.0857],
       device='cuda:2'), in_proj_covar=tensor([0.0159, 0.0171, 0.0208, 0.0253, 0.0205, 0.0162, 0.0169, 0.0174],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-22 16:39:46,856 INFO [train.py:894] (2/4) Epoch 5, batch 1050, loss[loss=0.2608, simple_loss=0.3356, pruned_loss=0.09299, over 18524.00 frames. ], tot_loss[loss=0.2508, simple_loss=0.3223, pruned_loss=0.08963, over 3695693.98 frames. ], batch size: 58, lr: 2.28e-02, grad_scale: 8.0
2022-12-22 16:40:18,178 INFO [zipformer.py:660] (2/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=15096.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 16:40:22,547 WARNING [train.py:1060] (2/4) Exclude cut with ID 2929-85685-0060-62364-0_sp0.9 from training. Duration: 21.361125
2022-12-22 16:40:29,979 WARNING [train.py:1060] (2/4) Exclude cut with ID 2929-85685-0079-62383-0_sp1.1 from training. Duration: 27.0318125
2022-12-22 16:40:39,232 WARNING [train.py:1060] (2/4) Exclude cut with ID 5622-44585-0006-90525-0_sp0.9 from training. Duration: 28.638875
2022-12-22 16:40:47,972 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 3.160e+02 5.503e+02 6.877e+02 7.962e+02 1.573e+03, threshold=1.375e+03, percent-clipped=2.0
2022-12-22 16:40:49,402 INFO [zipformer.py:660] (2/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=15117.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 16:40:55,288 WARNING [train.py:1060] (2/4) Exclude cut with ID 3340-169293-0054-76830-0 from training. Duration: 20.4
2022-12-22 16:41:03,201 INFO [train.py:894] (2/4) Epoch 5, batch 1100, loss[loss=0.2729, simple_loss=0.3451, pruned_loss=0.1003, over 18731.00 frames. ], tot_loss[loss=0.2499, simple_loss=0.3217, pruned_loss=0.08899, over 3700031.76 frames. ], batch size: 54, lr: 2.27e-02, grad_scale: 8.0
2022-12-22 16:41:27,808 WARNING [train.py:1060] (2/4) Exclude cut with ID 2929-85685-0071-62375-0 from training. Duration: 20.025
2022-12-22 16:41:29,183 WARNING [train.py:1060] (2/4) Exclude cut with ID 2364-131735-0112-64612-0_sp0.9 from training. Duration: 20.488875
2022-12-22 16:41:29,430 INFO [zipformer.py:660] (2/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=15143.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 16:41:33,516 WARNING [train.py:1060] (2/4) Exclude cut with ID 2929-85685-0079-62383-0 from training. Duration: 29.735
2022-12-22 16:42:15,247 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([2.6588, 2.2238, 1.7123, 0.9578, 3.3502, 2.7819, 2.0138, 1.5346],
       device='cuda:2'), covar=tensor([0.0384, 0.0493, 0.0816, 0.1105, 0.0078, 0.0339, 0.0746, 0.1411],
       device='cuda:2'), in_proj_covar=tensor([0.0115, 0.0106, 0.0131, 0.0120, 0.0071, 0.0114, 0.0134, 0.0144],
       device='cuda:2'), out_proj_covar=tensor([1.4558e-04, 1.3467e-04, 1.6049e-04, 1.4780e-04, 9.1160e-05, 1.3880e-04,
        1.6541e-04, 1.7703e-04], device='cuda:2')
2022-12-22 16:42:17,696 INFO [train.py:894] (2/4) Epoch 5, batch 1150, loss[loss=0.2516, simple_loss=0.3124, pruned_loss=0.09539, over 18522.00 frames. ], tot_loss[loss=0.2504, simple_loss=0.3224, pruned_loss=0.08917, over 3703361.94 frames. ], batch size: 47, lr: 2.27e-02, grad_scale: 8.0
2022-12-22 16:42:35,790 INFO [scaling.py:679] (2/4) Whitening: num_groups=8, num_channels=96, metric=1.28 vs. limit=2.0
2022-12-22 16:42:40,960 INFO [zipformer.py:660] (2/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=15191.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 16:42:47,360 INFO [zipformer.py:660] (2/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=15195.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 16:42:54,448 WARNING [train.py:1060] (2/4) Exclude cut with ID 7276-92427-0014-12983-0_sp0.9 from training. Duration: 21.3055625
2022-12-22 16:42:56,540 WARNING [train.py:1060] (2/4) Exclude cut with ID 1025-75365-0008-79168-0_sp0.9 from training. Duration: 22.0666875
2022-12-22 16:43:18,667 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 3.144e+02 5.609e+02 7.018e+02 8.970e+02 1.678e+03, threshold=1.404e+03, percent-clipped=4.0
2022-12-22 16:43:33,725 INFO [train.py:894] (2/4) Epoch 5, batch 1200, loss[loss=0.2848, simple_loss=0.3466, pruned_loss=0.1115, over 18513.00 frames. ], tot_loss[loss=0.2507, simple_loss=0.3231, pruned_loss=0.0892, over 3704872.58 frames. ], batch size: 58, lr: 2.27e-02, grad_scale: 8.0
2022-12-22 16:44:20,439 INFO [zipformer.py:660] (2/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=15256.0, num_to_drop=1, layers_to_drop={1}
2022-12-22 16:44:37,689 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([3.2134, 2.3592, 1.8372, 2.6797, 2.0889, 2.2285, 2.3948, 3.5009],
       device='cuda:2'), covar=tensor([0.1152, 0.1831, 0.1136, 0.2213, 0.2130, 0.0651, 0.1882, 0.0360],
       device='cuda:2'), in_proj_covar=tensor([0.0233, 0.0214, 0.0193, 0.0293, 0.0204, 0.0183, 0.0220, 0.0155],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0003, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-22 16:44:41,543 WARNING [train.py:1060] (2/4) Exclude cut with ID 2929-85685-0062-62366-0 from training. Duration: 20.26
2022-12-22 16:44:51,036 INFO [train.py:894] (2/4) Epoch 5, batch 1250, loss[loss=0.2627, simple_loss=0.3426, pruned_loss=0.09144, over 18563.00 frames. ], tot_loss[loss=0.2497, simple_loss=0.3222, pruned_loss=0.08863, over 3707341.95 frames. ], batch size: 57, lr: 2.26e-02, grad_scale: 8.0
2022-12-22 16:44:56,000 WARNING [train.py:1060] (2/4) Exclude cut with ID 5239-32139-0030-9324-0_sp0.9 from training. Duration: 21.3444375
2022-12-22 16:44:56,357 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.3849, 0.9527, 1.3890, 2.2983, 1.5490, 2.1988, 0.8319, 1.5113],
       device='cuda:2'), covar=tensor([0.1985, 0.2319, 0.1651, 0.0662, 0.1540, 0.1186, 0.2389, 0.1517],
       device='cuda:2'), in_proj_covar=tensor([0.0105, 0.0119, 0.0123, 0.0096, 0.0104, 0.0123, 0.0139, 0.0106],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0003, 0.0003, 0.0002, 0.0003, 0.0003, 0.0003, 0.0003],
       device='cuda:2')
2022-12-22 16:45:26,995 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.2969, 1.6815, 1.4449, 1.8223, 1.5938, 3.2331, 1.5119, 2.1763],
       device='cuda:2'), covar=tensor([0.3586, 0.1852, 0.1945, 0.1639, 0.1234, 0.0196, 0.1461, 0.0862],
       device='cuda:2'), in_proj_covar=tensor([0.0149, 0.0126, 0.0141, 0.0126, 0.0118, 0.0096, 0.0108, 0.0107],
       device='cuda:2'), out_proj_covar=tensor([0.0004, 0.0004, 0.0004, 0.0003, 0.0003, 0.0002, 0.0003, 0.0003],
       device='cuda:2')
2022-12-22 16:45:30,530 INFO [scaling.py:679] (2/4) Whitening: num_groups=1, num_channels=384, metric=4.74 vs. limit=5.0
2022-12-22 16:45:52,316 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 3.243e+02 5.439e+02 6.876e+02 8.678e+02 2.167e+03, threshold=1.375e+03, percent-clipped=2.0
2022-12-22 16:45:53,863 WARNING [train.py:1060] (2/4) Exclude cut with ID 497-129325-0061-62254-0_sp1.1 from training. Duration: 0.97725
2022-12-22 16:46:07,165 INFO [train.py:894] (2/4) Epoch 5, batch 1300, loss[loss=0.2419, simple_loss=0.3184, pruned_loss=0.0827, over 18579.00 frames. ], tot_loss[loss=0.2468, simple_loss=0.3198, pruned_loss=0.08688, over 3709554.04 frames. ], batch size: 51, lr: 2.26e-02, grad_scale: 8.0
2022-12-22 16:46:35,732 WARNING [train.py:1060] (2/4) Exclude cut with ID 7395-89880-0031-39906-0_sp0.9 from training. Duration: 22.97225
2022-12-22 16:47:06,761 WARNING [train.py:1060] (2/4) Exclude cut with ID 7395-89880-0047-39922-0_sp0.9 from training. Duration: 21.97775
2022-12-22 16:47:16,133 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.7466, 1.4752, 1.1771, 0.2194, 1.2507, 1.5402, 1.2129, 1.6306],
       device='cuda:2'), covar=tensor([0.0634, 0.0464, 0.0957, 0.1539, 0.0868, 0.1395, 0.1516, 0.0545],
       device='cuda:2'), in_proj_covar=tensor([0.0136, 0.0161, 0.0193, 0.0186, 0.0183, 0.0169, 0.0179, 0.0172],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-22 16:47:21,169 WARNING [train.py:1060] (2/4) Exclude cut with ID 1112-1043-0006-89194-0_sp0.9 from training. Duration: 21.8333125
2022-12-22 16:47:24,004 INFO [train.py:894] (2/4) Epoch 5, batch 1350, loss[loss=0.2478, simple_loss=0.3323, pruned_loss=0.08167, over 18597.00 frames. ], tot_loss[loss=0.2465, simple_loss=0.3195, pruned_loss=0.08671, over 3710080.95 frames. ], batch size: 56, lr: 2.26e-02, grad_scale: 8.0
2022-12-22 16:47:33,411 WARNING [train.py:1060] (2/4) Exclude cut with ID 1914-133440-0031-94921-0 from training. Duration: 20.47
2022-12-22 16:47:54,693 INFO [zipformer.py:660] (2/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=15396.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 16:48:24,645 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 3.202e+02 5.523e+02 6.743e+02 8.396e+02 3.038e+03, threshold=1.349e+03, percent-clipped=3.0
2022-12-22 16:48:35,631 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.7565, 1.0816, 1.9574, 3.1121, 2.1209, 2.2712, 0.9558, 2.1256],
       device='cuda:2'), covar=tensor([0.1794, 0.2234, 0.1569, 0.0589, 0.1286, 0.1469, 0.2518, 0.1246],
       device='cuda:2'), in_proj_covar=tensor([0.0102, 0.0115, 0.0118, 0.0093, 0.0101, 0.0120, 0.0134, 0.0103],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0003, 0.0003, 0.0002, 0.0003, 0.0003, 0.0003, 0.0003],
       device='cuda:2')
2022-12-22 16:48:38,846 WARNING [train.py:1060] (2/4) Exclude cut with ID 7395-89880-0037-39912-0_sp0.9 from training. Duration: 20.67225
2022-12-22 16:48:40,239 INFO [train.py:894] (2/4) Epoch 5, batch 1400, loss[loss=0.2586, simple_loss=0.3316, pruned_loss=0.09276, over 18667.00 frames. ], tot_loss[loss=0.2464, simple_loss=0.3197, pruned_loss=0.08655, over 3711805.30 frames. ], batch size: 62, lr: 2.25e-02, grad_scale: 8.0
2022-12-22 16:48:57,565 WARNING [train.py:1060] (2/4) Exclude cut with ID 1914-133440-0024-94914-0_sp0.9 from training. Duration: 25.2444375
2022-12-22 16:49:07,186 INFO [zipformer.py:660] (2/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=15444.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 16:49:21,182 WARNING [train.py:1060] (2/4) Exclude cut with ID 3340-169293-0021-76797-0_sp0.9 from training. Duration: 21.1445
2022-12-22 16:49:55,324 INFO [train.py:894] (2/4) Epoch 5, batch 1450, loss[loss=0.2098, simple_loss=0.2895, pruned_loss=0.06506, over 18429.00 frames. ], tot_loss[loss=0.2465, simple_loss=0.3193, pruned_loss=0.08689, over 3712886.25 frames. ], batch size: 48, lr: 2.25e-02, grad_scale: 8.0
2022-12-22 16:49:57,640 INFO [scaling.py:679] (2/4) Whitening: num_groups=1, num_channels=384, metric=3.28 vs. limit=5.0
2022-12-22 16:50:15,656 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.5677, 2.3248, 1.8708, 3.1746, 2.7399, 1.5222, 2.3009, 1.3174],
       device='cuda:2'), covar=tensor([0.2005, 0.1842, 0.1350, 0.0671, 0.1763, 0.1320, 0.1460, 0.1566],
       device='cuda:2'), in_proj_covar=tensor([0.0213, 0.0186, 0.0175, 0.0164, 0.0225, 0.0167, 0.0182, 0.0171],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-22 16:50:33,829 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.1415, 1.5097, 1.1657, 1.7769, 2.2596, 1.1987, 1.7686, 0.9421],
       device='cuda:2'), covar=tensor([0.2759, 0.2655, 0.2140, 0.1524, 0.1732, 0.2042, 0.1779, 0.2555],
       device='cuda:2'), in_proj_covar=tensor([0.0213, 0.0187, 0.0176, 0.0165, 0.0225, 0.0168, 0.0182, 0.0171],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-22 16:50:37,731 WARNING [train.py:1060] (2/4) Exclude cut with ID 2929-85685-0079-62383-0_sp0.9 from training. Duration: 33.038875
2022-12-22 16:50:56,240 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 3.500e+02 5.658e+02 7.097e+02 9.527e+02 2.232e+03, threshold=1.419e+03, percent-clipped=9.0
2022-12-22 16:51:11,642 INFO [train.py:894] (2/4) Epoch 5, batch 1500, loss[loss=0.2676, simple_loss=0.3392, pruned_loss=0.09798, over 18470.00 frames. ], tot_loss[loss=0.2467, simple_loss=0.3196, pruned_loss=0.08686, over 3713392.06 frames. ], batch size: 64, lr: 2.25e-02, grad_scale: 8.0
2022-12-22 16:51:16,177 WARNING [train.py:1060] (2/4) Exclude cut with ID 6426-64291-0000-16059-0_sp0.9 from training. Duration: 20.0944375
2022-12-22 16:51:31,144 WARNING [train.py:1060] (2/4) Exclude cut with ID 2929-85685-0044-62348-0_sp1.1 from training. Duration: 20.4
2022-12-22 16:51:37,882 INFO [scaling.py:679] (2/4) Whitening: num_groups=8, num_channels=192, metric=1.96 vs. limit=2.0
2022-12-22 16:51:38,550 WARNING [train.py:1060] (2/4) Exclude cut with ID 6330-62851-0022-91297-0 from training. Duration: 20.085
2022-12-22 16:51:49,103 WARNING [train.py:1060] (2/4) Exclude cut with ID 4860-13185-0032-76709-0_sp0.9 from training. Duration: 23.07775
2022-12-22 16:51:49,250 INFO [zipformer.py:660] (2/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=15551.0, num_to_drop=1, layers_to_drop={0}
2022-12-22 16:52:27,540 INFO [train.py:894] (2/4) Epoch 5, batch 1550, loss[loss=0.1992, simple_loss=0.279, pruned_loss=0.05968, over 18671.00 frames. ], tot_loss[loss=0.2456, simple_loss=0.3189, pruned_loss=0.08621, over 3713355.12 frames. ], batch size: 46, lr: 2.24e-02, grad_scale: 8.0
2022-12-22 16:52:39,029 WARNING [train.py:1060] (2/4) Exclude cut with ID 2929-85685-0044-62348-0_sp0.9 from training. Duration: 24.9333125
2022-12-22 16:53:25,399 WARNING [train.py:1060] (2/4) Exclude cut with ID 5118-111612-0016-124680-0_sp0.9 from training. Duration: 20.388875
2022-12-22 16:53:28,048 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 3.175e+02 5.012e+02 6.250e+02 7.299e+02 1.571e+03, threshold=1.250e+03, percent-clipped=1.0
2022-12-22 16:53:30,864 WARNING [train.py:1060] (2/4) Exclude cut with ID 432-122774-0017-62487-0_sp1.1 from training. Duration: 20.3590625
2022-12-22 16:53:43,429 INFO [train.py:894] (2/4) Epoch 5, batch 1600, loss[loss=0.2846, simple_loss=0.3479, pruned_loss=0.1107, over 18726.00 frames. ], tot_loss[loss=0.2456, simple_loss=0.3194, pruned_loss=0.08592, over 3714446.71 frames. ], batch size: 60, lr: 2.24e-02, grad_scale: 8.0
2022-12-22 16:54:11,124 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([2.9148, 5.0388, 4.8554, 2.3795, 5.0767, 3.7452, 0.9837, 3.7275],
       device='cuda:2'), covar=tensor([0.2013, 0.0596, 0.1174, 0.3793, 0.0584, 0.1004, 0.6224, 0.1487],
       device='cuda:2'), in_proj_covar=tensor([0.0125, 0.0101, 0.0143, 0.0118, 0.0105, 0.0097, 0.0143, 0.0106],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-22 16:54:40,081 WARNING [train.py:1060] (2/4) Exclude cut with ID 3557-8342-0013-54691-0_sp1.1 from training. Duration: 0.836375
2022-12-22 16:54:59,788 INFO [train.py:894] (2/4) Epoch 5, batch 1650, loss[loss=0.2693, simple_loss=0.3382, pruned_loss=0.1002, over 18513.00 frames. ], tot_loss[loss=0.2477, simple_loss=0.3205, pruned_loss=0.08742, over 3713573.82 frames. ], batch size: 52, lr: 2.24e-02, grad_scale: 8.0
2022-12-22 16:55:23,206 WARNING [train.py:1060] (2/4) Exclude cut with ID 8565-290391-0049-67394-0_sp0.9 from training. Duration: 21.3166875
2022-12-22 16:55:54,789 WARNING [train.py:1060] (2/4) Exclude cut with ID 6533-399-0029-104863-0_sp0.9 from training. Duration: 22.1055625
2022-12-22 16:56:01,467 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 4.196e+02 6.902e+02 8.534e+02 1.056e+03 2.303e+03, threshold=1.707e+03, percent-clipped=12.0
2022-12-22 16:56:06,050 WARNING [train.py:1060] (2/4) Exclude cut with ID 7699-105389-0094-26379-0_sp1.1 from training. Duration: 21.77725
2022-12-22 16:56:16,959 INFO [train.py:894] (2/4) Epoch 5, batch 1700, loss[loss=0.2707, simple_loss=0.3256, pruned_loss=0.1079, over 18415.00 frames. ], tot_loss[loss=0.2526, simple_loss=0.323, pruned_loss=0.09109, over 3713491.18 frames. ], batch size: 46, lr: 2.23e-02, grad_scale: 8.0
2022-12-22 16:56:26,145 WARNING [train.py:1060] (2/4) Exclude cut with ID 7255-291500-0005-134304-0_sp0.9 from training. Duration: 27.8166875
2022-12-22 16:56:38,303 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.0337, 0.6132, 0.7989, 0.9870, 1.0643, 1.1402, 1.0756, 0.9106],
       device='cuda:2'), covar=tensor([0.0439, 0.0400, 0.0736, 0.0369, 0.0344, 0.0434, 0.0356, 0.0406],
       device='cuda:2'), in_proj_covar=tensor([0.0076, 0.0107, 0.0125, 0.0121, 0.0102, 0.0088, 0.0080, 0.0112],
       device='cuda:2'), out_proj_covar=tensor([7.7692e-05, 1.0293e-04, 1.2736e-04, 1.1784e-04, 1.0479e-04, 8.4612e-05,
        7.9352e-05, 1.1053e-04], device='cuda:2')
2022-12-22 16:56:49,691 WARNING [train.py:1060] (2/4) Exclude cut with ID 7357-94126-0021-15852-0_sp1.1 from training. Duration: 22.5090625
2022-12-22 16:56:57,476 WARNING [train.py:1060] (2/4) Exclude cut with ID 7255-291500-0005-134304-0 from training. Duration: 25.035
2022-12-22 16:57:17,444 WARNING [train.py:1060] (2/4) Exclude cut with ID 774-127930-0014-10412-0_sp1.1 from training. Duration: 0.95
2022-12-22 16:57:32,044 INFO [train.py:894] (2/4) Epoch 5, batch 1750, loss[loss=0.2759, simple_loss=0.3475, pruned_loss=0.1021, over 18392.00 frames. ], tot_loss[loss=0.2585, simple_loss=0.3264, pruned_loss=0.0953, over 3713537.84 frames. ], batch size: 53, lr: 2.23e-02, grad_scale: 8.0
2022-12-22 16:57:34,864 WARNING [train.py:1060] (2/4) Exclude cut with ID 3033-130750-0096-55598-0_sp0.9 from training. Duration: 0.92225
2022-12-22 16:57:41,573 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([0.4712, 1.7247, 2.1404, 0.2169, 1.3486, 2.6047, 1.7903, 1.6926],
       device='cuda:2'), covar=tensor([0.0710, 0.0352, 0.0369, 0.0549, 0.0408, 0.0178, 0.0319, 0.0506],
       device='cuda:2'), in_proj_covar=tensor([0.0115, 0.0126, 0.0085, 0.0116, 0.0110, 0.0086, 0.0119, 0.0105],
       device='cuda:2'), out_proj_covar=tensor([1.1756e-04, 1.3048e-04, 8.8188e-05, 1.1719e-04, 1.1051e-04, 8.8550e-05,
        1.2436e-04, 1.0838e-04], device='cuda:2')
2022-12-22 16:57:48,123 INFO [scaling.py:679] (2/4) Whitening: num_groups=1, num_channels=384, metric=3.84 vs. limit=5.0
2022-12-22 16:58:04,886 WARNING [train.py:1060] (2/4) Exclude cut with ID 4511-76322-0006-80011-0 from training. Duration: 21.97
2022-12-22 16:58:22,969 WARNING [train.py:1060] (2/4) Exclude cut with ID 7492-105653-0055-62765-0_sp0.9 from training. Duration: 21.97225
2022-12-22 16:58:23,005 WARNING [train.py:1060] (2/4) Exclude cut with ID 453-131332-0000-47844-0_sp0.9 from training. Duration: 25.3333125
2022-12-22 16:58:23,309 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.3332, 0.8501, 0.7424, 1.0265, 1.6002, 0.6380, 1.0091, 1.3011],
       device='cuda:2'), covar=tensor([0.1760, 0.2463, 0.2466, 0.1885, 0.2204, 0.1770, 0.1769, 0.1869],
       device='cuda:2'), in_proj_covar=tensor([0.0093, 0.0104, 0.0129, 0.0103, 0.0112, 0.0095, 0.0097, 0.0100],
       device='cuda:2'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0002, 0.0003, 0.0003],
       device='cuda:2')
2022-12-22 16:58:32,744 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 4.623e+02 6.430e+02 8.235e+02 1.006e+03 2.020e+03, threshold=1.647e+03, percent-clipped=2.0
2022-12-22 16:58:36,779 WARNING [train.py:1060] (2/4) Exclude cut with ID 5172-29468-0015-19128-0_sp0.9 from training. Duration: 21.5055625
2022-12-22 16:58:46,285 WARNING [train.py:1060] (2/4) Exclude cut with ID 453-131332-0000-47844-0_sp1.1 from training. Duration: 20.72725
2022-12-22 16:58:47,851 INFO [train.py:894] (2/4) Epoch 5, batch 1800, loss[loss=0.272, simple_loss=0.3154, pruned_loss=0.1143, over 18597.00 frames. ], tot_loss[loss=0.2627, simple_loss=0.3284, pruned_loss=0.09853, over 3713034.14 frames. ], batch size: 41, lr: 2.23e-02, grad_scale: 8.0
2022-12-22 16:58:51,165 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.6489, 1.6994, 1.8039, 1.7080, 1.6813, 3.5338, 1.7203, 2.3700],
       device='cuda:2'), covar=tensor([0.3222, 0.1902, 0.1755, 0.1855, 0.1282, 0.0187, 0.1518, 0.0911],
       device='cuda:2'), in_proj_covar=tensor([0.0149, 0.0125, 0.0140, 0.0124, 0.0118, 0.0097, 0.0110, 0.0108],
       device='cuda:2'), out_proj_covar=tensor([0.0004, 0.0004, 0.0004, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003],
       device='cuda:2')
2022-12-22 16:58:57,864 INFO [scaling.py:679] (2/4) Whitening: num_groups=8, num_channels=96, metric=1.16 vs. limit=2.0
2022-12-22 16:59:17,006 WARNING [train.py:1060] (2/4) Exclude cut with ID 8631-249866-0030-130156-0_sp0.9 from training. Duration: 26.32775
2022-12-22 16:59:26,825 INFO [zipformer.py:660] (2/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=15851.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 16:59:40,265 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([2.8707, 2.2829, 2.1201, 0.9245, 2.0123, 2.3030, 1.6329, 1.9725],
       device='cuda:2'), covar=tensor([0.0495, 0.0501, 0.1054, 0.1564, 0.1175, 0.1009, 0.1341, 0.0880],
       device='cuda:2'), in_proj_covar=tensor([0.0141, 0.0162, 0.0196, 0.0187, 0.0185, 0.0170, 0.0182, 0.0177],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-22 16:59:52,673 WARNING [train.py:1060] (2/4) Exclude cut with ID 3867-173237-0077-144769-0 from training. Duration: 20.025
2022-12-22 16:59:57,137 WARNING [train.py:1060] (2/4) Exclude cut with ID 6709-74022-0004-86860-0_sp1.1 from training. Duration: 0.9409375
2022-12-22 16:59:57,147 WARNING [train.py:1060] (2/4) Exclude cut with ID 4757-1811-0023-62229-0_sp0.9 from training. Duration: 21.37775
2022-12-22 17:00:05,169 INFO [train.py:894] (2/4) Epoch 5, batch 1850, loss[loss=0.2369, simple_loss=0.291, pruned_loss=0.09139, over 18615.00 frames. ], tot_loss[loss=0.2675, simple_loss=0.3311, pruned_loss=0.102, over 3713672.25 frames. ], batch size: 41, lr: 2.22e-02, grad_scale: 8.0
2022-12-22 17:00:15,704 INFO [zipformer.py:660] (2/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=15883.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 17:00:19,776 WARNING [train.py:1060] (2/4) Exclude cut with ID 1250-135782-0004-25974-0_sp0.9 from training. Duration: 21.17225
2022-12-22 17:00:19,788 WARNING [train.py:1060] (2/4) Exclude cut with ID 7357-94126-0021-15852-0_sp0.9 from training. Duration: 27.511125
2022-12-22 17:00:39,107 INFO [zipformer.py:660] (2/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=15899.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 17:00:49,829 WARNING [train.py:1060] (2/4) Exclude cut with ID 453-131332-0000-47844-0 from training. Duration: 22.8
2022-12-22 17:00:54,676 WARNING [train.py:1060] (2/4) Exclude cut with ID 4964-30587-0040-44509-0 from training. Duration: 22.585
2022-12-22 17:01:04,854 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 3.750e+02 6.712e+02 8.382e+02 1.076e+03 1.765e+03, threshold=1.676e+03, percent-clipped=1.0
2022-12-22 17:01:19,845 INFO [train.py:894] (2/4) Epoch 5, batch 1900, loss[loss=0.2711, simple_loss=0.328, pruned_loss=0.1071, over 18373.00 frames. ], tot_loss[loss=0.2716, simple_loss=0.3335, pruned_loss=0.1048, over 3712911.65 frames. ], batch size: 46, lr: 2.22e-02, grad_scale: 8.0
2022-12-22 17:01:26,128 WARNING [train.py:1060] (2/4) Exclude cut with ID 6978-92210-0001-146967-0_sp0.9 from training. Duration: 22.0166875
2022-12-22 17:01:32,399 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([0.0575, 1.0964, 1.6673, 0.2853, 0.8197, 1.8527, 1.4551, 1.4403],
       device='cuda:2'), covar=tensor([0.0657, 0.0341, 0.0310, 0.0436, 0.0468, 0.0264, 0.0335, 0.0483],
       device='cuda:2'), in_proj_covar=tensor([0.0115, 0.0122, 0.0084, 0.0115, 0.0109, 0.0088, 0.0118, 0.0104],
       device='cuda:2'), out_proj_covar=tensor([1.1751e-04, 1.2670e-04, 8.6459e-05, 1.1653e-04, 1.0896e-04, 9.0462e-05,
        1.2303e-04, 1.0718e-04], device='cuda:2')
2022-12-22 17:01:43,607 WARNING [train.py:1060] (2/4) Exclude cut with ID 7255-291500-0003-134302-0_sp1.1 from training. Duration: 24.395375
2022-12-22 17:01:47,439 INFO [zipformer.py:660] (2/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=15944.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 17:01:49,906 WARNING [train.py:1060] (2/4) Exclude cut with ID 3488-85273-0017-41203-0_sp0.9 from training. Duration: 27.47775
2022-12-22 17:01:54,193 WARNING [train.py:1060] (2/4) Exclude cut with ID 432-122774-0017-62487-0_sp0.9 from training. Duration: 24.8833125
2022-12-22 17:01:57,107 WARNING [train.py:1060] (2/4) Exclude cut with ID 6758-72288-0033-108368-0 from training. Duration: 23.39
2022-12-22 17:02:01,814 WARNING [train.py:1060] (2/4) Exclude cut with ID 8291-282929-0007-12994-0_sp0.9 from training. Duration: 28.72225
2022-12-22 17:02:12,816 WARNING [train.py:1060] (2/4) Exclude cut with ID 585-294811-0110-133686-0_sp0.9 from training. Duration: 20.8944375
2022-12-22 17:02:27,104 WARNING [train.py:1060] (2/4) Exclude cut with ID 5796-66357-0007-116447-0_sp0.9 from training. Duration: 23.8444375
2022-12-22 17:02:31,061 INFO [scaling.py:679] (2/4) Whitening: num_groups=8, num_channels=96, metric=1.29 vs. limit=2.0
2022-12-22 17:02:37,328 INFO [train.py:894] (2/4) Epoch 5, batch 1950, loss[loss=0.2629, simple_loss=0.3206, pruned_loss=0.1026, over 18667.00 frames. ], tot_loss[loss=0.2735, simple_loss=0.3347, pruned_loss=0.1062, over 3713263.46 frames. ], batch size: 46, lr: 2.22e-02, grad_scale: 8.0
2022-12-22 17:02:52,267 WARNING [train.py:1060] (2/4) Exclude cut with ID 8291-282929-0007-12994-0 from training. Duration: 25.85
2022-12-22 17:02:52,278 WARNING [train.py:1060] (2/4) Exclude cut with ID 8291-282929-0023-13010-0 from training. Duration: 21.39
2022-12-22 17:03:03,341 WARNING [train.py:1060] (2/4) Exclude cut with ID 7357-94126-0014-15845-0 from training. Duration: 27.92
2022-12-22 17:03:08,892 INFO [scaling.py:679] (2/4) Whitening: num_groups=8, num_channels=96, metric=1.98 vs. limit=2.0
2022-12-22 17:03:18,665 INFO [zipformer.py:660] (2/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=16001.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 17:03:34,560 WARNING [train.py:1060] (2/4) Exclude cut with ID 8631-249866-0039-130165-0_sp0.9 from training. Duration: 20.661125
2022-12-22 17:03:42,602 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 4.362e+02 7.148e+02 8.465e+02 9.808e+02 1.967e+03, threshold=1.693e+03, percent-clipped=1.0
2022-12-22 17:03:57,082 INFO [train.py:894] (2/4) Epoch 5, batch 2000, loss[loss=0.2493, simple_loss=0.3165, pruned_loss=0.09106, over 18563.00 frames. ], tot_loss[loss=0.2735, simple_loss=0.334, pruned_loss=0.1065, over 3712834.25 frames. ], batch size: 49, lr: 2.21e-02, grad_scale: 8.0
2022-12-22 17:03:58,627 WARNING [train.py:1060] (2/4) Exclude cut with ID 7357-94126-0043-15874-0_sp0.9 from training. Duration: 20.07225
2022-12-22 17:04:06,005 WARNING [train.py:1060] (2/4) Exclude cut with ID 1085-156170-0017-128270-0 from training. Duration: 21.01
2022-12-22 17:04:16,533 INFO [scaling.py:679] (2/4) Whitening: num_groups=8, num_channels=96, metric=1.53 vs. limit=2.0
2022-12-22 17:04:48,095 INFO [scaling.py:679] (2/4) Whitening: num_groups=8, num_channels=96, metric=1.19 vs. limit=2.0
2022-12-22 17:04:51,962 INFO [zipformer.py:660] (2/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=16062.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 17:05:12,899 INFO [train.py:894] (2/4) Epoch 5, batch 2050, loss[loss=0.2825, simple_loss=0.3407, pruned_loss=0.1121, over 18535.00 frames. ], tot_loss[loss=0.2748, simple_loss=0.3342, pruned_loss=0.1077, over 3713217.55 frames. ], batch size: 97, lr: 2.21e-02, grad_scale: 8.0
2022-12-22 17:05:14,537 WARNING [train.py:1060] (2/4) Exclude cut with ID 2195-150901-0045-59933-0 from training. Duration: 20.65
2022-12-22 17:05:22,501 WARNING [train.py:1060] (2/4) Exclude cut with ID 5796-66357-0007-116447-0 from training. Duration: 21.46
2022-12-22 17:05:27,211 INFO [zipformer.py:660] (2/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=16085.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 17:06:04,183 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([2.6865, 3.4477, 3.5410, 1.7648, 3.3761, 2.4785, 0.7369, 2.3386],
       device='cuda:2'), covar=tensor([0.1637, 0.1020, 0.1383, 0.3358, 0.1061, 0.1249, 0.5608, 0.1754],
       device='cuda:2'), in_proj_covar=tensor([0.0121, 0.0105, 0.0147, 0.0119, 0.0109, 0.0100, 0.0146, 0.0108],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-22 17:06:08,785 WARNING [train.py:1060] (2/4) Exclude cut with ID 3557-8342-0013-54691-0 from training. Duration: 0.92
2022-12-22 17:06:13,337 WARNING [train.py:1060] (2/4) Exclude cut with ID 8291-282929-0023-13010-0_sp0.9 from training. Duration: 23.7666875
2022-12-22 17:06:14,712 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 3.991e+02 7.417e+02 9.427e+02 1.233e+03 2.937e+03, threshold=1.885e+03, percent-clipped=9.0
2022-12-22 17:06:30,466 INFO [train.py:894] (2/4) Epoch 5, batch 2100, loss[loss=0.3279, simple_loss=0.3679, pruned_loss=0.144, over 18653.00 frames. ], tot_loss[loss=0.276, simple_loss=0.3347, pruned_loss=0.1086, over 3714104.63 frames. ], batch size: 178, lr: 2.21e-02, grad_scale: 8.0
2022-12-22 17:06:46,171 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.7674, 1.3885, 2.2069, 3.0602, 2.0519, 2.3567, 0.7320, 1.9678],
       device='cuda:2'), covar=tensor([0.1947, 0.2039, 0.1528, 0.0629, 0.1497, 0.1625, 0.2949, 0.1441],
       device='cuda:2'), in_proj_covar=tensor([0.0105, 0.0118, 0.0122, 0.0097, 0.0106, 0.0125, 0.0139, 0.0105],
       device='cuda:2'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0002, 0.0003, 0.0003, 0.0003, 0.0003],
       device='cuda:2')
2022-12-22 17:06:49,564 WARNING [train.py:1060] (2/4) Exclude cut with ID 8544-281189-0060-101339-0_sp0.9 from training. Duration: 20.861125
2022-12-22 17:07:00,127 WARNING [train.py:1060] (2/4) Exclude cut with ID 3488-65654-0031-41259-0_sp0.9 from training. Duration: 22.711125
2022-12-22 17:07:02,507 INFO [zipformer.py:660] (2/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=16146.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 17:07:06,911 INFO [zipformer.py:660] (2/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=16149.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 17:07:39,730 WARNING [train.py:1060] (2/4) Exclude cut with ID 6951-79737-0043-132310-0_sp1.1 from training. Duration: 22.986375
2022-12-22 17:07:47,531 INFO [train.py:894] (2/4) Epoch 5, batch 2150, loss[loss=0.2893, simple_loss=0.354, pruned_loss=0.1123, over 18599.00 frames. ], tot_loss[loss=0.2767, simple_loss=0.3351, pruned_loss=0.1092, over 3715037.32 frames. ], batch size: 98, lr: 2.20e-02, grad_scale: 8.0
2022-12-22 17:07:54,880 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([5.8372, 4.9911, 5.1214, 5.6562, 5.2722, 5.1011, 5.6869, 1.6651],
       device='cuda:2'), covar=tensor([0.0467, 0.0427, 0.0388, 0.0529, 0.1027, 0.0729, 0.0340, 0.4113],
       device='cuda:2'), in_proj_covar=tensor([0.0227, 0.0172, 0.0164, 0.0157, 0.0234, 0.0187, 0.0186, 0.0217],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0001, 0.0001, 0.0001, 0.0002, 0.0001, 0.0002, 0.0002],
       device='cuda:2')
2022-12-22 17:07:57,602 WARNING [train.py:1060] (2/4) Exclude cut with ID 8040-260924-0003-80960-0_sp0.9 from training. Duration: 22.07225
2022-12-22 17:08:03,049 WARNING [train.py:1060] (2/4) Exclude cut with ID 7699-105389-0045-26330-0_sp0.9 from training. Duration: 20.3055625
2022-12-22 17:08:04,446 WARNING [train.py:1060] (2/4) Exclude cut with ID 6356-271890-0060-94317-0_sp0.9 from training. Duration: 20.72225
2022-12-22 17:08:11,301 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.4203, 1.3687, 1.0028, 0.6147, 1.7725, 1.4414, 1.3374, 1.0021],
       device='cuda:2'), covar=tensor([0.0452, 0.0465, 0.0716, 0.0859, 0.0202, 0.0407, 0.0568, 0.1177],
       device='cuda:2'), in_proj_covar=tensor([0.0124, 0.0114, 0.0139, 0.0128, 0.0076, 0.0114, 0.0142, 0.0155],
       device='cuda:2'), out_proj_covar=tensor([1.5548e-04, 1.4430e-04, 1.7115e-04, 1.5775e-04, 9.6625e-05, 1.3957e-04,
        1.7496e-04, 1.9035e-04], device='cuda:2')
2022-12-22 17:08:23,095 WARNING [train.py:1060] (2/4) Exclude cut with ID 3488-85273-0017-41203-0_sp1.1 from training. Duration: 22.4818125
2022-12-22 17:08:40,602 INFO [zipformer.py:660] (2/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=16210.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 17:08:49,800 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 3.943e+02 6.646e+02 8.265e+02 9.985e+02 1.734e+03, threshold=1.653e+03, percent-clipped=0.0
2022-12-22 17:08:49,861 WARNING [train.py:1060] (2/4) Exclude cut with ID 4964-30587-0040-44509-0_sp0.9 from training. Duration: 25.0944375
2022-12-22 17:08:54,281 WARNING [train.py:1060] (2/4) Exclude cut with ID 6533-399-0047-104881-0 from training. Duration: 21.515
2022-12-22 17:09:00,086 WARNING [train.py:1060] (2/4) Exclude cut with ID 7357-94126-0009-15840-0_sp0.9 from training. Duration: 27.02225
2022-12-22 17:09:04,587 INFO [train.py:894] (2/4) Epoch 5, batch 2200, loss[loss=0.283, simple_loss=0.3494, pruned_loss=0.1083, over 18469.00 frames. ], tot_loss[loss=0.2791, simple_loss=0.3364, pruned_loss=0.1109, over 3714670.02 frames. ], batch size: 54, lr: 2.20e-02, grad_scale: 8.0
2022-12-22 17:09:04,617 WARNING [train.py:1060] (2/4) Exclude cut with ID 432-122774-0010-62480-0_sp0.9 from training. Duration: 22.22225
2022-12-22 17:09:12,156 WARNING [train.py:1060] (2/4) Exclude cut with ID 4964-30587-0085-44554-0_sp0.9 from training. Duration: 20.85
2022-12-22 17:09:25,648 INFO [zipformer.py:660] (2/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=16239.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 17:09:46,971 WARNING [train.py:1060] (2/4) Exclude cut with ID 4295-39940-0007-92567-0 from training. Duration: 21.54
2022-12-22 17:09:51,403 WARNING [train.py:1060] (2/4) Exclude cut with ID 4964-30587-0040-44509-0_sp1.1 from training. Duration: 20.5318125
2022-12-22 17:09:52,176 INFO [scaling.py:679] (2/4) Whitening: num_groups=8, num_channels=96, metric=1.18 vs. limit=2.0
2022-12-22 17:10:02,289 WARNING [train.py:1060] (2/4) Exclude cut with ID 7255-291500-0012-134311-0_sp0.9 from training. Duration: 21.9333125
2022-12-22 17:10:05,778 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.4033, 1.1485, 1.8350, 1.1628, 1.4298, 1.4375, 1.1114, 1.7263],
       device='cuda:2'), covar=tensor([0.0827, 0.1416, 0.0869, 0.1082, 0.0699, 0.0871, 0.1853, 0.0466],
       device='cuda:2'), in_proj_covar=tensor([0.0194, 0.0178, 0.0204, 0.0190, 0.0180, 0.0204, 0.0197, 0.0173],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-22 17:10:22,609 INFO [train.py:894] (2/4) Epoch 5, batch 2250, loss[loss=0.3315, simple_loss=0.3699, pruned_loss=0.1465, over 18640.00 frames. ], tot_loss[loss=0.2804, simple_loss=0.3372, pruned_loss=0.1118, over 3714397.07 frames. ], batch size: 180, lr: 2.20e-02, grad_scale: 8.0
2022-12-22 17:10:30,355 INFO [scaling.py:679] (2/4) Whitening: num_groups=1, num_channels=384, metric=4.69 vs. limit=5.0
2022-12-22 17:10:50,969 WARNING [train.py:1060] (2/4) Exclude cut with ID 8631-249866-0025-130151-0_sp0.9 from training. Duration: 21.7944375
2022-12-22 17:11:01,708 WARNING [train.py:1060] (2/4) Exclude cut with ID 8291-282929-0002-12989-0_sp0.9 from training. Duration: 22.4666875
2022-12-22 17:11:09,609 WARNING [train.py:1060] (2/4) Exclude cut with ID 6121-9014-0076-24124-0 from training. Duration: 21.635
2022-12-22 17:11:14,100 WARNING [train.py:1060] (2/4) Exclude cut with ID 6121-9014-0076-24124-0_sp0.9 from training. Duration: 24.038875
2022-12-22 17:11:20,840 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.9016, 2.1980, 1.3600, 3.1813, 2.9882, 1.6164, 2.4018, 1.3805],
       device='cuda:2'), covar=tensor([0.1736, 0.1618, 0.1491, 0.0707, 0.1431, 0.1222, 0.1460, 0.1543],
       device='cuda:2'), in_proj_covar=tensor([0.0219, 0.0193, 0.0180, 0.0173, 0.0235, 0.0176, 0.0188, 0.0179],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-22 17:11:24,860 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 4.321e+02 6.545e+02 8.299e+02 1.005e+03 2.324e+03, threshold=1.660e+03, percent-clipped=4.0
2022-12-22 17:11:40,734 INFO [train.py:894] (2/4) Epoch 5, batch 2300, loss[loss=0.3048, simple_loss=0.3432, pruned_loss=0.1332, over 18667.00 frames. ], tot_loss[loss=0.2788, simple_loss=0.3359, pruned_loss=0.1109, over 3714718.64 frames. ], batch size: 48, lr: 2.20e-02, grad_scale: 8.0
2022-12-22 17:11:44,100 INFO [scaling.py:679] (2/4) Whitening: num_groups=8, num_channels=96, metric=1.23 vs. limit=2.0
2022-12-22 17:12:00,731 WARNING [train.py:1060] (2/4) Exclude cut with ID 7255-291500-0009-134308-0_sp1.1 from training. Duration: 21.786375
2022-12-22 17:12:13,623 WARNING [train.py:1060] (2/4) Exclude cut with ID 8291-282929-0002-12989-0 from training. Duration: 20.22
2022-12-22 17:12:27,044 INFO [zipformer.py:660] (2/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=16357.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 17:12:47,948 INFO [zipformer.py:660] (2/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=16371.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 17:12:55,725 INFO [train.py:894] (2/4) Epoch 5, batch 2350, loss[loss=0.2598, simple_loss=0.3266, pruned_loss=0.09649, over 18499.00 frames. ], tot_loss[loss=0.2782, simple_loss=0.3356, pruned_loss=0.1104, over 3714281.87 frames. ], batch size: 52, lr: 2.19e-02, grad_scale: 8.0
2022-12-22 17:13:56,201 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 4.206e+02 6.539e+02 8.779e+02 1.071e+03 2.079e+03, threshold=1.756e+03, percent-clipped=3.0
2022-12-22 17:14:11,765 INFO [train.py:894] (2/4) Epoch 5, batch 2400, loss[loss=0.2405, simple_loss=0.3105, pruned_loss=0.08528, over 18603.00 frames. ], tot_loss[loss=0.2781, simple_loss=0.3356, pruned_loss=0.1103, over 3714199.97 frames. ], batch size: 51, lr: 2.19e-02, grad_scale: 8.0
2022-12-22 17:14:13,338 WARNING [train.py:1060] (2/4) Exclude cut with ID 6951-79737-0043-132310-0 from training. Duration: 25.285
2022-12-22 17:14:18,473 INFO [scaling.py:679] (2/4) Whitening: num_groups=8, num_channels=96, metric=1.95 vs. limit=2.0
2022-12-22 17:14:21,044 INFO [zipformer.py:660] (2/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=16432.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 17:14:34,624 INFO [zipformer.py:660] (2/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=16441.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 17:15:17,009 WARNING [train.py:1060] (2/4) Exclude cut with ID 811-130148-0001-63453-0_sp0.9 from training. Duration: 20.861125
2022-12-22 17:15:27,552 INFO [train.py:894] (2/4) Epoch 5, batch 2450, loss[loss=0.3139, simple_loss=0.3512, pruned_loss=0.1383, over 18504.00 frames. ], tot_loss[loss=0.2771, simple_loss=0.3342, pruned_loss=0.11, over 3714051.88 frames. ], batch size: 44, lr: 2.19e-02, grad_scale: 8.0
2022-12-22 17:15:39,005 WARNING [train.py:1060] (2/4) Exclude cut with ID 6010-56788-0055-90261-0 from training. Duration: 20.88
2022-12-22 17:16:03,401 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([0.4762, 1.6625, 1.9816, 0.5353, 1.3915, 2.3203, 1.8721, 1.7695],
       device='cuda:2'), covar=tensor([0.0680, 0.0314, 0.0269, 0.0487, 0.0384, 0.0261, 0.0259, 0.0465],
       device='cuda:2'), in_proj_covar=tensor([0.0116, 0.0126, 0.0087, 0.0115, 0.0114, 0.0090, 0.0121, 0.0108],
       device='cuda:2'), out_proj_covar=tensor([1.1618e-04, 1.2833e-04, 8.8491e-05, 1.1476e-04, 1.1299e-04, 9.1469e-05,
        1.2598e-04, 1.1009e-04], device='cuda:2')
2022-12-22 17:16:10,357 WARNING [train.py:1060] (2/4) Exclude cut with ID 7357-94126-0045-15876-0_sp0.9 from training. Duration: 23.4166875
2022-12-22 17:16:12,197 INFO [zipformer.py:660] (2/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=16505.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 17:16:29,857 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 4.784e+02 7.165e+02 8.582e+02 1.131e+03 1.915e+03, threshold=1.716e+03, percent-clipped=1.0
2022-12-22 17:16:45,977 INFO [train.py:894] (2/4) Epoch 5, batch 2500, loss[loss=0.2641, simple_loss=0.3348, pruned_loss=0.09673, over 18583.00 frames. ], tot_loss[loss=0.2759, simple_loss=0.3335, pruned_loss=0.1091, over 3715062.28 frames. ], batch size: 51, lr: 2.18e-02, grad_scale: 8.0
2022-12-22 17:16:58,950 INFO [scaling.py:679] (2/4) Whitening: num_groups=8, num_channels=96, metric=1.51 vs. limit=2.0
2022-12-22 17:17:06,045 INFO [zipformer.py:660] (2/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=16539.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 17:17:27,456 WARNING [train.py:1060] (2/4) Exclude cut with ID 7255-291500-0006-134305-0 from training. Duration: 21.24
2022-12-22 17:17:27,470 WARNING [train.py:1060] (2/4) Exclude cut with ID 6533-399-0047-104881-0_sp0.9 from training. Duration: 23.9055625
2022-12-22 17:17:59,870 WARNING [train.py:1060] (2/4) Exclude cut with ID 6758-72288-0033-108368-0_sp0.9 from training. Duration: 25.988875
2022-12-22 17:18:02,739 INFO [train.py:894] (2/4) Epoch 5, batch 2550, loss[loss=0.2771, simple_loss=0.3403, pruned_loss=0.1069, over 18582.00 frames. ], tot_loss[loss=0.2753, simple_loss=0.3334, pruned_loss=0.1086, over 3715800.95 frames. ], batch size: 56, lr: 2.18e-02, grad_scale: 8.0
2022-12-22 17:18:08,909 WARNING [train.py:1060] (2/4) Exclude cut with ID 7255-291500-0001-134300-0_sp0.9 from training. Duration: 20.67225
2022-12-22 17:18:10,485 INFO [scaling.py:679] (2/4) Whitening: num_groups=8, num_channels=96, metric=1.37 vs. limit=2.0
2022-12-22 17:18:19,851 INFO [zipformer.py:660] (2/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=16587.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 17:18:34,123 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([3.0842, 2.2352, 1.2105, 2.6901, 2.0410, 1.6211, 2.1467, 3.4415],
       device='cuda:2'), covar=tensor([0.1463, 0.2273, 0.1766, 0.2571, 0.2498, 0.1121, 0.2416, 0.0474],
       device='cuda:2'), in_proj_covar=tensor([0.0248, 0.0226, 0.0204, 0.0309, 0.0213, 0.0193, 0.0230, 0.0163],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0003, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-22 17:18:55,675 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.4554, 0.7944, 1.2449, 1.2243, 1.4855, 1.5335, 1.5316, 1.0505],
       device='cuda:2'), covar=tensor([0.0294, 0.0358, 0.0486, 0.0288, 0.0286, 0.0317, 0.0235, 0.0351],
       device='cuda:2'), in_proj_covar=tensor([0.0074, 0.0103, 0.0122, 0.0120, 0.0100, 0.0087, 0.0077, 0.0111],
       device='cuda:2'), out_proj_covar=tensor([7.5094e-05, 9.8170e-05, 1.2273e-04, 1.1585e-04, 1.0154e-04, 8.2659e-05,
        7.5721e-05, 1.0852e-04], device='cuda:2')
2022-12-22 17:18:56,920 WARNING [train.py:1060] (2/4) Exclude cut with ID 3488-85273-0038-41224-0 from training. Duration: 20.34
2022-12-22 17:19:06,748 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 3.751e+02 6.596e+02 8.364e+02 1.015e+03 1.761e+03, threshold=1.673e+03, percent-clipped=1.0
2022-12-22 17:19:20,154 INFO [train.py:894] (2/4) Epoch 5, batch 2600, loss[loss=0.2249, simple_loss=0.2811, pruned_loss=0.0843, over 18604.00 frames. ], tot_loss[loss=0.2743, simple_loss=0.3328, pruned_loss=0.1079, over 3716208.20 frames. ], batch size: 41, lr: 2.18e-02, grad_scale: 8.0
2022-12-22 17:20:07,395 INFO [zipformer.py:660] (2/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=16657.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 17:20:14,403 WARNING [train.py:1060] (2/4) Exclude cut with ID 7357-94126-0026-15857-0_sp0.9 from training. Duration: 25.061125
2022-12-22 17:20:26,517 WARNING [train.py:1060] (2/4) Exclude cut with ID 3033-130750-0096-55598-0 from training. Duration: 0.83
2022-12-22 17:20:31,694 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.3942, 1.6290, 1.2144, 1.9150, 2.2737, 1.2565, 1.5287, 0.9808],
       device='cuda:2'), covar=tensor([0.2440, 0.2206, 0.2014, 0.1297, 0.1374, 0.1829, 0.1783, 0.2292],
       device='cuda:2'), in_proj_covar=tensor([0.0224, 0.0193, 0.0183, 0.0172, 0.0235, 0.0177, 0.0191, 0.0180],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-22 17:20:35,644 INFO [train.py:894] (2/4) Epoch 5, batch 2650, loss[loss=0.2498, simple_loss=0.3164, pruned_loss=0.09158, over 18479.00 frames. ], tot_loss[loss=0.2722, simple_loss=0.3312, pruned_loss=0.1066, over 3714860.88 frames. ], batch size: 50, lr: 2.17e-02, grad_scale: 8.0
2022-12-22 17:20:52,463 WARNING [train.py:1060] (2/4) Exclude cut with ID 3488-85273-0017-41203-0 from training. Duration: 24.73
2022-12-22 17:21:06,260 WARNING [train.py:1060] (2/4) Exclude cut with ID 7255-291500-0009-134308-0 from training. Duration: 23.965
2022-12-22 17:21:08,473 INFO [scaling.py:679] (2/4) Whitening: num_groups=8, num_channels=96, metric=1.17 vs. limit=2.0
2022-12-22 17:21:13,695 WARNING [train.py:1060] (2/4) Exclude cut with ID 6978-92210-0030-146996-0_sp0.9 from training. Duration: 22.088875
2022-12-22 17:21:19,940 INFO [zipformer.py:660] (2/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=16705.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 17:21:30,462 WARNING [train.py:1060] (2/4) Exclude cut with ID 7255-291500-0006-134305-0_sp0.9 from training. Duration: 23.6
2022-12-22 17:21:38,520 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 4.340e+02 6.537e+02 7.713e+02 9.830e+02 1.584e+03, threshold=1.543e+03, percent-clipped=0.0
2022-12-22 17:21:51,815 INFO [train.py:894] (2/4) Epoch 5, batch 2700, loss[loss=0.2581, simple_loss=0.3276, pruned_loss=0.09431, over 18456.00 frames. ], tot_loss[loss=0.2728, simple_loss=0.3323, pruned_loss=0.1066, over 3714828.33 frames. ], batch size: 54, lr: 2.17e-02, grad_scale: 8.0
2022-12-22 17:21:53,493 INFO [zipformer.py:660] (2/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=16727.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 17:22:15,953 INFO [zipformer.py:660] (2/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=16741.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 17:23:06,168 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.6794, 1.0408, 1.8947, 3.1323, 2.0601, 2.1076, 0.8898, 1.8566],
       device='cuda:2'), covar=tensor([0.1933, 0.2248, 0.1661, 0.0527, 0.1435, 0.1408, 0.2691, 0.1417],
       device='cuda:2'), in_proj_covar=tensor([0.0107, 0.0118, 0.0126, 0.0099, 0.0106, 0.0125, 0.0136, 0.0107],
       device='cuda:2'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0002, 0.0003, 0.0003, 0.0003, 0.0003],
       device='cuda:2')
2022-12-22 17:23:09,467 INFO [train.py:894] (2/4) Epoch 5, batch 2750, loss[loss=0.2759, simple_loss=0.3393, pruned_loss=0.1063, over 18664.00 frames. ], tot_loss[loss=0.2717, simple_loss=0.331, pruned_loss=0.1062, over 3713833.69 frames. ], batch size: 100, lr: 2.17e-02, grad_scale: 8.0
2022-12-22 17:23:12,359 WARNING [train.py:1060] (2/4) Exclude cut with ID 8291-282929-0024-13011-0 from training. Duration: 23.795
2022-12-22 17:23:29,378 WARNING [train.py:1060] (2/4) Exclude cut with ID 8631-249866-0030-130156-0_sp1.1 from training. Duration: 21.5409375
2022-12-22 17:23:29,535 INFO [zipformer.py:660] (2/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=16789.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 17:23:32,264 WARNING [train.py:1060] (2/4) Exclude cut with ID 6978-92210-0019-146985-0_sp0.9 from training. Duration: 24.97775
2022-12-22 17:23:41,533 WARNING [train.py:1060] (2/4) Exclude cut with ID 1085-156170-0017-128270-0_sp0.9 from training. Duration: 23.3444375
2022-12-22 17:23:54,546 INFO [zipformer.py:660] (2/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=16805.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 17:24:09,503 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.6366, 1.4725, 1.2203, 2.1901, 1.6608, 3.4778, 1.5440, 1.5728],
       device='cuda:2'), covar=tensor([0.0920, 0.1656, 0.1242, 0.0828, 0.1433, 0.0248, 0.1222, 0.1489],
       device='cuda:2'), in_proj_covar=tensor([0.0082, 0.0090, 0.0085, 0.0082, 0.0102, 0.0076, 0.0092, 0.0085],
       device='cuda:2'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003],
       device='cuda:2')
2022-12-22 17:24:11,943 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 4.184e+02 6.794e+02 8.297e+02 1.071e+03 3.389e+03, threshold=1.659e+03, percent-clipped=6.0
2022-12-22 17:24:11,958 WARNING [train.py:1060] (2/4) Exclude cut with ID 6010-56788-0055-90261-0_sp0.9 from training. Duration: 23.2
2022-12-22 17:24:16,733 WARNING [train.py:1060] (2/4) Exclude cut with ID 5653-46179-0060-117930-0_sp0.9 from training. Duration: 21.17225
2022-12-22 17:24:25,945 INFO [train.py:894] (2/4) Epoch 5, batch 2800, loss[loss=0.2464, simple_loss=0.3103, pruned_loss=0.09125, over 18466.00 frames. ], tot_loss[loss=0.2714, simple_loss=0.3308, pruned_loss=0.106, over 3713581.90 frames. ], batch size: 50, lr: 2.17e-02, grad_scale: 8.0
2022-12-22 17:24:38,451 WARNING [train.py:1060] (2/4) Exclude cut with ID 7255-291500-0008-134307-0_sp0.9 from training. Duration: 24.6555625
2022-12-22 17:25:07,684 INFO [zipformer.py:660] (2/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=16853.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 17:25:30,074 INFO [zipformer.py:660] (2/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=16868.0, num_to_drop=1, layers_to_drop={0}
2022-12-22 17:25:35,494 WARNING [train.py:1060] (2/4) Exclude cut with ID 3488-65654-0031-41259-0 from training. Duration: 20.44
2022-12-22 17:25:41,409 INFO [train.py:894] (2/4) Epoch 5, batch 2850, loss[loss=0.2921, simple_loss=0.3547, pruned_loss=0.1148, over 18622.00 frames. ], tot_loss[loss=0.2725, simple_loss=0.3318, pruned_loss=0.1066, over 3713615.89 frames. ], batch size: 99, lr: 2.16e-02, grad_scale: 8.0
2022-12-22 17:25:51,534 WARNING [train.py:1060] (2/4) Exclude cut with ID 6951-79737-0018-132285-0_sp0.9 from training. Duration: 23.45
2022-12-22 17:26:04,215 INFO [scaling.py:679] (2/4) Whitening: num_groups=8, num_channels=192, metric=1.93 vs. limit=2.0
2022-12-22 17:26:22,424 WARNING [train.py:1060] (2/4) Exclude cut with ID 6945-60535-0076-12784-0_sp0.9 from training. Duration: 20.52225
2022-12-22 17:26:27,189 INFO [zipformer.py:660] (2/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=16906.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 17:26:28,203 WARNING [train.py:1060] (2/4) Exclude cut with ID 7255-291500-0008-134307-0 from training. Duration: 22.19
2022-12-22 17:26:38,817 WARNING [train.py:1060] (2/4) Exclude cut with ID 7357-94126-0014-15845-0_sp1.1 from training. Duration: 25.3818125
2022-12-22 17:26:43,359 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 4.260e+02 6.499e+02 9.019e+02 1.112e+03 1.634e+03, threshold=1.804e+03, percent-clipped=0.0
2022-12-22 17:26:56,493 WARNING [train.py:1060] (2/4) Exclude cut with ID 6951-79737-0043-132310-0_sp0.9 from training. Duration: 28.0944375
2022-12-22 17:26:58,678 INFO [train.py:894] (2/4) Epoch 5, batch 2900, loss[loss=0.3051, simple_loss=0.3524, pruned_loss=0.1289, over 18608.00 frames. ], tot_loss[loss=0.2723, simple_loss=0.3313, pruned_loss=0.1067, over 3713762.36 frames. ], batch size: 176, lr: 2.16e-02, grad_scale: 8.0
2022-12-22 17:27:01,453 WARNING [train.py:1060] (2/4) Exclude cut with ID 2195-150901-0045-59933-0_sp0.9 from training. Duration: 22.9444375
2022-12-22 17:27:03,717 INFO [zipformer.py:660] (2/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=16929.0, num_to_drop=1, layers_to_drop={3}
2022-12-22 17:27:07,585 INFO [scaling.py:679] (2/4) Whitening: num_groups=8, num_channels=96, metric=1.28 vs. limit=2.0
2022-12-22 17:27:10,670 WARNING [train.py:1060] (2/4) Exclude cut with ID 8291-282929-0024-13011-0_sp1.1 from training. Duration: 21.6318125
2022-12-22 17:27:27,493 WARNING [train.py:1060] (2/4) Exclude cut with ID 8631-249866-0030-130156-0 from training. Duration: 23.695
2022-12-22 17:27:53,609 WARNING [train.py:1060] (2/4) Exclude cut with ID 7699-105389-0094-26379-0 from training. Duration: 23.955
2022-12-22 17:28:01,664 INFO [zipformer.py:660] (2/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=16967.0, num_to_drop=1, layers_to_drop={0}
2022-12-22 17:28:15,318 INFO [train.py:894] (2/4) Epoch 5, batch 2950, loss[loss=0.253, simple_loss=0.3104, pruned_loss=0.09781, over 18529.00 frames. ], tot_loss[loss=0.2705, simple_loss=0.3302, pruned_loss=0.1054, over 3714166.36 frames. ], batch size: 47, lr: 2.16e-02, grad_scale: 8.0
2022-12-22 17:28:24,922 WARNING [train.py:1060] (2/4) Exclude cut with ID 8291-282929-0024-13011-0_sp0.9 from training. Duration: 26.438875
2022-12-22 17:29:08,757 WARNING [train.py:1060] (2/4) Exclude cut with ID 7699-105389-0021-26306-0_sp0.9 from training. Duration: 21.2444375
2022-12-22 17:29:08,775 WARNING [train.py:1060] (2/4) Exclude cut with ID 7357-94126-0014-15845-0_sp0.9 from training. Duration: 31.02225
2022-12-22 17:29:18,056 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 4.179e+02 6.420e+02 8.025e+02 1.086e+03 2.995e+03, threshold=1.605e+03, percent-clipped=4.0
2022-12-22 17:29:20,982 WARNING [train.py:1060] (2/4) Exclude cut with ID 432-122774-0017-62487-0 from training. Duration: 22.395
2022-12-22 17:29:30,961 INFO [train.py:894] (2/4) Epoch 5, batch 3000, loss[loss=0.233, simple_loss=0.2979, pruned_loss=0.08402, over 18629.00 frames. ], tot_loss[loss=0.2703, simple_loss=0.3301, pruned_loss=0.1052, over 3714620.57 frames. ], batch size: 45, lr: 2.15e-02, grad_scale: 8.0
2022-12-22 17:29:30,961 INFO [train.py:919] (2/4) Computing validation loss
2022-12-22 17:29:42,386 INFO [train.py:928] (2/4) Epoch 5, validation: loss=0.2011, simple_loss=0.2985, pruned_loss=0.05191, over 944034.00 frames. 
2022-12-22 17:29:42,387 INFO [train.py:929] (2/4) Maximum memory allocated so far is 24320MB
2022-12-22 17:29:44,511 INFO [zipformer.py:660] (2/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=17027.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 17:29:48,686 WARNING [train.py:1060] (2/4) Exclude cut with ID 7357-94126-0045-15876-0 from training. Duration: 21.075
2022-12-22 17:29:51,870 WARNING [train.py:1060] (2/4) Exclude cut with ID 6482-98857-0025-147532-0_sp0.9 from training. Duration: 20.0055625
2022-12-22 17:29:53,252 WARNING [train.py:1060] (2/4) Exclude cut with ID 6951-79737-0037-132304-0_sp0.9 from training. Duration: 22.05
2022-12-22 17:29:53,264 WARNING [train.py:1060] (2/4) Exclude cut with ID 7255-291500-0003-134302-0 from training. Duration: 26.8349375
2022-12-22 17:29:54,544 INFO [scaling.py:679] (2/4) Whitening: num_groups=8, num_channels=192, metric=1.87 vs. limit=2.0
2022-12-22 17:29:56,381 WARNING [train.py:1060] (2/4) Exclude cut with ID 7357-94126-0009-15840-0_sp1.1 from training. Duration: 22.1090625
2022-12-22 17:30:03,139 WARNING [train.py:1060] (2/4) Exclude cut with ID 7699-105389-0094-26379-0_sp0.9 from training. Duration: 26.6166875
2022-12-22 17:30:21,257 WARNING [train.py:1060] (2/4) Exclude cut with ID 2046-178027-0000-53705-0_sp0.9 from training. Duration: 20.3055625
2022-12-22 17:30:43,953 WARNING [train.py:1060] (2/4) Exclude cut with ID 7205-50138-0008-5373-0_sp0.9 from training. Duration: 20.7
2022-12-22 17:30:57,714 INFO [zipformer.py:660] (2/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=17075.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 17:30:59,055 INFO [train.py:894] (2/4) Epoch 5, batch 3050, loss[loss=0.2986, simple_loss=0.3451, pruned_loss=0.1261, over 18712.00 frames. ], tot_loss[loss=0.2714, simple_loss=0.3309, pruned_loss=0.1059, over 3714600.28 frames. ], batch size: 52, lr: 2.15e-02, grad_scale: 8.0
2022-12-22 17:31:28,556 WARNING [train.py:1060] (2/4) Exclude cut with ID 6978-92210-0019-146985-0 from training. Duration: 22.48
2022-12-22 17:31:46,133 WARNING [train.py:1060] (2/4) Exclude cut with ID 7255-291500-0003-134302-0_sp0.9 from training. Duration: 29.816625
2022-12-22 17:32:01,345 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 4.119e+02 6.435e+02 7.611e+02 9.477e+02 1.701e+03, threshold=1.522e+03, percent-clipped=2.0
2022-12-22 17:32:05,902 WARNING [train.py:1060] (2/4) Exclude cut with ID 7255-291500-0005-134304-0_sp1.1 from training. Duration: 22.7590625
2022-12-22 17:32:10,281 WARNING [train.py:1060] (2/4) Exclude cut with ID 7357-94126-0026-15857-0 from training. Duration: 22.555
2022-12-22 17:32:14,808 INFO [train.py:894] (2/4) Epoch 5, batch 3100, loss[loss=0.3292, simple_loss=0.3667, pruned_loss=0.1459, over 18453.00 frames. ], tot_loss[loss=0.2708, simple_loss=0.3305, pruned_loss=0.1055, over 3714385.64 frames. ], batch size: 50, lr: 2.15e-02, grad_scale: 8.0
2022-12-22 17:32:31,968 WARNING [train.py:1060] (2/4) Exclude cut with ID 1250-135782-0005-25975-0_sp0.9 from training. Duration: 21.688875
2022-12-22 17:33:04,182 WARNING [train.py:1060] (2/4) Exclude cut with ID 3488-85273-0038-41224-0_sp0.9 from training. Duration: 22.6
2022-12-22 17:33:30,779 INFO [train.py:894] (2/4) Epoch 5, batch 3150, loss[loss=0.2621, simple_loss=0.3258, pruned_loss=0.09918, over 18706.00 frames. ], tot_loss[loss=0.2698, simple_loss=0.3296, pruned_loss=0.105, over 3714435.89 frames. ], batch size: 50, lr: 2.15e-02, grad_scale: 8.0
2022-12-22 17:33:43,555 WARNING [train.py:1060] (2/4) Exclude cut with ID 7357-94126-0009-15840-0 from training. Duration: 24.32
2022-12-22 17:33:50,984 INFO [zipformer.py:660] (2/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=17189.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 17:34:33,621 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 4.890e+02 6.624e+02 8.139e+02 1.082e+03 2.327e+03, threshold=1.628e+03, percent-clipped=7.0
2022-12-22 17:34:39,909 WARNING [train.py:1060] (2/4) Exclude cut with ID 8291-276745-0093-13116-0_sp0.9 from training. Duration: 21.061125
2022-12-22 17:34:45,177 INFO [zipformer.py:660] (2/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=17224.0, num_to_drop=1, layers_to_drop={3}
2022-12-22 17:34:48,216 INFO [train.py:894] (2/4) Epoch 5, batch 3200, loss[loss=0.2805, simple_loss=0.3204, pruned_loss=0.1203, over 18537.00 frames. ], tot_loss[loss=0.2706, simple_loss=0.3302, pruned_loss=0.1055, over 3715664.00 frames. ], batch size: 44, lr: 2.14e-02, grad_scale: 8.0
2022-12-22 17:34:55,572 WARNING [train.py:1060] (2/4) Exclude cut with ID 7357-94126-0024-15855-0_sp0.9 from training. Duration: 20.32225
2022-12-22 17:35:07,882 WARNING [train.py:1060] (2/4) Exclude cut with ID 3033-130750-0096-55598-0_sp1.1 from training. Duration: 0.7545625
2022-12-22 17:35:23,348 WARNING [train.py:1060] (2/4) Exclude cut with ID 4295-39940-0007-92567-0_sp0.9 from training. Duration: 23.9333125
2022-12-22 17:35:25,306 INFO [zipformer.py:660] (2/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=17250.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 17:35:43,712 INFO [zipformer.py:660] (2/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=17262.0, num_to_drop=1, layers_to_drop={2}
2022-12-22 17:35:56,054 WARNING [train.py:1060] (2/4) Exclude cut with ID 7255-291500-0008-134307-0_sp1.1 from training. Duration: 20.17275
2022-12-22 17:36:01,878 WARNING [train.py:1060] (2/4) Exclude cut with ID 6978-92210-0019-146985-0_sp1.1 from training. Duration: 20.436375
2022-12-22 17:36:04,733 INFO [train.py:894] (2/4) Epoch 5, batch 3250, loss[loss=0.3042, simple_loss=0.3547, pruned_loss=0.1268, over 18504.00 frames. ], tot_loss[loss=0.2686, simple_loss=0.3285, pruned_loss=0.1043, over 3714961.61 frames. ], batch size: 58, lr: 2.14e-02, grad_scale: 8.0
2022-12-22 17:37:07,497 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 3.347e+02 7.021e+02 8.751e+02 1.059e+03 1.719e+03, threshold=1.750e+03, percent-clipped=2.0
2022-12-22 17:37:20,741 INFO [train.py:894] (2/4) Epoch 5, batch 3300, loss[loss=0.2484, simple_loss=0.3122, pruned_loss=0.09225, over 18702.00 frames. ], tot_loss[loss=0.2693, simple_loss=0.3296, pruned_loss=0.1045, over 3715839.96 frames. ], batch size: 50, lr: 2.14e-02, grad_scale: 8.0
2022-12-22 17:37:27,094 WARNING [train.py:1060] (2/4) Exclude cut with ID 4234-40345-0022-142709-0_sp0.9 from training. Duration: 23.1055625
2022-12-22 17:37:28,313 WARNING [train.py:1060] (2/4) Exclude cut with ID 8291-282929-0007-12994-0_sp1.1 from training. Duration: 23.5
2022-12-22 17:37:40,434 WARNING [train.py:1060] (2/4) Exclude cut with ID 7255-291500-0009-134308-0_sp0.9 from training. Duration: 26.62775
2022-12-22 17:37:50,749 WARNING [train.py:1060] (2/4) Exclude cut with ID 6951-79737-0018-132285-0 from training. Duration: 21.105
2022-12-22 17:37:55,731 WARNING [train.py:1060] (2/4) Exclude cut with ID 4511-76322-0006-80011-0_sp0.9 from training. Duration: 24.411125
2022-12-22 17:38:15,893 INFO [zipformer.py:660] (2/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=17362.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 17:38:22,975 WARNING [train.py:1060] (2/4) Exclude cut with ID 6758-72288-0033-108368-0_sp1.1 from training. Duration: 21.263625
2022-12-22 17:38:37,315 INFO [train.py:894] (2/4) Epoch 5, batch 3350, loss[loss=0.2824, simple_loss=0.3441, pruned_loss=0.1104, over 18695.00 frames. ], tot_loss[loss=0.27, simple_loss=0.3303, pruned_loss=0.1049, over 3716113.17 frames. ], batch size: 62, lr: 2.13e-02, grad_scale: 8.0
2022-12-22 17:38:48,353 INFO [zipformer.py:660] (2/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=17383.0, num_to_drop=1, layers_to_drop={1}
2022-12-22 17:38:53,760 WARNING [train.py:1060] (2/4) Exclude cut with ID 4234-40345-0022-142709-0 from training. Duration: 20.795
2022-12-22 17:39:05,956 WARNING [train.py:1060] (2/4) Exclude cut with ID 7357-94126-0021-15852-0 from training. Duration: 24.76
2022-12-22 17:39:05,977 WARNING [train.py:1060] (2/4) Exclude cut with ID 3867-173237-0077-144769-0_sp0.9 from training. Duration: 22.25
2022-12-22 17:39:32,739 WARNING [train.py:1060] (2/4) Exclude cut with ID 7357-94126-0026-15857-0_sp1.1 from training. Duration: 20.5045625
2022-12-22 17:39:39,975 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 3.521e+02 6.812e+02 8.775e+02 1.065e+03 2.100e+03, threshold=1.755e+03, percent-clipped=2.0
2022-12-22 17:39:45,655 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.5048, 1.5027, 0.9885, 1.7977, 1.4959, 3.2622, 1.2095, 1.4075],
       device='cuda:2'), covar=tensor([0.1000, 0.1712, 0.1507, 0.0992, 0.1593, 0.0281, 0.1464, 0.1696],
       device='cuda:2'), in_proj_covar=tensor([0.0081, 0.0089, 0.0084, 0.0083, 0.0101, 0.0076, 0.0090, 0.0084],
       device='cuda:2'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0003, 0.0004, 0.0003, 0.0003, 0.0003],
       device='cuda:2')
2022-12-22 17:39:50,569 INFO [zipformer.py:660] (2/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=17423.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 17:39:54,350 INFO [train.py:894] (2/4) Epoch 5, batch 3400, loss[loss=0.3074, simple_loss=0.3583, pruned_loss=0.1283, over 18475.00 frames. ], tot_loss[loss=0.2687, simple_loss=0.3291, pruned_loss=0.1041, over 3715192.69 frames. ], batch size: 64, lr: 2.13e-02, grad_scale: 8.0
2022-12-22 17:40:21,377 INFO [zipformer.py:660] (2/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=17444.0, num_to_drop=1, layers_to_drop={3}
2022-12-22 17:41:06,282 INFO [train.py:894] (2/4) Epoch 5, batch 3450, loss[loss=0.3213, simple_loss=0.367, pruned_loss=0.1378, over 18582.00 frames. ], tot_loss[loss=0.2682, simple_loss=0.3285, pruned_loss=0.104, over 3714893.40 frames. ], batch size: 56, lr: 2.13e-02, grad_scale: 8.0
2022-12-22 17:41:08,156 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([0.3912, 1.6302, 2.2275, 0.3430, 1.2972, 2.5696, 1.7370, 1.7169],
       device='cuda:2'), covar=tensor([0.0802, 0.0474, 0.0281, 0.0559, 0.0434, 0.0271, 0.0376, 0.0599],
       device='cuda:2'), in_proj_covar=tensor([0.0123, 0.0135, 0.0091, 0.0119, 0.0116, 0.0093, 0.0126, 0.0113],
       device='cuda:2'), out_proj_covar=tensor([1.2148e-04, 1.3463e-04, 9.1523e-05, 1.1725e-04, 1.1323e-04, 9.3058e-05,
        1.2827e-04, 1.1365e-04], device='cuda:2')
2022-12-22 17:42:04,841 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 4.179e+02 6.499e+02 8.083e+02 1.067e+03 3.667e+03, threshold=1.617e+03, percent-clipped=5.0
2022-12-22 17:42:16,666 INFO [zipformer.py:660] (2/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=17524.0, num_to_drop=1, layers_to_drop={0}
2022-12-22 17:42:19,496 INFO [train.py:894] (2/4) Epoch 5, batch 3500, loss[loss=0.348, simple_loss=0.3733, pruned_loss=0.1613, over 18633.00 frames. ], tot_loss[loss=0.2699, simple_loss=0.3298, pruned_loss=0.1051, over 3714526.34 frames. ], batch size: 175, lr: 2.13e-02, grad_scale: 8.0
2022-12-22 17:42:40,455 WARNING [train.py:1060] (2/4) Exclude cut with ID 7859-102521-0017-7548-0_sp1.1 from training. Duration: 22.2954375
2022-12-22 17:42:51,005 INFO [train.py:894] (2/4) Epoch 6, batch 0, loss[loss=0.2329, simple_loss=0.3075, pruned_loss=0.07911, over 18400.00 frames. ], tot_loss[loss=0.2329, simple_loss=0.3075, pruned_loss=0.07911, over 18400.00 frames. ], batch size: 46, lr: 1.98e-02, grad_scale: 8.0
2022-12-22 17:42:51,005 INFO [train.py:919] (2/4) Computing validation loss
2022-12-22 17:43:02,448 INFO [train.py:928] (2/4) Epoch 6, validation: loss=0.2017, simple_loss=0.2993, pruned_loss=0.05207, over 944034.00 frames. 
2022-12-22 17:43:02,448 INFO [train.py:929] (2/4) Maximum memory allocated so far is 24320MB
2022-12-22 17:43:18,273 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.3093, 1.2708, 1.6756, 0.8874, 1.4041, 1.4009, 1.1198, 1.6535],
       device='cuda:2'), covar=tensor([0.1029, 0.1571, 0.1071, 0.1521, 0.0852, 0.1058, 0.2378, 0.0642],
       device='cuda:2'), in_proj_covar=tensor([0.0196, 0.0183, 0.0200, 0.0191, 0.0185, 0.0207, 0.0199, 0.0176],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-22 17:43:22,700 INFO [zipformer.py:660] (2/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=17545.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 17:43:49,517 INFO [zipformer.py:660] (2/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=17562.0, num_to_drop=1, layers_to_drop={0}
2022-12-22 17:43:52,171 WARNING [train.py:1060] (2/4) Exclude cut with ID 298-126791-0067-24026-0_sp0.9 from training. Duration: 21.438875
2022-12-22 17:43:58,035 WARNING [train.py:1060] (2/4) Exclude cut with ID 5652-39938-0025-23684-0_sp0.9 from training. Duration: 22.2055625
2022-12-22 17:44:04,562 INFO [zipformer.py:660] (2/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=17572.0, num_to_drop=1, layers_to_drop={1}
2022-12-22 17:44:18,705 INFO [train.py:894] (2/4) Epoch 6, batch 50, loss[loss=0.2404, simple_loss=0.3257, pruned_loss=0.0776, over 18679.00 frames. ], tot_loss[loss=0.2403, simple_loss=0.3141, pruned_loss=0.08326, over 837444.95 frames. ], batch size: 60, lr: 1.98e-02, grad_scale: 8.0
2022-12-22 17:44:25,517 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([2.2545, 2.0119, 1.4524, 2.2678, 1.8017, 1.7381, 1.8090, 2.5003],
       device='cuda:2'), covar=tensor([0.1479, 0.2211, 0.1293, 0.2078, 0.2302, 0.0816, 0.2135, 0.0477],
       device='cuda:2'), in_proj_covar=tensor([0.0253, 0.0234, 0.0205, 0.0313, 0.0223, 0.0197, 0.0234, 0.0166],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0003, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-22 17:44:52,776 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([0.0926, 1.1668, 1.5604, 0.2175, 0.8248, 1.7938, 1.5791, 1.3476],
       device='cuda:2'), covar=tensor([0.0554, 0.0277, 0.0294, 0.0402, 0.0381, 0.0243, 0.0241, 0.0517],
       device='cuda:2'), in_proj_covar=tensor([0.0119, 0.0133, 0.0091, 0.0117, 0.0115, 0.0092, 0.0124, 0.0112],
       device='cuda:2'), out_proj_covar=tensor([1.1701e-04, 1.3253e-04, 9.1196e-05, 1.1504e-04, 1.1170e-04, 9.1238e-05,
        1.2523e-04, 1.1250e-04], device='cuda:2')
2022-12-22 17:45:01,189 INFO [zipformer.py:660] (2/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=17610.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 17:45:11,045 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 2.839e+02 4.470e+02 5.623e+02 8.118e+02 1.969e+03, threshold=1.125e+03, percent-clipped=4.0
2022-12-22 17:45:34,093 INFO [train.py:894] (2/4) Epoch 6, batch 100, loss[loss=0.2389, simple_loss=0.3156, pruned_loss=0.08108, over 18530.00 frames. ], tot_loss[loss=0.2377, simple_loss=0.3124, pruned_loss=0.08151, over 1474503.15 frames. ], batch size: 55, lr: 1.98e-02, grad_scale: 8.0
2022-12-22 17:45:51,187 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([0.0523, 0.8397, 1.1155, 0.0398, 0.5524, 1.2423, 1.1752, 1.1472],
       device='cuda:2'), covar=tensor([0.0636, 0.0319, 0.0306, 0.0502, 0.0450, 0.0320, 0.0283, 0.0492],
       device='cuda:2'), in_proj_covar=tensor([0.0118, 0.0132, 0.0091, 0.0117, 0.0115, 0.0093, 0.0123, 0.0112],
       device='cuda:2'), out_proj_covar=tensor([1.1643e-04, 1.3169e-04, 9.0921e-05, 1.1488e-04, 1.1176e-04, 9.2235e-05,
        1.2465e-04, 1.1244e-04], device='cuda:2')
2022-12-22 17:45:55,716 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([2.2335, 2.2679, 1.6030, 2.5886, 2.3904, 1.9722, 2.9329, 2.4605],
       device='cuda:2'), covar=tensor([0.0717, 0.1252, 0.1947, 0.1581, 0.1178, 0.0693, 0.0709, 0.0733],
       device='cuda:2'), in_proj_covar=tensor([0.0167, 0.0175, 0.0216, 0.0264, 0.0213, 0.0174, 0.0181, 0.0183],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-22 17:46:51,278 INFO [train.py:894] (2/4) Epoch 6, batch 150, loss[loss=0.2349, simple_loss=0.3087, pruned_loss=0.0806, over 18665.00 frames. ], tot_loss[loss=0.2345, simple_loss=0.3103, pruned_loss=0.07934, over 1970592.26 frames. ], batch size: 48, lr: 1.98e-02, grad_scale: 8.0
2022-12-22 17:47:00,249 WARNING [train.py:1060] (2/4) Exclude cut with ID 7859-102521-0017-7548-0 from training. Duration: 24.525
2022-12-22 17:47:05,077 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([2.7985, 2.1458, 1.6423, 0.9566, 1.7077, 1.9808, 1.4909, 1.9409],
       device='cuda:2'), covar=tensor([0.0565, 0.0665, 0.1492, 0.1799, 0.1663, 0.1456, 0.1604, 0.1087],
       device='cuda:2'), in_proj_covar=tensor([0.0144, 0.0171, 0.0202, 0.0191, 0.0193, 0.0176, 0.0190, 0.0184],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-22 17:47:16,359 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([2.1448, 1.9004, 2.2275, 1.2630, 2.3886, 2.3224, 1.4080, 2.4750],
       device='cuda:2'), covar=tensor([0.0990, 0.1450, 0.1227, 0.1941, 0.0637, 0.0965, 0.2139, 0.0500],
       device='cuda:2'), in_proj_covar=tensor([0.0191, 0.0178, 0.0195, 0.0188, 0.0179, 0.0203, 0.0195, 0.0170],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-22 17:47:17,689 INFO [zipformer.py:660] (2/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=17700.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 17:47:34,256 WARNING [train.py:1060] (2/4) Exclude cut with ID 3699-47246-0007-3408-0_sp0.9 from training. Duration: 20.26675
2022-12-22 17:47:43,063 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 3.077e+02 5.092e+02 6.350e+02 7.853e+02 1.179e+03, threshold=1.270e+03, percent-clipped=2.0
2022-12-22 17:47:44,713 INFO [zipformer.py:660] (2/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=17718.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 17:47:47,750 WARNING [train.py:1060] (2/4) Exclude cut with ID 7859-102521-0017-7548-0_sp0.9 from training. Duration: 27.25
2022-12-22 17:48:03,302 INFO [zipformer.py:660] (2/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=17730.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 17:48:06,100 INFO [train.py:894] (2/4) Epoch 6, batch 200, loss[loss=0.268, simple_loss=0.3363, pruned_loss=0.09983, over 18633.00 frames. ], tot_loss[loss=0.2352, simple_loss=0.3112, pruned_loss=0.07959, over 2356676.54 frames. ], batch size: 69, lr: 1.97e-02, grad_scale: 8.0
2022-12-22 17:48:17,180 INFO [zipformer.py:660] (2/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=17739.0, num_to_drop=1, layers_to_drop={3}
2022-12-22 17:48:49,553 INFO [zipformer.py:660] (2/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=17761.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 17:48:59,693 WARNING [train.py:1060] (2/4) Exclude cut with ID 6426-64292-0017-15984-0 from training. Duration: 21.68
2022-12-22 17:49:10,479 WARNING [train.py:1060] (2/4) Exclude cut with ID 4278-13270-0007-59342-0 from training. Duration: 21.6300625
2022-12-22 17:49:21,680 INFO [train.py:894] (2/4) Epoch 6, batch 250, loss[loss=0.2376, simple_loss=0.3173, pruned_loss=0.07893, over 18490.00 frames. ], tot_loss[loss=0.2329, simple_loss=0.3083, pruned_loss=0.07876, over 2657127.07 frames. ], batch size: 54, lr: 1.97e-02, grad_scale: 8.0
2022-12-22 17:49:32,154 WARNING [train.py:1060] (2/4) Exclude cut with ID 4278-13270-0007-59342-0_sp0.9 from training. Duration: 24.033375
2022-12-22 17:49:35,419 INFO [zipformer.py:660] (2/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=17791.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 17:50:14,416 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 3.256e+02 5.169e+02 5.967e+02 7.942e+02 1.897e+03, threshold=1.193e+03, percent-clipped=6.0
2022-12-22 17:50:29,194 INFO [scaling.py:679] (2/4) Whitening: num_groups=8, num_channels=192, metric=1.88 vs. limit=2.0
2022-12-22 17:50:30,534 WARNING [train.py:1060] (2/4) Exclude cut with ID 4278-13270-0009-59344-0 from training. Duration: 22.905
2022-12-22 17:50:31,902 WARNING [train.py:1060] (2/4) Exclude cut with ID 5622-44585-0006-90525-0_sp1.1 from training. Duration: 23.4318125
2022-12-22 17:50:37,535 INFO [train.py:894] (2/4) Epoch 6, batch 300, loss[loss=0.2071, simple_loss=0.2734, pruned_loss=0.07037, over 18634.00 frames. ], tot_loss[loss=0.233, simple_loss=0.3083, pruned_loss=0.07879, over 2891162.88 frames. ], batch size: 41, lr: 1.97e-02, grad_scale: 8.0
2022-12-22 17:50:57,039 INFO [zipformer.py:660] (2/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=17845.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 17:51:18,846 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([2.6017, 3.7163, 3.6309, 1.6743, 3.7811, 2.7079, 0.5230, 2.4973],
       device='cuda:2'), covar=tensor([0.1588, 0.0672, 0.1186, 0.3210, 0.0716, 0.1017, 0.5571, 0.1531],
       device='cuda:2'), in_proj_covar=tensor([0.0120, 0.0104, 0.0142, 0.0112, 0.0106, 0.0099, 0.0141, 0.0106],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-22 17:51:53,306 INFO [train.py:894] (2/4) Epoch 6, batch 350, loss[loss=0.2402, simple_loss=0.3217, pruned_loss=0.07937, over 18523.00 frames. ], tot_loss[loss=0.2346, simple_loss=0.3094, pruned_loss=0.07992, over 3073336.10 frames. ], batch size: 55, lr: 1.96e-02, grad_scale: 8.0
2022-12-22 17:52:09,672 INFO [zipformer.py:660] (2/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=17893.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 17:52:30,039 WARNING [train.py:1060] (2/4) Exclude cut with ID 4278-13270-0009-59344-0_sp1.1 from training. Duration: 20.82275
2022-12-22 17:52:31,533 WARNING [train.py:1060] (2/4) Exclude cut with ID 4278-13270-0009-59344-0_sp0.9 from training. Duration: 25.45
2022-12-22 17:52:46,760 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 3.444e+02 5.136e+02 6.255e+02 7.350e+02 1.917e+03, threshold=1.251e+03, percent-clipped=5.0
2022-12-22 17:52:53,729 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([3.7649, 3.2376, 3.1780, 3.6511, 3.3742, 3.3288, 3.9226, 1.1948],
       device='cuda:2'), covar=tensor([0.0709, 0.0608, 0.0653, 0.0581, 0.1405, 0.0908, 0.0564, 0.3980],
       device='cuda:2'), in_proj_covar=tensor([0.0226, 0.0167, 0.0161, 0.0152, 0.0228, 0.0185, 0.0185, 0.0214],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0001, 0.0001, 0.0001, 0.0002, 0.0001, 0.0002, 0.0002],
       device='cuda:2')
2022-12-22 17:53:10,026 INFO [train.py:894] (2/4) Epoch 6, batch 400, loss[loss=0.2429, simple_loss=0.33, pruned_loss=0.07787, over 18575.00 frames. ], tot_loss[loss=0.2367, simple_loss=0.3109, pruned_loss=0.08128, over 3216161.29 frames. ], batch size: 56, lr: 1.96e-02, grad_scale: 8.0
2022-12-22 17:53:32,448 WARNING [train.py:1060] (2/4) Exclude cut with ID 5622-44585-0006-90525-0 from training. Duration: 25.775
2022-12-22 17:53:52,412 WARNING [train.py:1060] (2/4) Exclude cut with ID 2929-85685-0071-62375-0_sp0.9 from training. Duration: 22.25
2022-12-22 17:53:52,749 INFO [zipformer.py:660] (2/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=17960.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 17:54:20,706 WARNING [train.py:1060] (2/4) Exclude cut with ID 3972-170212-0014-23379-0 from training. Duration: 26.205
2022-12-22 17:54:26,347 INFO [train.py:894] (2/4) Epoch 6, batch 450, loss[loss=0.2702, simple_loss=0.3388, pruned_loss=0.1008, over 18605.00 frames. ], tot_loss[loss=0.2379, simple_loss=0.3124, pruned_loss=0.08166, over 3326288.50 frames. ], batch size: 78, lr: 1.96e-02, grad_scale: 8.0
2022-12-22 17:54:37,896 WARNING [train.py:1060] (2/4) Exclude cut with ID 5239-32139-0047-9341-0_sp0.9 from training. Duration: 30.1555625
2022-12-22 17:54:43,719 WARNING [train.py:1060] (2/4) Exclude cut with ID 1265-135635-0050-6781-0_sp0.9 from training. Duration: 21.8333125
2022-12-22 17:54:52,578 WARNING [train.py:1060] (2/4) Exclude cut with ID 1914-133440-0024-94914-0_sp1.1 from training. Duration: 20.6545625
2022-12-22 17:55:08,661 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([3.3765, 3.2600, 1.6881, 2.1211, 4.3546, 4.1555, 2.9786, 2.1650],
       device='cuda:2'), covar=tensor([0.0375, 0.0325, 0.0793, 0.0698, 0.0041, 0.0213, 0.0548, 0.0830],
       device='cuda:2'), in_proj_covar=tensor([0.0119, 0.0111, 0.0130, 0.0120, 0.0074, 0.0115, 0.0138, 0.0149],
       device='cuda:2'), out_proj_covar=tensor([1.4771e-04, 1.3903e-04, 1.5946e-04, 1.4841e-04, 9.4993e-05, 1.4071e-04,
        1.7058e-04, 1.8375e-04], device='cuda:2')
2022-12-22 17:55:21,990 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 3.240e+02 5.677e+02 6.582e+02 7.577e+02 1.434e+03, threshold=1.316e+03, percent-clipped=3.0
2022-12-22 17:55:23,918 INFO [zipformer.py:660] (2/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=18018.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 17:55:28,401 INFO [zipformer.py:660] (2/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=18021.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 17:55:41,907 WARNING [train.py:1060] (2/4) Exclude cut with ID 7395-89880-0045-39920-0_sp0.9 from training. Duration: 20.52225
2022-12-22 17:55:44,507 INFO [train.py:894] (2/4) Epoch 6, batch 500, loss[loss=0.2961, simple_loss=0.3677, pruned_loss=0.1123, over 18698.00 frames. ], tot_loss[loss=0.2383, simple_loss=0.3128, pruned_loss=0.08196, over 3411550.06 frames. ], batch size: 62, lr: 1.96e-02, grad_scale: 8.0
2022-12-22 17:55:55,070 INFO [zipformer.py:660] (2/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=18039.0, num_to_drop=1, layers_to_drop={2}
2022-12-22 17:56:02,926 WARNING [train.py:1060] (2/4) Exclude cut with ID 3972-170212-0014-23379-0_sp0.9 from training. Duration: 29.1166875
2022-12-22 17:56:07,635 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.4989, 1.0749, 0.6949, 1.0315, 1.7975, 0.6009, 1.3393, 1.3865],
       device='cuda:2'), covar=tensor([0.1620, 0.2149, 0.2328, 0.1738, 0.1719, 0.1706, 0.1502, 0.1797],
       device='cuda:2'), in_proj_covar=tensor([0.0092, 0.0106, 0.0126, 0.0099, 0.0110, 0.0093, 0.0097, 0.0101],
       device='cuda:2'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003],
       device='cuda:2')
2022-12-22 17:56:21,311 INFO [zipformer.py:660] (2/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=18056.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 17:56:36,518 INFO [zipformer.py:660] (2/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=18066.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 17:56:59,283 WARNING [train.py:1060] (2/4) Exclude cut with ID 543-133211-0007-59831-0_sp0.9 from training. Duration: 21.388875
2022-12-22 17:57:00,621 INFO [train.py:894] (2/4) Epoch 6, batch 550, loss[loss=0.2315, simple_loss=0.3144, pruned_loss=0.07427, over 18496.00 frames. ], tot_loss[loss=0.2405, simple_loss=0.3147, pruned_loss=0.08309, over 3479040.42 frames. ], batch size: 52, lr: 1.95e-02, grad_scale: 8.0
2022-12-22 17:57:06,916 INFO [zipformer.py:660] (2/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=18086.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 17:57:07,457 INFO [scaling.py:679] (2/4) Whitening: num_groups=1, num_channels=384, metric=3.11 vs. limit=5.0
2022-12-22 17:57:08,314 INFO [zipformer.py:660] (2/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=18087.0, num_to_drop=1, layers_to_drop={1}
2022-12-22 17:57:37,416 WARNING [train.py:1060] (2/4) Exclude cut with ID 1914-133440-0024-94914-0 from training. Duration: 22.72
2022-12-22 17:57:38,914 WARNING [train.py:1060] (2/4) Exclude cut with ID 1914-133440-0031-94921-0_sp0.9 from training. Duration: 22.7444375
2022-12-22 17:57:54,189 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 3.507e+02 5.216e+02 6.559e+02 8.218e+02 2.484e+03, threshold=1.312e+03, percent-clipped=5.0
2022-12-22 17:58:17,191 INFO [train.py:894] (2/4) Epoch 6, batch 600, loss[loss=0.2385, simple_loss=0.3143, pruned_loss=0.08133, over 18574.00 frames. ], tot_loss[loss=0.2398, simple_loss=0.3144, pruned_loss=0.08263, over 3531597.01 frames. ], batch size: 56, lr: 1.95e-02, grad_scale: 8.0
2022-12-22 17:58:23,401 WARNING [train.py:1060] (2/4) Exclude cut with ID 4133-6541-0027-40495-0_sp1.1 from training. Duration: 0.9681875
2022-12-22 17:58:28,318 WARNING [train.py:1060] (2/4) Exclude cut with ID 6330-62851-0022-91297-0_sp0.9 from training. Duration: 22.3166875
2022-12-22 17:58:32,948 WARNING [train.py:1060] (2/4) Exclude cut with ID 543-133212-0015-59917-0_sp0.9 from training. Duration: 21.8166875
2022-12-22 17:59:21,322 INFO [zipformer.py:660] (2/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=18174.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 17:59:22,950 INFO [zipformer.py:660] (2/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=18175.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 17:59:30,934 INFO [scaling.py:679] (2/4) Whitening: num_groups=1, num_channels=384, metric=4.24 vs. limit=5.0
2022-12-22 17:59:32,811 INFO [train.py:894] (2/4) Epoch 6, batch 650, loss[loss=0.2125, simple_loss=0.2886, pruned_loss=0.06817, over 18537.00 frames. ], tot_loss[loss=0.2406, simple_loss=0.315, pruned_loss=0.08309, over 3571546.94 frames. ], batch size: 47, lr: 1.95e-02, grad_scale: 8.0
2022-12-22 18:00:14,585 WARNING [train.py:1060] (2/4) Exclude cut with ID 4957-30119-0041-23990-0_sp0.9 from training. Duration: 20.22775
2022-12-22 18:00:27,697 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 3.151e+02 5.694e+02 6.622e+02 8.175e+02 1.499e+03, threshold=1.324e+03, percent-clipped=1.0
2022-12-22 18:00:33,234 INFO [scaling.py:679] (2/4) Whitening: num_groups=1, num_channels=384, metric=4.90 vs. limit=5.0
2022-12-22 18:00:49,605 INFO [train.py:894] (2/4) Epoch 6, batch 700, loss[loss=0.2231, simple_loss=0.291, pruned_loss=0.0776, over 18701.00 frames. ], tot_loss[loss=0.2402, simple_loss=0.3145, pruned_loss=0.08295, over 3603647.02 frames. ], batch size: 46, lr: 1.95e-02, grad_scale: 8.0
2022-12-22 18:00:54,948 INFO [zipformer.py:660] (2/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=18235.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 18:00:56,440 INFO [zipformer.py:660] (2/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=18236.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 18:01:00,418 WARNING [train.py:1060] (2/4) Exclude cut with ID 5239-32139-0047-9341-0_sp1.1 from training. Duration: 24.67275
2022-12-22 18:01:27,521 WARNING [train.py:1060] (2/4) Exclude cut with ID 3082-165428-0081-50734-0_sp0.9 from training. Duration: 21.8055625
2022-12-22 18:01:52,361 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.6166, 1.1887, 0.7750, 1.2381, 1.6864, 1.2954, 1.5926, 1.7631],
       device='cuda:2'), covar=tensor([0.1630, 0.2158, 0.2750, 0.1700, 0.2005, 0.1527, 0.1397, 0.1505],
       device='cuda:2'), in_proj_covar=tensor([0.0092, 0.0106, 0.0126, 0.0100, 0.0110, 0.0094, 0.0097, 0.0100],
       device='cuda:2'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003],
       device='cuda:2')
2022-12-22 18:02:05,108 INFO [train.py:894] (2/4) Epoch 6, batch 750, loss[loss=0.203, simple_loss=0.2881, pruned_loss=0.05894, over 18673.00 frames. ], tot_loss[loss=0.2405, simple_loss=0.3155, pruned_loss=0.08276, over 3627869.49 frames. ], batch size: 48, lr: 1.94e-02, grad_scale: 8.0
2022-12-22 18:02:05,166 WARNING [train.py:1060] (2/4) Exclude cut with ID 3340-169293-0054-76830-0_sp0.9 from training. Duration: 22.6666875
2022-12-22 18:02:07,967 INFO [scaling.py:679] (2/4) Whitening: num_groups=8, num_channels=96, metric=1.68 vs. limit=2.0
2022-12-22 18:02:25,743 INFO [scaling.py:679] (2/4) Whitening: num_groups=8, num_channels=96, metric=1.23 vs. limit=2.0
2022-12-22 18:02:40,126 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.6465, 1.7319, 1.2044, 1.8360, 1.5764, 1.5074, 1.5132, 1.7681],
       device='cuda:2'), covar=tensor([0.1636, 0.1906, 0.1324, 0.1722, 0.2013, 0.0782, 0.1812, 0.0590],
       device='cuda:2'), in_proj_covar=tensor([0.0253, 0.0237, 0.0208, 0.0315, 0.0226, 0.0195, 0.0237, 0.0168],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0003, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-22 18:02:57,205 INFO [zipformer.py:660] (2/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=18316.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 18:02:58,473 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 2.968e+02 5.188e+02 6.272e+02 8.631e+02 1.940e+03, threshold=1.254e+03, percent-clipped=5.0
2022-12-22 18:03:09,633 WARNING [train.py:1060] (2/4) Exclude cut with ID 2411-132532-0017-82279-0_sp1.1 from training. Duration: 0.9681875
2022-12-22 18:03:20,391 INFO [train.py:894] (2/4) Epoch 6, batch 800, loss[loss=0.2107, simple_loss=0.292, pruned_loss=0.06464, over 18679.00 frames. ], tot_loss[loss=0.2404, simple_loss=0.3156, pruned_loss=0.08259, over 3646762.28 frames. ], batch size: 48, lr: 1.94e-02, grad_scale: 8.0
2022-12-22 18:03:36,003 WARNING [train.py:1060] (2/4) Exclude cut with ID 6330-62850-0007-91323-0 from training. Duration: 22.485
2022-12-22 18:03:56,467 INFO [zipformer.py:660] (2/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=18356.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 18:04:15,038 WARNING [train.py:1060] (2/4) Exclude cut with ID 3972-170212-0014-23379-0_sp1.1 from training. Duration: 23.82275
2022-12-22 18:04:19,692 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.6869, 0.5700, 1.4772, 1.3599, 1.7447, 1.5540, 1.2542, 1.1517],
       device='cuda:2'), covar=tensor([0.1043, 0.1601, 0.1320, 0.1249, 0.0802, 0.0538, 0.1348, 0.0679],
       device='cuda:2'), in_proj_covar=tensor([0.0212, 0.0246, 0.0220, 0.0240, 0.0223, 0.0200, 0.0238, 0.0191],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-22 18:04:27,983 WARNING [train.py:1060] (2/4) Exclude cut with ID 4860-13185-0032-76709-0 from training. Duration: 20.77
2022-12-22 18:04:35,184 INFO [train.py:894] (2/4) Epoch 6, batch 850, loss[loss=0.2348, simple_loss=0.3148, pruned_loss=0.07735, over 18566.00 frames. ], tot_loss[loss=0.2393, simple_loss=0.3143, pruned_loss=0.08213, over 3661586.24 frames. ], batch size: 57, lr: 1.94e-02, grad_scale: 8.0
2022-12-22 18:04:35,198 WARNING [train.py:1060] (2/4) Exclude cut with ID 6426-64292-0017-15984-0_sp0.9 from training. Duration: 24.088875
2022-12-22 18:04:41,619 INFO [zipformer.py:660] (2/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=18386.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 18:05:02,340 WARNING [train.py:1060] (2/4) Exclude cut with ID 6330-62850-0007-91323-0_sp1.1 from training. Duration: 20.4409375
2022-12-22 18:05:09,319 INFO [zipformer.py:660] (2/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=18404.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 18:05:29,138 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 3.354e+02 5.112e+02 6.131e+02 8.014e+02 1.589e+03, threshold=1.226e+03, percent-clipped=3.0
2022-12-22 18:05:51,626 INFO [train.py:894] (2/4) Epoch 6, batch 900, loss[loss=0.2146, simple_loss=0.288, pruned_loss=0.07064, over 18668.00 frames. ], tot_loss[loss=0.2383, simple_loss=0.3137, pruned_loss=0.08142, over 3672957.21 frames. ], batch size: 41, lr: 1.94e-02, grad_scale: 8.0
2022-12-22 18:05:54,910 INFO [zipformer.py:660] (2/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=18434.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 18:06:22,677 WARNING [train.py:1060] (2/4) Exclude cut with ID 2929-85685-0062-62366-0_sp0.9 from training. Duration: 22.511125
2022-12-22 18:06:22,701 WARNING [train.py:1060] (2/4) Exclude cut with ID 7395-89880-0031-39906-0 from training. Duration: 20.675
2022-12-22 18:07:07,803 INFO [train.py:894] (2/4) Epoch 6, batch 950, loss[loss=0.2402, simple_loss=0.3171, pruned_loss=0.08171, over 18680.00 frames. ], tot_loss[loss=0.2371, simple_loss=0.3127, pruned_loss=0.08075, over 3682045.47 frames. ], batch size: 60, lr: 1.93e-02, grad_scale: 8.0
2022-12-22 18:07:20,638 INFO [zipformer.py:660] (2/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=18490.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 18:07:41,917 INFO [scaling.py:679] (2/4) Whitening: num_groups=8, num_channels=192, metric=1.88 vs. limit=2.0
2022-12-22 18:07:47,176 INFO [zipformer.py:660] (2/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=18508.0, num_to_drop=1, layers_to_drop={1}
2022-12-22 18:08:00,283 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 3.265e+02 5.277e+02 6.294e+02 8.257e+02 2.016e+03, threshold=1.259e+03, percent-clipped=4.0
2022-12-22 18:08:06,517 WARNING [train.py:1060] (2/4) Exclude cut with ID 6330-62850-0007-91323-0_sp0.9 from training. Duration: 24.9833125
2022-12-22 18:08:19,596 INFO [zipformer.py:660] (2/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=18530.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 18:08:21,858 INFO [zipformer.py:660] (2/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=18531.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 18:08:23,116 INFO [train.py:894] (2/4) Epoch 6, batch 1000, loss[loss=0.2417, simple_loss=0.3213, pruned_loss=0.08101, over 18632.00 frames. ], tot_loss[loss=0.2382, simple_loss=0.3135, pruned_loss=0.08141, over 3688496.50 frames. ], batch size: 53, lr: 1.93e-02, grad_scale: 8.0
2022-12-22 18:08:35,648 WARNING [train.py:1060] (2/4) Exclude cut with ID 5239-32139-0047-9341-0 from training. Duration: 27.14
2022-12-22 18:08:51,256 INFO [zipformer.py:660] (2/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=18551.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 18:08:53,076 WARNING [train.py:1060] (2/4) Exclude cut with ID 2929-85685-0044-62348-0 from training. Duration: 22.44
2022-12-22 18:09:03,378 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.8054, 1.2368, 0.6303, 1.2095, 1.8887, 1.4091, 1.3287, 1.9511],
       device='cuda:2'), covar=tensor([0.1675, 0.2206, 0.2984, 0.1804, 0.1918, 0.1527, 0.1588, 0.1512],
       device='cuda:2'), in_proj_covar=tensor([0.0094, 0.0106, 0.0127, 0.0102, 0.0110, 0.0094, 0.0097, 0.0100],
       device='cuda:2'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003],
       device='cuda:2')
2022-12-22 18:09:18,876 INFO [zipformer.py:660] (2/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=18569.0, num_to_drop=1, layers_to_drop={0}
2022-12-22 18:09:36,968 INFO [train.py:894] (2/4) Epoch 6, batch 1050, loss[loss=0.2148, simple_loss=0.3066, pruned_loss=0.06156, over 18481.00 frames. ], tot_loss[loss=0.2376, simple_loss=0.313, pruned_loss=0.08111, over 3694373.63 frames. ], batch size: 54, lr: 1.93e-02, grad_scale: 8.0
2022-12-22 18:10:09,785 WARNING [train.py:1060] (2/4) Exclude cut with ID 2929-85685-0060-62364-0_sp0.9 from training. Duration: 21.361125
2022-12-22 18:10:15,372 WARNING [train.py:1060] (2/4) Exclude cut with ID 2929-85685-0079-62383-0_sp1.1 from training. Duration: 27.0318125
2022-12-22 18:10:25,697 INFO [scaling.py:679] (2/4) Whitening: num_groups=8, num_channels=192, metric=1.91 vs. limit=2.0
2022-12-22 18:10:26,286 WARNING [train.py:1060] (2/4) Exclude cut with ID 5622-44585-0006-90525-0_sp0.9 from training. Duration: 28.638875
2022-12-22 18:10:29,170 INFO [zipformer.py:660] (2/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=18616.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 18:10:30,228 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 3.146e+02 5.294e+02 6.925e+02 9.252e+02 1.661e+03, threshold=1.385e+03, percent-clipped=6.0
2022-12-22 18:10:42,050 WARNING [train.py:1060] (2/4) Exclude cut with ID 3340-169293-0054-76830-0 from training. Duration: 20.4
2022-12-22 18:10:52,470 INFO [train.py:894] (2/4) Epoch 6, batch 1100, loss[loss=0.2526, simple_loss=0.337, pruned_loss=0.08414, over 18616.00 frames. ], tot_loss[loss=0.2391, simple_loss=0.3144, pruned_loss=0.08191, over 3698119.74 frames. ], batch size: 77, lr: 1.93e-02, grad_scale: 16.0
2022-12-22 18:11:15,798 WARNING [train.py:1060] (2/4) Exclude cut with ID 2929-85685-0071-62375-0 from training. Duration: 20.025
2022-12-22 18:11:15,812 WARNING [train.py:1060] (2/4) Exclude cut with ID 2364-131735-0112-64612-0_sp0.9 from training. Duration: 20.488875
2022-12-22 18:11:16,057 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([4.5658, 3.8790, 3.7953, 4.4839, 4.0948, 4.0655, 4.6706, 1.1829],
       device='cuda:2'), covar=tensor([0.0478, 0.0472, 0.0485, 0.0399, 0.1165, 0.0847, 0.0414, 0.4484],
       device='cuda:2'), in_proj_covar=tensor([0.0229, 0.0166, 0.0162, 0.0152, 0.0228, 0.0188, 0.0186, 0.0219],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0001, 0.0001, 0.0001, 0.0002, 0.0001, 0.0002, 0.0002],
       device='cuda:2')
2022-12-22 18:11:21,872 WARNING [train.py:1060] (2/4) Exclude cut with ID 2929-85685-0079-62383-0 from training. Duration: 29.735
2022-12-22 18:11:40,277 INFO [zipformer.py:660] (2/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=18664.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 18:11:49,564 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([0.0401, 0.7572, 1.1406, 0.0935, 0.6171, 1.2482, 1.1853, 1.1702],
       device='cuda:2'), covar=tensor([0.0663, 0.0339, 0.0288, 0.0426, 0.0448, 0.0305, 0.0261, 0.0444],
       device='cuda:2'), in_proj_covar=tensor([0.0119, 0.0129, 0.0089, 0.0114, 0.0116, 0.0093, 0.0121, 0.0111],
       device='cuda:2'), out_proj_covar=tensor([1.1548e-04, 1.2680e-04, 8.7306e-05, 1.1086e-04, 1.1177e-04, 9.0641e-05,
        1.2058e-04, 1.0793e-04], device='cuda:2')
2022-12-22 18:11:50,783 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.6389, 1.0916, 0.7729, 1.3105, 1.7560, 1.4496, 1.2993, 1.8076],
       device='cuda:2'), covar=tensor([0.2033, 0.2800, 0.2939, 0.2083, 0.2049, 0.1787, 0.1804, 0.2094],
       device='cuda:2'), in_proj_covar=tensor([0.0093, 0.0106, 0.0127, 0.0102, 0.0110, 0.0095, 0.0098, 0.0100],
       device='cuda:2'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003],
       device='cuda:2')
2022-12-22 18:11:55,067 INFO [zipformer.py:660] (2/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=18674.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 18:12:06,854 INFO [train.py:894] (2/4) Epoch 6, batch 1150, loss[loss=0.2385, simple_loss=0.3081, pruned_loss=0.08447, over 18545.00 frames. ], tot_loss[loss=0.2388, simple_loss=0.3143, pruned_loss=0.0816, over 3702353.58 frames. ], batch size: 47, lr: 1.93e-02, grad_scale: 16.0
2022-12-22 18:12:19,987 INFO [scaling.py:679] (2/4) Whitening: num_groups=8, num_channels=96, metric=1.41 vs. limit=2.0
2022-12-22 18:12:42,286 WARNING [train.py:1060] (2/4) Exclude cut with ID 7276-92427-0014-12983-0_sp0.9 from training. Duration: 21.3055625
2022-12-22 18:12:43,694 WARNING [train.py:1060] (2/4) Exclude cut with ID 1025-75365-0008-79168-0_sp0.9 from training. Duration: 22.0666875
2022-12-22 18:12:59,392 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 3.797e+02 5.337e+02 6.606e+02 8.749e+02 2.715e+03, threshold=1.321e+03, percent-clipped=4.0
2022-12-22 18:13:21,629 INFO [train.py:894] (2/4) Epoch 6, batch 1200, loss[loss=0.216, simple_loss=0.3023, pruned_loss=0.06489, over 18500.00 frames. ], tot_loss[loss=0.2367, simple_loss=0.3126, pruned_loss=0.08045, over 3703876.17 frames. ], batch size: 52, lr: 1.92e-02, grad_scale: 16.0
2022-12-22 18:13:26,758 INFO [zipformer.py:660] (2/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=18735.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 18:14:32,220 WARNING [train.py:1060] (2/4) Exclude cut with ID 2929-85685-0062-62366-0 from training. Duration: 20.26
2022-12-22 18:14:38,456 INFO [train.py:894] (2/4) Epoch 6, batch 1250, loss[loss=0.2497, simple_loss=0.3279, pruned_loss=0.0858, over 18619.00 frames. ], tot_loss[loss=0.236, simple_loss=0.3116, pruned_loss=0.08018, over 3705448.84 frames. ], batch size: 53, lr: 1.92e-02, grad_scale: 16.0
2022-12-22 18:14:40,767 INFO [scaling.py:679] (2/4) Whitening: num_groups=8, num_channels=96, metric=1.20 vs. limit=2.0
2022-12-22 18:14:44,487 WARNING [train.py:1060] (2/4) Exclude cut with ID 5239-32139-0030-9324-0_sp0.9 from training. Duration: 21.3444375
2022-12-22 18:15:32,035 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 3.330e+02 4.900e+02 6.256e+02 8.296e+02 1.790e+03, threshold=1.251e+03, percent-clipped=3.0
2022-12-22 18:15:39,447 WARNING [train.py:1060] (2/4) Exclude cut with ID 497-129325-0061-62254-0_sp1.1 from training. Duration: 0.97725
2022-12-22 18:15:52,915 INFO [zipformer.py:660] (2/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=18830.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 18:15:54,486 INFO [zipformer.py:660] (2/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=18831.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 18:15:55,637 INFO [train.py:894] (2/4) Epoch 6, batch 1300, loss[loss=0.2557, simple_loss=0.3181, pruned_loss=0.09661, over 18594.00 frames. ], tot_loss[loss=0.2352, simple_loss=0.3109, pruned_loss=0.07979, over 3707968.57 frames. ], batch size: 51, lr: 1.92e-02, grad_scale: 16.0
2022-12-22 18:16:16,778 INFO [zipformer.py:660] (2/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=18846.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 18:16:22,823 WARNING [train.py:1060] (2/4) Exclude cut with ID 7395-89880-0031-39906-0_sp0.9 from training. Duration: 22.97225
2022-12-22 18:16:23,362 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.2509, 1.6442, 1.3128, 2.1883, 2.1881, 1.4534, 1.3671, 1.0949],
       device='cuda:2'), covar=tensor([0.1848, 0.1610, 0.1367, 0.0783, 0.1110, 0.1142, 0.1562, 0.1440],
       device='cuda:2'), in_proj_covar=tensor([0.0223, 0.0195, 0.0186, 0.0175, 0.0235, 0.0176, 0.0194, 0.0176],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-22 18:16:43,933 INFO [zipformer.py:660] (2/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=18864.0, num_to_drop=1, layers_to_drop={3}
2022-12-22 18:16:53,974 WARNING [train.py:1060] (2/4) Exclude cut with ID 7395-89880-0047-39922-0_sp0.9 from training. Duration: 21.97775
2022-12-22 18:17:05,346 INFO [zipformer.py:660] (2/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=18878.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 18:17:06,666 INFO [zipformer.py:660] (2/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=18879.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 18:17:07,944 WARNING [train.py:1060] (2/4) Exclude cut with ID 1112-1043-0006-89194-0_sp0.9 from training. Duration: 21.8333125
2022-12-22 18:17:09,485 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.7600, 1.0282, 1.8542, 2.9836, 2.1678, 2.4694, 0.6047, 1.9985],
       device='cuda:2'), covar=tensor([0.1978, 0.2596, 0.1767, 0.0720, 0.1471, 0.1356, 0.3002, 0.1558],
       device='cuda:2'), in_proj_covar=tensor([0.0104, 0.0114, 0.0123, 0.0098, 0.0104, 0.0124, 0.0133, 0.0106],
       device='cuda:2'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0002, 0.0003, 0.0003, 0.0003, 0.0003],
       device='cuda:2')
2022-12-22 18:17:10,449 INFO [train.py:894] (2/4) Epoch 6, batch 1350, loss[loss=0.2378, simple_loss=0.3111, pruned_loss=0.08222, over 18544.00 frames. ], tot_loss[loss=0.2358, simple_loss=0.3117, pruned_loss=0.07994, over 3709740.94 frames. ], batch size: 55, lr: 1.92e-02, grad_scale: 8.0
2022-12-22 18:17:18,048 WARNING [train.py:1060] (2/4) Exclude cut with ID 1914-133440-0031-94921-0 from training. Duration: 20.47
2022-12-22 18:18:04,571 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 3.035e+02 5.335e+02 6.297e+02 7.950e+02 1.792e+03, threshold=1.259e+03, percent-clipped=3.0
2022-12-22 18:18:22,339 WARNING [train.py:1060] (2/4) Exclude cut with ID 7395-89880-0037-39912-0_sp0.9 from training. Duration: 20.67225
2022-12-22 18:18:24,031 INFO [zipformer.py:660] (2/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=18930.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 18:18:26,870 INFO [train.py:894] (2/4) Epoch 6, batch 1400, loss[loss=0.1973, simple_loss=0.2703, pruned_loss=0.06212, over 18620.00 frames. ], tot_loss[loss=0.2354, simple_loss=0.3111, pruned_loss=0.07985, over 3710801.54 frames. ], batch size: 45, lr: 1.91e-02, grad_scale: 8.0
2022-12-22 18:18:40,692 WARNING [train.py:1060] (2/4) Exclude cut with ID 1914-133440-0024-94914-0_sp0.9 from training. Duration: 25.2444375
2022-12-22 18:18:50,909 INFO [scaling.py:679] (2/4) Whitening: num_groups=8, num_channels=192, metric=2.00 vs. limit=2.0
2022-12-22 18:19:00,380 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.7882, 1.2324, 0.7371, 1.4387, 2.0624, 1.4204, 1.5593, 1.8866],
       device='cuda:2'), covar=tensor([0.1659, 0.2186, 0.3005, 0.1660, 0.1770, 0.1580, 0.1598, 0.1623],
       device='cuda:2'), in_proj_covar=tensor([0.0095, 0.0107, 0.0129, 0.0103, 0.0112, 0.0097, 0.0101, 0.0101],
       device='cuda:2'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003],
       device='cuda:2')
2022-12-22 18:19:04,175 WARNING [train.py:1060] (2/4) Exclude cut with ID 3340-169293-0021-76797-0_sp0.9 from training. Duration: 21.1445
2022-12-22 18:19:42,952 INFO [train.py:894] (2/4) Epoch 6, batch 1450, loss[loss=0.2414, simple_loss=0.3252, pruned_loss=0.07884, over 18693.00 frames. ], tot_loss[loss=0.2349, simple_loss=0.3106, pruned_loss=0.07965, over 3712004.61 frames. ], batch size: 60, lr: 1.91e-02, grad_scale: 8.0
2022-12-22 18:19:57,311 INFO [zipformer.py:660] (2/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=18991.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 18:20:19,056 WARNING [train.py:1060] (2/4) Exclude cut with ID 2929-85685-0079-62383-0_sp0.9 from training. Duration: 33.038875
2022-12-22 18:20:37,420 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 3.579e+02 4.874e+02 5.865e+02 6.786e+02 3.300e+03, threshold=1.173e+03, percent-clipped=4.0
2022-12-22 18:20:56,573 INFO [zipformer.py:660] (2/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=19030.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 18:20:57,923 WARNING [train.py:1060] (2/4) Exclude cut with ID 6426-64291-0000-16059-0_sp0.9 from training. Duration: 20.0944375
2022-12-22 18:20:59,352 INFO [train.py:894] (2/4) Epoch 6, batch 1500, loss[loss=0.243, simple_loss=0.3207, pruned_loss=0.08264, over 18472.00 frames. ], tot_loss[loss=0.2342, simple_loss=0.3103, pruned_loss=0.07902, over 3713125.28 frames. ], batch size: 54, lr: 1.91e-02, grad_scale: 8.0
2022-12-22 18:21:12,828 WARNING [train.py:1060] (2/4) Exclude cut with ID 2929-85685-0044-62348-0_sp1.1 from training. Duration: 20.4
2022-12-22 18:21:20,091 WARNING [train.py:1060] (2/4) Exclude cut with ID 6330-62851-0022-91297-0 from training. Duration: 20.085
2022-12-22 18:21:28,220 INFO [scaling.py:679] (2/4) Whitening: num_groups=8, num_channels=192, metric=1.85 vs. limit=2.0
2022-12-22 18:21:30,204 WARNING [train.py:1060] (2/4) Exclude cut with ID 4860-13185-0032-76709-0_sp0.9 from training. Duration: 23.07775
2022-12-22 18:22:14,904 INFO [train.py:894] (2/4) Epoch 6, batch 1550, loss[loss=0.2535, simple_loss=0.3316, pruned_loss=0.08777, over 18384.00 frames. ], tot_loss[loss=0.2336, simple_loss=0.3098, pruned_loss=0.07871, over 3713084.87 frames. ], batch size: 53, lr: 1.91e-02, grad_scale: 8.0
2022-12-22 18:22:17,588 WARNING [train.py:1060] (2/4) Exclude cut with ID 2929-85685-0044-62348-0_sp0.9 from training. Duration: 24.9333125
2022-12-22 18:22:51,726 INFO [scaling.py:679] (2/4) Whitening: num_groups=1, num_channels=384, metric=5.04 vs. limit=5.0
2022-12-22 18:22:55,414 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([3.3469, 2.3211, 1.7210, 2.9251, 2.0563, 2.2928, 2.3116, 3.5736],
       device='cuda:2'), covar=tensor([0.1161, 0.2208, 0.1271, 0.2012, 0.2422, 0.0705, 0.2058, 0.0372],
       device='cuda:2'), in_proj_covar=tensor([0.0248, 0.0234, 0.0203, 0.0310, 0.0224, 0.0194, 0.0235, 0.0164],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0003, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-22 18:22:58,376 INFO [scaling.py:679] (2/4) Whitening: num_groups=1, num_channels=384, metric=2.71 vs. limit=5.0
2022-12-22 18:23:04,285 WARNING [train.py:1060] (2/4) Exclude cut with ID 5118-111612-0016-124680-0_sp0.9 from training. Duration: 20.388875
2022-12-22 18:23:08,351 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 3.609e+02 5.320e+02 6.098e+02 7.401e+02 2.797e+03, threshold=1.220e+03, percent-clipped=4.0
2022-12-22 18:23:09,976 WARNING [train.py:1060] (2/4) Exclude cut with ID 432-122774-0017-62487-0_sp1.1 from training. Duration: 20.3590625
2022-12-22 18:23:25,943 INFO [zipformer.py:660] (2/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=19129.0, num_to_drop=1, layers_to_drop={1}
2022-12-22 18:23:29,943 INFO [train.py:894] (2/4) Epoch 6, batch 1600, loss[loss=0.1896, simple_loss=0.2595, pruned_loss=0.05979, over 18613.00 frames. ], tot_loss[loss=0.2325, simple_loss=0.3091, pruned_loss=0.07801, over 3713535.39 frames. ], batch size: 41, lr: 1.90e-02, grad_scale: 8.0
2022-12-22 18:23:51,383 INFO [zipformer.py:660] (2/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=19146.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 18:24:17,821 INFO [zipformer.py:660] (2/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=19164.0, num_to_drop=1, layers_to_drop={2}
2022-12-22 18:24:20,433 WARNING [train.py:1060] (2/4) Exclude cut with ID 3557-8342-0013-54691-0_sp1.1 from training. Duration: 0.836375
2022-12-22 18:24:29,634 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([2.6736, 3.3981, 3.5100, 1.4324, 3.5210, 2.4997, 0.8120, 2.3033],
       device='cuda:2'), covar=tensor([0.1773, 0.1048, 0.1395, 0.3630, 0.1012, 0.1171, 0.5398, 0.1673],
       device='cuda:2'), in_proj_covar=tensor([0.0123, 0.0101, 0.0141, 0.0112, 0.0104, 0.0094, 0.0138, 0.0104],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-22 18:24:46,279 INFO [train.py:894] (2/4) Epoch 6, batch 1650, loss[loss=0.206, simple_loss=0.2754, pruned_loss=0.06834, over 18532.00 frames. ], tot_loss[loss=0.2356, simple_loss=0.3108, pruned_loss=0.08017, over 3713528.26 frames. ], batch size: 44, lr: 1.90e-02, grad_scale: 8.0
2022-12-22 18:24:53,965 INFO [scaling.py:679] (2/4) Whitening: num_groups=8, num_channels=96, metric=1.52 vs. limit=2.0
2022-12-22 18:24:59,003 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.6218, 1.5892, 1.3058, 1.9909, 1.5118, 3.4032, 1.5975, 1.4920],
       device='cuda:2'), covar=tensor([0.1249, 0.2348, 0.1539, 0.1174, 0.1900, 0.0285, 0.1784, 0.2081],
       device='cuda:2'), in_proj_covar=tensor([0.0079, 0.0088, 0.0084, 0.0082, 0.0101, 0.0074, 0.0090, 0.0081],
       device='cuda:2'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0003, 0.0004, 0.0003, 0.0003, 0.0003],
       device='cuda:2')
2022-12-22 18:24:59,063 INFO [zipformer.py:660] (2/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=19190.0, num_to_drop=1, layers_to_drop={0}
2022-12-22 18:25:04,719 INFO [zipformer.py:660] (2/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=19194.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 18:25:06,038 WARNING [train.py:1060] (2/4) Exclude cut with ID 8565-290391-0049-67394-0_sp0.9 from training. Duration: 21.3166875
2022-12-22 18:25:18,744 INFO [scaling.py:679] (2/4) Whitening: num_groups=1, num_channels=384, metric=3.81 vs. limit=5.0
2022-12-22 18:25:31,111 INFO [zipformer.py:660] (2/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=19212.0, num_to_drop=1, layers_to_drop={1}
2022-12-22 18:25:35,152 WARNING [train.py:1060] (2/4) Exclude cut with ID 6533-399-0029-104863-0_sp0.9 from training. Duration: 22.1055625
2022-12-22 18:25:40,079 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 3.440e+02 5.588e+02 6.972e+02 9.071e+02 2.121e+03, threshold=1.394e+03, percent-clipped=5.0
2022-12-22 18:25:44,726 WARNING [train.py:1060] (2/4) Exclude cut with ID 7699-105389-0094-26379-0_sp1.1 from training. Duration: 21.77725
2022-12-22 18:25:47,763 INFO [scaling.py:679] (2/4) Whitening: num_groups=8, num_channels=192, metric=1.88 vs. limit=2.0
2022-12-22 18:26:02,485 INFO [train.py:894] (2/4) Epoch 6, batch 1700, loss[loss=0.2601, simple_loss=0.3324, pruned_loss=0.09394, over 18652.00 frames. ], tot_loss[loss=0.2411, simple_loss=0.3141, pruned_loss=0.08404, over 3713138.05 frames. ], batch size: 98, lr: 1.90e-02, grad_scale: 8.0
2022-12-22 18:26:05,360 WARNING [train.py:1060] (2/4) Exclude cut with ID 7255-291500-0005-134304-0_sp0.9 from training. Duration: 27.8166875
2022-12-22 18:26:09,428 INFO [scaling.py:679] (2/4) Whitening: num_groups=8, num_channels=192, metric=1.87 vs. limit=2.0
2022-12-22 18:26:28,957 WARNING [train.py:1060] (2/4) Exclude cut with ID 7357-94126-0021-15852-0_sp1.1 from training. Duration: 22.5090625
2022-12-22 18:26:33,279 WARNING [train.py:1060] (2/4) Exclude cut with ID 7255-291500-0005-134304-0 from training. Duration: 25.035
2022-12-22 18:26:52,403 WARNING [train.py:1060] (2/4) Exclude cut with ID 774-127930-0014-10412-0_sp1.1 from training. Duration: 0.95
2022-12-22 18:27:10,368 WARNING [train.py:1060] (2/4) Exclude cut with ID 3033-130750-0096-55598-0_sp0.9 from training. Duration: 0.92225
2022-12-22 18:27:18,391 INFO [train.py:894] (2/4) Epoch 6, batch 1750, loss[loss=0.2606, simple_loss=0.3165, pruned_loss=0.1023, over 18383.00 frames. ], tot_loss[loss=0.2453, simple_loss=0.3164, pruned_loss=0.08708, over 3713492.15 frames. ], batch size: 46, lr: 1.90e-02, grad_scale: 8.0
2022-12-22 18:27:24,629 INFO [zipformer.py:660] (2/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=19286.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 18:27:35,122 WARNING [train.py:1060] (2/4) Exclude cut with ID 4511-76322-0006-80011-0 from training. Duration: 21.97
2022-12-22 18:27:53,753 WARNING [train.py:1060] (2/4) Exclude cut with ID 7492-105653-0055-62765-0_sp0.9 from training. Duration: 21.97225
2022-12-22 18:27:55,031 WARNING [train.py:1060] (2/4) Exclude cut with ID 453-131332-0000-47844-0_sp0.9 from training. Duration: 25.3333125
2022-12-22 18:28:06,737 WARNING [train.py:1060] (2/4) Exclude cut with ID 5172-29468-0015-19128-0_sp0.9 from training. Duration: 21.5055625
2022-12-22 18:28:10,406 INFO [scaling.py:679] (2/4) Whitening: num_groups=8, num_channels=192, metric=1.85 vs. limit=2.0
2022-12-22 18:28:12,634 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 4.198e+02 6.146e+02 7.824e+02 9.741e+02 2.877e+03, threshold=1.565e+03, percent-clipped=5.0
2022-12-22 18:28:16,334 WARNING [train.py:1060] (2/4) Exclude cut with ID 453-131332-0000-47844-0_sp1.1 from training. Duration: 20.72725
2022-12-22 18:28:31,700 INFO [zipformer.py:660] (2/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=19330.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 18:28:34,155 INFO [train.py:894] (2/4) Epoch 6, batch 1800, loss[loss=0.258, simple_loss=0.3289, pruned_loss=0.09351, over 18462.00 frames. ], tot_loss[loss=0.25, simple_loss=0.319, pruned_loss=0.09055, over 3713248.72 frames. ], batch size: 50, lr: 1.89e-02, grad_scale: 8.0
2022-12-22 18:28:46,206 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.7683, 1.0961, 0.6648, 1.2868, 2.0170, 1.1681, 1.5403, 1.9855],
       device='cuda:2'), covar=tensor([0.1671, 0.2294, 0.2903, 0.1645, 0.1787, 0.1761, 0.1499, 0.1514],
       device='cuda:2'), in_proj_covar=tensor([0.0098, 0.0108, 0.0130, 0.0103, 0.0117, 0.0099, 0.0102, 0.0102],
       device='cuda:2'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003],
       device='cuda:2')
2022-12-22 18:28:49,225 WARNING [train.py:1060] (2/4) Exclude cut with ID 8631-249866-0030-130156-0_sp0.9 from training. Duration: 26.32775
2022-12-22 18:29:20,917 WARNING [train.py:1060] (2/4) Exclude cut with ID 3867-173237-0077-144769-0 from training. Duration: 20.025
2022-12-22 18:29:27,467 WARNING [train.py:1060] (2/4) Exclude cut with ID 6709-74022-0004-86860-0_sp1.1 from training. Duration: 0.9409375
2022-12-22 18:29:27,473 WARNING [train.py:1060] (2/4) Exclude cut with ID 4757-1811-0023-62229-0_sp0.9 from training. Duration: 21.37775
2022-12-22 18:29:44,349 INFO [zipformer.py:660] (2/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=19378.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 18:29:46,959 WARNING [train.py:1060] (2/4) Exclude cut with ID 1250-135782-0004-25974-0_sp0.9 from training. Duration: 21.17225
2022-12-22 18:29:46,967 WARNING [train.py:1060] (2/4) Exclude cut with ID 7357-94126-0021-15852-0_sp0.9 from training. Duration: 27.511125
2022-12-22 18:29:50,072 INFO [train.py:894] (2/4) Epoch 6, batch 1850, loss[loss=0.2612, simple_loss=0.3126, pruned_loss=0.1048, over 18614.00 frames. ], tot_loss[loss=0.2533, simple_loss=0.3209, pruned_loss=0.09287, over 3714399.40 frames. ], batch size: 45, lr: 1.89e-02, grad_scale: 8.0
2022-12-22 18:30:07,722 INFO [zipformer.py:660] (2/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=19394.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 18:30:21,997 WARNING [train.py:1060] (2/4) Exclude cut with ID 453-131332-0000-47844-0 from training. Duration: 22.8
2022-12-22 18:30:26,461 WARNING [train.py:1060] (2/4) Exclude cut with ID 4964-30587-0040-44509-0 from training. Duration: 22.585
2022-12-22 18:30:28,326 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.6829, 0.5830, 1.4722, 1.3421, 1.6168, 1.3966, 1.2264, 1.1015],
       device='cuda:2'), covar=tensor([0.0823, 0.1373, 0.1036, 0.0963, 0.0700, 0.0515, 0.1012, 0.0594],
       device='cuda:2'), in_proj_covar=tensor([0.0211, 0.0246, 0.0221, 0.0240, 0.0223, 0.0199, 0.0235, 0.0191],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-22 18:30:44,799 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 4.043e+02 6.226e+02 7.501e+02 1.005e+03 2.297e+03, threshold=1.500e+03, percent-clipped=4.0
2022-12-22 18:30:57,207 WARNING [train.py:1060] (2/4) Exclude cut with ID 6978-92210-0001-146967-0_sp0.9 from training. Duration: 22.0166875
2022-12-22 18:31:05,990 INFO [train.py:894] (2/4) Epoch 6, batch 1900, loss[loss=0.2757, simple_loss=0.3367, pruned_loss=0.1074, over 18715.00 frames. ], tot_loss[loss=0.255, simple_loss=0.3211, pruned_loss=0.09441, over 3714300.31 frames. ], batch size: 52, lr: 1.89e-02, grad_scale: 8.0
2022-12-22 18:31:13,574 WARNING [train.py:1060] (2/4) Exclude cut with ID 7255-291500-0003-134302-0_sp1.1 from training. Duration: 24.395375
2022-12-22 18:31:20,831 WARNING [train.py:1060] (2/4) Exclude cut with ID 3488-85273-0017-41203-0_sp0.9 from training. Duration: 27.47775
2022-12-22 18:31:25,274 WARNING [train.py:1060] (2/4) Exclude cut with ID 432-122774-0017-62487-0_sp0.9 from training. Duration: 24.8833125
2022-12-22 18:31:30,650 WARNING [train.py:1060] (2/4) Exclude cut with ID 6758-72288-0033-108368-0 from training. Duration: 23.39
2022-12-22 18:31:35,148 WARNING [train.py:1060] (2/4) Exclude cut with ID 8291-282929-0007-12994-0_sp0.9 from training. Duration: 28.72225
2022-12-22 18:31:41,354 INFO [zipformer.py:660] (2/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=19455.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 18:31:43,874 WARNING [train.py:1060] (2/4) Exclude cut with ID 585-294811-0110-133686-0_sp0.9 from training. Duration: 20.8944375
2022-12-22 18:31:59,326 WARNING [train.py:1060] (2/4) Exclude cut with ID 5796-66357-0007-116447-0_sp0.9 from training. Duration: 23.8444375
2022-12-22 18:32:12,920 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([0.9936, 2.2050, 2.8303, 0.4527, 2.2159, 3.1436, 1.9322, 2.5351],
       device='cuda:2'), covar=tensor([0.0711, 0.0452, 0.0367, 0.0546, 0.0481, 0.0273, 0.0423, 0.0734],
       device='cuda:2'), in_proj_covar=tensor([0.0124, 0.0137, 0.0095, 0.0121, 0.0127, 0.0099, 0.0128, 0.0121],
       device='cuda:2'), out_proj_covar=tensor([1.1825e-04, 1.3375e-04, 9.2136e-05, 1.1511e-04, 1.2150e-04, 9.5656e-05,
        1.2532e-04, 1.1674e-04], device='cuda:2')
2022-12-22 18:32:22,700 INFO [train.py:894] (2/4) Epoch 6, batch 1950, loss[loss=0.2661, simple_loss=0.3223, pruned_loss=0.1049, over 18681.00 frames. ], tot_loss[loss=0.2582, simple_loss=0.323, pruned_loss=0.09672, over 3714309.33 frames. ], batch size: 50, lr: 1.89e-02, grad_scale: 8.0
2022-12-22 18:32:22,738 WARNING [train.py:1060] (2/4) Exclude cut with ID 8291-282929-0007-12994-0 from training. Duration: 25.85
2022-12-22 18:32:22,747 WARNING [train.py:1060] (2/4) Exclude cut with ID 8291-282929-0023-13010-0 from training. Duration: 21.39
2022-12-22 18:32:27,206 INFO [zipformer.py:660] (2/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=19485.0, num_to_drop=1, layers_to_drop={2}
2022-12-22 18:32:32,855 WARNING [train.py:1060] (2/4) Exclude cut with ID 7357-94126-0014-15845-0 from training. Duration: 27.92
2022-12-22 18:33:01,886 WARNING [train.py:1060] (2/4) Exclude cut with ID 8631-249866-0039-130165-0_sp0.9 from training. Duration: 20.661125
2022-12-22 18:33:16,721 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 3.731e+02 6.433e+02 7.824e+02 1.097e+03 2.201e+03, threshold=1.565e+03, percent-clipped=9.0
2022-12-22 18:33:24,442 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([2.1557, 1.4437, 1.9820, 1.9231, 2.1194, 1.7577, 1.8641, 1.3868],
       device='cuda:2'), covar=tensor([0.0872, 0.1427, 0.1034, 0.1102, 0.0747, 0.0517, 0.1194, 0.0646],
       device='cuda:2'), in_proj_covar=tensor([0.0211, 0.0247, 0.0219, 0.0239, 0.0223, 0.0198, 0.0236, 0.0192],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-22 18:33:26,934 WARNING [train.py:1060] (2/4) Exclude cut with ID 7357-94126-0043-15874-0_sp0.9 from training. Duration: 20.07225
2022-12-22 18:33:32,692 WARNING [train.py:1060] (2/4) Exclude cut with ID 1085-156170-0017-128270-0 from training. Duration: 21.01
2022-12-22 18:33:37,265 INFO [train.py:894] (2/4) Epoch 6, batch 2000, loss[loss=0.2508, simple_loss=0.3137, pruned_loss=0.09399, over 18598.00 frames. ], tot_loss[loss=0.2604, simple_loss=0.3242, pruned_loss=0.09831, over 3714086.48 frames. ], batch size: 51, lr: 1.89e-02, grad_scale: 8.0
2022-12-22 18:34:01,738 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([2.4909, 2.1685, 1.5760, 0.6047, 1.5975, 1.9691, 1.4497, 1.9343],
       device='cuda:2'), covar=tensor([0.0544, 0.0380, 0.0965, 0.1445, 0.1074, 0.1139, 0.1440, 0.0623],
       device='cuda:2'), in_proj_covar=tensor([0.0149, 0.0171, 0.0196, 0.0189, 0.0195, 0.0175, 0.0191, 0.0179],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-22 18:34:23,878 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.8911, 1.5874, 1.5666, 2.3784, 1.7361, 3.3855, 1.6376, 1.6104],
       device='cuda:2'), covar=tensor([0.0994, 0.1781, 0.1358, 0.0868, 0.1634, 0.0340, 0.1342, 0.1589],
       device='cuda:2'), in_proj_covar=tensor([0.0078, 0.0088, 0.0083, 0.0083, 0.0100, 0.0074, 0.0089, 0.0081],
       device='cuda:2'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0003, 0.0004, 0.0003, 0.0003, 0.0003],
       device='cuda:2')
2022-12-22 18:34:40,491 WARNING [train.py:1060] (2/4) Exclude cut with ID 2195-150901-0045-59933-0 from training. Duration: 20.65
2022-12-22 18:34:47,610 WARNING [train.py:1060] (2/4) Exclude cut with ID 5796-66357-0007-116447-0 from training. Duration: 21.46
2022-12-22 18:34:53,400 INFO [train.py:894] (2/4) Epoch 6, batch 2050, loss[loss=0.2231, simple_loss=0.2887, pruned_loss=0.07875, over 18621.00 frames. ], tot_loss[loss=0.2612, simple_loss=0.3245, pruned_loss=0.09895, over 3713706.07 frames. ], batch size: 41, lr: 1.88e-02, grad_scale: 8.0
2022-12-22 18:35:00,299 INFO [zipformer.py:660] (2/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=19586.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 18:35:05,079 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.8071, 1.4623, 1.0826, 0.1467, 1.1450, 1.5360, 1.1796, 1.5819],
       device='cuda:2'), covar=tensor([0.0549, 0.0421, 0.0811, 0.1375, 0.0945, 0.1279, 0.1375, 0.0547],
       device='cuda:2'), in_proj_covar=tensor([0.0148, 0.0169, 0.0194, 0.0187, 0.0192, 0.0174, 0.0188, 0.0177],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-22 18:35:09,989 INFO [scaling.py:679] (2/4) Whitening: num_groups=1, num_channels=384, metric=3.77 vs. limit=5.0
2022-12-22 18:35:30,346 INFO [scaling.py:679] (2/4) Whitening: num_groups=1, num_channels=384, metric=2.98 vs. limit=5.0
2022-12-22 18:35:34,114 WARNING [train.py:1060] (2/4) Exclude cut with ID 3557-8342-0013-54691-0 from training. Duration: 0.92
2022-12-22 18:35:41,595 WARNING [train.py:1060] (2/4) Exclude cut with ID 8291-282929-0023-13010-0_sp0.9 from training. Duration: 23.7666875
2022-12-22 18:35:48,884 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 4.006e+02 6.032e+02 7.436e+02 8.945e+02 2.067e+03, threshold=1.487e+03, percent-clipped=2.0
2022-12-22 18:35:49,292 INFO [zipformer.py:660] (2/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=19618.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 18:36:09,625 INFO [train.py:894] (2/4) Epoch 6, batch 2100, loss[loss=0.2597, simple_loss=0.3282, pruned_loss=0.09559, over 18679.00 frames. ], tot_loss[loss=0.263, simple_loss=0.3256, pruned_loss=0.1002, over 3712872.21 frames. ], batch size: 60, lr: 1.88e-02, grad_scale: 8.0
2022-12-22 18:36:13,552 INFO [zipformer.py:660] (2/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=19634.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 18:36:14,010 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.7563, 0.5687, 1.6768, 1.4602, 1.7887, 1.6049, 1.3662, 1.1934],
       device='cuda:2'), covar=tensor([0.0977, 0.1516, 0.1232, 0.1061, 0.0793, 0.0545, 0.1084, 0.0693],
       device='cuda:2'), in_proj_covar=tensor([0.0217, 0.0251, 0.0225, 0.0244, 0.0227, 0.0203, 0.0241, 0.0198],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-22 18:36:19,385 WARNING [train.py:1060] (2/4) Exclude cut with ID 8544-281189-0060-101339-0_sp0.9 from training. Duration: 20.861125
2022-12-22 18:36:30,137 WARNING [train.py:1060] (2/4) Exclude cut with ID 3488-65654-0031-41259-0_sp0.9 from training. Duration: 22.711125
2022-12-22 18:36:44,380 INFO [scaling.py:679] (2/4) Whitening: num_groups=8, num_channels=192, metric=1.96 vs. limit=2.0
2022-12-22 18:37:10,869 WARNING [train.py:1060] (2/4) Exclude cut with ID 6951-79737-0043-132310-0_sp1.1 from training. Duration: 22.986375
2022-12-22 18:37:17,485 INFO [scaling.py:679] (2/4) Whitening: num_groups=1, num_channels=384, metric=4.77 vs. limit=5.0
2022-12-22 18:37:21,955 INFO [zipformer.py:660] (2/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=19679.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 18:37:25,760 INFO [train.py:894] (2/4) Epoch 6, batch 2150, loss[loss=0.2743, simple_loss=0.3423, pruned_loss=0.1031, over 18492.00 frames. ], tot_loss[loss=0.2642, simple_loss=0.3263, pruned_loss=0.101, over 3713263.60 frames. ], batch size: 58, lr: 1.88e-02, grad_scale: 8.0
2022-12-22 18:37:28,724 WARNING [train.py:1060] (2/4) Exclude cut with ID 8040-260924-0003-80960-0_sp0.9 from training. Duration: 22.07225
2022-12-22 18:37:32,332 WARNING [train.py:1060] (2/4) Exclude cut with ID 7699-105389-0045-26330-0_sp0.9 from training. Duration: 20.3055625
2022-12-22 18:37:35,168 WARNING [train.py:1060] (2/4) Exclude cut with ID 6356-271890-0060-94317-0_sp0.9 from training. Duration: 20.72225
2022-12-22 18:37:45,485 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([2.0440, 1.4034, 1.5853, 1.9050, 2.0134, 1.8417, 2.0997, 1.1065],
       device='cuda:2'), covar=tensor([0.1392, 0.2022, 0.1734, 0.1590, 0.1183, 0.0741, 0.1562, 0.1018],
       device='cuda:2'), in_proj_covar=tensor([0.0215, 0.0249, 0.0223, 0.0241, 0.0225, 0.0200, 0.0240, 0.0196],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-22 18:37:56,547 WARNING [train.py:1060] (2/4) Exclude cut with ID 3488-85273-0017-41203-0_sp1.1 from training. Duration: 22.4818125
2022-12-22 18:38:21,384 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 3.545e+02 5.984e+02 7.689e+02 9.435e+02 1.935e+03, threshold=1.538e+03, percent-clipped=2.0
2022-12-22 18:38:22,842 WARNING [train.py:1060] (2/4) Exclude cut with ID 4964-30587-0040-44509-0_sp0.9 from training. Duration: 25.0944375
2022-12-22 18:38:26,061 WARNING [train.py:1060] (2/4) Exclude cut with ID 6533-399-0047-104881-0 from training. Duration: 21.515
2022-12-22 18:38:31,966 WARNING [train.py:1060] (2/4) Exclude cut with ID 7357-94126-0009-15840-0_sp0.9 from training. Duration: 27.02225
2022-12-22 18:38:36,976 WARNING [train.py:1060] (2/4) Exclude cut with ID 432-122774-0010-62480-0_sp0.9 from training. Duration: 22.22225
2022-12-22 18:38:43,679 INFO [train.py:894] (2/4) Epoch 6, batch 2200, loss[loss=0.2424, simple_loss=0.3139, pruned_loss=0.08545, over 18638.00 frames. ], tot_loss[loss=0.2632, simple_loss=0.3254, pruned_loss=0.1005, over 3713343.93 frames. ], batch size: 97, lr: 1.88e-02, grad_scale: 8.0
2022-12-22 18:38:45,219 WARNING [train.py:1060] (2/4) Exclude cut with ID 4964-30587-0085-44554-0_sp0.9 from training. Duration: 20.85
2022-12-22 18:38:56,494 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.9403, 1.5870, 1.2862, 1.5217, 1.8957, 1.6329, 1.6418, 1.8805],
       device='cuda:2'), covar=tensor([0.1281, 0.1779, 0.2034, 0.1366, 0.1795, 0.1373, 0.1249, 0.1294],
       device='cuda:2'), in_proj_covar=tensor([0.0094, 0.0106, 0.0127, 0.0103, 0.0114, 0.0096, 0.0101, 0.0100],
       device='cuda:2'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003],
       device='cuda:2')
2022-12-22 18:39:12,344 INFO [zipformer.py:660] (2/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=19750.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 18:39:18,149 WARNING [train.py:1060] (2/4) Exclude cut with ID 4295-39940-0007-92567-0 from training. Duration: 21.54
2022-12-22 18:39:22,314 WARNING [train.py:1060] (2/4) Exclude cut with ID 4964-30587-0040-44509-0_sp1.1 from training. Duration: 20.5318125
2022-12-22 18:39:30,479 WARNING [train.py:1060] (2/4) Exclude cut with ID 7255-291500-0012-134311-0_sp0.9 from training. Duration: 21.9333125
2022-12-22 18:39:33,161 INFO [scaling.py:679] (2/4) Whitening: num_groups=8, num_channels=96, metric=1.59 vs. limit=2.0
2022-12-22 18:39:59,928 INFO [train.py:894] (2/4) Epoch 6, batch 2250, loss[loss=0.2735, simple_loss=0.3201, pruned_loss=0.1135, over 18365.00 frames. ], tot_loss[loss=0.2628, simple_loss=0.325, pruned_loss=0.1003, over 3713091.99 frames. ], batch size: 46, lr: 1.87e-02, grad_scale: 8.0
2022-12-22 18:40:04,878 INFO [zipformer.py:660] (2/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=19785.0, num_to_drop=1, layers_to_drop={2}
2022-12-22 18:40:14,394 INFO [scaling.py:679] (2/4) Whitening: num_groups=8, num_channels=96, metric=1.34 vs. limit=2.0
2022-12-22 18:40:19,194 WARNING [train.py:1060] (2/4) Exclude cut with ID 8631-249866-0025-130151-0_sp0.9 from training. Duration: 21.7944375
2022-12-22 18:40:33,363 WARNING [train.py:1060] (2/4) Exclude cut with ID 8291-282929-0002-12989-0_sp0.9 from training. Duration: 22.4666875
2022-12-22 18:40:41,906 WARNING [train.py:1060] (2/4) Exclude cut with ID 6121-9014-0076-24124-0 from training. Duration: 21.635
2022-12-22 18:40:47,233 WARNING [train.py:1060] (2/4) Exclude cut with ID 6121-9014-0076-24124-0_sp0.9 from training. Duration: 24.038875
2022-12-22 18:40:53,007 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 4.367e+02 6.231e+02 7.152e+02 9.491e+02 2.745e+03, threshold=1.430e+03, percent-clipped=1.0
2022-12-22 18:41:14,826 INFO [train.py:894] (2/4) Epoch 6, batch 2300, loss[loss=0.284, simple_loss=0.3477, pruned_loss=0.1101, over 18714.00 frames. ], tot_loss[loss=0.2609, simple_loss=0.3231, pruned_loss=0.09934, over 3713521.26 frames. ], batch size: 54, lr: 1.87e-02, grad_scale: 8.0
2022-12-22 18:41:15,274 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([2.8223, 2.3236, 1.9188, 0.8622, 1.9874, 2.2377, 1.5461, 1.9398],
       device='cuda:2'), covar=tensor([0.0546, 0.0515, 0.1080, 0.1563, 0.1191, 0.1080, 0.1343, 0.0805],
       device='cuda:2'), in_proj_covar=tensor([0.0150, 0.0175, 0.0198, 0.0194, 0.0197, 0.0178, 0.0193, 0.0182],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-22 18:41:17,427 INFO [zipformer.py:660] (2/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=19833.0, num_to_drop=1, layers_to_drop={0}
2022-12-22 18:41:31,198 WARNING [train.py:1060] (2/4) Exclude cut with ID 7255-291500-0009-134308-0_sp1.1 from training. Duration: 21.786375
2022-12-22 18:41:42,438 WARNING [train.py:1060] (2/4) Exclude cut with ID 8291-282929-0002-12989-0 from training. Duration: 20.22
2022-12-22 18:42:31,654 INFO [train.py:894] (2/4) Epoch 6, batch 2350, loss[loss=0.2163, simple_loss=0.2827, pruned_loss=0.07494, over 18412.00 frames. ], tot_loss[loss=0.2601, simple_loss=0.3223, pruned_loss=0.0989, over 3713836.88 frames. ], batch size: 42, lr: 1.87e-02, grad_scale: 8.0
2022-12-22 18:42:38,010 INFO [scaling.py:679] (2/4) Whitening: num_groups=1, num_channels=384, metric=3.38 vs. limit=5.0
2022-12-22 18:43:26,584 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 2.986e+02 6.436e+02 8.082e+02 9.463e+02 1.888e+03, threshold=1.616e+03, percent-clipped=4.0
2022-12-22 18:43:45,652 WARNING [train.py:1060] (2/4) Exclude cut with ID 6951-79737-0043-132310-0 from training. Duration: 25.285
2022-12-22 18:43:48,413 INFO [train.py:894] (2/4) Epoch 6, batch 2400, loss[loss=0.264, simple_loss=0.327, pruned_loss=0.1005, over 18596.00 frames. ], tot_loss[loss=0.2601, simple_loss=0.3223, pruned_loss=0.0989, over 3714324.64 frames. ], batch size: 57, lr: 1.87e-02, grad_scale: 8.0
2022-12-22 18:44:18,745 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([5.9802, 5.1350, 5.3863, 5.6875, 5.4177, 5.3130, 5.8221, 1.7012],
       device='cuda:2'), covar=tensor([0.0491, 0.0413, 0.0331, 0.0584, 0.1123, 0.0773, 0.0327, 0.3889],
       device='cuda:2'), in_proj_covar=tensor([0.0251, 0.0179, 0.0175, 0.0171, 0.0251, 0.0202, 0.0203, 0.0225],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0001, 0.0001, 0.0001, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-22 18:44:49,865 WARNING [train.py:1060] (2/4) Exclude cut with ID 811-130148-0001-63453-0_sp0.9 from training. Duration: 20.861125
2022-12-22 18:44:53,607 INFO [zipformer.py:660] (2/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=19974.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 18:45:05,739 INFO [train.py:894] (2/4) Epoch 6, batch 2450, loss[loss=0.2736, simple_loss=0.335, pruned_loss=0.1061, over 18490.00 frames. ], tot_loss[loss=0.2589, simple_loss=0.3215, pruned_loss=0.0982, over 3713655.46 frames. ], batch size: 77, lr: 1.87e-02, grad_scale: 8.0
2022-12-22 18:45:11,579 WARNING [train.py:1060] (2/4) Exclude cut with ID 6010-56788-0055-90261-0 from training. Duration: 20.88
2022-12-22 18:45:29,175 INFO [scaling.py:679] (2/4) Whitening: num_groups=8, num_channels=192, metric=1.99 vs. limit=2.0
2022-12-22 18:45:47,107 WARNING [train.py:1060] (2/4) Exclude cut with ID 7357-94126-0045-15876-0_sp0.9 from training. Duration: 23.4166875
2022-12-22 18:46:04,282 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 3.728e+02 6.177e+02 7.986e+02 9.402e+02 1.900e+03, threshold=1.597e+03, percent-clipped=3.0
2022-12-22 18:46:08,262 INFO [zipformer.py:660] (2/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=20020.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 18:46:25,448 INFO [train.py:894] (2/4) Epoch 6, batch 2500, loss[loss=0.2182, simple_loss=0.3021, pruned_loss=0.06718, over 18604.00 frames. ], tot_loss[loss=0.2571, simple_loss=0.3198, pruned_loss=0.09716, over 3713505.04 frames. ], batch size: 51, lr: 1.86e-02, grad_scale: 8.0
2022-12-22 18:46:34,011 INFO [scaling.py:679] (2/4) Whitening: num_groups=8, num_channels=192, metric=1.98 vs. limit=2.0
2022-12-22 18:46:52,539 INFO [zipformer.py:660] (2/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=20050.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 18:47:07,553 WARNING [train.py:1060] (2/4) Exclude cut with ID 7255-291500-0006-134305-0 from training. Duration: 21.24
2022-12-22 18:47:08,958 WARNING [train.py:1060] (2/4) Exclude cut with ID 6533-399-0047-104881-0_sp0.9 from training. Duration: 23.9055625
2022-12-22 18:47:28,206 INFO [scaling.py:679] (2/4) Whitening: num_groups=1, num_channels=384, metric=3.40 vs. limit=5.0
2022-12-22 18:47:39,271 INFO [scaling.py:679] (2/4) Whitening: num_groups=8, num_channels=96, metric=1.67 vs. limit=2.0
2022-12-22 18:47:40,047 INFO [zipformer.py:660] (2/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=20081.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 18:47:41,026 INFO [train.py:894] (2/4) Epoch 6, batch 2550, loss[loss=0.2747, simple_loss=0.3414, pruned_loss=0.104, over 18680.00 frames. ], tot_loss[loss=0.2569, simple_loss=0.3197, pruned_loss=0.09708, over 3713552.04 frames. ], batch size: 60, lr: 1.86e-02, grad_scale: 8.0
2022-12-22 18:47:43,008 WARNING [train.py:1060] (2/4) Exclude cut with ID 6758-72288-0033-108368-0_sp0.9 from training. Duration: 25.988875
2022-12-22 18:47:50,551 WARNING [train.py:1060] (2/4) Exclude cut with ID 7255-291500-0001-134300-0_sp0.9 from training. Duration: 20.67225
2022-12-22 18:47:52,486 INFO [zipformer.py:660] (2/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=20089.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 18:48:06,650 INFO [zipformer.py:660] (2/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=20098.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 18:48:28,721 INFO [zipformer.py:660] (2/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=20112.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 18:48:37,932 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 4.210e+02 7.103e+02 8.652e+02 9.800e+02 3.308e+03, threshold=1.730e+03, percent-clipped=5.0
2022-12-22 18:48:39,506 WARNING [train.py:1060] (2/4) Exclude cut with ID 3488-85273-0038-41224-0 from training. Duration: 20.34
2022-12-22 18:48:56,802 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.4611, 1.4672, 1.5299, 1.5207, 1.3287, 2.9488, 1.4787, 2.0923],
       device='cuda:2'), covar=tensor([0.3169, 0.2003, 0.1797, 0.1807, 0.1210, 0.0262, 0.1364, 0.0850],
       device='cuda:2'), in_proj_covar=tensor([0.0148, 0.0124, 0.0136, 0.0125, 0.0110, 0.0100, 0.0108, 0.0104],
       device='cuda:2'), out_proj_covar=tensor([0.0004, 0.0004, 0.0004, 0.0004, 0.0003, 0.0003, 0.0004, 0.0003],
       device='cuda:2')
2022-12-22 18:48:59,237 INFO [train.py:894] (2/4) Epoch 6, batch 2600, loss[loss=0.2295, simple_loss=0.3064, pruned_loss=0.0763, over 18586.00 frames. ], tot_loss[loss=0.2576, simple_loss=0.3206, pruned_loss=0.09736, over 3714013.50 frames. ], batch size: 56, lr: 1.86e-02, grad_scale: 8.0
2022-12-22 18:49:20,037 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([2.1999, 1.2769, 0.8139, 1.7142, 2.1973, 1.7156, 2.0015, 2.2922],
       device='cuda:2'), covar=tensor([0.1614, 0.2424, 0.2756, 0.1742, 0.1777, 0.1537, 0.1567, 0.1701],
       device='cuda:2'), in_proj_covar=tensor([0.0096, 0.0106, 0.0127, 0.0103, 0.0113, 0.0096, 0.0100, 0.0101],
       device='cuda:2'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003],
       device='cuda:2')
2022-12-22 18:49:25,897 INFO [zipformer.py:660] (2/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=20150.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 18:49:51,153 WARNING [train.py:1060] (2/4) Exclude cut with ID 7357-94126-0026-15857-0_sp0.9 from training. Duration: 25.061125
2022-12-22 18:50:01,415 INFO [zipformer.py:660] (2/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=20173.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 18:50:02,446 WARNING [train.py:1060] (2/4) Exclude cut with ID 3033-130750-0096-55598-0 from training. Duration: 0.83
2022-12-22 18:50:14,688 INFO [train.py:894] (2/4) Epoch 6, batch 2650, loss[loss=0.2558, simple_loss=0.3216, pruned_loss=0.09495, over 18471.00 frames. ], tot_loss[loss=0.2573, simple_loss=0.3209, pruned_loss=0.09685, over 3713388.49 frames. ], batch size: 54, lr: 1.86e-02, grad_scale: 8.0
2022-12-22 18:50:28,819 WARNING [train.py:1060] (2/4) Exclude cut with ID 3488-85273-0017-41203-0 from training. Duration: 24.73
2022-12-22 18:50:40,417 WARNING [train.py:1060] (2/4) Exclude cut with ID 7255-291500-0009-134308-0 from training. Duration: 23.965
2022-12-22 18:50:48,927 WARNING [train.py:1060] (2/4) Exclude cut with ID 6978-92210-0030-146996-0_sp0.9 from training. Duration: 22.088875
2022-12-22 18:50:49,948 INFO [zipformer.py:660] (2/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=20204.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 18:51:04,929 WARNING [train.py:1060] (2/4) Exclude cut with ID 7255-291500-0006-134305-0_sp0.9 from training. Duration: 23.6
2022-12-22 18:51:10,963 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 4.129e+02 6.870e+02 8.496e+02 1.069e+03 1.947e+03, threshold=1.699e+03, percent-clipped=3.0
2022-12-22 18:51:31,608 INFO [train.py:894] (2/4) Epoch 6, batch 2700, loss[loss=0.2167, simple_loss=0.2828, pruned_loss=0.07526, over 18510.00 frames. ], tot_loss[loss=0.2568, simple_loss=0.3202, pruned_loss=0.09667, over 3713540.22 frames. ], batch size: 44, lr: 1.85e-02, grad_scale: 8.0
2022-12-22 18:52:22,487 INFO [zipformer.py:660] (2/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=20265.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 18:52:36,235 INFO [zipformer.py:660] (2/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=20274.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 18:52:48,470 INFO [train.py:894] (2/4) Epoch 6, batch 2750, loss[loss=0.2302, simple_loss=0.3034, pruned_loss=0.07848, over 18603.00 frames. ], tot_loss[loss=0.2573, simple_loss=0.3206, pruned_loss=0.097, over 3713451.99 frames. ], batch size: 56, lr: 1.85e-02, grad_scale: 8.0
2022-12-22 18:52:48,471 WARNING [train.py:1060] (2/4) Exclude cut with ID 8291-282929-0024-13011-0 from training. Duration: 23.795
2022-12-22 18:52:48,861 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([2.2108, 2.4795, 1.4868, 1.3330, 3.1671, 2.5196, 2.0035, 1.5128],
       device='cuda:2'), covar=tensor([0.0395, 0.0310, 0.0664, 0.0708, 0.0087, 0.0323, 0.0503, 0.0793],
       device='cuda:2'), in_proj_covar=tensor([0.0119, 0.0117, 0.0133, 0.0122, 0.0078, 0.0117, 0.0137, 0.0152],
       device='cuda:2'), out_proj_covar=tensor([1.4853e-04, 1.4681e-04, 1.6233e-04, 1.5131e-04, 9.9679e-05, 1.4123e-04,
        1.6898e-04, 1.8808e-04], device='cuda:2')
2022-12-22 18:53:03,718 WARNING [train.py:1060] (2/4) Exclude cut with ID 8631-249866-0030-130156-0_sp1.1 from training. Duration: 21.5409375
2022-12-22 18:53:06,904 WARNING [train.py:1060] (2/4) Exclude cut with ID 6978-92210-0019-146985-0_sp0.9 from training. Duration: 24.97775
2022-12-22 18:53:19,701 WARNING [train.py:1060] (2/4) Exclude cut with ID 1085-156170-0017-128270-0_sp0.9 from training. Duration: 23.3444375
2022-12-22 18:53:44,335 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 4.398e+02 6.276e+02 7.992e+02 1.109e+03 2.849e+03, threshold=1.598e+03, percent-clipped=4.0
2022-12-22 18:53:45,873 WARNING [train.py:1060] (2/4) Exclude cut with ID 6010-56788-0055-90261-0_sp0.9 from training. Duration: 23.2
2022-12-22 18:53:51,275 INFO [zipformer.py:660] (2/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=20322.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 18:53:52,638 WARNING [train.py:1060] (2/4) Exclude cut with ID 5653-46179-0060-117930-0_sp0.9 from training. Duration: 21.17225
2022-12-22 18:53:55,885 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.3261, 1.2220, 1.7115, 1.2248, 1.4511, 1.4486, 1.1742, 1.7390],
       device='cuda:2'), covar=tensor([0.0982, 0.1618, 0.0993, 0.1243, 0.0778, 0.0994, 0.2272, 0.0512],
       device='cuda:2'), in_proj_covar=tensor([0.0199, 0.0185, 0.0194, 0.0189, 0.0177, 0.0206, 0.0196, 0.0175],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-22 18:54:05,729 INFO [train.py:894] (2/4) Epoch 6, batch 2800, loss[loss=0.2705, simple_loss=0.3389, pruned_loss=0.101, over 18481.00 frames. ], tot_loss[loss=0.2566, simple_loss=0.32, pruned_loss=0.09658, over 3713164.20 frames. ], batch size: 58, lr: 1.85e-02, grad_scale: 8.0
2022-12-22 18:54:12,099 WARNING [train.py:1060] (2/4) Exclude cut with ID 7255-291500-0008-134307-0_sp0.9 from training. Duration: 24.6555625
2022-12-22 18:54:15,537 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([2.4709, 1.7692, 2.2415, 2.4166, 2.2990, 2.1048, 2.1619, 1.4202],
       device='cuda:2'), covar=tensor([0.1037, 0.1795, 0.1226, 0.1363, 0.0856, 0.0576, 0.1520, 0.0738],
       device='cuda:2'), in_proj_covar=tensor([0.0218, 0.0253, 0.0226, 0.0245, 0.0228, 0.0204, 0.0243, 0.0197],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-22 18:54:55,614 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.4532, 1.8689, 1.2431, 2.5199, 2.4556, 1.4754, 1.5343, 1.1438],
       device='cuda:2'), covar=tensor([0.1850, 0.1662, 0.1578, 0.0770, 0.1351, 0.1147, 0.1874, 0.1532],
       device='cuda:2'), in_proj_covar=tensor([0.0229, 0.0203, 0.0193, 0.0178, 0.0247, 0.0183, 0.0204, 0.0183],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-22 18:55:07,551 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([2.0048, 1.8483, 2.0563, 1.2047, 2.0741, 2.0882, 1.4717, 2.6209],
       device='cuda:2'), covar=tensor([0.1023, 0.1491, 0.1370, 0.1988, 0.0802, 0.1169, 0.2153, 0.0439],
       device='cuda:2'), in_proj_covar=tensor([0.0199, 0.0184, 0.0195, 0.0190, 0.0177, 0.0205, 0.0195, 0.0174],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-22 18:55:08,619 WARNING [train.py:1060] (2/4) Exclude cut with ID 3488-65654-0031-41259-0 from training. Duration: 20.44
2022-12-22 18:55:09,022 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([2.2712, 1.9016, 1.9123, 1.7367, 1.9974, 2.8047, 2.4238, 1.8256],
       device='cuda:2'), covar=tensor([0.0434, 0.0371, 0.0454, 0.0339, 0.0333, 0.0298, 0.0476, 0.0403],
       device='cuda:2'), in_proj_covar=tensor([0.0078, 0.0111, 0.0133, 0.0118, 0.0103, 0.0094, 0.0081, 0.0116],
       device='cuda:2'), out_proj_covar=tensor([7.4587e-05, 1.0263e-04, 1.2786e-04, 1.1076e-04, 9.9529e-05, 8.4647e-05,
        7.5250e-05, 1.0800e-04], device='cuda:2')
2022-12-22 18:55:13,482 INFO [zipformer.py:660] (2/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=20376.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 18:55:22,167 INFO [train.py:894] (2/4) Epoch 6, batch 2850, loss[loss=0.2558, simple_loss=0.3095, pruned_loss=0.101, over 18383.00 frames. ], tot_loss[loss=0.2572, simple_loss=0.3204, pruned_loss=0.09698, over 3712647.57 frames. ], batch size: 46, lr: 1.85e-02, grad_scale: 8.0
2022-12-22 18:55:23,877 WARNING [train.py:1060] (2/4) Exclude cut with ID 6951-79737-0018-132285-0_sp0.9 from training. Duration: 23.45
2022-12-22 18:55:31,487 INFO [zipformer.py:660] (2/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=20388.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 18:55:53,463 WARNING [train.py:1060] (2/4) Exclude cut with ID 6945-60535-0076-12784-0_sp0.9 from training. Duration: 20.52225
2022-12-22 18:56:00,840 WARNING [train.py:1060] (2/4) Exclude cut with ID 7255-291500-0008-134307-0 from training. Duration: 22.19
2022-12-22 18:56:11,896 WARNING [train.py:1060] (2/4) Exclude cut with ID 7357-94126-0014-15845-0_sp1.1 from training. Duration: 25.3818125
2022-12-22 18:56:18,483 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 4.122e+02 6.167e+02 7.330e+02 9.180e+02 5.006e+03, threshold=1.466e+03, percent-clipped=5.0
2022-12-22 18:56:18,990 INFO [zipformer.py:660] (2/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=20418.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 18:56:29,199 WARNING [train.py:1060] (2/4) Exclude cut with ID 6951-79737-0043-132310-0_sp0.9 from training. Duration: 28.0944375
2022-12-22 18:56:35,039 WARNING [train.py:1060] (2/4) Exclude cut with ID 2195-150901-0045-59933-0_sp0.9 from training. Duration: 22.9444375
2022-12-22 18:56:39,025 INFO [train.py:894] (2/4) Epoch 6, batch 2900, loss[loss=0.2535, simple_loss=0.3316, pruned_loss=0.08769, over 18716.00 frames. ], tot_loss[loss=0.256, simple_loss=0.3193, pruned_loss=0.09634, over 3712776.69 frames. ], batch size: 78, lr: 1.85e-02, grad_scale: 8.0
2022-12-22 18:56:41,934 WARNING [train.py:1060] (2/4) Exclude cut with ID 8291-282929-0024-13011-0_sp1.1 from training. Duration: 21.6318125
2022-12-22 18:56:44,607 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.7636, 0.5488, 1.6212, 1.5420, 1.7504, 1.5787, 1.4601, 1.1975],
       device='cuda:2'), covar=tensor([0.1015, 0.1628, 0.1260, 0.1066, 0.0816, 0.0571, 0.1102, 0.0717],
       device='cuda:2'), in_proj_covar=tensor([0.0220, 0.0257, 0.0227, 0.0246, 0.0229, 0.0206, 0.0244, 0.0199],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-22 18:56:59,266 WARNING [train.py:1060] (2/4) Exclude cut with ID 8631-249866-0030-130156-0 from training. Duration: 23.695
2022-12-22 18:56:59,366 INFO [zipformer.py:660] (2/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=20445.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 18:57:05,543 INFO [zipformer.py:660] (2/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=20449.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 18:57:25,522 WARNING [train.py:1060] (2/4) Exclude cut with ID 7699-105389-0094-26379-0 from training. Duration: 23.955
2022-12-22 18:57:33,893 INFO [zipformer.py:660] (2/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=20468.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 18:57:50,575 INFO [zipformer.py:660] (2/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=20479.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 18:57:54,439 INFO [train.py:894] (2/4) Epoch 6, batch 2950, loss[loss=0.2623, simple_loss=0.3229, pruned_loss=0.1008, over 18491.00 frames. ], tot_loss[loss=0.2569, simple_loss=0.3204, pruned_loss=0.09672, over 3713550.13 frames. ], batch size: 77, lr: 1.84e-02, grad_scale: 8.0
2022-12-22 18:58:01,499 WARNING [train.py:1060] (2/4) Exclude cut with ID 8291-282929-0024-13011-0_sp0.9 from training. Duration: 26.438875
2022-12-22 18:58:30,576 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.9150, 0.8840, 1.7350, 1.6112, 1.8419, 1.7241, 1.5526, 1.2218],
       device='cuda:2'), covar=tensor([0.1075, 0.1733, 0.1320, 0.1204, 0.0912, 0.0581, 0.1260, 0.0741],
       device='cuda:2'), in_proj_covar=tensor([0.0218, 0.0255, 0.0227, 0.0245, 0.0229, 0.0204, 0.0245, 0.0199],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-22 18:58:48,754 WARNING [train.py:1060] (2/4) Exclude cut with ID 7699-105389-0021-26306-0_sp0.9 from training. Duration: 21.2444375
2022-12-22 18:58:48,779 WARNING [train.py:1060] (2/4) Exclude cut with ID 7357-94126-0014-15845-0_sp0.9 from training. Duration: 31.02225
2022-12-22 18:58:50,128 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 3.777e+02 5.957e+02 7.468e+02 9.949e+02 2.082e+03, threshold=1.494e+03, percent-clipped=7.0
2022-12-22 18:58:58,441 WARNING [train.py:1060] (2/4) Exclude cut with ID 432-122774-0017-62487-0 from training. Duration: 22.395
2022-12-22 18:59:10,908 INFO [train.py:894] (2/4) Epoch 6, batch 3000, loss[loss=0.2741, simple_loss=0.3351, pruned_loss=0.1066, over 18536.00 frames. ], tot_loss[loss=0.2581, simple_loss=0.3213, pruned_loss=0.09749, over 3713979.53 frames. ], batch size: 55, lr: 1.84e-02, grad_scale: 8.0
2022-12-22 18:59:10,908 INFO [train.py:919] (2/4) Computing validation loss
2022-12-22 18:59:17,848 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([3.3281, 2.7324, 2.5843, 1.5173, 2.4205, 2.3273, 1.8398, 2.4814],
       device='cuda:2'), covar=tensor([0.0726, 0.0582, 0.1275, 0.1770, 0.1811, 0.1424, 0.1633, 0.1052],
       device='cuda:2'), in_proj_covar=tensor([0.0158, 0.0178, 0.0205, 0.0195, 0.0202, 0.0182, 0.0199, 0.0185],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-22 18:59:22,156 INFO [train.py:928] (2/4) Epoch 6, validation: loss=0.1911, simple_loss=0.2898, pruned_loss=0.04619, over 944034.00 frames. 
2022-12-22 18:59:22,157 INFO [train.py:929] (2/4) Maximum memory allocated so far is 24320MB
2022-12-22 18:59:28,876 WARNING [train.py:1060] (2/4) Exclude cut with ID 7357-94126-0045-15876-0 from training. Duration: 21.075
2022-12-22 18:59:35,108 WARNING [train.py:1060] (2/4) Exclude cut with ID 6482-98857-0025-147532-0_sp0.9 from training. Duration: 20.0055625
2022-12-22 18:59:35,119 WARNING [train.py:1060] (2/4) Exclude cut with ID 6951-79737-0037-132304-0_sp0.9 from training. Duration: 22.05
2022-12-22 18:59:35,130 WARNING [train.py:1060] (2/4) Exclude cut with ID 7255-291500-0003-134302-0 from training. Duration: 26.8349375
2022-12-22 18:59:36,938 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([2.2558, 2.1767, 1.7311, 1.4676, 2.9898, 2.5107, 1.9041, 1.3618],
       device='cuda:2'), covar=tensor([0.0352, 0.0359, 0.0548, 0.0660, 0.0119, 0.0304, 0.0556, 0.0967],
       device='cuda:2'), in_proj_covar=tensor([0.0117, 0.0114, 0.0129, 0.0117, 0.0075, 0.0114, 0.0135, 0.0150],
       device='cuda:2'), out_proj_covar=tensor([1.4555e-04, 1.4324e-04, 1.5829e-04, 1.4500e-04, 9.6017e-05, 1.3800e-04,
        1.6562e-04, 1.8484e-04], device='cuda:2')
2022-12-22 18:59:38,029 WARNING [train.py:1060] (2/4) Exclude cut with ID 7357-94126-0009-15840-0_sp1.1 from training. Duration: 22.1090625
2022-12-22 18:59:44,106 WARNING [train.py:1060] (2/4) Exclude cut with ID 7699-105389-0094-26379-0_sp0.9 from training. Duration: 26.6166875
2022-12-22 19:00:02,745 WARNING [train.py:1060] (2/4) Exclude cut with ID 2046-178027-0000-53705-0_sp0.9 from training. Duration: 20.3055625
2022-12-22 19:00:06,090 INFO [zipformer.py:660] (2/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=20560.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 19:00:26,312 WARNING [train.py:1060] (2/4) Exclude cut with ID 7205-50138-0008-5373-0_sp0.9 from training. Duration: 20.7
2022-12-22 19:00:40,407 INFO [train.py:894] (2/4) Epoch 6, batch 3050, loss[loss=0.2875, simple_loss=0.3482, pruned_loss=0.1134, over 18698.00 frames. ], tot_loss[loss=0.2562, simple_loss=0.3199, pruned_loss=0.09625, over 3714502.01 frames. ], batch size: 52, lr: 1.84e-02, grad_scale: 8.0
2022-12-22 19:00:49,764 INFO [zipformer.py:660] (2/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=20588.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 19:01:10,565 WARNING [train.py:1060] (2/4) Exclude cut with ID 6978-92210-0019-146985-0 from training. Duration: 22.48
2022-12-22 19:01:26,402 WARNING [train.py:1060] (2/4) Exclude cut with ID 7255-291500-0003-134302-0_sp0.9 from training. Duration: 29.816625
2022-12-22 19:01:36,229 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 4.602e+02 6.371e+02 7.791e+02 9.358e+02 2.068e+03, threshold=1.558e+03, percent-clipped=2.0
2022-12-22 19:01:46,755 WARNING [train.py:1060] (2/4) Exclude cut with ID 7255-291500-0005-134304-0_sp1.1 from training. Duration: 22.7590625
2022-12-22 19:01:47,063 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.8107, 1.4626, 1.2793, 1.9779, 1.7179, 3.4190, 1.4427, 1.5090],
       device='cuda:2'), covar=tensor([0.1025, 0.2021, 0.1453, 0.1103, 0.1581, 0.0250, 0.1558, 0.1729],
       device='cuda:2'), in_proj_covar=tensor([0.0080, 0.0088, 0.0083, 0.0084, 0.0099, 0.0074, 0.0091, 0.0081],
       device='cuda:2'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0003, 0.0004, 0.0003, 0.0004, 0.0003],
       device='cuda:2')
2022-12-22 19:01:51,635 WARNING [train.py:1060] (2/4) Exclude cut with ID 7357-94126-0026-15857-0 from training. Duration: 22.555
2022-12-22 19:01:57,504 INFO [train.py:894] (2/4) Epoch 6, batch 3100, loss[loss=0.2976, simple_loss=0.3541, pruned_loss=0.1206, over 18565.00 frames. ], tot_loss[loss=0.2567, simple_loss=0.3202, pruned_loss=0.09658, over 3714080.54 frames. ], batch size: 77, lr: 1.84e-02, grad_scale: 8.0
2022-12-22 19:02:11,319 WARNING [train.py:1060] (2/4) Exclude cut with ID 1250-135782-0005-25975-0_sp0.9 from training. Duration: 21.688875
2022-12-22 19:02:23,590 INFO [zipformer.py:660] (2/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=20649.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 19:02:47,436 WARNING [train.py:1060] (2/4) Exclude cut with ID 3488-85273-0038-41224-0_sp0.9 from training. Duration: 22.6
2022-12-22 19:03:04,013 INFO [zipformer.py:660] (2/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=20676.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 19:03:10,869 INFO [scaling.py:679] (2/4) Whitening: num_groups=8, num_channels=192, metric=2.03 vs. limit=2.0
2022-12-22 19:03:12,818 INFO [train.py:894] (2/4) Epoch 6, batch 3150, loss[loss=0.2601, simple_loss=0.3284, pruned_loss=0.09587, over 18728.00 frames. ], tot_loss[loss=0.2571, simple_loss=0.3205, pruned_loss=0.09683, over 3713342.39 frames. ], batch size: 54, lr: 1.84e-02, grad_scale: 8.0
2022-12-22 19:03:23,607 WARNING [train.py:1060] (2/4) Exclude cut with ID 7357-94126-0009-15840-0 from training. Duration: 24.32
2022-12-22 19:04:08,115 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 3.940e+02 6.143e+02 7.454e+02 8.942e+02 3.821e+03, threshold=1.491e+03, percent-clipped=3.0
2022-12-22 19:04:13,539 INFO [scaling.py:679] (2/4) Whitening: num_groups=8, num_channels=96, metric=1.42 vs. limit=2.0
2022-12-22 19:04:17,329 INFO [zipformer.py:660] (2/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=20724.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 19:04:20,051 WARNING [train.py:1060] (2/4) Exclude cut with ID 8291-276745-0093-13116-0_sp0.9 from training. Duration: 21.061125
2022-12-22 19:04:29,663 INFO [train.py:894] (2/4) Epoch 6, batch 3200, loss[loss=0.2406, simple_loss=0.3116, pruned_loss=0.08479, over 18563.00 frames. ], tot_loss[loss=0.257, simple_loss=0.3206, pruned_loss=0.09666, over 3714276.47 frames. ], batch size: 49, lr: 1.83e-02, grad_scale: 8.0
2022-12-22 19:04:34,429 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.9770, 1.7843, 1.4421, 1.2709, 2.5626, 2.1496, 1.7674, 1.3717],
       device='cuda:2'), covar=tensor([0.0342, 0.0347, 0.0501, 0.0624, 0.0129, 0.0282, 0.0458, 0.0841],
       device='cuda:2'), in_proj_covar=tensor([0.0119, 0.0115, 0.0132, 0.0121, 0.0077, 0.0115, 0.0138, 0.0150],
       device='cuda:2'), out_proj_covar=tensor([1.4866e-04, 1.4419e-04, 1.6192e-04, 1.4974e-04, 9.8005e-05, 1.3957e-04,
        1.6884e-04, 1.8546e-04], device='cuda:2')
2022-12-22 19:04:35,460 WARNING [train.py:1060] (2/4) Exclude cut with ID 7357-94126-0024-15855-0_sp0.9 from training. Duration: 20.32225
2022-12-22 19:04:48,099 INFO [zipformer.py:660] (2/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=20744.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 19:04:49,755 WARNING [train.py:1060] (2/4) Exclude cut with ID 3033-130750-0096-55598-0_sp1.1 from training. Duration: 0.7545625
2022-12-22 19:04:49,946 INFO [zipformer.py:660] (2/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=20745.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 19:05:01,016 WARNING [train.py:1060] (2/4) Exclude cut with ID 4295-39940-0007-92567-0_sp0.9 from training. Duration: 23.9333125
2022-12-22 19:05:25,560 INFO [zipformer.py:660] (2/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=20768.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 19:05:25,751 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([2.3641, 1.9254, 1.4500, 2.2194, 1.7488, 1.6852, 1.8619, 2.3412],
       device='cuda:2'), covar=tensor([0.1549, 0.2327, 0.1402, 0.2296, 0.2435, 0.0907, 0.1999, 0.0581],
       device='cuda:2'), in_proj_covar=tensor([0.0268, 0.0247, 0.0215, 0.0332, 0.0234, 0.0206, 0.0245, 0.0177],
       device='cuda:2'), out_proj_covar=tensor([0.0003, 0.0002, 0.0002, 0.0003, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-22 19:05:34,260 INFO [zipformer.py:660] (2/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=20774.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 19:05:37,176 WARNING [train.py:1060] (2/4) Exclude cut with ID 7255-291500-0008-134307-0_sp1.1 from training. Duration: 20.17275
2022-12-22 19:05:39,469 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([2.9872, 3.1072, 1.4470, 1.6910, 4.1758, 4.0568, 2.6457, 2.1285],
       device='cuda:2'), covar=tensor([0.0413, 0.0285, 0.0724, 0.0702, 0.0057, 0.0201, 0.0528, 0.0723],
       device='cuda:2'), in_proj_covar=tensor([0.0119, 0.0114, 0.0131, 0.0121, 0.0076, 0.0113, 0.0136, 0.0148],
       device='cuda:2'), out_proj_covar=tensor([1.4858e-04, 1.4316e-04, 1.6049e-04, 1.4891e-04, 9.6412e-05, 1.3719e-04,
        1.6667e-04, 1.8301e-04], device='cuda:2')
2022-12-22 19:05:43,421 WARNING [train.py:1060] (2/4) Exclude cut with ID 6978-92210-0019-146985-0_sp1.1 from training. Duration: 20.436375
2022-12-22 19:05:46,187 INFO [train.py:894] (2/4) Epoch 6, batch 3250, loss[loss=0.2501, simple_loss=0.3229, pruned_loss=0.08868, over 18572.00 frames. ], tot_loss[loss=0.2568, simple_loss=0.3208, pruned_loss=0.09638, over 3713813.61 frames. ], batch size: 57, lr: 1.83e-02, grad_scale: 4.0
2022-12-22 19:06:03,697 INFO [zipformer.py:660] (2/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=20793.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 19:06:40,255 INFO [zipformer.py:660] (2/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=20816.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 19:06:44,639 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 3.773e+02 6.050e+02 7.541e+02 9.973e+02 3.613e+03, threshold=1.508e+03, percent-clipped=8.0
2022-12-22 19:07:04,321 INFO [train.py:894] (2/4) Epoch 6, batch 3300, loss[loss=0.2579, simple_loss=0.3345, pruned_loss=0.09059, over 18669.00 frames. ], tot_loss[loss=0.2556, simple_loss=0.3201, pruned_loss=0.09554, over 3713665.00 frames. ], batch size: 69, lr: 1.83e-02, grad_scale: 4.0
2022-12-22 19:07:05,553 WARNING [train.py:1060] (2/4) Exclude cut with ID 4234-40345-0022-142709-0_sp0.9 from training. Duration: 23.1055625
2022-12-22 19:07:07,028 WARNING [train.py:1060] (2/4) Exclude cut with ID 8291-282929-0007-12994-0_sp1.1 from training. Duration: 23.5
2022-12-22 19:07:17,657 WARNING [train.py:1060] (2/4) Exclude cut with ID 7255-291500-0009-134308-0_sp0.9 from training. Duration: 26.62775
2022-12-22 19:07:31,440 WARNING [train.py:1060] (2/4) Exclude cut with ID 6951-79737-0018-132285-0 from training. Duration: 21.105
2022-12-22 19:07:35,857 WARNING [train.py:1060] (2/4) Exclude cut with ID 4511-76322-0006-80011-0_sp0.9 from training. Duration: 24.411125
2022-12-22 19:07:46,454 INFO [zipformer.py:660] (2/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=20860.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 19:08:05,300 WARNING [train.py:1060] (2/4) Exclude cut with ID 6758-72288-0033-108368-0_sp1.1 from training. Duration: 21.263625
2022-12-22 19:08:19,489 INFO [train.py:894] (2/4) Epoch 6, batch 3350, loss[loss=0.2036, simple_loss=0.2842, pruned_loss=0.06154, over 18530.00 frames. ], tot_loss[loss=0.2551, simple_loss=0.3195, pruned_loss=0.09536, over 3714427.18 frames. ], batch size: 47, lr: 1.83e-02, grad_scale: 4.0
2022-12-22 19:08:35,958 WARNING [train.py:1060] (2/4) Exclude cut with ID 4234-40345-0022-142709-0 from training. Duration: 20.795
2022-12-22 19:08:46,086 WARNING [train.py:1060] (2/4) Exclude cut with ID 7357-94126-0021-15852-0 from training. Duration: 24.76
2022-12-22 19:08:46,102 WARNING [train.py:1060] (2/4) Exclude cut with ID 3867-173237-0077-144769-0_sp0.9 from training. Duration: 22.25
2022-12-22 19:09:00,344 INFO [zipformer.py:660] (2/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=20908.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 19:09:10,531 WARNING [train.py:1060] (2/4) Exclude cut with ID 7357-94126-0026-15857-0_sp1.1 from training. Duration: 20.5045625
2022-12-22 19:09:16,418 INFO [scaling.py:679] (2/4) Whitening: num_groups=8, num_channels=96, metric=1.39 vs. limit=2.0
2022-12-22 19:09:16,850 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 3.707e+02 6.289e+02 7.876e+02 9.972e+02 1.794e+03, threshold=1.575e+03, percent-clipped=4.0
2022-12-22 19:09:37,391 INFO [train.py:894] (2/4) Epoch 6, batch 3400, loss[loss=0.2568, simple_loss=0.3167, pruned_loss=0.09849, over 18700.00 frames. ], tot_loss[loss=0.256, simple_loss=0.3199, pruned_loss=0.09604, over 3714136.70 frames. ], batch size: 50, lr: 1.83e-02, grad_scale: 4.0
2022-12-22 19:09:55,167 INFO [zipformer.py:660] (2/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=20944.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 19:10:12,373 INFO [scaling.py:679] (2/4) Whitening: num_groups=8, num_channels=96, metric=1.20 vs. limit=2.0
2022-12-22 19:10:50,665 INFO [train.py:894] (2/4) Epoch 6, batch 3450, loss[loss=0.236, simple_loss=0.304, pruned_loss=0.08398, over 18402.00 frames. ], tot_loss[loss=0.2564, simple_loss=0.3203, pruned_loss=0.09622, over 3714323.19 frames. ], batch size: 53, lr: 1.82e-02, grad_scale: 4.0
2022-12-22 19:11:13,172 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([2.4884, 2.9350, 3.0316, 1.9121, 2.8066, 2.3528, 1.3108, 2.2854],
       device='cuda:2'), covar=tensor([0.2040, 0.1003, 0.1226, 0.2451, 0.1142, 0.0967, 0.3762, 0.1396],
       device='cuda:2'), in_proj_covar=tensor([0.0125, 0.0110, 0.0147, 0.0114, 0.0112, 0.0099, 0.0141, 0.0108],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-22 19:11:19,523 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.4938, 1.4339, 1.2923, 1.0996, 1.8521, 1.5725, 1.4469, 0.9857],
       device='cuda:2'), covar=tensor([0.0296, 0.0287, 0.0421, 0.0489, 0.0140, 0.0251, 0.0356, 0.0802],
       device='cuda:2'), in_proj_covar=tensor([0.0120, 0.0115, 0.0132, 0.0121, 0.0077, 0.0114, 0.0136, 0.0151],
       device='cuda:2'), out_proj_covar=tensor([1.4947e-04, 1.4391e-04, 1.6195e-04, 1.5003e-04, 9.7767e-05, 1.3794e-04,
        1.6620e-04, 1.8731e-04], device='cuda:2')
2022-12-22 19:11:44,684 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 3.589e+02 6.787e+02 7.956e+02 1.031e+03 2.201e+03, threshold=1.591e+03, percent-clipped=5.0
2022-12-22 19:12:04,662 INFO [train.py:894] (2/4) Epoch 6, batch 3500, loss[loss=0.3285, simple_loss=0.3731, pruned_loss=0.142, over 18595.00 frames. ], tot_loss[loss=0.2598, simple_loss=0.3232, pruned_loss=0.0982, over 3715576.95 frames. ], batch size: 172, lr: 1.82e-02, grad_scale: 4.0
2022-12-22 19:12:24,861 WARNING [train.py:1060] (2/4) Exclude cut with ID 7859-102521-0017-7548-0_sp1.1 from training. Duration: 22.2954375
2022-12-22 19:12:35,558 INFO [train.py:894] (2/4) Epoch 7, batch 0, loss[loss=0.2195, simple_loss=0.2849, pruned_loss=0.07702, over 18530.00 frames. ], tot_loss[loss=0.2195, simple_loss=0.2849, pruned_loss=0.07702, over 18530.00 frames. ], batch size: 44, lr: 1.71e-02, grad_scale: 8.0
2022-12-22 19:12:35,558 INFO [train.py:919] (2/4) Computing validation loss
2022-12-22 19:12:47,233 INFO [train.py:928] (2/4) Epoch 7, validation: loss=0.1926, simple_loss=0.291, pruned_loss=0.0471, over 944034.00 frames. 
2022-12-22 19:12:47,234 INFO [train.py:929] (2/4) Maximum memory allocated so far is 24320MB
2022-12-22 19:12:56,935 INFO [zipformer.py:660] (2/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=21044.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 19:13:09,207 INFO [zipformer.py:660] (2/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=21052.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 19:13:39,797 WARNING [train.py:1060] (2/4) Exclude cut with ID 298-126791-0067-24026-0_sp0.9 from training. Duration: 21.438875
2022-12-22 19:13:41,441 INFO [zipformer.py:660] (2/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=21074.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 19:13:43,895 WARNING [train.py:1060] (2/4) Exclude cut with ID 5652-39938-0025-23684-0_sp0.9 from training. Duration: 22.2055625
2022-12-22 19:14:00,027 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.5015, 1.3727, 1.2011, 1.7431, 1.5026, 3.2729, 1.3553, 1.4330],
       device='cuda:2'), covar=tensor([0.0986, 0.1765, 0.1284, 0.0963, 0.1482, 0.0206, 0.1312, 0.1494],
       device='cuda:2'), in_proj_covar=tensor([0.0081, 0.0090, 0.0084, 0.0084, 0.0099, 0.0075, 0.0092, 0.0083],
       device='cuda:2'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0003, 0.0004, 0.0003, 0.0004, 0.0003],
       device='cuda:2')
2022-12-22 19:14:01,169 INFO [train.py:894] (2/4) Epoch 7, batch 50, loss[loss=0.2274, simple_loss=0.3117, pruned_loss=0.0716, over 18730.00 frames. ], tot_loss[loss=0.2347, simple_loss=0.3108, pruned_loss=0.07928, over 838275.90 frames. ], batch size: 54, lr: 1.70e-02, grad_scale: 8.0
2022-12-22 19:14:07,955 INFO [zipformer.py:660] (2/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=21092.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 19:14:40,457 INFO [zipformer.py:660] (2/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=21113.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 19:14:48,457 INFO [zipformer.py:660] (2/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=21118.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 19:14:49,494 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 3.520e+02 5.183e+02 6.375e+02 7.921e+02 1.523e+03, threshold=1.275e+03, percent-clipped=0.0
2022-12-22 19:14:53,939 INFO [zipformer.py:660] (2/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=21122.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 19:15:17,781 INFO [train.py:894] (2/4) Epoch 7, batch 100, loss[loss=0.2461, simple_loss=0.3227, pruned_loss=0.08479, over 18523.00 frames. ], tot_loss[loss=0.2267, simple_loss=0.3039, pruned_loss=0.07475, over 1474782.49 frames. ], batch size: 58, lr: 1.70e-02, grad_scale: 8.0
2022-12-22 19:16:20,289 INFO [zipformer.py:660] (2/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=21179.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 19:16:31,929 INFO [scaling.py:679] (2/4) Whitening: num_groups=1, num_channels=384, metric=3.25 vs. limit=5.0
2022-12-22 19:16:33,904 INFO [train.py:894] (2/4) Epoch 7, batch 150, loss[loss=0.1948, simple_loss=0.278, pruned_loss=0.05579, over 18590.00 frames. ], tot_loss[loss=0.2228, simple_loss=0.3006, pruned_loss=0.0725, over 1970707.08 frames. ], batch size: 45, lr: 1.70e-02, grad_scale: 4.0
2022-12-22 19:16:41,822 WARNING [train.py:1060] (2/4) Exclude cut with ID 7859-102521-0017-7548-0 from training. Duration: 24.525
2022-12-22 19:17:16,258 WARNING [train.py:1060] (2/4) Exclude cut with ID 3699-47246-0007-3408-0_sp0.9 from training. Duration: 20.26675
2022-12-22 19:17:22,375 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 2.317e+02 5.094e+02 6.367e+02 7.897e+02 2.757e+03, threshold=1.273e+03, percent-clipped=4.0
2022-12-22 19:17:28,275 WARNING [train.py:1060] (2/4) Exclude cut with ID 7859-102521-0017-7548-0_sp0.9 from training. Duration: 27.25
2022-12-22 19:17:38,122 INFO [scaling.py:679] (2/4) Whitening: num_groups=8, num_channels=96, metric=1.43 vs. limit=2.0
2022-12-22 19:17:49,905 INFO [train.py:894] (2/4) Epoch 7, batch 200, loss[loss=0.2606, simple_loss=0.3344, pruned_loss=0.09341, over 18713.00 frames. ], tot_loss[loss=0.2205, simple_loss=0.2983, pruned_loss=0.07137, over 2357152.76 frames. ], batch size: 52, lr: 1.70e-02, grad_scale: 4.0
2022-12-22 19:18:00,286 INFO [zipformer.py:660] (2/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=21244.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 19:18:41,658 WARNING [train.py:1060] (2/4) Exclude cut with ID 6426-64292-0017-15984-0 from training. Duration: 21.68
2022-12-22 19:18:42,410 INFO [scaling.py:679] (2/4) Whitening: num_groups=1, num_channels=384, metric=2.90 vs. limit=5.0
2022-12-22 19:18:53,648 WARNING [train.py:1060] (2/4) Exclude cut with ID 4278-13270-0007-59342-0 from training. Duration: 21.6300625
2022-12-22 19:19:06,068 INFO [train.py:894] (2/4) Epoch 7, batch 250, loss[loss=0.2078, simple_loss=0.3, pruned_loss=0.05786, over 18534.00 frames. ], tot_loss[loss=0.2213, simple_loss=0.2998, pruned_loss=0.0714, over 2658281.36 frames. ], batch size: 58, lr: 1.70e-02, grad_scale: 4.0
2022-12-22 19:19:13,094 INFO [zipformer.py:660] (2/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=21292.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 19:19:18,654 WARNING [train.py:1060] (2/4) Exclude cut with ID 4278-13270-0007-59342-0_sp0.9 from training. Duration: 24.033375
2022-12-22 19:19:43,239 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.5263, 1.5335, 1.5385, 1.5007, 1.5365, 3.5657, 1.6671, 2.3480],
       device='cuda:2'), covar=tensor([0.3242, 0.1971, 0.1906, 0.2016, 0.1239, 0.0160, 0.1429, 0.0798],
       device='cuda:2'), in_proj_covar=tensor([0.0148, 0.0125, 0.0135, 0.0125, 0.0110, 0.0100, 0.0106, 0.0102],
       device='cuda:2'), out_proj_covar=tensor([0.0004, 0.0004, 0.0004, 0.0004, 0.0003, 0.0003, 0.0004, 0.0003],
       device='cuda:2')
2022-12-22 19:19:43,257 INFO [zipformer.py:660] (2/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=21312.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 19:19:54,507 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 3.081e+02 4.850e+02 5.968e+02 7.648e+02 1.525e+03, threshold=1.194e+03, percent-clipped=4.0
2022-12-22 19:20:06,485 INFO [zipformer.py:660] (2/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=21328.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 19:20:15,567 WARNING [train.py:1060] (2/4) Exclude cut with ID 4278-13270-0009-59344-0 from training. Duration: 22.905
2022-12-22 19:20:17,083 WARNING [train.py:1060] (2/4) Exclude cut with ID 5622-44585-0006-90525-0_sp1.1 from training. Duration: 23.4318125
2022-12-22 19:20:21,443 INFO [train.py:894] (2/4) Epoch 7, batch 300, loss[loss=0.2059, simple_loss=0.2846, pruned_loss=0.06357, over 18684.00 frames. ], tot_loss[loss=0.2208, simple_loss=0.2993, pruned_loss=0.07114, over 2892042.75 frames. ], batch size: 50, lr: 1.70e-02, grad_scale: 4.0
2022-12-22 19:21:14,001 INFO [zipformer.py:660] (2/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=21373.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 19:21:36,820 INFO [train.py:894] (2/4) Epoch 7, batch 350, loss[loss=0.1906, simple_loss=0.2655, pruned_loss=0.0579, over 18529.00 frames. ], tot_loss[loss=0.2228, simple_loss=0.3008, pruned_loss=0.07239, over 3074159.42 frames. ], batch size: 44, lr: 1.69e-02, grad_scale: 4.0
2022-12-22 19:21:39,145 INFO [zipformer.py:660] (2/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=21389.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 19:21:42,598 INFO [scaling.py:679] (2/4) Whitening: num_groups=8, num_channels=96, metric=1.46 vs. limit=2.0
2022-12-22 19:21:53,133 INFO [scaling.py:679] (2/4) Whitening: num_groups=8, num_channels=192, metric=2.15 vs. limit=2.0
2022-12-22 19:22:07,399 INFO [zipformer.py:660] (2/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=21408.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 19:22:14,456 WARNING [train.py:1060] (2/4) Exclude cut with ID 4278-13270-0009-59344-0_sp1.1 from training. Duration: 20.82275
2022-12-22 19:22:14,498 WARNING [train.py:1060] (2/4) Exclude cut with ID 4278-13270-0009-59344-0_sp0.9 from training. Duration: 25.45
2022-12-22 19:22:25,362 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 3.397e+02 5.112e+02 6.223e+02 8.119e+02 3.223e+03, threshold=1.245e+03, percent-clipped=4.0
2022-12-22 19:22:53,013 INFO [train.py:894] (2/4) Epoch 7, batch 400, loss[loss=0.2373, simple_loss=0.3149, pruned_loss=0.07983, over 18499.00 frames. ], tot_loss[loss=0.2242, simple_loss=0.3019, pruned_loss=0.07321, over 3216762.76 frames. ], batch size: 52, lr: 1.69e-02, grad_scale: 8.0
2022-12-22 19:23:02,935 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([2.5226, 2.1883, 1.5786, 2.5657, 1.9370, 1.9039, 2.1756, 2.7328],
       device='cuda:2'), covar=tensor([0.1344, 0.2133, 0.1253, 0.2015, 0.2284, 0.0748, 0.1884, 0.0474],
       device='cuda:2'), in_proj_covar=tensor([0.0262, 0.0243, 0.0213, 0.0321, 0.0232, 0.0203, 0.0244, 0.0171],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0003, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-22 19:23:13,666 WARNING [train.py:1060] (2/4) Exclude cut with ID 5622-44585-0006-90525-0 from training. Duration: 25.775
2022-12-22 19:23:28,331 INFO [scaling.py:679] (2/4) Whitening: num_groups=8, num_channels=96, metric=1.25 vs. limit=2.0
2022-12-22 19:23:35,408 WARNING [train.py:1060] (2/4) Exclude cut with ID 2929-85685-0071-62375-0_sp0.9 from training. Duration: 22.25
2022-12-22 19:23:47,400 INFO [zipformer.py:660] (2/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=21474.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 19:24:05,058 WARNING [train.py:1060] (2/4) Exclude cut with ID 3972-170212-0014-23379-0 from training. Duration: 26.205
2022-12-22 19:24:09,419 INFO [train.py:894] (2/4) Epoch 7, batch 450, loss[loss=0.2291, simple_loss=0.3095, pruned_loss=0.07429, over 18628.00 frames. ], tot_loss[loss=0.2259, simple_loss=0.3032, pruned_loss=0.07427, over 3326607.10 frames. ], batch size: 53, lr: 1.69e-02, grad_scale: 8.0
2022-12-22 19:24:21,344 WARNING [train.py:1060] (2/4) Exclude cut with ID 5239-32139-0047-9341-0_sp0.9 from training. Duration: 30.1555625
2022-12-22 19:24:25,377 WARNING [train.py:1060] (2/4) Exclude cut with ID 1265-135635-0050-6781-0_sp0.9 from training. Duration: 21.8333125
2022-12-22 19:24:32,448 INFO [scaling.py:679] (2/4) Whitening: num_groups=8, num_channels=192, metric=1.86 vs. limit=2.0
2022-12-22 19:24:34,893 WARNING [train.py:1060] (2/4) Exclude cut with ID 1914-133440-0024-94914-0_sp1.1 from training. Duration: 20.6545625
2022-12-22 19:24:57,402 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 2.897e+02 4.831e+02 6.096e+02 7.649e+02 1.561e+03, threshold=1.219e+03, percent-clipped=2.0
2022-12-22 19:25:01,265 INFO [scaling.py:679] (2/4) Whitening: num_groups=1, num_channels=384, metric=4.25 vs. limit=5.0
2022-12-22 19:25:21,668 WARNING [train.py:1060] (2/4) Exclude cut with ID 7395-89880-0045-39920-0_sp0.9 from training. Duration: 20.52225
2022-12-22 19:25:24,463 INFO [train.py:894] (2/4) Epoch 7, batch 500, loss[loss=0.2029, simple_loss=0.2781, pruned_loss=0.0638, over 18608.00 frames. ], tot_loss[loss=0.2262, simple_loss=0.3034, pruned_loss=0.07447, over 3412062.52 frames. ], batch size: 45, lr: 1.69e-02, grad_scale: 8.0
2022-12-22 19:25:39,258 WARNING [train.py:1060] (2/4) Exclude cut with ID 3972-170212-0014-23379-0_sp0.9 from training. Duration: 29.1166875
2022-12-22 19:26:11,534 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.7194, 1.2905, 1.3027, 1.2844, 1.6678, 1.9204, 1.9542, 1.1723],
       device='cuda:2'), covar=tensor([0.0385, 0.0275, 0.0496, 0.0284, 0.0221, 0.0307, 0.0229, 0.0300],
       device='cuda:2'), in_proj_covar=tensor([0.0079, 0.0108, 0.0132, 0.0120, 0.0104, 0.0095, 0.0081, 0.0111],
       device='cuda:2'), out_proj_covar=tensor([7.4664e-05, 9.8614e-05, 1.2525e-04, 1.1066e-04, 9.9290e-05, 8.5069e-05,
        7.3646e-05, 1.0187e-04], device='cuda:2')
2022-12-22 19:26:38,647 WARNING [train.py:1060] (2/4) Exclude cut with ID 543-133211-0007-59831-0_sp0.9 from training. Duration: 21.388875
2022-12-22 19:26:40,248 INFO [train.py:894] (2/4) Epoch 7, batch 550, loss[loss=0.2122, simple_loss=0.279, pruned_loss=0.07271, over 18471.00 frames. ], tot_loss[loss=0.2275, simple_loss=0.3044, pruned_loss=0.07534, over 3478569.21 frames. ], batch size: 43, lr: 1.69e-02, grad_scale: 8.0
2022-12-22 19:26:51,039 INFO [zipformer.py:660] (2/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=21595.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 19:27:06,339 INFO [scaling.py:679] (2/4) Whitening: num_groups=8, num_channels=192, metric=1.94 vs. limit=2.0
2022-12-22 19:27:15,472 WARNING [train.py:1060] (2/4) Exclude cut with ID 1914-133440-0024-94914-0 from training. Duration: 22.72
2022-12-22 19:27:15,505 WARNING [train.py:1060] (2/4) Exclude cut with ID 1914-133440-0031-94921-0_sp0.9 from training. Duration: 22.7444375
2022-12-22 19:27:18,738 INFO [zipformer.py:660] (2/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=21613.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 19:27:28,905 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 3.471e+02 5.026e+02 6.152e+02 7.291e+02 2.360e+03, threshold=1.230e+03, percent-clipped=3.0
2022-12-22 19:27:57,037 INFO [train.py:894] (2/4) Epoch 7, batch 600, loss[loss=0.2062, simple_loss=0.2943, pruned_loss=0.05902, over 18415.00 frames. ], tot_loss[loss=0.2279, simple_loss=0.3049, pruned_loss=0.07549, over 3531836.69 frames. ], batch size: 48, lr: 1.68e-02, grad_scale: 8.0
2022-12-22 19:27:58,521 WARNING [train.py:1060] (2/4) Exclude cut with ID 4133-6541-0027-40495-0_sp1.1 from training. Duration: 0.9681875
2022-12-22 19:28:01,604 WARNING [train.py:1060] (2/4) Exclude cut with ID 6330-62851-0022-91297-0_sp0.9 from training. Duration: 22.3166875
2022-12-22 19:28:09,006 WARNING [train.py:1060] (2/4) Exclude cut with ID 543-133212-0015-59917-0_sp0.9 from training. Duration: 21.8166875
2022-12-22 19:28:24,645 INFO [zipformer.py:660] (2/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=21656.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 19:28:42,092 INFO [zipformer.py:660] (2/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=21668.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 19:28:43,945 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([2.1788, 1.3899, 1.8841, 1.9194, 2.0552, 2.0313, 2.0539, 1.3231],
       device='cuda:2'), covar=tensor([0.1145, 0.1937, 0.1404, 0.1465, 0.0952, 0.0536, 0.1503, 0.0720],
       device='cuda:2'), in_proj_covar=tensor([0.0223, 0.0258, 0.0230, 0.0253, 0.0238, 0.0208, 0.0251, 0.0200],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-22 19:28:45,998 INFO [scaling.py:679] (2/4) Whitening: num_groups=8, num_channels=192, metric=1.96 vs. limit=2.0
2022-12-22 19:28:50,988 INFO [zipformer.py:660] (2/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=21674.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 19:29:07,177 INFO [zipformer.py:660] (2/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=21684.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 19:29:13,215 INFO [train.py:894] (2/4) Epoch 7, batch 650, loss[loss=0.2575, simple_loss=0.3324, pruned_loss=0.09132, over 18704.00 frames. ], tot_loss[loss=0.2288, simple_loss=0.3057, pruned_loss=0.07601, over 3571913.12 frames. ], batch size: 60, lr: 1.68e-02, grad_scale: 8.0
2022-12-22 19:29:45,300 INFO [zipformer.py:660] (2/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=21708.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 19:29:56,863 WARNING [train.py:1060] (2/4) Exclude cut with ID 4957-30119-0041-23990-0_sp0.9 from training. Duration: 20.22775
2022-12-22 19:30:02,250 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 3.404e+02 4.992e+02 6.426e+02 7.914e+02 1.249e+03, threshold=1.285e+03, percent-clipped=1.0
2022-12-22 19:30:30,274 INFO [train.py:894] (2/4) Epoch 7, batch 700, loss[loss=0.2289, simple_loss=0.3136, pruned_loss=0.07209, over 18571.00 frames. ], tot_loss[loss=0.229, simple_loss=0.3057, pruned_loss=0.07621, over 3603238.06 frames. ], batch size: 57, lr: 1.68e-02, grad_scale: 8.0
2022-12-22 19:30:42,013 WARNING [train.py:1060] (2/4) Exclude cut with ID 5239-32139-0047-9341-0_sp1.1 from training. Duration: 24.67275
2022-12-22 19:30:57,383 INFO [zipformer.py:660] (2/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=21756.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 19:31:07,768 INFO [zipformer.py:660] (2/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=21763.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 19:31:10,229 WARNING [train.py:1060] (2/4) Exclude cut with ID 3082-165428-0081-50734-0_sp0.9 from training. Duration: 21.8055625
2022-12-22 19:31:24,456 INFO [zipformer.py:660] (2/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=21774.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 19:31:45,977 INFO [train.py:894] (2/4) Epoch 7, batch 750, loss[loss=0.2164, simple_loss=0.3022, pruned_loss=0.06528, over 18726.00 frames. ], tot_loss[loss=0.2286, simple_loss=0.3056, pruned_loss=0.07584, over 3628538.55 frames. ], batch size: 54, lr: 1.68e-02, grad_scale: 8.0
2022-12-22 19:31:47,689 WARNING [train.py:1060] (2/4) Exclude cut with ID 3340-169293-0054-76830-0_sp0.9 from training. Duration: 22.6666875
2022-12-22 19:32:19,483 INFO [scaling.py:679] (2/4) Whitening: num_groups=8, num_channels=96, metric=1.91 vs. limit=2.0
2022-12-22 19:32:34,810 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 3.290e+02 4.818e+02 5.879e+02 7.436e+02 2.235e+03, threshold=1.176e+03, percent-clipped=3.0
2022-12-22 19:32:38,393 INFO [zipformer.py:660] (2/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=21822.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 19:32:42,085 INFO [zipformer.py:660] (2/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=21824.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 19:32:49,542 WARNING [train.py:1060] (2/4) Exclude cut with ID 2411-132532-0017-82279-0_sp1.1 from training. Duration: 0.9681875
2022-12-22 19:33:02,677 INFO [train.py:894] (2/4) Epoch 7, batch 800, loss[loss=0.1866, simple_loss=0.2724, pruned_loss=0.05044, over 18388.00 frames. ], tot_loss[loss=0.2279, simple_loss=0.305, pruned_loss=0.07541, over 3647531.35 frames. ], batch size: 46, lr: 1.68e-02, grad_scale: 8.0
2022-12-22 19:33:16,781 WARNING [train.py:1060] (2/4) Exclude cut with ID 6330-62850-0007-91323-0 from training. Duration: 22.485
2022-12-22 19:33:21,668 INFO [zipformer.py:660] (2/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=21850.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 19:33:54,508 WARNING [train.py:1060] (2/4) Exclude cut with ID 3972-170212-0014-23379-0_sp1.1 from training. Duration: 23.82275
2022-12-22 19:34:08,642 WARNING [train.py:1060] (2/4) Exclude cut with ID 4860-13185-0032-76709-0 from training. Duration: 20.77
2022-12-22 19:34:09,330 INFO [scaling.py:679] (2/4) Whitening: num_groups=1, num_channels=384, metric=2.75 vs. limit=5.0
2022-12-22 19:34:14,754 WARNING [train.py:1060] (2/4) Exclude cut with ID 6426-64292-0017-15984-0_sp0.9 from training. Duration: 24.088875
2022-12-22 19:34:19,347 INFO [train.py:894] (2/4) Epoch 7, batch 850, loss[loss=0.2692, simple_loss=0.3438, pruned_loss=0.0973, over 18579.00 frames. ], tot_loss[loss=0.2276, simple_loss=0.3049, pruned_loss=0.07515, over 3661299.89 frames. ], batch size: 57, lr: 1.67e-02, grad_scale: 8.0
2022-12-22 19:34:46,517 WARNING [train.py:1060] (2/4) Exclude cut with ID 6330-62850-0007-91323-0_sp1.1 from training. Duration: 20.4409375
2022-12-22 19:34:54,156 INFO [zipformer.py:660] (2/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=21911.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 19:35:07,550 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 3.100e+02 5.077e+02 6.142e+02 7.793e+02 1.798e+03, threshold=1.228e+03, percent-clipped=5.0
2022-12-22 19:35:34,838 INFO [train.py:894] (2/4) Epoch 7, batch 900, loss[loss=0.2216, simple_loss=0.3065, pruned_loss=0.06836, over 18511.00 frames. ], tot_loss[loss=0.2264, simple_loss=0.3039, pruned_loss=0.07447, over 3673572.61 frames. ], batch size: 52, lr: 1.67e-02, grad_scale: 8.0
2022-12-22 19:35:54,491 INFO [zipformer.py:660] (2/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=21951.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 19:35:54,670 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.9867, 1.2883, 0.6076, 1.3904, 2.0696, 1.4627, 1.8083, 2.2863],
       device='cuda:2'), covar=tensor([0.1496, 0.2037, 0.2806, 0.1591, 0.1698, 0.1463, 0.1367, 0.1279],
       device='cuda:2'), in_proj_covar=tensor([0.0093, 0.0104, 0.0124, 0.0100, 0.0111, 0.0094, 0.0097, 0.0098],
       device='cuda:2'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003],
       device='cuda:2')
2022-12-22 19:36:04,121 WARNING [train.py:1060] (2/4) Exclude cut with ID 2929-85685-0062-62366-0_sp0.9 from training. Duration: 22.511125
2022-12-22 19:36:05,552 WARNING [train.py:1060] (2/4) Exclude cut with ID 7395-89880-0031-39906-0 from training. Duration: 20.675
2022-12-22 19:36:19,559 INFO [zipformer.py:660] (2/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=21968.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 19:36:20,758 INFO [zipformer.py:660] (2/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=21969.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 19:36:22,444 INFO [zipformer.py:660] (2/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=21970.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 19:36:44,717 INFO [zipformer.py:660] (2/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=21984.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 19:36:50,567 INFO [train.py:894] (2/4) Epoch 7, batch 950, loss[loss=0.2248, simple_loss=0.3106, pruned_loss=0.06952, over 18579.00 frames. ], tot_loss[loss=0.2267, simple_loss=0.3044, pruned_loss=0.07454, over 3682273.13 frames. ], batch size: 51, lr: 1.67e-02, grad_scale: 8.0
2022-12-22 19:37:17,481 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.8103, 1.3826, 1.3606, 1.8957, 1.4678, 3.4439, 1.2988, 1.6139],
       device='cuda:2'), covar=tensor([0.0885, 0.1805, 0.1198, 0.1055, 0.1638, 0.0233, 0.1378, 0.1541],
       device='cuda:2'), in_proj_covar=tensor([0.0078, 0.0087, 0.0081, 0.0083, 0.0096, 0.0074, 0.0088, 0.0082],
       device='cuda:2'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0003, 0.0004, 0.0003, 0.0003, 0.0003],
       device='cuda:2')
2022-12-22 19:37:36,855 INFO [zipformer.py:660] (2/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=22016.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 19:37:42,503 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 2.707e+02 4.665e+02 5.685e+02 7.143e+02 1.158e+03, threshold=1.137e+03, percent-clipped=0.0
2022-12-22 19:37:51,821 WARNING [train.py:1060] (2/4) Exclude cut with ID 6330-62850-0007-91323-0_sp0.9 from training. Duration: 24.9833125
2022-12-22 19:37:59,559 INFO [zipformer.py:660] (2/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=22031.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 19:38:01,074 INFO [zipformer.py:660] (2/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=22032.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 19:38:10,119 INFO [train.py:894] (2/4) Epoch 7, batch 1000, loss[loss=0.2389, simple_loss=0.3237, pruned_loss=0.07701, over 18599.00 frames. ], tot_loss[loss=0.2266, simple_loss=0.3044, pruned_loss=0.07444, over 3688981.36 frames. ], batch size: 77, lr: 1.67e-02, grad_scale: 8.0
2022-12-22 19:38:20,477 WARNING [train.py:1060] (2/4) Exclude cut with ID 5239-32139-0047-9341-0 from training. Duration: 27.14
2022-12-22 19:38:34,543 WARNING [train.py:1060] (2/4) Exclude cut with ID 2929-85685-0044-62348-0 from training. Duration: 22.44
2022-12-22 19:39:26,394 INFO [train.py:894] (2/4) Epoch 7, batch 1050, loss[loss=0.233, simple_loss=0.3141, pruned_loss=0.07596, over 18594.00 frames. ], tot_loss[loss=0.2281, simple_loss=0.3054, pruned_loss=0.07543, over 3694544.29 frames. ], batch size: 56, lr: 1.67e-02, grad_scale: 8.0
2022-12-22 19:39:34,491 INFO [scaling.py:679] (2/4) Whitening: num_groups=8, num_channels=96, metric=1.41 vs. limit=2.0
2022-12-22 19:39:43,039 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([2.1570, 2.1665, 1.2461, 2.3651, 2.3296, 1.9830, 3.1450, 2.1269],
       device='cuda:2'), covar=tensor([0.0788, 0.1475, 0.2444, 0.1613, 0.1424, 0.0785, 0.0759, 0.1018],
       device='cuda:2'), in_proj_covar=tensor([0.0169, 0.0182, 0.0222, 0.0266, 0.0213, 0.0172, 0.0190, 0.0184],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-22 19:39:53,374 WARNING [train.py:1060] (2/4) Exclude cut with ID 2929-85685-0060-62364-0_sp0.9 from training. Duration: 21.361125
2022-12-22 19:40:01,069 WARNING [train.py:1060] (2/4) Exclude cut with ID 2929-85685-0079-62383-0_sp1.1 from training. Duration: 27.0318125
2022-12-22 19:40:07,323 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([3.4485, 2.9849, 3.0743, 3.3165, 3.1126, 3.1420, 3.5431, 1.6535],
       device='cuda:2'), covar=tensor([0.0662, 0.0557, 0.0527, 0.0652, 0.1170, 0.0837, 0.0653, 0.3134],
       device='cuda:2'), in_proj_covar=tensor([0.0244, 0.0174, 0.0174, 0.0170, 0.0241, 0.0197, 0.0195, 0.0225],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0001, 0.0001, 0.0001, 0.0002, 0.0001, 0.0002, 0.0002],
       device='cuda:2')
2022-12-22 19:40:09,355 WARNING [train.py:1060] (2/4) Exclude cut with ID 5622-44585-0006-90525-0_sp0.9 from training. Duration: 28.638875
2022-12-22 19:40:13,879 INFO [zipformer.py:660] (2/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=22119.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 19:40:15,138 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 3.278e+02 5.032e+02 6.199e+02 7.170e+02 1.495e+03, threshold=1.240e+03, percent-clipped=4.0
2022-12-22 19:40:25,140 WARNING [train.py:1060] (2/4) Exclude cut with ID 3340-169293-0054-76830-0 from training. Duration: 20.4
2022-12-22 19:40:29,969 INFO [zipformer.py:660] (2/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=22130.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 19:40:42,010 INFO [train.py:894] (2/4) Epoch 7, batch 1100, loss[loss=0.2128, simple_loss=0.2942, pruned_loss=0.06572, over 18465.00 frames. ], tot_loss[loss=0.2274, simple_loss=0.305, pruned_loss=0.07489, over 3698584.94 frames. ], batch size: 50, lr: 1.67e-02, grad_scale: 8.0
2022-12-22 19:40:58,252 WARNING [train.py:1060] (2/4) Exclude cut with ID 2929-85685-0071-62375-0 from training. Duration: 20.025
2022-12-22 19:40:59,758 WARNING [train.py:1060] (2/4) Exclude cut with ID 2364-131735-0112-64612-0_sp0.9 from training. Duration: 20.488875
2022-12-22 19:41:04,479 WARNING [train.py:1060] (2/4) Exclude cut with ID 2929-85685-0079-62383-0 from training. Duration: 29.735
2022-12-22 19:41:40,919 INFO [zipformer.py:660] (2/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=22176.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 19:41:58,847 INFO [train.py:894] (2/4) Epoch 7, batch 1150, loss[loss=0.2334, simple_loss=0.3108, pruned_loss=0.07793, over 18472.00 frames. ], tot_loss[loss=0.2266, simple_loss=0.3045, pruned_loss=0.07441, over 3701401.70 frames. ], batch size: 54, lr: 1.66e-02, grad_scale: 8.0
2022-12-22 19:42:03,756 INFO [zipformer.py:660] (2/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=22191.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 19:42:24,344 WARNING [train.py:1060] (2/4) Exclude cut with ID 7276-92427-0014-12983-0_sp0.9 from training. Duration: 21.3055625
2022-12-22 19:42:26,404 WARNING [train.py:1060] (2/4) Exclude cut with ID 1025-75365-0008-79168-0_sp0.9 from training. Duration: 22.0666875
2022-12-22 19:42:26,509 INFO [zipformer.py:660] (2/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=22206.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 19:42:40,404 INFO [zipformer.py:660] (2/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=22215.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 19:42:47,743 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 3.336e+02 4.819e+02 6.126e+02 7.928e+02 2.801e+03, threshold=1.225e+03, percent-clipped=6.0
2022-12-22 19:43:14,512 INFO [zipformer.py:660] (2/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=22237.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 19:43:14,523 INFO [zipformer.py:660] (2/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=22237.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 19:43:15,638 INFO [train.py:894] (2/4) Epoch 7, batch 1200, loss[loss=0.2, simple_loss=0.2733, pruned_loss=0.06335, over 18586.00 frames. ], tot_loss[loss=0.2253, simple_loss=0.3031, pruned_loss=0.07376, over 3704591.74 frames. ], batch size: 41, lr: 1.66e-02, grad_scale: 8.0
2022-12-22 19:43:34,860 INFO [zipformer.py:660] (2/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=22251.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 19:43:35,426 INFO [scaling.py:679] (2/4) Whitening: num_groups=1, num_channels=384, metric=3.81 vs. limit=5.0
2022-12-22 19:44:02,636 INFO [zipformer.py:660] (2/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=22269.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 19:44:13,705 INFO [zipformer.py:660] (2/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=22276.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 19:44:16,032 WARNING [train.py:1060] (2/4) Exclude cut with ID 2929-85685-0062-62366-0 from training. Duration: 20.26
2022-12-22 19:44:28,718 WARNING [train.py:1060] (2/4) Exclude cut with ID 5239-32139-0030-9324-0_sp0.9 from training. Duration: 21.3444375
2022-12-22 19:44:30,247 INFO [train.py:894] (2/4) Epoch 7, batch 1250, loss[loss=0.1975, simple_loss=0.2755, pruned_loss=0.05972, over 18709.00 frames. ], tot_loss[loss=0.2249, simple_loss=0.303, pruned_loss=0.07336, over 3705984.95 frames. ], batch size: 46, lr: 1.66e-02, grad_scale: 8.0
2022-12-22 19:44:44,942 INFO [zipformer.py:660] (2/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=22298.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 19:44:46,115 INFO [zipformer.py:660] (2/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=22299.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 19:44:51,948 INFO [scaling.py:679] (2/4) Whitening: num_groups=8, num_channels=96, metric=1.53 vs. limit=2.0
2022-12-22 19:45:14,645 INFO [zipformer.py:660] (2/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=22317.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 19:45:18,806 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 3.340e+02 4.996e+02 6.161e+02 7.748e+02 1.648e+03, threshold=1.232e+03, percent-clipped=4.0
2022-12-22 19:45:25,231 WARNING [train.py:1060] (2/4) Exclude cut with ID 497-129325-0061-62254-0_sp1.1 from training. Duration: 0.97725
2022-12-22 19:45:28,507 INFO [zipformer.py:660] (2/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=22326.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 19:45:36,030 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.5476, 1.1139, 1.2715, 1.1513, 1.6170, 1.8326, 1.8389, 1.2124],
       device='cuda:2'), covar=tensor([0.0322, 0.0327, 0.0484, 0.0286, 0.0236, 0.0285, 0.0248, 0.0298],
       device='cuda:2'), in_proj_covar=tensor([0.0078, 0.0107, 0.0131, 0.0117, 0.0100, 0.0093, 0.0082, 0.0111],
       device='cuda:2'), out_proj_covar=tensor([7.2907e-05, 9.7002e-05, 1.2405e-04, 1.0750e-04, 9.5157e-05, 8.1806e-05,
        7.4384e-05, 1.0037e-04], device='cuda:2')
2022-12-22 19:45:45,639 INFO [train.py:894] (2/4) Epoch 7, batch 1300, loss[loss=0.2061, simple_loss=0.2776, pruned_loss=0.06737, over 18597.00 frames. ], tot_loss[loss=0.225, simple_loss=0.3031, pruned_loss=0.07343, over 3709201.49 frames. ], batch size: 45, lr: 1.66e-02, grad_scale: 8.0
2022-12-22 19:46:08,660 WARNING [train.py:1060] (2/4) Exclude cut with ID 7395-89880-0031-39906-0_sp0.9 from training. Duration: 22.97225
2022-12-22 19:46:39,104 WARNING [train.py:1060] (2/4) Exclude cut with ID 7395-89880-0047-39922-0_sp0.9 from training. Duration: 21.97775
2022-12-22 19:46:53,730 WARNING [train.py:1060] (2/4) Exclude cut with ID 1112-1043-0006-89194-0_sp0.9 from training. Duration: 21.8333125
2022-12-22 19:46:57,068 INFO [zipformer.py:660] (2/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=22385.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 19:47:00,755 INFO [train.py:894] (2/4) Epoch 7, batch 1350, loss[loss=0.2714, simple_loss=0.3423, pruned_loss=0.1002, over 18644.00 frames. ], tot_loss[loss=0.2244, simple_loss=0.3024, pruned_loss=0.07317, over 3710092.38 frames. ], batch size: 78, lr: 1.66e-02, grad_scale: 8.0
2022-12-22 19:47:03,427 WARNING [train.py:1060] (2/4) Exclude cut with ID 1914-133440-0031-94921-0 from training. Duration: 20.47
2022-12-22 19:47:47,933 INFO [zipformer.py:660] (2/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=22419.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 19:47:49,173 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 2.961e+02 4.936e+02 5.995e+02 7.767e+02 1.533e+03, threshold=1.199e+03, percent-clipped=4.0
2022-12-22 19:48:10,090 WARNING [train.py:1060] (2/4) Exclude cut with ID 7395-89880-0037-39912-0_sp0.9 from training. Duration: 20.67225
2022-12-22 19:48:15,711 INFO [train.py:894] (2/4) Epoch 7, batch 1400, loss[loss=0.2082, simple_loss=0.2784, pruned_loss=0.069, over 18405.00 frames. ], tot_loss[loss=0.2246, simple_loss=0.3026, pruned_loss=0.07332, over 3710780.49 frames. ], batch size: 42, lr: 1.66e-02, grad_scale: 8.0
2022-12-22 19:48:28,222 WARNING [train.py:1060] (2/4) Exclude cut with ID 1914-133440-0024-94914-0_sp0.9 from training. Duration: 25.2444375
2022-12-22 19:48:28,609 INFO [zipformer.py:660] (2/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=22446.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 19:48:52,362 WARNING [train.py:1060] (2/4) Exclude cut with ID 3340-169293-0021-76797-0_sp0.9 from training. Duration: 21.1445
2022-12-22 19:49:01,409 INFO [zipformer.py:660] (2/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=22467.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 19:49:19,954 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.2885, 1.6332, 1.2251, 2.0559, 2.0584, 1.4255, 1.4113, 1.1037],
       device='cuda:2'), covar=tensor([0.1940, 0.1791, 0.1548, 0.0892, 0.1401, 0.1246, 0.1695, 0.1626],
       device='cuda:2'), in_proj_covar=tensor([0.0228, 0.0202, 0.0194, 0.0179, 0.0243, 0.0185, 0.0200, 0.0182],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-22 19:49:29,870 INFO [zipformer.py:660] (2/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=22486.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 19:49:32,419 INFO [train.py:894] (2/4) Epoch 7, batch 1450, loss[loss=0.1936, simple_loss=0.2764, pruned_loss=0.05539, over 18387.00 frames. ], tot_loss[loss=0.2239, simple_loss=0.3022, pruned_loss=0.07279, over 3710770.28 frames. ], batch size: 46, lr: 1.65e-02, grad_scale: 8.0
2022-12-22 19:49:59,534 INFO [zipformer.py:660] (2/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=22506.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 19:50:10,443 WARNING [train.py:1060] (2/4) Exclude cut with ID 2929-85685-0079-62383-0_sp0.9 from training. Duration: 33.038875
2022-12-22 19:50:21,665 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 2.790e+02 4.574e+02 5.514e+02 7.111e+02 1.060e+03, threshold=1.103e+03, percent-clipped=0.0
2022-12-22 19:50:35,064 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([2.4192, 1.6614, 0.9347, 1.7141, 2.5070, 1.6036, 2.4970, 2.4977],
       device='cuda:2'), covar=tensor([0.1610, 0.2163, 0.2634, 0.1671, 0.1653, 0.1571, 0.1295, 0.1570],
       device='cuda:2'), in_proj_covar=tensor([0.0093, 0.0104, 0.0123, 0.0100, 0.0111, 0.0093, 0.0098, 0.0098],
       device='cuda:2'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003],
       device='cuda:2')
2022-12-22 19:50:39,564 INFO [zipformer.py:660] (2/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=22532.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 19:50:46,413 WARNING [train.py:1060] (2/4) Exclude cut with ID 6426-64291-0000-16059-0_sp0.9 from training. Duration: 20.0944375
2022-12-22 19:50:47,905 INFO [train.py:894] (2/4) Epoch 7, batch 1500, loss[loss=0.2421, simple_loss=0.3171, pruned_loss=0.08356, over 18724.00 frames. ], tot_loss[loss=0.224, simple_loss=0.3023, pruned_loss=0.07286, over 3711143.11 frames. ], batch size: 54, lr: 1.65e-02, grad_scale: 8.0
2022-12-22 19:51:01,900 WARNING [train.py:1060] (2/4) Exclude cut with ID 2929-85685-0044-62348-0_sp1.1 from training. Duration: 20.4
2022-12-22 19:51:09,230 WARNING [train.py:1060] (2/4) Exclude cut with ID 6330-62851-0022-91297-0 from training. Duration: 20.085
2022-12-22 19:51:12,167 INFO [zipformer.py:660] (2/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=22554.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 19:51:22,453 WARNING [train.py:1060] (2/4) Exclude cut with ID 4860-13185-0032-76709-0_sp0.9 from training. Duration: 23.07775
2022-12-22 19:51:39,098 INFO [zipformer.py:660] (2/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=22571.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 19:52:03,706 INFO [train.py:894] (2/4) Epoch 7, batch 1550, loss[loss=0.2414, simple_loss=0.3245, pruned_loss=0.07917, over 18558.00 frames. ], tot_loss[loss=0.2243, simple_loss=0.3026, pruned_loss=0.07297, over 3710853.13 frames. ], batch size: 55, lr: 1.65e-02, grad_scale: 8.0
2022-12-22 19:52:05,447 WARNING [train.py:1060] (2/4) Exclude cut with ID 2929-85685-0044-62348-0_sp0.9 from training. Duration: 24.9333125
2022-12-22 19:52:12,267 INFO [zipformer.py:660] (2/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=22593.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 19:52:54,996 WARNING [train.py:1060] (2/4) Exclude cut with ID 5118-111612-0016-124680-0_sp0.9 from training. Duration: 20.388875
2022-12-22 19:52:56,545 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 3.322e+02 5.232e+02 6.595e+02 8.156e+02 1.283e+03, threshold=1.319e+03, percent-clipped=2.0
2022-12-22 19:53:02,269 WARNING [train.py:1060] (2/4) Exclude cut with ID 432-122774-0017-62487-0_sp1.1 from training. Duration: 20.3590625
2022-12-22 19:53:05,150 INFO [zipformer.py:660] (2/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=22626.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 19:53:21,282 INFO [train.py:894] (2/4) Epoch 7, batch 1600, loss[loss=0.2542, simple_loss=0.3307, pruned_loss=0.08884, over 18447.00 frames. ], tot_loss[loss=0.2252, simple_loss=0.3032, pruned_loss=0.07359, over 3710465.18 frames. ], batch size: 64, lr: 1.65e-02, grad_scale: 8.0
2022-12-22 19:53:23,196 INFO [zipformer.py:660] (2/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=22639.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 19:53:28,723 INFO [scaling.py:679] (2/4) Whitening: num_groups=8, num_channels=192, metric=1.99 vs. limit=2.0
2022-12-22 19:54:06,712 WARNING [train.py:1060] (2/4) Exclude cut with ID 3557-8342-0013-54691-0_sp1.1 from training. Duration: 0.836375
2022-12-22 19:54:16,993 INFO [zipformer.py:660] (2/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=22674.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 19:54:37,273 INFO [train.py:894] (2/4) Epoch 7, batch 1650, loss[loss=0.2269, simple_loss=0.301, pruned_loss=0.07643, over 18701.00 frames. ], tot_loss[loss=0.2277, simple_loss=0.3047, pruned_loss=0.0754, over 3711526.89 frames. ], batch size: 50, lr: 1.65e-02, grad_scale: 8.0
2022-12-22 19:54:37,742 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.4938, 1.6292, 1.5199, 1.5584, 1.4809, 3.3018, 1.6621, 1.9188],
       device='cuda:2'), covar=tensor([0.3219, 0.1846, 0.1848, 0.1974, 0.1279, 0.0208, 0.1398, 0.0922],
       device='cuda:2'), in_proj_covar=tensor([0.0146, 0.0125, 0.0135, 0.0124, 0.0110, 0.0100, 0.0104, 0.0101],
       device='cuda:2'), out_proj_covar=tensor([0.0004, 0.0004, 0.0004, 0.0004, 0.0003, 0.0003, 0.0004, 0.0003],
       device='cuda:2')
2022-12-22 19:54:51,043 WARNING [train.py:1060] (2/4) Exclude cut with ID 8565-290391-0049-67394-0_sp0.9 from training. Duration: 21.3166875
2022-12-22 19:54:55,825 INFO [zipformer.py:660] (2/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=22700.0, num_to_drop=1, layers_to_drop={0}
2022-12-22 19:55:22,658 WARNING [train.py:1060] (2/4) Exclude cut with ID 6533-399-0029-104863-0_sp0.9 from training. Duration: 22.1055625
2022-12-22 19:55:26,846 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 3.320e+02 5.398e+02 6.482e+02 7.925e+02 1.712e+03, threshold=1.296e+03, percent-clipped=2.0
2022-12-22 19:55:32,297 WARNING [train.py:1060] (2/4) Exclude cut with ID 7699-105389-0094-26379-0_sp1.1 from training. Duration: 21.77725
2022-12-22 19:55:43,600 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.2937, 0.9514, 1.3720, 2.0648, 1.3581, 2.1668, 0.7169, 1.4877],
       device='cuda:2'), covar=tensor([0.2031, 0.1992, 0.1570, 0.0773, 0.1670, 0.0976, 0.2250, 0.1517],
       device='cuda:2'), in_proj_covar=tensor([0.0106, 0.0115, 0.0128, 0.0111, 0.0105, 0.0131, 0.0132, 0.0109],
       device='cuda:2'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003],
       device='cuda:2')
2022-12-22 19:55:53,842 INFO [train.py:894] (2/4) Epoch 7, batch 1700, loss[loss=0.3757, simple_loss=0.4015, pruned_loss=0.1749, over 18657.00 frames. ], tot_loss[loss=0.2328, simple_loss=0.3078, pruned_loss=0.0789, over 3712172.29 frames. ], batch size: 181, lr: 1.64e-02, grad_scale: 8.0
2022-12-22 19:55:53,876 WARNING [train.py:1060] (2/4) Exclude cut with ID 7255-291500-0005-134304-0_sp0.9 from training. Duration: 27.8166875
2022-12-22 19:55:58,547 INFO [zipformer.py:660] (2/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=22741.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 19:56:02,998 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.5976, 1.0722, 1.9545, 2.9436, 1.9425, 2.1403, 0.5380, 1.7923],
       device='cuda:2'), covar=tensor([0.1734, 0.1870, 0.1417, 0.0526, 0.1313, 0.1322, 0.2576, 0.1463],
       device='cuda:2'), in_proj_covar=tensor([0.0106, 0.0115, 0.0126, 0.0110, 0.0105, 0.0130, 0.0132, 0.0109],
       device='cuda:2'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003],
       device='cuda:2')
2022-12-22 19:56:16,566 WARNING [train.py:1060] (2/4) Exclude cut with ID 7357-94126-0021-15852-0_sp1.1 from training. Duration: 22.5090625
2022-12-22 19:56:24,435 WARNING [train.py:1060] (2/4) Exclude cut with ID 7255-291500-0005-134304-0 from training. Duration: 25.035
2022-12-22 19:56:32,425 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.7042, 1.7437, 1.2418, 1.7614, 1.5612, 1.5550, 1.5228, 1.8591],
       device='cuda:2'), covar=tensor([0.1588, 0.2041, 0.1384, 0.1879, 0.2035, 0.0845, 0.1816, 0.0571],
       device='cuda:2'), in_proj_covar=tensor([0.0260, 0.0248, 0.0213, 0.0321, 0.0235, 0.0201, 0.0244, 0.0171],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0003, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-22 19:56:41,725 WARNING [train.py:1060] (2/4) Exclude cut with ID 774-127930-0014-10412-0_sp1.1 from training. Duration: 0.95
2022-12-22 19:57:02,215 WARNING [train.py:1060] (2/4) Exclude cut with ID 3033-130750-0096-55598-0_sp0.9 from training. Duration: 0.92225
2022-12-22 19:57:06,980 INFO [zipformer.py:660] (2/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=22786.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 19:57:09,506 INFO [train.py:894] (2/4) Epoch 7, batch 1750, loss[loss=0.233, simple_loss=0.2921, pruned_loss=0.08689, over 18470.00 frames. ], tot_loss[loss=0.2363, simple_loss=0.3094, pruned_loss=0.08161, over 3712584.48 frames. ], batch size: 43, lr: 1.64e-02, grad_scale: 8.0
2022-12-22 19:57:18,653 INFO [zipformer.py:660] (2/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=22794.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 19:57:27,798 WARNING [train.py:1060] (2/4) Exclude cut with ID 4511-76322-0006-80011-0 from training. Duration: 21.97
2022-12-22 19:57:48,461 WARNING [train.py:1060] (2/4) Exclude cut with ID 7492-105653-0055-62765-0_sp0.9 from training. Duration: 21.97225
2022-12-22 19:57:48,496 WARNING [train.py:1060] (2/4) Exclude cut with ID 453-131332-0000-47844-0_sp0.9 from training. Duration: 25.3333125
2022-12-22 19:57:58,563 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 3.602e+02 6.360e+02 7.801e+02 9.201e+02 2.109e+03, threshold=1.560e+03, percent-clipped=7.0
2022-12-22 19:58:00,207 WARNING [train.py:1060] (2/4) Exclude cut with ID 5172-29468-0015-19128-0_sp0.9 from training. Duration: 21.5055625
2022-12-22 19:58:06,229 INFO [zipformer.py:660] (2/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=22825.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 19:58:07,433 WARNING [train.py:1060] (2/4) Exclude cut with ID 453-131332-0000-47844-0_sp1.1 from training. Duration: 20.72725
2022-12-22 19:58:12,196 INFO [scaling.py:679] (2/4) Whitening: num_groups=8, num_channels=192, metric=1.97 vs. limit=2.0
2022-12-22 19:58:17,314 INFO [zipformer.py:660] (2/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=22832.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 19:58:20,138 INFO [zipformer.py:660] (2/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=22834.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 19:58:26,161 INFO [train.py:894] (2/4) Epoch 7, batch 1800, loss[loss=0.1958, simple_loss=0.267, pruned_loss=0.06225, over 18514.00 frames. ], tot_loss[loss=0.2396, simple_loss=0.3109, pruned_loss=0.08418, over 3713021.53 frames. ], batch size: 44, lr: 1.64e-02, grad_scale: 8.0
2022-12-22 19:58:39,380 WARNING [train.py:1060] (2/4) Exclude cut with ID 8631-249866-0030-130156-0_sp0.9 from training. Duration: 26.32775
2022-12-22 19:58:54,850 INFO [zipformer.py:660] (2/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=22855.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 19:59:13,688 WARNING [train.py:1060] (2/4) Exclude cut with ID 3867-173237-0077-144769-0 from training. Duration: 20.025
2022-12-22 19:59:20,398 WARNING [train.py:1060] (2/4) Exclude cut with ID 6709-74022-0004-86860-0_sp1.1 from training. Duration: 0.9409375
2022-12-22 19:59:20,408 WARNING [train.py:1060] (2/4) Exclude cut with ID 4757-1811-0023-62229-0_sp0.9 from training. Duration: 21.37775
2022-12-22 19:59:20,613 INFO [zipformer.py:660] (2/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=22871.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 19:59:34,091 INFO [zipformer.py:660] (2/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=22880.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 19:59:42,784 WARNING [train.py:1060] (2/4) Exclude cut with ID 1250-135782-0004-25974-0_sp0.9 from training. Duration: 21.17225
2022-12-22 19:59:42,792 WARNING [train.py:1060] (2/4) Exclude cut with ID 7357-94126-0021-15852-0_sp0.9 from training. Duration: 27.511125
2022-12-22 19:59:43,324 INFO [zipformer.py:660] (2/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=22886.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 19:59:45,881 INFO [train.py:894] (2/4) Epoch 7, batch 1850, loss[loss=0.2905, simple_loss=0.3444, pruned_loss=0.1183, over 18479.00 frames. ], tot_loss[loss=0.2423, simple_loss=0.312, pruned_loss=0.0863, over 3712757.95 frames. ], batch size: 54, lr: 1.64e-02, grad_scale: 8.0
2022-12-22 19:59:54,163 INFO [zipformer.py:660] (2/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=22893.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 20:00:15,926 WARNING [train.py:1060] (2/4) Exclude cut with ID 453-131332-0000-47844-0 from training. Duration: 22.8
2022-12-22 20:00:20,667 WARNING [train.py:1060] (2/4) Exclude cut with ID 4964-30587-0040-44509-0 from training. Duration: 22.585
2022-12-22 20:00:34,572 INFO [zipformer.py:660] (2/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=22919.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 20:00:35,823 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 3.355e+02 5.680e+02 7.296e+02 9.676e+02 2.623e+03, threshold=1.459e+03, percent-clipped=2.0
2022-12-22 20:00:51,922 WARNING [train.py:1060] (2/4) Exclude cut with ID 6978-92210-0001-146967-0_sp0.9 from training. Duration: 22.0166875
2022-12-22 20:01:02,713 INFO [train.py:894] (2/4) Epoch 7, batch 1900, loss[loss=0.2341, simple_loss=0.3087, pruned_loss=0.07977, over 18719.00 frames. ], tot_loss[loss=0.2454, simple_loss=0.3134, pruned_loss=0.08876, over 3712732.72 frames. ], batch size: 52, lr: 1.64e-02, grad_scale: 8.0
2022-12-22 20:01:07,088 INFO [zipformer.py:660] (2/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=22941.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 20:01:08,565 WARNING [train.py:1060] (2/4) Exclude cut with ID 7255-291500-0003-134302-0_sp1.1 from training. Duration: 24.395375
2022-12-22 20:01:14,429 INFO [scaling.py:679] (2/4) Whitening: num_groups=8, num_channels=96, metric=1.41 vs. limit=2.0
2022-12-22 20:01:16,815 WARNING [train.py:1060] (2/4) Exclude cut with ID 3488-85273-0017-41203-0_sp0.9 from training. Duration: 27.47775
2022-12-22 20:01:19,722 WARNING [train.py:1060] (2/4) Exclude cut with ID 432-122774-0017-62487-0_sp0.9 from training. Duration: 24.8833125
2022-12-22 20:01:22,693 WARNING [train.py:1060] (2/4) Exclude cut with ID 6758-72288-0033-108368-0 from training. Duration: 23.39
2022-12-22 20:01:29,491 WARNING [train.py:1060] (2/4) Exclude cut with ID 8291-282929-0007-12994-0_sp0.9 from training. Duration: 28.72225
2022-12-22 20:01:38,092 WARNING [train.py:1060] (2/4) Exclude cut with ID 585-294811-0110-133686-0_sp0.9 from training. Duration: 20.8944375
2022-12-22 20:01:52,333 WARNING [train.py:1060] (2/4) Exclude cut with ID 5796-66357-0007-116447-0_sp0.9 from training. Duration: 23.8444375
2022-12-22 20:02:17,663 WARNING [train.py:1060] (2/4) Exclude cut with ID 8291-282929-0007-12994-0 from training. Duration: 25.85
2022-12-22 20:02:17,669 WARNING [train.py:1060] (2/4) Exclude cut with ID 8291-282929-0023-13010-0 from training. Duration: 21.39
2022-12-22 20:02:18,902 INFO [train.py:894] (2/4) Epoch 7, batch 1950, loss[loss=0.3248, simple_loss=0.3772, pruned_loss=0.1363, over 18469.00 frames. ], tot_loss[loss=0.2485, simple_loss=0.3153, pruned_loss=0.09086, over 3712896.57 frames. ], batch size: 64, lr: 1.64e-02, grad_scale: 8.0
2022-12-22 20:02:28,357 WARNING [train.py:1060] (2/4) Exclude cut with ID 7357-94126-0014-15845-0 from training. Duration: 27.92
2022-12-22 20:02:31,560 INFO [zipformer.py:660] (2/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=22995.0, num_to_drop=1, layers_to_drop={2}
2022-12-22 20:02:58,718 WARNING [train.py:1060] (2/4) Exclude cut with ID 8631-249866-0039-130165-0_sp0.9 from training. Duration: 20.661125
2022-12-22 20:03:08,545 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 4.141e+02 5.951e+02 7.381e+02 9.078e+02 1.762e+03, threshold=1.476e+03, percent-clipped=3.0
2022-12-22 20:03:22,456 WARNING [train.py:1060] (2/4) Exclude cut with ID 7357-94126-0043-15874-0_sp0.9 from training. Duration: 20.07225
2022-12-22 20:03:31,671 WARNING [train.py:1060] (2/4) Exclude cut with ID 1085-156170-0017-128270-0 from training. Duration: 21.01
2022-12-22 20:03:35,795 INFO [train.py:894] (2/4) Epoch 7, batch 2000, loss[loss=0.2164, simple_loss=0.2924, pruned_loss=0.07017, over 18728.00 frames. ], tot_loss[loss=0.25, simple_loss=0.3161, pruned_loss=0.09198, over 3712872.50 frames. ], batch size: 54, lr: 1.63e-02, grad_scale: 8.0
2022-12-22 20:03:41,260 INFO [zipformer.py:660] (2/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=23041.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 20:04:38,378 WARNING [train.py:1060] (2/4) Exclude cut with ID 2195-150901-0045-59933-0 from training. Duration: 20.65
2022-12-22 20:04:48,225 WARNING [train.py:1060] (2/4) Exclude cut with ID 5796-66357-0007-116447-0 from training. Duration: 21.46
2022-12-22 20:04:52,428 INFO [train.py:894] (2/4) Epoch 7, batch 2050, loss[loss=0.2416, simple_loss=0.3126, pruned_loss=0.08529, over 18450.00 frames. ], tot_loss[loss=0.2501, simple_loss=0.3157, pruned_loss=0.09225, over 3712645.90 frames. ], batch size: 50, lr: 1.63e-02, grad_scale: 8.0
2022-12-22 20:04:54,128 INFO [zipformer.py:660] (2/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=23089.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 20:05:24,247 INFO [zipformer.py:660] (2/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=23108.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 20:05:34,033 WARNING [train.py:1060] (2/4) Exclude cut with ID 3557-8342-0013-54691-0 from training. Duration: 0.92
2022-12-22 20:05:39,992 WARNING [train.py:1060] (2/4) Exclude cut with ID 8291-282929-0023-13010-0_sp0.9 from training. Duration: 23.7666875
2022-12-22 20:05:41,554 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 3.719e+02 5.993e+02 7.128e+02 8.967e+02 1.938e+03, threshold=1.426e+03, percent-clipped=4.0
2022-12-22 20:06:08,562 INFO [train.py:894] (2/4) Epoch 7, batch 2100, loss[loss=0.2751, simple_loss=0.3213, pruned_loss=0.1144, over 18541.00 frames. ], tot_loss[loss=0.2498, simple_loss=0.3156, pruned_loss=0.09197, over 3714257.44 frames. ], batch size: 44, lr: 1.63e-02, grad_scale: 8.0
2022-12-22 20:06:19,721 WARNING [train.py:1060] (2/4) Exclude cut with ID 8544-281189-0060-101339-0_sp0.9 from training. Duration: 20.861125
2022-12-22 20:06:27,304 INFO [zipformer.py:660] (2/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=23150.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 20:06:28,906 WARNING [train.py:1060] (2/4) Exclude cut with ID 3488-65654-0031-41259-0_sp0.9 from training. Duration: 22.711125
2022-12-22 20:06:55,960 INFO [zipformer.py:660] (2/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=23169.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 20:07:11,682 WARNING [train.py:1060] (2/4) Exclude cut with ID 6951-79737-0043-132310-0_sp1.1 from training. Duration: 22.986375
2022-12-22 20:07:14,733 INFO [zipformer.py:660] (2/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=23181.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 20:07:25,736 INFO [train.py:894] (2/4) Epoch 7, batch 2150, loss[loss=0.286, simple_loss=0.3338, pruned_loss=0.1191, over 18659.00 frames. ], tot_loss[loss=0.2497, simple_loss=0.3151, pruned_loss=0.09215, over 3713620.30 frames. ], batch size: 178, lr: 1.63e-02, grad_scale: 16.0
2022-12-22 20:07:27,272 WARNING [train.py:1060] (2/4) Exclude cut with ID 8040-260924-0003-80960-0_sp0.9 from training. Duration: 22.07225
2022-12-22 20:07:31,864 WARNING [train.py:1060] (2/4) Exclude cut with ID 7699-105389-0045-26330-0_sp0.9 from training. Duration: 20.3055625
2022-12-22 20:07:33,621 WARNING [train.py:1060] (2/4) Exclude cut with ID 6356-271890-0060-94317-0_sp0.9 from training. Duration: 20.72225
2022-12-22 20:07:52,349 WARNING [train.py:1060] (2/4) Exclude cut with ID 3488-85273-0017-41203-0_sp1.1 from training. Duration: 22.4818125
2022-12-22 20:08:10,190 INFO [zipformer.py:660] (2/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=23217.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 20:08:13,985 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 4.135e+02 6.077e+02 7.648e+02 9.193e+02 2.501e+03, threshold=1.530e+03, percent-clipped=6.0
2022-12-22 20:08:17,453 WARNING [train.py:1060] (2/4) Exclude cut with ID 4964-30587-0040-44509-0_sp0.9 from training. Duration: 25.0944375
2022-12-22 20:08:21,955 WARNING [train.py:1060] (2/4) Exclude cut with ID 6533-399-0047-104881-0 from training. Duration: 21.515
2022-12-22 20:08:27,932 INFO [zipformer.py:660] (2/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=23229.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 20:08:29,007 WARNING [train.py:1060] (2/4) Exclude cut with ID 7357-94126-0009-15840-0_sp0.9 from training. Duration: 27.02225
2022-12-22 20:08:33,992 WARNING [train.py:1060] (2/4) Exclude cut with ID 432-122774-0010-62480-0_sp0.9 from training. Duration: 22.22225
2022-12-22 20:08:40,965 INFO [train.py:894] (2/4) Epoch 7, batch 2200, loss[loss=0.2309, simple_loss=0.3005, pruned_loss=0.08058, over 18720.00 frames. ], tot_loss[loss=0.2499, simple_loss=0.3154, pruned_loss=0.0922, over 3713950.00 frames. ], batch size: 52, lr: 1.63e-02, grad_scale: 16.0
2022-12-22 20:08:42,548 WARNING [train.py:1060] (2/4) Exclude cut with ID 4964-30587-0085-44554-0_sp0.9 from training. Duration: 20.85
2022-12-22 20:09:14,823 WARNING [train.py:1060] (2/4) Exclude cut with ID 4295-39940-0007-92567-0 from training. Duration: 21.54
2022-12-22 20:09:20,685 WARNING [train.py:1060] (2/4) Exclude cut with ID 4964-30587-0040-44509-0_sp1.1 from training. Duration: 20.5318125
2022-12-22 20:09:22,738 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([2.1670, 1.8994, 1.5787, 2.3936, 1.9793, 4.5721, 2.0167, 2.0678],
       device='cuda:2'), covar=tensor([0.0853, 0.1723, 0.1191, 0.0933, 0.1476, 0.0187, 0.1256, 0.1478],
       device='cuda:2'), in_proj_covar=tensor([0.0078, 0.0086, 0.0081, 0.0082, 0.0098, 0.0075, 0.0088, 0.0080],
       device='cuda:2'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0003, 0.0004, 0.0003, 0.0004, 0.0003],
       device='cuda:2')
2022-12-22 20:09:30,764 WARNING [train.py:1060] (2/4) Exclude cut with ID 7255-291500-0012-134311-0_sp0.9 from training. Duration: 21.9333125
2022-12-22 20:09:43,445 INFO [zipformer.py:660] (2/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=23278.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 20:09:44,869 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.6564, 1.6023, 1.0813, 1.5600, 1.7337, 1.5062, 2.1131, 1.7157],
       device='cuda:2'), covar=tensor([0.0942, 0.1398, 0.2464, 0.1644, 0.1703, 0.0879, 0.0945, 0.1097],
       device='cuda:2'), in_proj_covar=tensor([0.0173, 0.0186, 0.0224, 0.0277, 0.0222, 0.0178, 0.0196, 0.0188],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-22 20:09:48,387 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([0.7140, 1.8795, 2.1359, 0.7133, 1.5575, 2.4276, 1.7799, 1.8263],
       device='cuda:2'), covar=tensor([0.0591, 0.0292, 0.0246, 0.0384, 0.0309, 0.0261, 0.0245, 0.0470],
       device='cuda:2'), in_proj_covar=tensor([0.0129, 0.0146, 0.0101, 0.0120, 0.0131, 0.0108, 0.0131, 0.0127],
       device='cuda:2'), out_proj_covar=tensor([1.1638e-04, 1.3458e-04, 9.2959e-05, 1.0879e-04, 1.2015e-04, 9.9720e-05,
        1.2248e-04, 1.1720e-04], device='cuda:2')
2022-12-22 20:09:58,311 INFO [train.py:894] (2/4) Epoch 7, batch 2250, loss[loss=0.2223, simple_loss=0.2932, pruned_loss=0.0757, over 18463.00 frames. ], tot_loss[loss=0.2503, simple_loss=0.3158, pruned_loss=0.09244, over 3713588.20 frames. ], batch size: 50, lr: 1.63e-02, grad_scale: 16.0
2022-12-22 20:10:01,550 INFO [zipformer.py:660] (2/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=23290.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 20:10:09,200 INFO [zipformer.py:660] (2/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=23295.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 20:10:18,244 WARNING [train.py:1060] (2/4) Exclude cut with ID 8631-249866-0025-130151-0_sp0.9 from training. Duration: 21.7944375
2022-12-22 20:10:30,213 WARNING [train.py:1060] (2/4) Exclude cut with ID 8291-282929-0002-12989-0_sp0.9 from training. Duration: 22.4666875
2022-12-22 20:10:37,736 WARNING [train.py:1060] (2/4) Exclude cut with ID 6121-9014-0076-24124-0 from training. Duration: 21.635
2022-12-22 20:10:43,609 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.0310, 2.9261, 3.2959, 0.3512, 2.5755, 3.5678, 2.0302, 3.0757],
       device='cuda:2'), covar=tensor([0.0778, 0.0312, 0.0244, 0.0526, 0.0424, 0.0203, 0.0400, 0.0449],
       device='cuda:2'), in_proj_covar=tensor([0.0128, 0.0145, 0.0102, 0.0120, 0.0131, 0.0107, 0.0131, 0.0128],
       device='cuda:2'), out_proj_covar=tensor([1.1553e-04, 1.3344e-04, 9.3114e-05, 1.0846e-04, 1.1953e-04, 9.8745e-05,
        1.2189e-04, 1.1773e-04], device='cuda:2')
2022-12-22 20:10:46,142 WARNING [train.py:1060] (2/4) Exclude cut with ID 6121-9014-0076-24124-0_sp0.9 from training. Duration: 24.038875
2022-12-22 20:10:47,625 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 3.847e+02 6.548e+02 7.944e+02 9.504e+02 1.304e+03, threshold=1.589e+03, percent-clipped=0.0
2022-12-22 20:10:48,164 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([2.0337, 2.4296, 1.1693, 2.7895, 2.7607, 2.0390, 3.6715, 2.2334],
       device='cuda:2'), covar=tensor([0.0781, 0.1437, 0.2401, 0.1731, 0.1374, 0.0795, 0.0728, 0.0944],
       device='cuda:2'), in_proj_covar=tensor([0.0174, 0.0188, 0.0226, 0.0277, 0.0224, 0.0179, 0.0198, 0.0189],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-22 20:11:00,111 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.2426, 1.1687, 1.7753, 1.1685, 1.3358, 1.4727, 1.1320, 1.7593],
       device='cuda:2'), covar=tensor([0.0949, 0.1610, 0.0873, 0.1131, 0.0763, 0.0837, 0.2076, 0.0430],
       device='cuda:2'), in_proj_covar=tensor([0.0201, 0.0189, 0.0197, 0.0185, 0.0180, 0.0203, 0.0197, 0.0172],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-22 20:11:14,340 INFO [train.py:894] (2/4) Epoch 7, batch 2300, loss[loss=0.2435, simple_loss=0.3148, pruned_loss=0.08612, over 18511.00 frames. ], tot_loss[loss=0.2484, simple_loss=0.3143, pruned_loss=0.09127, over 3714123.20 frames. ], batch size: 52, lr: 1.62e-02, grad_scale: 16.0
2022-12-22 20:11:22,210 INFO [zipformer.py:660] (2/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=23343.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 20:11:27,067 WARNING [train.py:1060] (2/4) Exclude cut with ID 7255-291500-0009-134308-0_sp1.1 from training. Duration: 21.786375
2022-12-22 20:11:37,277 WARNING [train.py:1060] (2/4) Exclude cut with ID 8291-282929-0002-12989-0 from training. Duration: 20.22
2022-12-22 20:12:16,765 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([2.0658, 1.9814, 1.3885, 2.1204, 1.7568, 1.7143, 1.7399, 2.0439],
       device='cuda:2'), covar=tensor([0.1540, 0.2055, 0.1408, 0.1880, 0.2344, 0.0798, 0.2007, 0.0607],
       device='cuda:2'), in_proj_covar=tensor([0.0265, 0.0249, 0.0217, 0.0327, 0.0239, 0.0207, 0.0247, 0.0178],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0003, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-22 20:12:17,056 INFO [scaling.py:679] (2/4) Whitening: num_groups=8, num_channels=96, metric=1.37 vs. limit=2.0
2022-12-22 20:12:31,220 INFO [train.py:894] (2/4) Epoch 7, batch 2350, loss[loss=0.2696, simple_loss=0.3283, pruned_loss=0.1054, over 18678.00 frames. ], tot_loss[loss=0.2497, simple_loss=0.3152, pruned_loss=0.09207, over 3715302.65 frames. ], batch size: 179, lr: 1.62e-02, grad_scale: 16.0
2022-12-22 20:13:20,059 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 4.063e+02 6.275e+02 7.475e+02 9.020e+02 1.417e+03, threshold=1.495e+03, percent-clipped=0.0
2022-12-22 20:13:35,291 INFO [scaling.py:679] (2/4) Whitening: num_groups=8, num_channels=96, metric=1.25 vs. limit=2.0
2022-12-22 20:13:41,849 WARNING [train.py:1060] (2/4) Exclude cut with ID 6951-79737-0043-132310-0 from training. Duration: 25.285
2022-12-22 20:13:46,621 INFO [train.py:894] (2/4) Epoch 7, batch 2400, loss[loss=0.261, simple_loss=0.3319, pruned_loss=0.09502, over 18701.00 frames. ], tot_loss[loss=0.2479, simple_loss=0.3136, pruned_loss=0.09112, over 3714461.77 frames. ], batch size: 54, lr: 1.62e-02, grad_scale: 16.0
2022-12-22 20:14:04,995 INFO [zipformer.py:660] (2/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=23450.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 20:14:27,862 INFO [zipformer.py:660] (2/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=23464.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 20:14:50,107 WARNING [train.py:1060] (2/4) Exclude cut with ID 811-130148-0001-63453-0_sp0.9 from training. Duration: 20.861125
2022-12-22 20:14:53,347 INFO [zipformer.py:660] (2/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=23481.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 20:15:03,523 INFO [train.py:894] (2/4) Epoch 7, batch 2450, loss[loss=0.267, simple_loss=0.337, pruned_loss=0.09853, over 18641.00 frames. ], tot_loss[loss=0.2476, simple_loss=0.3137, pruned_loss=0.0907, over 3714621.40 frames. ], batch size: 62, lr: 1.62e-02, grad_scale: 16.0
2022-12-22 20:15:03,859 INFO [zipformer.py:660] (2/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=23488.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 20:15:10,926 WARNING [train.py:1060] (2/4) Exclude cut with ID 6010-56788-0055-90261-0 from training. Duration: 20.88
2022-12-22 20:15:18,665 INFO [zipformer.py:660] (2/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=23498.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 20:15:43,989 WARNING [train.py:1060] (2/4) Exclude cut with ID 7357-94126-0045-15876-0_sp0.9 from training. Duration: 23.4166875
2022-12-22 20:15:52,506 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 3.340e+02 5.776e+02 7.206e+02 8.413e+02 2.583e+03, threshold=1.441e+03, percent-clipped=4.0
2022-12-22 20:16:05,566 INFO [zipformer.py:660] (2/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=23529.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 20:16:19,235 INFO [train.py:894] (2/4) Epoch 7, batch 2500, loss[loss=0.2616, simple_loss=0.3176, pruned_loss=0.1028, over 18707.00 frames. ], tot_loss[loss=0.2489, simple_loss=0.3146, pruned_loss=0.0916, over 3714705.10 frames. ], batch size: 50, lr: 1.62e-02, grad_scale: 16.0
2022-12-22 20:16:31,867 INFO [scaling.py:679] (2/4) Whitening: num_groups=8, num_channels=96, metric=1.39 vs. limit=2.0
2022-12-22 20:16:35,340 INFO [zipformer.py:660] (2/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=23549.0, num_to_drop=1, layers_to_drop={0}
2022-12-22 20:17:03,489 WARNING [train.py:1060] (2/4) Exclude cut with ID 7255-291500-0006-134305-0 from training. Duration: 21.24
2022-12-22 20:17:03,504 WARNING [train.py:1060] (2/4) Exclude cut with ID 6533-399-0047-104881-0_sp0.9 from training. Duration: 23.9055625
2022-12-22 20:17:12,264 INFO [zipformer.py:660] (2/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=23573.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 20:17:30,609 INFO [zipformer.py:660] (2/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=23585.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 20:17:34,797 INFO [train.py:894] (2/4) Epoch 7, batch 2550, loss[loss=0.2689, simple_loss=0.333, pruned_loss=0.1024, over 18717.00 frames. ], tot_loss[loss=0.2484, simple_loss=0.3138, pruned_loss=0.09149, over 3714461.46 frames. ], batch size: 52, lr: 1.62e-02, grad_scale: 16.0
2022-12-22 20:17:36,230 WARNING [train.py:1060] (2/4) Exclude cut with ID 6758-72288-0033-108368-0_sp0.9 from training. Duration: 25.988875
2022-12-22 20:17:45,005 WARNING [train.py:1060] (2/4) Exclude cut with ID 7255-291500-0001-134300-0_sp0.9 from training. Duration: 20.67225
2022-12-22 20:18:23,692 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 4.388e+02 6.581e+02 8.072e+02 9.783e+02 1.636e+03, threshold=1.614e+03, percent-clipped=3.0
2022-12-22 20:18:33,007 WARNING [train.py:1060] (2/4) Exclude cut with ID 3488-85273-0038-41224-0 from training. Duration: 20.34
2022-12-22 20:18:50,192 INFO [train.py:894] (2/4) Epoch 7, batch 2600, loss[loss=0.2181, simple_loss=0.2847, pruned_loss=0.07571, over 18679.00 frames. ], tot_loss[loss=0.2473, simple_loss=0.3132, pruned_loss=0.09072, over 3714558.47 frames. ], batch size: 46, lr: 1.61e-02, grad_scale: 16.0
2022-12-22 20:19:19,605 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([3.8395, 3.3875, 3.4354, 3.7250, 3.4781, 3.4885, 3.8822, 2.0666],
       device='cuda:2'), covar=tensor([0.0639, 0.0489, 0.0561, 0.0675, 0.1231, 0.0968, 0.0644, 0.3104],
       device='cuda:2'), in_proj_covar=tensor([0.0263, 0.0184, 0.0185, 0.0183, 0.0257, 0.0214, 0.0212, 0.0235],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0001, 0.0001, 0.0001, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-22 20:19:45,528 WARNING [train.py:1060] (2/4) Exclude cut with ID 7357-94126-0026-15857-0_sp0.9 from training. Duration: 25.061125
2022-12-22 20:19:55,362 WARNING [train.py:1060] (2/4) Exclude cut with ID 3033-130750-0096-55598-0 from training. Duration: 0.83
2022-12-22 20:20:06,002 INFO [train.py:894] (2/4) Epoch 7, batch 2650, loss[loss=0.2472, simple_loss=0.3164, pruned_loss=0.08906, over 18624.00 frames. ], tot_loss[loss=0.2474, simple_loss=0.313, pruned_loss=0.09087, over 3714773.49 frames. ], batch size: 96, lr: 1.61e-02, grad_scale: 16.0
2022-12-22 20:20:20,662 WARNING [train.py:1060] (2/4) Exclude cut with ID 3488-85273-0017-41203-0 from training. Duration: 24.73
2022-12-22 20:20:34,400 WARNING [train.py:1060] (2/4) Exclude cut with ID 7255-291500-0009-134308-0 from training. Duration: 23.965
2022-12-22 20:20:40,600 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([2.2948, 1.7528, 1.6432, 1.4692, 2.1426, 2.4333, 2.2922, 1.8138],
       device='cuda:2'), covar=tensor([0.0271, 0.0356, 0.0404, 0.0314, 0.0307, 0.0319, 0.0293, 0.0340],
       device='cuda:2'), in_proj_covar=tensor([0.0079, 0.0112, 0.0132, 0.0120, 0.0101, 0.0096, 0.0083, 0.0112],
       device='cuda:2'), out_proj_covar=tensor([7.2068e-05, 1.0077e-04, 1.2335e-04, 1.0811e-04, 9.4317e-05, 8.4094e-05,
        7.4524e-05, 9.9652e-05], device='cuda:2')
2022-12-22 20:20:41,816 WARNING [train.py:1060] (2/4) Exclude cut with ID 6978-92210-0030-146996-0_sp0.9 from training. Duration: 22.088875
2022-12-22 20:20:51,185 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([2.0040, 1.7580, 2.0212, 1.1443, 2.0032, 2.0538, 1.4004, 2.4493],
       device='cuda:2'), covar=tensor([0.0849, 0.1361, 0.1150, 0.1594, 0.0676, 0.0956, 0.2020, 0.0415],
       device='cuda:2'), in_proj_covar=tensor([0.0199, 0.0188, 0.0197, 0.0189, 0.0181, 0.0204, 0.0200, 0.0174],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-22 20:20:55,653 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 3.767e+02 5.981e+02 7.717e+02 9.257e+02 2.451e+03, threshold=1.543e+03, percent-clipped=2.0
2022-12-22 20:20:58,602 WARNING [train.py:1060] (2/4) Exclude cut with ID 7255-291500-0006-134305-0_sp0.9 from training. Duration: 23.6
2022-12-22 20:21:22,068 INFO [train.py:894] (2/4) Epoch 7, batch 2700, loss[loss=0.2896, simple_loss=0.3531, pruned_loss=0.113, over 18574.00 frames. ], tot_loss[loss=0.2495, simple_loss=0.3148, pruned_loss=0.09213, over 3713969.91 frames. ], batch size: 56, lr: 1.61e-02, grad_scale: 16.0
2022-12-22 20:22:01,278 INFO [zipformer.py:660] (2/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=23764.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 20:22:36,643 WARNING [train.py:1060] (2/4) Exclude cut with ID 8291-282929-0024-13011-0 from training. Duration: 23.795
2022-12-22 20:22:38,055 INFO [train.py:894] (2/4) Epoch 7, batch 2750, loss[loss=0.2064, simple_loss=0.2713, pruned_loss=0.07075, over 18478.00 frames. ], tot_loss[loss=0.2488, simple_loss=0.3142, pruned_loss=0.09174, over 3714087.05 frames. ], batch size: 43, lr: 1.61e-02, grad_scale: 16.0
2022-12-22 20:22:54,977 WARNING [train.py:1060] (2/4) Exclude cut with ID 8631-249866-0030-130156-0_sp1.1 from training. Duration: 21.5409375
2022-12-22 20:22:58,114 WARNING [train.py:1060] (2/4) Exclude cut with ID 6978-92210-0019-146985-0_sp0.9 from training. Duration: 24.97775
2022-12-22 20:23:08,785 WARNING [train.py:1060] (2/4) Exclude cut with ID 1085-156170-0017-128270-0_sp0.9 from training. Duration: 23.3444375
2022-12-22 20:23:15,226 INFO [zipformer.py:660] (2/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=23812.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 20:23:20,247 INFO [scaling.py:679] (2/4) Whitening: num_groups=8, num_channels=192, metric=1.97 vs. limit=2.0
2022-12-22 20:23:26,611 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 3.804e+02 6.137e+02 7.582e+02 9.579e+02 1.896e+03, threshold=1.516e+03, percent-clipped=2.0
2022-12-22 20:23:28,429 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([4.7568, 4.0930, 4.0918, 4.6180, 4.2317, 4.2447, 4.8154, 1.3561],
       device='cuda:2'), covar=tensor([0.0509, 0.0508, 0.0526, 0.0513, 0.1229, 0.0727, 0.0343, 0.4444],
       device='cuda:2'), in_proj_covar=tensor([0.0266, 0.0184, 0.0185, 0.0184, 0.0258, 0.0214, 0.0211, 0.0238],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0001, 0.0001, 0.0001, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-22 20:23:35,180 WARNING [train.py:1060] (2/4) Exclude cut with ID 6010-56788-0055-90261-0_sp0.9 from training. Duration: 23.2
2022-12-22 20:23:41,525 WARNING [train.py:1060] (2/4) Exclude cut with ID 5653-46179-0060-117930-0_sp0.9 from training. Duration: 21.17225
2022-12-22 20:23:56,092 INFO [train.py:894] (2/4) Epoch 7, batch 2800, loss[loss=0.2387, simple_loss=0.3072, pruned_loss=0.08506, over 18685.00 frames. ], tot_loss[loss=0.2467, simple_loss=0.3123, pruned_loss=0.09058, over 3714245.27 frames. ], batch size: 50, lr: 1.61e-02, grad_scale: 16.0
2022-12-22 20:24:02,323 WARNING [train.py:1060] (2/4) Exclude cut with ID 7255-291500-0008-134307-0_sp0.9 from training. Duration: 24.6555625
2022-12-22 20:24:05,372 INFO [zipformer.py:660] (2/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=23844.0, num_to_drop=1, layers_to_drop={1}
2022-12-22 20:24:10,490 INFO [scaling.py:679] (2/4) Whitening: num_groups=8, num_channels=96, metric=1.23 vs. limit=2.0
2022-12-22 20:24:41,284 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([2.4990, 2.6001, 1.9998, 1.2324, 3.2060, 2.7351, 2.1725, 1.7304],
       device='cuda:2'), covar=tensor([0.0329, 0.0275, 0.0505, 0.0723, 0.0101, 0.0262, 0.0509, 0.0789],
       device='cuda:2'), in_proj_covar=tensor([0.0118, 0.0111, 0.0129, 0.0119, 0.0079, 0.0112, 0.0134, 0.0147],
       device='cuda:2'), out_proj_covar=tensor([1.4669e-04, 1.3889e-04, 1.5753e-04, 1.4596e-04, 9.9343e-05, 1.3583e-04,
        1.6440e-04, 1.8086e-04], device='cuda:2')
2022-12-22 20:24:49,019 INFO [zipformer.py:660] (2/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=23873.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 20:24:58,919 WARNING [train.py:1060] (2/4) Exclude cut with ID 3488-65654-0031-41259-0 from training. Duration: 20.44
2022-12-22 20:25:08,238 INFO [zipformer.py:660] (2/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=23885.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 20:25:12,097 INFO [train.py:894] (2/4) Epoch 7, batch 2850, loss[loss=0.2245, simple_loss=0.2968, pruned_loss=0.07612, over 18606.00 frames. ], tot_loss[loss=0.2468, simple_loss=0.313, pruned_loss=0.09029, over 3714529.79 frames. ], batch size: 69, lr: 1.61e-02, grad_scale: 16.0
2022-12-22 20:25:13,803 WARNING [train.py:1060] (2/4) Exclude cut with ID 6951-79737-0018-132285-0_sp0.9 from training. Duration: 23.45
2022-12-22 20:25:21,445 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.3137, 1.7059, 1.2470, 2.0615, 2.1865, 1.4271, 1.5109, 1.1498],
       device='cuda:2'), covar=tensor([0.1977, 0.1697, 0.1608, 0.0836, 0.1142, 0.1124, 0.1519, 0.1452],
       device='cuda:2'), in_proj_covar=tensor([0.0229, 0.0200, 0.0195, 0.0179, 0.0243, 0.0183, 0.0202, 0.0184],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-22 20:25:29,414 INFO [scaling.py:679] (2/4) Whitening: num_groups=1, num_channels=384, metric=4.89 vs. limit=5.0
2022-12-22 20:25:44,882 WARNING [train.py:1060] (2/4) Exclude cut with ID 6945-60535-0076-12784-0_sp0.9 from training. Duration: 20.52225
2022-12-22 20:25:52,489 WARNING [train.py:1060] (2/4) Exclude cut with ID 7255-291500-0008-134307-0 from training. Duration: 22.19
2022-12-22 20:26:00,488 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 4.104e+02 6.657e+02 8.527e+02 1.024e+03 2.094e+03, threshold=1.705e+03, percent-clipped=3.0
2022-12-22 20:26:02,357 INFO [zipformer.py:660] (2/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=23921.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 20:26:03,609 WARNING [train.py:1060] (2/4) Exclude cut with ID 7357-94126-0014-15845-0_sp1.1 from training. Duration: 25.3818125
2022-12-22 20:26:19,270 WARNING [train.py:1060] (2/4) Exclude cut with ID 6951-79737-0043-132310-0_sp0.9 from training. Duration: 28.0944375
2022-12-22 20:26:19,967 INFO [scaling.py:679] (2/4) Whitening: num_groups=8, num_channels=192, metric=1.83 vs. limit=2.0
2022-12-22 20:26:20,838 INFO [zipformer.py:660] (2/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=23933.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 20:26:23,886 WARNING [train.py:1060] (2/4) Exclude cut with ID 2195-150901-0045-59933-0_sp0.9 from training. Duration: 22.9444375
2022-12-22 20:26:25,627 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([2.4248, 1.2307, 0.6226, 1.8022, 2.3085, 1.5406, 1.6803, 2.1164],
       device='cuda:2'), covar=tensor([0.1932, 0.3051, 0.3332, 0.2040, 0.2002, 0.2134, 0.1957, 0.2215],
       device='cuda:2'), in_proj_covar=tensor([0.0090, 0.0102, 0.0122, 0.0099, 0.0111, 0.0092, 0.0098, 0.0095],
       device='cuda:2'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003],
       device='cuda:2')
2022-12-22 20:26:28,193 INFO [train.py:894] (2/4) Epoch 7, batch 2900, loss[loss=0.219, simple_loss=0.2907, pruned_loss=0.07359, over 18704.00 frames. ], tot_loss[loss=0.2479, simple_loss=0.3135, pruned_loss=0.09111, over 3714987.76 frames. ], batch size: 50, lr: 1.60e-02, grad_scale: 16.0
2022-12-22 20:26:31,801 WARNING [train.py:1060] (2/4) Exclude cut with ID 8291-282929-0024-13011-0_sp1.1 from training. Duration: 21.6318125
2022-12-22 20:26:49,759 WARNING [train.py:1060] (2/4) Exclude cut with ID 8631-249866-0030-130156-0 from training. Duration: 23.695
2022-12-22 20:27:15,063 WARNING [train.py:1060] (2/4) Exclude cut with ID 7699-105389-0094-26379-0 from training. Duration: 23.955
2022-12-22 20:27:44,170 INFO [train.py:894] (2/4) Epoch 7, batch 2950, loss[loss=0.2392, simple_loss=0.3103, pruned_loss=0.08399, over 18530.00 frames. ], tot_loss[loss=0.2482, simple_loss=0.3134, pruned_loss=0.09144, over 3714593.07 frames. ], batch size: 55, lr: 1.60e-02, grad_scale: 16.0
2022-12-22 20:27:50,180 WARNING [train.py:1060] (2/4) Exclude cut with ID 8291-282929-0024-13011-0_sp0.9 from training. Duration: 26.438875
2022-12-22 20:28:36,256 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 3.403e+02 6.654e+02 7.554e+02 9.176e+02 1.909e+03, threshold=1.511e+03, percent-clipped=1.0
2022-12-22 20:28:40,038 WARNING [train.py:1060] (2/4) Exclude cut with ID 7699-105389-0021-26306-0_sp0.9 from training. Duration: 21.2444375
2022-12-22 20:28:41,425 WARNING [train.py:1060] (2/4) Exclude cut with ID 7357-94126-0014-15845-0_sp0.9 from training. Duration: 31.02225
2022-12-22 20:28:49,973 WARNING [train.py:1060] (2/4) Exclude cut with ID 432-122774-0017-62487-0 from training. Duration: 22.395
2022-12-22 20:29:02,746 INFO [zipformer.py:660] (2/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=24037.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 20:29:03,976 INFO [train.py:894] (2/4) Epoch 7, batch 3000, loss[loss=0.2863, simple_loss=0.3423, pruned_loss=0.1152, over 18665.00 frames. ], tot_loss[loss=0.2467, simple_loss=0.3122, pruned_loss=0.09065, over 3714447.58 frames. ], batch size: 179, lr: 1.60e-02, grad_scale: 16.0
2022-12-22 20:29:03,976 INFO [train.py:919] (2/4) Computing validation loss
2022-12-22 20:29:14,915 INFO [train.py:928] (2/4) Epoch 7, validation: loss=0.187, simple_loss=0.2863, pruned_loss=0.04389, over 944034.00 frames. 
2022-12-22 20:29:14,916 INFO [train.py:929] (2/4) Maximum memory allocated so far is 24680MB
2022-12-22 20:29:16,406 WARNING [train.py:1060] (2/4) Exclude cut with ID 7357-94126-0045-15876-0 from training. Duration: 21.075
2022-12-22 20:29:21,296 WARNING [train.py:1060] (2/4) Exclude cut with ID 6482-98857-0025-147532-0_sp0.9 from training. Duration: 20.0055625
2022-12-22 20:29:22,901 WARNING [train.py:1060] (2/4) Exclude cut with ID 6951-79737-0037-132304-0_sp0.9 from training. Duration: 22.05
2022-12-22 20:29:22,914 WARNING [train.py:1060] (2/4) Exclude cut with ID 7255-291500-0003-134302-0 from training. Duration: 26.8349375
2022-12-22 20:29:26,000 WARNING [train.py:1060] (2/4) Exclude cut with ID 7357-94126-0009-15840-0_sp1.1 from training. Duration: 22.1090625
2022-12-22 20:29:27,982 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.7781, 0.6711, 1.5420, 1.4399, 1.7285, 1.5967, 1.4189, 1.2429],
       device='cuda:2'), covar=tensor([0.1243, 0.1946, 0.1550, 0.1452, 0.0982, 0.0625, 0.1454, 0.0781],
       device='cuda:2'), in_proj_covar=tensor([0.0233, 0.0267, 0.0239, 0.0264, 0.0246, 0.0217, 0.0266, 0.0210],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-22 20:29:33,693 WARNING [train.py:1060] (2/4) Exclude cut with ID 7699-105389-0094-26379-0_sp0.9 from training. Duration: 26.6166875
2022-12-22 20:29:53,027 WARNING [train.py:1060] (2/4) Exclude cut with ID 2046-178027-0000-53705-0_sp0.9 from training. Duration: 20.3055625
2022-12-22 20:29:53,664 INFO [scaling.py:679] (2/4) Whitening: num_groups=1, num_channels=384, metric=3.11 vs. limit=5.0
2022-12-22 20:30:15,401 WARNING [train.py:1060] (2/4) Exclude cut with ID 7205-50138-0008-5373-0_sp0.9 from training. Duration: 20.7
2022-12-22 20:30:18,012 INFO [scaling.py:679] (2/4) Whitening: num_groups=8, num_channels=96, metric=1.27 vs. limit=2.0
2022-12-22 20:30:30,102 INFO [train.py:894] (2/4) Epoch 7, batch 3050, loss[loss=0.2743, simple_loss=0.3283, pruned_loss=0.1102, over 18587.00 frames. ], tot_loss[loss=0.2456, simple_loss=0.3116, pruned_loss=0.08984, over 3714224.67 frames. ], batch size: 49, lr: 1.60e-02, grad_scale: 16.0
2022-12-22 20:30:46,496 INFO [zipformer.py:660] (2/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=24098.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 20:30:57,192 WARNING [train.py:1060] (2/4) Exclude cut with ID 6978-92210-0019-146985-0 from training. Duration: 22.48
2022-12-22 20:31:13,329 WARNING [train.py:1060] (2/4) Exclude cut with ID 7255-291500-0003-134302-0_sp0.9 from training. Duration: 29.816625
2022-12-22 20:31:19,113 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 4.185e+02 6.628e+02 8.234e+02 1.025e+03 2.088e+03, threshold=1.647e+03, percent-clipped=1.0
2022-12-22 20:31:34,766 WARNING [train.py:1060] (2/4) Exclude cut with ID 7255-291500-0005-134304-0_sp1.1 from training. Duration: 22.7590625
2022-12-22 20:31:39,758 WARNING [train.py:1060] (2/4) Exclude cut with ID 7357-94126-0026-15857-0 from training. Duration: 22.555
2022-12-22 20:31:46,789 INFO [train.py:894] (2/4) Epoch 7, batch 3100, loss[loss=0.2456, simple_loss=0.3103, pruned_loss=0.09049, over 18558.00 frames. ], tot_loss[loss=0.2461, simple_loss=0.3121, pruned_loss=0.08998, over 3714779.12 frames. ], batch size: 77, lr: 1.60e-02, grad_scale: 16.0
2022-12-22 20:31:56,179 INFO [zipformer.py:660] (2/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=24144.0, num_to_drop=1, layers_to_drop={2}
2022-12-22 20:32:00,126 WARNING [train.py:1060] (2/4) Exclude cut with ID 1250-135782-0005-25975-0_sp0.9 from training. Duration: 21.688875
2022-12-22 20:32:34,337 WARNING [train.py:1060] (2/4) Exclude cut with ID 3488-85273-0038-41224-0_sp0.9 from training. Duration: 22.6
2022-12-22 20:33:02,557 INFO [train.py:894] (2/4) Epoch 7, batch 3150, loss[loss=0.2335, simple_loss=0.3144, pruned_loss=0.07635, over 18693.00 frames. ], tot_loss[loss=0.2467, simple_loss=0.3126, pruned_loss=0.09042, over 3714601.38 frames. ], batch size: 62, lr: 1.60e-02, grad_scale: 8.0
2022-12-22 20:33:09,075 INFO [zipformer.py:660] (2/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=24192.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 20:33:13,830 WARNING [train.py:1060] (2/4) Exclude cut with ID 7357-94126-0009-15840-0 from training. Duration: 24.32
2022-12-22 20:33:53,925 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 4.219e+02 6.067e+02 7.224e+02 8.656e+02 2.684e+03, threshold=1.445e+03, percent-clipped=3.0
2022-12-22 20:34:14,208 WARNING [train.py:1060] (2/4) Exclude cut with ID 8291-276745-0093-13116-0_sp0.9 from training. Duration: 21.061125
2022-12-22 20:34:20,624 INFO [train.py:894] (2/4) Epoch 7, batch 3200, loss[loss=0.2125, simple_loss=0.2783, pruned_loss=0.07337, over 18453.00 frames. ], tot_loss[loss=0.2445, simple_loss=0.311, pruned_loss=0.089, over 3714982.46 frames. ], batch size: 42, lr: 1.60e-02, grad_scale: 8.0
2022-12-22 20:34:28,804 WARNING [train.py:1060] (2/4) Exclude cut with ID 7357-94126-0024-15855-0_sp0.9 from training. Duration: 20.32225
2022-12-22 20:34:42,376 WARNING [train.py:1060] (2/4) Exclude cut with ID 3033-130750-0096-55598-0_sp1.1 from training. Duration: 0.7545625
2022-12-22 20:34:57,941 WARNING [train.py:1060] (2/4) Exclude cut with ID 4295-39940-0007-92567-0_sp0.9 from training. Duration: 23.9333125
2022-12-22 20:35:15,268 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([4.3877, 3.7920, 3.8540, 4.2713, 3.9749, 3.9754, 4.5044, 1.8069],
       device='cuda:2'), covar=tensor([0.0636, 0.0535, 0.0561, 0.0700, 0.1196, 0.0817, 0.0434, 0.3754],
       device='cuda:2'), in_proj_covar=tensor([0.0268, 0.0188, 0.0190, 0.0191, 0.0261, 0.0220, 0.0217, 0.0240],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0001, 0.0001, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-22 20:35:27,739 WARNING [train.py:1060] (2/4) Exclude cut with ID 7255-291500-0008-134307-0_sp1.1 from training. Duration: 20.17275
2022-12-22 20:35:33,827 WARNING [train.py:1060] (2/4) Exclude cut with ID 6978-92210-0019-146985-0_sp1.1 from training. Duration: 20.436375
2022-12-22 20:35:39,170 INFO [train.py:894] (2/4) Epoch 7, batch 3250, loss[loss=0.2532, simple_loss=0.3132, pruned_loss=0.09665, over 18428.00 frames. ], tot_loss[loss=0.2439, simple_loss=0.3109, pruned_loss=0.08849, over 3714737.95 frames. ], batch size: 48, lr: 1.59e-02, grad_scale: 8.0
2022-12-22 20:35:40,898 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.5639, 1.0129, 1.5803, 2.5218, 1.7407, 2.2263, 1.0752, 1.6632],
       device='cuda:2'), covar=tensor([0.1754, 0.2034, 0.1625, 0.0744, 0.1434, 0.1129, 0.2171, 0.1496],
       device='cuda:2'), in_proj_covar=tensor([0.0104, 0.0116, 0.0126, 0.0114, 0.0106, 0.0131, 0.0132, 0.0110],
       device='cuda:2'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003],
       device='cuda:2')
2022-12-22 20:36:27,318 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 4.060e+02 5.677e+02 7.047e+02 8.775e+02 1.756e+03, threshold=1.409e+03, percent-clipped=1.0
2022-12-22 20:36:53,152 INFO [train.py:894] (2/4) Epoch 7, batch 3300, loss[loss=0.2166, simple_loss=0.2804, pruned_loss=0.07638, over 18546.00 frames. ], tot_loss[loss=0.2433, simple_loss=0.3103, pruned_loss=0.08818, over 3714090.52 frames. ], batch size: 44, lr: 1.59e-02, grad_scale: 8.0
2022-12-22 20:36:56,256 WARNING [train.py:1060] (2/4) Exclude cut with ID 4234-40345-0022-142709-0_sp0.9 from training. Duration: 23.1055625
2022-12-22 20:36:57,605 WARNING [train.py:1060] (2/4) Exclude cut with ID 8291-282929-0007-12994-0_sp1.1 from training. Duration: 23.5
2022-12-22 20:37:05,666 INFO [scaling.py:679] (2/4) Whitening: num_groups=1, num_channels=384, metric=4.82 vs. limit=5.0
2022-12-22 20:37:08,020 WARNING [train.py:1060] (2/4) Exclude cut with ID 7255-291500-0009-134308-0_sp0.9 from training. Duration: 26.62775
2022-12-22 20:37:22,383 WARNING [train.py:1060] (2/4) Exclude cut with ID 6951-79737-0018-132285-0 from training. Duration: 21.105
2022-12-22 20:37:26,430 WARNING [train.py:1060] (2/4) Exclude cut with ID 4511-76322-0006-80011-0_sp0.9 from training. Duration: 24.411125
2022-12-22 20:37:48,738 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.7712, 1.1510, 0.6473, 1.3502, 2.0322, 1.2486, 1.4439, 1.8385],
       device='cuda:2'), covar=tensor([0.1803, 0.2402, 0.2954, 0.1673, 0.1988, 0.1771, 0.1602, 0.1582],
       device='cuda:2'), in_proj_covar=tensor([0.0093, 0.0106, 0.0125, 0.0101, 0.0114, 0.0095, 0.0100, 0.0098],
       device='cuda:2'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003],
       device='cuda:2')
2022-12-22 20:37:52,868 WARNING [train.py:1060] (2/4) Exclude cut with ID 6758-72288-0033-108368-0_sp1.1 from training. Duration: 21.263625
2022-12-22 20:38:09,180 INFO [train.py:894] (2/4) Epoch 7, batch 3350, loss[loss=0.2289, simple_loss=0.296, pruned_loss=0.08091, over 18696.00 frames. ], tot_loss[loss=0.2443, simple_loss=0.311, pruned_loss=0.08882, over 3714650.74 frames. ], batch size: 46, lr: 1.59e-02, grad_scale: 8.0
2022-12-22 20:38:15,993 INFO [zipformer.py:660] (2/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=24393.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 20:38:23,431 WARNING [train.py:1060] (2/4) Exclude cut with ID 4234-40345-0022-142709-0 from training. Duration: 20.795
2022-12-22 20:38:35,046 WARNING [train.py:1060] (2/4) Exclude cut with ID 7357-94126-0021-15852-0 from training. Duration: 24.76
2022-12-22 20:38:35,061 WARNING [train.py:1060] (2/4) Exclude cut with ID 3867-173237-0077-144769-0_sp0.9 from training. Duration: 22.25
2022-12-22 20:39:00,409 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 3.754e+02 6.178e+02 8.136e+02 9.455e+02 2.840e+03, threshold=1.627e+03, percent-clipped=7.0
2022-12-22 20:39:00,495 WARNING [train.py:1060] (2/4) Exclude cut with ID 7357-94126-0026-15857-0_sp1.1 from training. Duration: 20.5045625
2022-12-22 20:39:25,575 INFO [train.py:894] (2/4) Epoch 7, batch 3400, loss[loss=0.2524, simple_loss=0.3231, pruned_loss=0.09084, over 18710.00 frames. ], tot_loss[loss=0.2445, simple_loss=0.3109, pruned_loss=0.08901, over 3714691.78 frames. ], batch size: 65, lr: 1.59e-02, grad_scale: 8.0
2022-12-22 20:40:37,525 INFO [train.py:894] (2/4) Epoch 7, batch 3450, loss[loss=0.2576, simple_loss=0.3261, pruned_loss=0.0945, over 18684.00 frames. ], tot_loss[loss=0.244, simple_loss=0.3109, pruned_loss=0.08852, over 3715730.76 frames. ], batch size: 60, lr: 1.59e-02, grad_scale: 8.0
2022-12-22 20:40:41,984 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([2.1495, 1.2663, 1.7097, 1.8605, 2.0228, 1.8234, 1.8243, 1.3925],
       device='cuda:2'), covar=tensor([0.1230, 0.2016, 0.1575, 0.1540, 0.0975, 0.0643, 0.1581, 0.0791],
       device='cuda:2'), in_proj_covar=tensor([0.0230, 0.0267, 0.0239, 0.0265, 0.0249, 0.0216, 0.0265, 0.0210],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-22 20:41:24,513 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 4.034e+02 6.206e+02 8.000e+02 1.008e+03 2.086e+03, threshold=1.600e+03, percent-clipped=2.0
2022-12-22 20:41:50,849 INFO [train.py:894] (2/4) Epoch 7, batch 3500, loss[loss=0.263, simple_loss=0.3324, pruned_loss=0.0968, over 18636.00 frames. ], tot_loss[loss=0.2445, simple_loss=0.3116, pruned_loss=0.08866, over 3714912.40 frames. ], batch size: 169, lr: 1.59e-02, grad_scale: 8.0
2022-12-22 20:42:11,777 WARNING [train.py:1060] (2/4) Exclude cut with ID 7859-102521-0017-7548-0_sp1.1 from training. Duration: 22.2954375
2022-12-22 20:42:24,853 INFO [train.py:894] (2/4) Epoch 8, batch 0, loss[loss=0.2283, simple_loss=0.3232, pruned_loss=0.06669, over 18717.00 frames. ], tot_loss[loss=0.2283, simple_loss=0.3232, pruned_loss=0.06669, over 18717.00 frames. ], batch size: 54, lr: 1.49e-02, grad_scale: 8.0
2022-12-22 20:42:24,853 INFO [train.py:919] (2/4) Computing validation loss
2022-12-22 20:42:35,896 INFO [train.py:928] (2/4) Epoch 8, validation: loss=0.1867, simple_loss=0.286, pruned_loss=0.04371, over 944034.00 frames. 
2022-12-22 20:42:35,897 INFO [train.py:929] (2/4) Maximum memory allocated so far is 24680MB
2022-12-22 20:42:40,706 INFO [zipformer.py:660] (2/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=24547.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 20:43:13,705 INFO [zipformer.py:660] (2/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=24569.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 20:43:28,795 WARNING [train.py:1060] (2/4) Exclude cut with ID 298-126791-0067-24026-0_sp0.9 from training. Duration: 21.438875
2022-12-22 20:43:34,836 WARNING [train.py:1060] (2/4) Exclude cut with ID 5652-39938-0025-23684-0_sp0.9 from training. Duration: 22.2055625
2022-12-22 20:43:51,691 INFO [train.py:894] (2/4) Epoch 8, batch 50, loss[loss=0.1984, simple_loss=0.2742, pruned_loss=0.06134, over 18426.00 frames. ], tot_loss[loss=0.2238, simple_loss=0.3036, pruned_loss=0.072, over 836827.97 frames. ], batch size: 48, lr: 1.49e-02, grad_scale: 8.0
2022-12-22 20:44:14,288 INFO [zipformer.py:660] (2/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=24608.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 20:44:33,406 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 2.554e+02 4.593e+02 5.489e+02 6.811e+02 1.722e+03, threshold=1.098e+03, percent-clipped=1.0
2022-12-22 20:44:46,934 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([2.5975, 3.8665, 3.9346, 1.8897, 4.0816, 3.1070, 0.8560, 2.7889],
       device='cuda:2'), covar=tensor([0.2006, 0.0804, 0.1307, 0.3322, 0.0671, 0.0909, 0.5187, 0.1533],
       device='cuda:2'), in_proj_covar=tensor([0.0133, 0.0115, 0.0151, 0.0121, 0.0116, 0.0104, 0.0142, 0.0112],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-22 20:44:47,120 INFO [zipformer.py:660] (2/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=24630.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 20:45:07,743 INFO [train.py:894] (2/4) Epoch 8, batch 100, loss[loss=0.2389, simple_loss=0.3138, pruned_loss=0.08194, over 18679.00 frames. ], tot_loss[loss=0.2199, simple_loss=0.2997, pruned_loss=0.07004, over 1475050.56 frames. ], batch size: 60, lr: 1.49e-02, grad_scale: 8.0
2022-12-22 20:46:21,672 INFO [zipformer.py:660] (2/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=24693.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 20:46:22,852 INFO [train.py:894] (2/4) Epoch 8, batch 150, loss[loss=0.2419, simple_loss=0.3133, pruned_loss=0.0853, over 18576.00 frames. ], tot_loss[loss=0.2208, simple_loss=0.3001, pruned_loss=0.07071, over 1971350.61 frames. ], batch size: 57, lr: 1.49e-02, grad_scale: 8.0
2022-12-22 20:46:33,789 WARNING [train.py:1060] (2/4) Exclude cut with ID 7859-102521-0017-7548-0 from training. Duration: 24.525
2022-12-22 20:47:04,372 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 2.865e+02 4.829e+02 5.972e+02 7.031e+02 2.033e+03, threshold=1.194e+03, percent-clipped=7.0
2022-12-22 20:47:05,990 WARNING [train.py:1060] (2/4) Exclude cut with ID 3699-47246-0007-3408-0_sp0.9 from training. Duration: 20.26675
2022-12-22 20:47:16,916 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([0.5163, 1.7404, 2.0786, 1.1018, 1.3447, 2.3531, 1.9063, 1.7123],
       device='cuda:2'), covar=tensor([0.0654, 0.0295, 0.0256, 0.0324, 0.0372, 0.0245, 0.0230, 0.0461],
       device='cuda:2'), in_proj_covar=tensor([0.0127, 0.0141, 0.0099, 0.0121, 0.0129, 0.0107, 0.0130, 0.0125],
       device='cuda:2'), out_proj_covar=tensor([1.1304e-04, 1.2898e-04, 8.8126e-05, 1.0764e-04, 1.1561e-04, 9.7561e-05,
        1.1967e-04, 1.1324e-04], device='cuda:2')
2022-12-22 20:47:19,105 WARNING [train.py:1060] (2/4) Exclude cut with ID 7859-102521-0017-7548-0_sp0.9 from training. Duration: 27.25
2022-12-22 20:47:35,597 INFO [zipformer.py:660] (2/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=24741.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 20:47:39,681 INFO [train.py:894] (2/4) Epoch 8, batch 200, loss[loss=0.2269, simple_loss=0.316, pruned_loss=0.06888, over 18697.00 frames. ], tot_loss[loss=0.217, simple_loss=0.2966, pruned_loss=0.06873, over 2357663.23 frames. ], batch size: 98, lr: 1.49e-02, grad_scale: 8.0
2022-12-22 20:48:12,281 INFO [zipformer.py:660] (2/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=24765.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 20:48:33,139 WARNING [train.py:1060] (2/4) Exclude cut with ID 6426-64292-0017-15984-0 from training. Duration: 21.68
2022-12-22 20:48:41,847 INFO [scaling.py:679] (2/4) Whitening: num_groups=8, num_channels=96, metric=1.97 vs. limit=2.0
2022-12-22 20:48:44,278 WARNING [train.py:1060] (2/4) Exclude cut with ID 4278-13270-0007-59342-0 from training. Duration: 21.6300625
2022-12-22 20:48:56,688 INFO [train.py:894] (2/4) Epoch 8, batch 250, loss[loss=0.224, simple_loss=0.3075, pruned_loss=0.07026, over 18697.00 frames. ], tot_loss[loss=0.2156, simple_loss=0.2957, pruned_loss=0.06778, over 2658804.11 frames. ], batch size: 65, lr: 1.49e-02, grad_scale: 8.0
2022-12-22 20:49:07,439 INFO [scaling.py:679] (2/4) Whitening: num_groups=8, num_channels=192, metric=1.94 vs. limit=2.0
2022-12-22 20:49:09,705 WARNING [train.py:1060] (2/4) Exclude cut with ID 4278-13270-0007-59342-0_sp0.9 from training. Duration: 24.033375
2022-12-22 20:49:37,318 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 3.500e+02 4.749e+02 6.003e+02 7.011e+02 1.781e+03, threshold=1.201e+03, percent-clipped=2.0
2022-12-22 20:49:45,344 INFO [zipformer.py:660] (2/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=24826.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 20:50:06,373 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([0.9929, 2.4473, 3.2585, 0.9344, 2.6601, 3.7599, 2.4259, 2.8952],
       device='cuda:2'), covar=tensor([0.0741, 0.0375, 0.0207, 0.0431, 0.0336, 0.0136, 0.0302, 0.0489],
       device='cuda:2'), in_proj_covar=tensor([0.0125, 0.0141, 0.0099, 0.0120, 0.0128, 0.0106, 0.0129, 0.0125],
       device='cuda:2'), out_proj_covar=tensor([1.1100e-04, 1.2844e-04, 8.8549e-05, 1.0619e-04, 1.1440e-04, 9.6704e-05,
        1.1849e-04, 1.1319e-04], device='cuda:2')
2022-12-22 20:50:08,815 WARNING [train.py:1060] (2/4) Exclude cut with ID 4278-13270-0009-59344-0 from training. Duration: 22.905
2022-12-22 20:50:10,098 WARNING [train.py:1060] (2/4) Exclude cut with ID 5622-44585-0006-90525-0_sp1.1 from training. Duration: 23.4318125
2022-12-22 20:50:13,373 INFO [train.py:894] (2/4) Epoch 8, batch 300, loss[loss=0.2119, simple_loss=0.294, pruned_loss=0.06489, over 18538.00 frames. ], tot_loss[loss=0.2143, simple_loss=0.2943, pruned_loss=0.06714, over 2891772.43 frames. ], batch size: 55, lr: 1.49e-02, grad_scale: 8.0
2022-12-22 20:50:19,945 INFO [zipformer.py:660] (2/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=24848.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 20:51:29,560 INFO [train.py:894] (2/4) Epoch 8, batch 350, loss[loss=0.1952, simple_loss=0.2688, pruned_loss=0.06082, over 18531.00 frames. ], tot_loss[loss=0.2157, simple_loss=0.2948, pruned_loss=0.06833, over 3073507.32 frames. ], batch size: 44, lr: 1.48e-02, grad_scale: 8.0
2022-12-22 20:51:42,475 INFO [zipformer.py:660] (2/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=24903.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 20:51:51,437 INFO [zipformer.py:660] (2/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=24909.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 20:52:06,330 WARNING [train.py:1060] (2/4) Exclude cut with ID 4278-13270-0009-59344-0_sp1.1 from training. Duration: 20.82275
2022-12-22 20:52:07,657 WARNING [train.py:1060] (2/4) Exclude cut with ID 4278-13270-0009-59344-0_sp0.9 from training. Duration: 25.45
2022-12-22 20:52:08,981 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 2.882e+02 4.524e+02 5.830e+02 7.124e+02 1.429e+03, threshold=1.166e+03, percent-clipped=4.0
2022-12-22 20:52:15,222 INFO [zipformer.py:660] (2/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=24925.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 20:52:45,004 INFO [train.py:894] (2/4) Epoch 8, batch 400, loss[loss=0.2196, simple_loss=0.3029, pruned_loss=0.06818, over 18730.00 frames. ], tot_loss[loss=0.2179, simple_loss=0.297, pruned_loss=0.06938, over 3216434.27 frames. ], batch size: 52, lr: 1.48e-02, grad_scale: 8.0
2022-12-22 20:53:06,830 WARNING [train.py:1060] (2/4) Exclude cut with ID 5622-44585-0006-90525-0 from training. Duration: 25.775
2022-12-22 20:53:29,571 WARNING [train.py:1060] (2/4) Exclude cut with ID 2929-85685-0071-62375-0_sp0.9 from training. Duration: 22.25
2022-12-22 20:53:59,113 WARNING [train.py:1060] (2/4) Exclude cut with ID 3972-170212-0014-23379-0 from training. Duration: 26.205
2022-12-22 20:54:00,509 INFO [train.py:894] (2/4) Epoch 8, batch 450, loss[loss=0.2494, simple_loss=0.3225, pruned_loss=0.08811, over 18582.00 frames. ], tot_loss[loss=0.2178, simple_loss=0.2964, pruned_loss=0.06962, over 3325611.40 frames. ], batch size: 57, lr: 1.48e-02, grad_scale: 8.0
2022-12-22 20:54:02,284 INFO [zipformer.py:660] (2/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=24995.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 20:54:16,641 WARNING [train.py:1060] (2/4) Exclude cut with ID 5239-32139-0047-9341-0_sp0.9 from training. Duration: 30.1555625
2022-12-22 20:54:21,187 WARNING [train.py:1060] (2/4) Exclude cut with ID 1265-135635-0050-6781-0_sp0.9 from training. Duration: 21.8333125
2022-12-22 20:54:29,434 WARNING [train.py:1060] (2/4) Exclude cut with ID 1914-133440-0024-94914-0_sp1.1 from training. Duration: 20.6545625
2022-12-22 20:54:41,152 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 3.289e+02 4.744e+02 5.823e+02 7.043e+02 1.775e+03, threshold=1.165e+03, percent-clipped=2.0
2022-12-22 20:55:11,438 WARNING [train.py:1060] (2/4) Exclude cut with ID 7395-89880-0045-39920-0_sp0.9 from training. Duration: 20.52225
2022-12-22 20:55:17,407 INFO [train.py:894] (2/4) Epoch 8, batch 500, loss[loss=0.2066, simple_loss=0.2916, pruned_loss=0.06076, over 18722.00 frames. ], tot_loss[loss=0.2199, simple_loss=0.2984, pruned_loss=0.0707, over 3411977.46 frames. ], batch size: 52, lr: 1.48e-02, grad_scale: 8.0
2022-12-22 20:55:17,828 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([0.3605, 1.5444, 1.9070, 0.6920, 1.1088, 2.2333, 1.8158, 1.5906],
       device='cuda:2'), covar=tensor([0.0616, 0.0348, 0.0346, 0.0368, 0.0386, 0.0264, 0.0206, 0.0495],
       device='cuda:2'), in_proj_covar=tensor([0.0125, 0.0143, 0.0101, 0.0121, 0.0130, 0.0108, 0.0131, 0.0127],
       device='cuda:2'), out_proj_covar=tensor([1.1127e-04, 1.2946e-04, 9.0566e-05, 1.0675e-04, 1.1693e-04, 9.8110e-05,
        1.2039e-04, 1.1565e-04], device='cuda:2')
2022-12-22 20:55:32,368 WARNING [train.py:1060] (2/4) Exclude cut with ID 3972-170212-0014-23379-0_sp0.9 from training. Duration: 29.1166875
2022-12-22 20:55:35,740 INFO [zipformer.py:660] (2/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=25056.0, num_to_drop=1, layers_to_drop={1}
2022-12-22 20:56:31,308 WARNING [train.py:1060] (2/4) Exclude cut with ID 543-133211-0007-59831-0_sp0.9 from training. Duration: 21.388875
2022-12-22 20:56:34,191 INFO [train.py:894] (2/4) Epoch 8, batch 550, loss[loss=0.2174, simple_loss=0.2959, pruned_loss=0.0695, over 18694.00 frames. ], tot_loss[loss=0.2205, simple_loss=0.2991, pruned_loss=0.07099, over 3478990.80 frames. ], batch size: 50, lr: 1.48e-02, grad_scale: 8.0
2022-12-22 20:57:09,733 WARNING [train.py:1060] (2/4) Exclude cut with ID 1914-133440-0024-94914-0 from training. Duration: 22.72
2022-12-22 20:57:11,229 WARNING [train.py:1060] (2/4) Exclude cut with ID 1914-133440-0031-94921-0_sp0.9 from training. Duration: 22.7444375
2022-12-22 20:57:14,378 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 3.133e+02 4.717e+02 6.040e+02 7.385e+02 1.198e+03, threshold=1.208e+03, percent-clipped=2.0
2022-12-22 20:57:14,606 INFO [zipformer.py:660] (2/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=25121.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 20:57:50,254 INFO [train.py:894] (2/4) Epoch 8, batch 600, loss[loss=0.2267, simple_loss=0.3105, pruned_loss=0.07146, over 18640.00 frames. ], tot_loss[loss=0.2207, simple_loss=0.2991, pruned_loss=0.07115, over 3531162.89 frames. ], batch size: 53, lr: 1.48e-02, grad_scale: 8.0
2022-12-22 20:57:54,336 WARNING [train.py:1060] (2/4) Exclude cut with ID 4133-6541-0027-40495-0_sp1.1 from training. Duration: 0.9681875
2022-12-22 20:57:58,882 WARNING [train.py:1060] (2/4) Exclude cut with ID 6330-62851-0022-91297-0_sp0.9 from training. Duration: 22.3166875
2022-12-22 20:58:03,147 WARNING [train.py:1060] (2/4) Exclude cut with ID 543-133212-0015-59917-0_sp0.9 from training. Duration: 21.8166875
2022-12-22 20:58:14,218 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([0.4838, 1.6535, 2.1069, 1.0711, 1.3983, 2.3964, 2.0005, 1.7922],
       device='cuda:2'), covar=tensor([0.0695, 0.0324, 0.0258, 0.0320, 0.0381, 0.0253, 0.0205, 0.0607],
       device='cuda:2'), in_proj_covar=tensor([0.0127, 0.0145, 0.0102, 0.0122, 0.0131, 0.0108, 0.0133, 0.0129],
       device='cuda:2'), out_proj_covar=tensor([1.1335e-04, 1.3117e-04, 9.1193e-05, 1.0777e-04, 1.1715e-04, 9.7942e-05,
        1.2174e-04, 1.1713e-04], device='cuda:2')
2022-12-22 20:58:56,693 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.6094, 1.5373, 1.0418, 1.5348, 1.5870, 1.4801, 2.1856, 1.7256],
       device='cuda:2'), covar=tensor([0.0899, 0.1584, 0.2700, 0.1661, 0.1795, 0.0928, 0.0932, 0.1157],
       device='cuda:2'), in_proj_covar=tensor([0.0170, 0.0188, 0.0230, 0.0273, 0.0222, 0.0177, 0.0200, 0.0190],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-22 20:59:03,009 INFO [scaling.py:679] (2/4) Whitening: num_groups=8, num_channels=96, metric=1.32 vs. limit=2.0
2022-12-22 20:59:04,819 INFO [train.py:894] (2/4) Epoch 8, batch 650, loss[loss=0.2448, simple_loss=0.3137, pruned_loss=0.08794, over 18538.00 frames. ], tot_loss[loss=0.2203, simple_loss=0.2986, pruned_loss=0.07093, over 3571282.06 frames. ], batch size: 47, lr: 1.48e-02, grad_scale: 8.0
2022-12-22 20:59:18,229 INFO [zipformer.py:660] (2/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=25203.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 20:59:19,460 INFO [zipformer.py:660] (2/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=25204.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 20:59:45,820 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 3.325e+02 5.178e+02 6.418e+02 7.876e+02 2.415e+03, threshold=1.284e+03, percent-clipped=3.0
2022-12-22 20:59:48,717 WARNING [train.py:1060] (2/4) Exclude cut with ID 4957-30119-0041-23990-0_sp0.9 from training. Duration: 20.22775
2022-12-22 20:59:51,920 INFO [zipformer.py:660] (2/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=25225.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 21:00:21,378 INFO [train.py:894] (2/4) Epoch 8, batch 700, loss[loss=0.281, simple_loss=0.3451, pruned_loss=0.1085, over 18682.00 frames. ], tot_loss[loss=0.2204, simple_loss=0.2993, pruned_loss=0.07077, over 3603719.96 frames. ], batch size: 181, lr: 1.47e-02, grad_scale: 8.0
2022-12-22 21:00:30,253 WARNING [train.py:1060] (2/4) Exclude cut with ID 5239-32139-0047-9341-0_sp1.1 from training. Duration: 24.67275
2022-12-22 21:00:31,774 INFO [zipformer.py:660] (2/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=25251.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 21:00:36,988 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([0.4086, 1.4066, 1.7381, 0.5804, 1.1040, 1.9115, 1.6580, 1.4954],
       device='cuda:2'), covar=tensor([0.0622, 0.0337, 0.0284, 0.0410, 0.0395, 0.0351, 0.0229, 0.0531],
       device='cuda:2'), in_proj_covar=tensor([0.0127, 0.0144, 0.0102, 0.0122, 0.0129, 0.0109, 0.0133, 0.0127],
       device='cuda:2'), out_proj_covar=tensor([1.1322e-04, 1.2995e-04, 9.0447e-05, 1.0768e-04, 1.1557e-04, 9.8554e-05,
        1.2188e-04, 1.1503e-04], device='cuda:2')
2022-12-22 21:00:58,867 WARNING [train.py:1060] (2/4) Exclude cut with ID 3082-165428-0081-50734-0_sp0.9 from training. Duration: 21.8055625
2022-12-22 21:01:00,919 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.3459, 1.2102, 1.5484, 1.0424, 1.3129, 1.3885, 1.1776, 1.6487],
       device='cuda:2'), covar=tensor([0.0835, 0.1540, 0.0972, 0.1128, 0.0791, 0.0848, 0.2059, 0.0465],
       device='cuda:2'), in_proj_covar=tensor([0.0197, 0.0197, 0.0200, 0.0187, 0.0182, 0.0207, 0.0204, 0.0177],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-22 21:01:05,034 INFO [zipformer.py:660] (2/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=25273.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 21:01:17,621 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.2062, 1.3928, 1.0572, 1.6032, 1.7784, 1.2995, 1.0569, 1.0645],
       device='cuda:2'), covar=tensor([0.2127, 0.1905, 0.1755, 0.1200, 0.1204, 0.1292, 0.1824, 0.1646],
       device='cuda:2'), in_proj_covar=tensor([0.0229, 0.0203, 0.0197, 0.0182, 0.0245, 0.0184, 0.0203, 0.0185],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-22 21:01:36,661 INFO [train.py:894] (2/4) Epoch 8, batch 750, loss[loss=0.1903, simple_loss=0.277, pruned_loss=0.05175, over 18545.00 frames. ], tot_loss[loss=0.2205, simple_loss=0.2994, pruned_loss=0.07082, over 3628400.62 frames. ], batch size: 47, lr: 1.47e-02, grad_scale: 8.0
2022-12-22 21:01:38,035 WARNING [train.py:1060] (2/4) Exclude cut with ID 3340-169293-0054-76830-0_sp0.9 from training. Duration: 22.6666875
2022-12-22 21:02:17,917 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 2.591e+02 5.085e+02 6.172e+02 8.301e+02 1.456e+03, threshold=1.234e+03, percent-clipped=2.0
2022-12-22 21:02:40,162 WARNING [train.py:1060] (2/4) Exclude cut with ID 2411-132532-0017-82279-0_sp1.1 from training. Duration: 0.9681875
2022-12-22 21:02:51,220 INFO [scaling.py:679] (2/4) Whitening: num_groups=8, num_channels=192, metric=1.93 vs. limit=2.0
2022-12-22 21:02:53,245 INFO [train.py:894] (2/4) Epoch 8, batch 800, loss[loss=0.2129, simple_loss=0.301, pruned_loss=0.06238, over 18589.00 frames. ], tot_loss[loss=0.2198, simple_loss=0.2987, pruned_loss=0.07047, over 3647823.46 frames. ], batch size: 57, lr: 1.47e-02, grad_scale: 8.0
2022-12-22 21:03:04,123 INFO [zipformer.py:660] (2/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=25351.0, num_to_drop=1, layers_to_drop={0}
2022-12-22 21:03:05,385 WARNING [train.py:1060] (2/4) Exclude cut with ID 6330-62850-0007-91323-0 from training. Duration: 22.485
2022-12-22 21:03:41,205 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([2.1385, 2.1389, 1.2913, 2.1495, 2.2574, 1.9237, 3.1148, 2.1482],
       device='cuda:2'), covar=tensor([0.0771, 0.1533, 0.2520, 0.1907, 0.1536, 0.0779, 0.0805, 0.1044],
       device='cuda:2'), in_proj_covar=tensor([0.0169, 0.0187, 0.0227, 0.0275, 0.0219, 0.0177, 0.0202, 0.0189],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-22 21:03:43,846 WARNING [train.py:1060] (2/4) Exclude cut with ID 3972-170212-0014-23379-0_sp1.1 from training. Duration: 23.82275
2022-12-22 21:03:56,217 WARNING [train.py:1060] (2/4) Exclude cut with ID 4860-13185-0032-76709-0 from training. Duration: 20.77
2022-12-22 21:04:03,689 WARNING [train.py:1060] (2/4) Exclude cut with ID 6426-64292-0017-15984-0_sp0.9 from training. Duration: 24.088875
2022-12-22 21:04:09,290 INFO [train.py:894] (2/4) Epoch 8, batch 850, loss[loss=0.2064, simple_loss=0.2889, pruned_loss=0.06198, over 18696.00 frames. ], tot_loss[loss=0.2198, simple_loss=0.299, pruned_loss=0.07033, over 3662955.39 frames. ], batch size: 50, lr: 1.47e-02, grad_scale: 8.0
2022-12-22 21:04:17,707 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.7620, 1.8722, 1.7466, 1.9122, 1.7593, 4.8454, 2.3108, 2.8592],
       device='cuda:2'), covar=tensor([0.4208, 0.2625, 0.2426, 0.2417, 0.1293, 0.0131, 0.1432, 0.0915],
       device='cuda:2'), in_proj_covar=tensor([0.0145, 0.0124, 0.0135, 0.0124, 0.0108, 0.0101, 0.0103, 0.0101],
       device='cuda:2'), out_proj_covar=tensor([0.0005, 0.0004, 0.0004, 0.0004, 0.0003, 0.0003, 0.0004, 0.0003],
       device='cuda:2')
2022-12-22 21:04:34,843 WARNING [train.py:1060] (2/4) Exclude cut with ID 6330-62850-0007-91323-0_sp1.1 from training. Duration: 20.4409375
2022-12-22 21:04:42,808 INFO [scaling.py:679] (2/4) Whitening: num_groups=8, num_channels=96, metric=1.40 vs. limit=2.0
2022-12-22 21:04:51,069 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 3.528e+02 4.741e+02 5.831e+02 7.300e+02 1.630e+03, threshold=1.166e+03, percent-clipped=1.0
2022-12-22 21:04:51,421 INFO [zipformer.py:660] (2/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=25421.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 21:05:01,979 INFO [zipformer.py:660] (2/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=25428.0, num_to_drop=1, layers_to_drop={0}
2022-12-22 21:05:02,141 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([2.0496, 2.2563, 1.2138, 2.1601, 2.1142, 1.9141, 2.9717, 2.2081],
       device='cuda:2'), covar=tensor([0.0882, 0.1390, 0.2773, 0.2062, 0.1774, 0.0908, 0.1016, 0.1047],
       device='cuda:2'), in_proj_covar=tensor([0.0172, 0.0189, 0.0229, 0.0277, 0.0221, 0.0180, 0.0204, 0.0192],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-22 21:05:25,259 INFO [train.py:894] (2/4) Epoch 8, batch 900, loss[loss=0.209, simple_loss=0.2802, pruned_loss=0.06895, over 18532.00 frames. ], tot_loss[loss=0.2201, simple_loss=0.2995, pruned_loss=0.07036, over 3674391.42 frames. ], batch size: 47, lr: 1.47e-02, grad_scale: 8.0
2022-12-22 21:05:32,112 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.6638, 1.5885, 1.8114, 1.2490, 1.7684, 1.7742, 1.3560, 2.1517],
       device='cuda:2'), covar=tensor([0.0987, 0.1538, 0.1111, 0.1371, 0.0779, 0.1008, 0.2056, 0.0417],
       device='cuda:2'), in_proj_covar=tensor([0.0195, 0.0193, 0.0195, 0.0185, 0.0180, 0.0202, 0.0199, 0.0173],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-22 21:05:52,637 WARNING [train.py:1060] (2/4) Exclude cut with ID 2929-85685-0062-62366-0_sp0.9 from training. Duration: 22.511125
2022-12-22 21:05:54,122 WARNING [train.py:1060] (2/4) Exclude cut with ID 7395-89880-0031-39906-0 from training. Duration: 20.675
2022-12-22 21:06:03,830 INFO [zipformer.py:660] (2/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=25469.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 21:06:19,321 INFO [scaling.py:679] (2/4) Whitening: num_groups=1, num_channels=384, metric=2.96 vs. limit=5.0
2022-12-22 21:06:34,040 INFO [zipformer.py:660] (2/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=25489.0, num_to_drop=1, layers_to_drop={0}
2022-12-22 21:06:41,152 INFO [train.py:894] (2/4) Epoch 8, batch 950, loss[loss=0.2022, simple_loss=0.2951, pruned_loss=0.05469, over 18513.00 frames. ], tot_loss[loss=0.2194, simple_loss=0.2991, pruned_loss=0.06991, over 3682395.69 frames. ], batch size: 55, lr: 1.47e-02, grad_scale: 8.0
2022-12-22 21:06:56,942 INFO [zipformer.py:660] (2/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=25504.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 21:07:22,403 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 2.658e+02 4.840e+02 5.615e+02 6.973e+02 1.516e+03, threshold=1.123e+03, percent-clipped=4.0
2022-12-22 21:07:31,702 WARNING [train.py:1060] (2/4) Exclude cut with ID 6330-62850-0007-91323-0_sp0.9 from training. Duration: 24.9833125
2022-12-22 21:07:51,305 INFO [scaling.py:679] (2/4) Whitening: num_groups=8, num_channels=96, metric=1.20 vs. limit=2.0
2022-12-22 21:07:56,196 INFO [train.py:894] (2/4) Epoch 8, batch 1000, loss[loss=0.2161, simple_loss=0.3068, pruned_loss=0.06268, over 18720.00 frames. ], tot_loss[loss=0.2192, simple_loss=0.2989, pruned_loss=0.06973, over 3689512.30 frames. ], batch size: 54, lr: 1.47e-02, grad_scale: 8.0
2022-12-22 21:08:04,843 WARNING [train.py:1060] (2/4) Exclude cut with ID 5239-32139-0047-9341-0 from training. Duration: 27.14
2022-12-22 21:08:09,522 INFO [zipformer.py:660] (2/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=25552.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 21:08:19,594 WARNING [train.py:1060] (2/4) Exclude cut with ID 2929-85685-0044-62348-0 from training. Duration: 22.44
2022-12-22 21:08:57,535 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.9111, 1.5518, 1.5708, 1.7138, 1.7182, 1.9918, 2.0264, 1.4582],
       device='cuda:2'), covar=tensor([0.0340, 0.0248, 0.0368, 0.0204, 0.0206, 0.0308, 0.0197, 0.0239],
       device='cuda:2'), in_proj_covar=tensor([0.0082, 0.0111, 0.0134, 0.0120, 0.0104, 0.0098, 0.0083, 0.0109],
       device='cuda:2'), out_proj_covar=tensor([7.3715e-05, 9.8777e-05, 1.2421e-04, 1.0720e-04, 9.5014e-05, 8.4992e-05,
        7.4227e-05, 9.5810e-05], device='cuda:2')
2022-12-22 21:09:11,417 INFO [train.py:894] (2/4) Epoch 8, batch 1050, loss[loss=0.2343, simple_loss=0.3185, pruned_loss=0.0751, over 18589.00 frames. ], tot_loss[loss=0.2183, simple_loss=0.2981, pruned_loss=0.06925, over 3694768.34 frames. ], batch size: 98, lr: 1.46e-02, grad_scale: 8.0
2022-12-22 21:09:38,511 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.3600, 1.6433, 1.2244, 2.0470, 2.1930, 1.4908, 1.2161, 1.1700],
       device='cuda:2'), covar=tensor([0.2010, 0.1690, 0.1650, 0.0942, 0.1207, 0.1156, 0.1893, 0.1515],
       device='cuda:2'), in_proj_covar=tensor([0.0231, 0.0204, 0.0196, 0.0181, 0.0247, 0.0183, 0.0204, 0.0184],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-22 21:09:39,467 WARNING [train.py:1060] (2/4) Exclude cut with ID 2929-85685-0060-62364-0_sp0.9 from training. Duration: 21.361125
2022-12-22 21:09:45,437 WARNING [train.py:1060] (2/4) Exclude cut with ID 2929-85685-0079-62383-0_sp1.1 from training. Duration: 27.0318125
2022-12-22 21:09:52,802 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 2.730e+02 4.743e+02 5.690e+02 7.080e+02 2.001e+03, threshold=1.138e+03, percent-clipped=6.0
2022-12-22 21:09:54,288 WARNING [train.py:1060] (2/4) Exclude cut with ID 5622-44585-0006-90525-0_sp0.9 from training. Duration: 28.638875
2022-12-22 21:10:10,657 WARNING [train.py:1060] (2/4) Exclude cut with ID 3340-169293-0054-76830-0 from training. Duration: 20.4
2022-12-22 21:10:26,940 INFO [train.py:894] (2/4) Epoch 8, batch 1100, loss[loss=0.2181, simple_loss=0.2948, pruned_loss=0.07072, over 18460.00 frames. ], tot_loss[loss=0.2187, simple_loss=0.2982, pruned_loss=0.06963, over 3698770.86 frames. ], batch size: 50, lr: 1.46e-02, grad_scale: 8.0
2022-12-22 21:10:38,532 INFO [zipformer.py:660] (2/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=25651.0, num_to_drop=1, layers_to_drop={2}
2022-12-22 21:10:42,863 WARNING [train.py:1060] (2/4) Exclude cut with ID 2929-85685-0071-62375-0 from training. Duration: 20.025
2022-12-22 21:10:44,161 WARNING [train.py:1060] (2/4) Exclude cut with ID 2364-131735-0112-64612-0_sp0.9 from training. Duration: 20.488875
2022-12-22 21:10:49,212 WARNING [train.py:1060] (2/4) Exclude cut with ID 2929-85685-0079-62383-0 from training. Duration: 29.735
2022-12-22 21:11:22,164 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([2.3556, 2.2241, 1.7164, 1.1209, 2.7026, 2.4629, 1.9150, 1.5250],
       device='cuda:2'), covar=tensor([0.0278, 0.0266, 0.0451, 0.0718, 0.0126, 0.0235, 0.0474, 0.0750],
       device='cuda:2'), in_proj_covar=tensor([0.0121, 0.0110, 0.0128, 0.0122, 0.0081, 0.0115, 0.0135, 0.0150],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0001, 0.0002, 0.0002, 0.0001, 0.0001, 0.0002, 0.0002],
       device='cuda:2')
2022-12-22 21:11:22,436 INFO [scaling.py:679] (2/4) Whitening: num_groups=8, num_channels=96, metric=1.24 vs. limit=2.0
2022-12-22 21:11:29,346 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.6654, 1.8723, 1.3324, 2.0955, 1.8149, 3.4704, 1.5151, 1.5527],
       device='cuda:2'), covar=tensor([0.0950, 0.1568, 0.1265, 0.0905, 0.1318, 0.0241, 0.1326, 0.1580],
       device='cuda:2'), in_proj_covar=tensor([0.0078, 0.0086, 0.0079, 0.0081, 0.0094, 0.0075, 0.0088, 0.0081],
       device='cuda:2'), out_proj_covar=tensor([0.0003, 0.0004, 0.0003, 0.0003, 0.0004, 0.0003, 0.0004, 0.0003],
       device='cuda:2')
2022-12-22 21:11:43,076 INFO [train.py:894] (2/4) Epoch 8, batch 1150, loss[loss=0.1985, simple_loss=0.2837, pruned_loss=0.0566, over 18434.00 frames. ], tot_loss[loss=0.2181, simple_loss=0.2983, pruned_loss=0.06899, over 3702490.19 frames. ], batch size: 50, lr: 1.46e-02, grad_scale: 8.0
2022-12-22 21:11:51,108 INFO [zipformer.py:660] (2/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=25699.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 21:12:02,722 INFO [scaling.py:679] (2/4) Whitening: num_groups=8, num_channels=96, metric=1.24 vs. limit=2.0
2022-12-22 21:12:12,607 WARNING [train.py:1060] (2/4) Exclude cut with ID 7276-92427-0014-12983-0_sp0.9 from training. Duration: 21.3055625
2022-12-22 21:12:14,092 WARNING [train.py:1060] (2/4) Exclude cut with ID 1025-75365-0008-79168-0_sp0.9 from training. Duration: 22.0666875
2022-12-22 21:12:25,034 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 2.727e+02 4.430e+02 5.330e+02 7.922e+02 1.445e+03, threshold=1.066e+03, percent-clipped=6.0
2022-12-22 21:12:43,898 INFO [scaling.py:679] (2/4) Whitening: num_groups=8, num_channels=96, metric=1.67 vs. limit=2.0
2022-12-22 21:12:59,948 INFO [train.py:894] (2/4) Epoch 8, batch 1200, loss[loss=0.2066, simple_loss=0.2807, pruned_loss=0.06623, over 18426.00 frames. ], tot_loss[loss=0.2163, simple_loss=0.2968, pruned_loss=0.06791, over 3704207.59 frames. ], batch size: 42, lr: 1.46e-02, grad_scale: 8.0
2022-12-22 21:13:27,609 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([2.6141, 2.5532, 2.1603, 1.0098, 1.9039, 2.1168, 1.7231, 2.1368],
       device='cuda:2'), covar=tensor([0.0633, 0.0420, 0.1063, 0.1578, 0.1335, 0.1232, 0.1362, 0.0814],
       device='cuda:2'), in_proj_covar=tensor([0.0154, 0.0172, 0.0197, 0.0190, 0.0199, 0.0182, 0.0194, 0.0183],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-22 21:14:01,358 INFO [zipformer.py:660] (2/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=25784.0, num_to_drop=1, layers_to_drop={2}
2022-12-22 21:14:02,546 WARNING [train.py:1060] (2/4) Exclude cut with ID 2929-85685-0062-62366-0 from training. Duration: 20.26
2022-12-22 21:14:15,457 WARNING [train.py:1060] (2/4) Exclude cut with ID 5239-32139-0030-9324-0_sp0.9 from training. Duration: 21.3444375
2022-12-22 21:14:16,946 INFO [train.py:894] (2/4) Epoch 8, batch 1250, loss[loss=0.2091, simple_loss=0.2779, pruned_loss=0.07013, over 18580.00 frames. ], tot_loss[loss=0.2158, simple_loss=0.2962, pruned_loss=0.06771, over 3705520.91 frames. ], batch size: 45, lr: 1.46e-02, grad_scale: 8.0
2022-12-22 21:14:23,778 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.4830, 1.3842, 1.2822, 2.0722, 1.7049, 3.4181, 1.2149, 1.5162],
       device='cuda:2'), covar=tensor([0.1014, 0.1853, 0.1171, 0.0873, 0.1351, 0.0218, 0.1496, 0.1613],
       device='cuda:2'), in_proj_covar=tensor([0.0078, 0.0086, 0.0078, 0.0080, 0.0094, 0.0074, 0.0088, 0.0081],
       device='cuda:2'), out_proj_covar=tensor([0.0003, 0.0004, 0.0003, 0.0003, 0.0004, 0.0003, 0.0004, 0.0003],
       device='cuda:2')
2022-12-22 21:14:58,385 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 2.631e+02 4.769e+02 5.676e+02 7.211e+02 2.091e+03, threshold=1.135e+03, percent-clipped=3.0
2022-12-22 21:15:12,172 WARNING [train.py:1060] (2/4) Exclude cut with ID 497-129325-0061-62254-0_sp1.1 from training. Duration: 0.97725
2022-12-22 21:15:33,884 INFO [train.py:894] (2/4) Epoch 8, batch 1300, loss[loss=0.2182, simple_loss=0.304, pruned_loss=0.06623, over 18638.00 frames. ], tot_loss[loss=0.2164, simple_loss=0.2967, pruned_loss=0.06804, over 3706660.40 frames. ], batch size: 53, lr: 1.46e-02, grad_scale: 8.0
2022-12-22 21:15:54,853 WARNING [train.py:1060] (2/4) Exclude cut with ID 7395-89880-0031-39906-0_sp0.9 from training. Duration: 22.97225
2022-12-22 21:15:55,205 INFO [zipformer.py:660] (2/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=25858.0, num_to_drop=1, layers_to_drop={0}
2022-12-22 21:16:30,499 WARNING [train.py:1060] (2/4) Exclude cut with ID 7395-89880-0047-39922-0_sp0.9 from training. Duration: 21.97775
2022-12-22 21:16:43,461 WARNING [train.py:1060] (2/4) Exclude cut with ID 1112-1043-0006-89194-0_sp0.9 from training. Duration: 21.8333125
2022-12-22 21:16:49,566 INFO [train.py:894] (2/4) Epoch 8, batch 1350, loss[loss=0.2394, simple_loss=0.3062, pruned_loss=0.08631, over 18554.00 frames. ], tot_loss[loss=0.2168, simple_loss=0.2969, pruned_loss=0.06831, over 3707542.19 frames. ], batch size: 49, lr: 1.46e-02, grad_scale: 8.0
2022-12-22 21:16:54,071 WARNING [train.py:1060] (2/4) Exclude cut with ID 1914-133440-0031-94921-0 from training. Duration: 20.47
2022-12-22 21:17:23,646 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([2.8152, 2.7379, 2.0530, 1.9786, 3.1149, 3.1297, 2.7138, 2.2540],
       device='cuda:2'), covar=tensor([0.0251, 0.0233, 0.0466, 0.0487, 0.0113, 0.0180, 0.0338, 0.0477],
       device='cuda:2'), in_proj_covar=tensor([0.0119, 0.0109, 0.0126, 0.0119, 0.0080, 0.0113, 0.0132, 0.0146],
       device='cuda:2'), out_proj_covar=tensor([0.0001, 0.0001, 0.0002, 0.0001, 0.0001, 0.0001, 0.0002, 0.0002],
       device='cuda:2')
2022-12-22 21:17:26,963 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([2.2200, 1.5108, 1.8446, 1.8802, 2.1451, 2.0271, 1.9918, 1.4840],
       device='cuda:2'), covar=tensor([0.1417, 0.2260, 0.1719, 0.1782, 0.1137, 0.0634, 0.1878, 0.0822],
       device='cuda:2'), in_proj_covar=tensor([0.0235, 0.0271, 0.0240, 0.0272, 0.0255, 0.0222, 0.0273, 0.0208],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-22 21:17:28,248 INFO [zipformer.py:660] (2/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=25919.0, num_to_drop=1, layers_to_drop={1}
2022-12-22 21:17:30,638 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 2.888e+02 4.729e+02 5.749e+02 7.465e+02 1.908e+03, threshold=1.150e+03, percent-clipped=5.0
2022-12-22 21:17:45,852 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([2.0697, 1.8891, 2.1052, 1.0939, 2.1626, 2.1170, 1.4459, 2.5797],
       device='cuda:2'), covar=tensor([0.1024, 0.1498, 0.1240, 0.1903, 0.0755, 0.1125, 0.2126, 0.0447],
       device='cuda:2'), in_proj_covar=tensor([0.0201, 0.0197, 0.0198, 0.0188, 0.0183, 0.0209, 0.0206, 0.0176],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-22 21:17:58,193 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.6222, 0.8846, 1.3000, 1.3161, 1.5579, 1.6857, 1.7093, 1.1475],
       device='cuda:2'), covar=tensor([0.0369, 0.0303, 0.0412, 0.0237, 0.0205, 0.0304, 0.0196, 0.0269],
       device='cuda:2'), in_proj_covar=tensor([0.0081, 0.0110, 0.0130, 0.0118, 0.0103, 0.0099, 0.0083, 0.0109],
       device='cuda:2'), out_proj_covar=tensor([7.2421e-05, 9.7626e-05, 1.2031e-04, 1.0556e-04, 9.4581e-05, 8.5112e-05,
        7.4025e-05, 9.5585e-05], device='cuda:2')
2022-12-22 21:18:00,819 WARNING [train.py:1060] (2/4) Exclude cut with ID 7395-89880-0037-39912-0_sp0.9 from training. Duration: 20.67225
2022-12-22 21:18:05,199 INFO [train.py:894] (2/4) Epoch 8, batch 1400, loss[loss=0.1828, simple_loss=0.2693, pruned_loss=0.04811, over 18694.00 frames. ], tot_loss[loss=0.2163, simple_loss=0.2964, pruned_loss=0.06807, over 3710092.21 frames. ], batch size: 46, lr: 1.45e-02, grad_scale: 8.0
2022-12-22 21:18:20,501 WARNING [train.py:1060] (2/4) Exclude cut with ID 1914-133440-0024-94914-0_sp0.9 from training. Duration: 25.2444375
2022-12-22 21:18:45,901 WARNING [train.py:1060] (2/4) Exclude cut with ID 3340-169293-0021-76797-0_sp0.9 from training. Duration: 21.1445
2022-12-22 21:19:03,633 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.1158, 2.5271, 3.0795, 0.8878, 2.5607, 3.5527, 2.2595, 3.0041],
       device='cuda:2'), covar=tensor([0.0666, 0.0312, 0.0233, 0.0411, 0.0353, 0.0174, 0.0330, 0.0378],
       device='cuda:2'), in_proj_covar=tensor([0.0126, 0.0144, 0.0102, 0.0121, 0.0129, 0.0109, 0.0133, 0.0129],
       device='cuda:2'), out_proj_covar=tensor([1.1058e-04, 1.2941e-04, 9.0128e-05, 1.0593e-04, 1.1448e-04, 9.8529e-05,
        1.2043e-04, 1.1580e-04], device='cuda:2')
2022-12-22 21:19:20,508 INFO [train.py:894] (2/4) Epoch 8, batch 1450, loss[loss=0.2066, simple_loss=0.2758, pruned_loss=0.06873, over 18447.00 frames. ], tot_loss[loss=0.2163, simple_loss=0.2966, pruned_loss=0.06795, over 3711208.04 frames. ], batch size: 42, lr: 1.45e-02, grad_scale: 8.0
2022-12-22 21:19:45,164 INFO [scaling.py:679] (2/4) Whitening: num_groups=8, num_channels=192, metric=1.92 vs. limit=2.0
2022-12-22 21:19:52,053 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.7222, 1.4682, 1.2860, 2.0101, 1.6306, 3.2764, 1.4319, 1.7399],
       device='cuda:2'), covar=tensor([0.0831, 0.1652, 0.1217, 0.0870, 0.1409, 0.0232, 0.1317, 0.1408],
       device='cuda:2'), in_proj_covar=tensor([0.0079, 0.0086, 0.0080, 0.0081, 0.0094, 0.0074, 0.0087, 0.0080],
       device='cuda:2'), out_proj_covar=tensor([0.0003, 0.0004, 0.0003, 0.0003, 0.0004, 0.0003, 0.0004, 0.0003],
       device='cuda:2')
2022-12-22 21:20:02,258 WARNING [train.py:1060] (2/4) Exclude cut with ID 2929-85685-0079-62383-0_sp0.9 from training. Duration: 33.038875
2022-12-22 21:20:05,119 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 3.601e+02 4.725e+02 5.598e+02 6.886e+02 1.154e+03, threshold=1.120e+03, percent-clipped=1.0
2022-12-22 21:20:10,060 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.6992, 2.2496, 1.3348, 2.7700, 2.8716, 1.6041, 1.9983, 1.3729],
       device='cuda:2'), covar=tensor([0.2000, 0.1542, 0.1647, 0.0880, 0.1552, 0.1263, 0.1724, 0.1594],
       device='cuda:2'), in_proj_covar=tensor([0.0231, 0.0205, 0.0195, 0.0179, 0.0245, 0.0184, 0.0199, 0.0182],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-22 21:20:39,978 INFO [train.py:894] (2/4) Epoch 8, batch 1500, loss[loss=0.2049, simple_loss=0.2975, pruned_loss=0.05614, over 18533.00 frames. ], tot_loss[loss=0.2171, simple_loss=0.2975, pruned_loss=0.06835, over 3712204.74 frames. ], batch size: 55, lr: 1.45e-02, grad_scale: 8.0
2022-12-22 21:20:41,474 WARNING [train.py:1060] (2/4) Exclude cut with ID 6426-64291-0000-16059-0_sp0.9 from training. Duration: 20.0944375
2022-12-22 21:20:57,742 WARNING [train.py:1060] (2/4) Exclude cut with ID 2929-85685-0044-62348-0_sp1.1 from training. Duration: 20.4
2022-12-22 21:21:03,315 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.9632, 2.2213, 1.2150, 2.4880, 2.1923, 2.0160, 3.1242, 2.1234],
       device='cuda:2'), covar=tensor([0.0754, 0.1405, 0.2405, 0.1563, 0.1473, 0.0723, 0.0735, 0.0897],
       device='cuda:2'), in_proj_covar=tensor([0.0170, 0.0187, 0.0227, 0.0274, 0.0219, 0.0176, 0.0201, 0.0189],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-22 21:21:07,572 WARNING [train.py:1060] (2/4) Exclude cut with ID 6330-62851-0022-91297-0 from training. Duration: 20.085
2022-12-22 21:21:17,829 WARNING [train.py:1060] (2/4) Exclude cut with ID 4860-13185-0032-76709-0_sp0.9 from training. Duration: 23.07775
2022-12-22 21:21:40,647 INFO [zipformer.py:660] (2/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=26084.0, num_to_drop=1, layers_to_drop={1}
2022-12-22 21:21:55,690 INFO [train.py:894] (2/4) Epoch 8, batch 1550, loss[loss=0.2526, simple_loss=0.3161, pruned_loss=0.09456, over 18569.00 frames. ], tot_loss[loss=0.2166, simple_loss=0.2969, pruned_loss=0.06816, over 3712724.98 frames. ], batch size: 49, lr: 1.45e-02, grad_scale: 8.0
2022-12-22 21:22:05,247 WARNING [train.py:1060] (2/4) Exclude cut with ID 2929-85685-0044-62348-0_sp0.9 from training. Duration: 24.9333125
2022-12-22 21:22:08,783 INFO [zipformer.py:660] (2/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=26102.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 21:22:36,911 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 3.030e+02 4.809e+02 6.108e+02 7.519e+02 1.643e+03, threshold=1.222e+03, percent-clipped=3.0
2022-12-22 21:22:51,624 WARNING [train.py:1060] (2/4) Exclude cut with ID 5118-111612-0016-124680-0_sp0.9 from training. Duration: 20.388875
2022-12-22 21:22:53,320 INFO [zipformer.py:660] (2/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=26132.0, num_to_drop=1, layers_to_drop={0}
2022-12-22 21:22:57,950 WARNING [train.py:1060] (2/4) Exclude cut with ID 432-122774-0017-62487-0_sp1.1 from training. Duration: 20.3590625
2022-12-22 21:23:11,907 INFO [train.py:894] (2/4) Epoch 8, batch 1600, loss[loss=0.2191, simple_loss=0.2908, pruned_loss=0.07375, over 18485.00 frames. ], tot_loss[loss=0.2168, simple_loss=0.2971, pruned_loss=0.06825, over 3712850.01 frames. ], batch size: 43, lr: 1.45e-02, grad_scale: 8.0
2022-12-22 21:23:30,128 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([5.6287, 4.8114, 4.8727, 5.5887, 5.0864, 5.0795, 5.6760, 1.4596],
       device='cuda:2'), covar=tensor([0.0565, 0.0476, 0.0418, 0.0600, 0.1178, 0.0805, 0.0350, 0.4597],
       device='cuda:2'), in_proj_covar=tensor([0.0263, 0.0180, 0.0182, 0.0188, 0.0253, 0.0211, 0.0210, 0.0233],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0001, 0.0001, 0.0001, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-22 21:23:41,475 INFO [zipformer.py:660] (2/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=26163.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 21:23:48,689 INFO [zipformer.py:660] (2/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=26168.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 21:24:07,974 WARNING [train.py:1060] (2/4) Exclude cut with ID 3557-8342-0013-54691-0_sp1.1 from training. Duration: 0.836375
2022-12-22 21:24:27,778 INFO [train.py:894] (2/4) Epoch 8, batch 1650, loss[loss=0.2494, simple_loss=0.3244, pruned_loss=0.08725, over 18574.00 frames. ], tot_loss[loss=0.2179, simple_loss=0.2976, pruned_loss=0.06915, over 3712590.88 frames. ], batch size: 57, lr: 1.45e-02, grad_scale: 16.0
2022-12-22 21:24:50,728 WARNING [train.py:1060] (2/4) Exclude cut with ID 8565-290391-0049-67394-0_sp0.9 from training. Duration: 21.3166875
2022-12-22 21:24:53,671 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([2.9553, 5.2235, 4.8933, 2.3336, 5.3849, 3.9822, 0.9132, 3.8238],
       device='cuda:2'), covar=tensor([0.1753, 0.0818, 0.1043, 0.3993, 0.0654, 0.0858, 0.5884, 0.1388],
       device='cuda:2'), in_proj_covar=tensor([0.0130, 0.0111, 0.0144, 0.0117, 0.0116, 0.0102, 0.0140, 0.0107],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-22 21:24:59,607 INFO [zipformer.py:660] (2/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=26214.0, num_to_drop=1, layers_to_drop={1}
2022-12-22 21:25:09,420 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 3.083e+02 5.122e+02 6.330e+02 7.747e+02 1.750e+03, threshold=1.266e+03, percent-clipped=7.0
2022-12-22 21:25:20,483 WARNING [train.py:1060] (2/4) Exclude cut with ID 6533-399-0029-104863-0_sp0.9 from training. Duration: 22.1055625
2022-12-22 21:25:22,364 INFO [zipformer.py:660] (2/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=26229.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 21:25:31,642 WARNING [train.py:1060] (2/4) Exclude cut with ID 7699-105389-0094-26379-0_sp1.1 from training. Duration: 21.77725
2022-12-22 21:25:45,445 INFO [train.py:894] (2/4) Epoch 8, batch 1700, loss[loss=0.2397, simple_loss=0.3062, pruned_loss=0.08662, over 18565.00 frames. ], tot_loss[loss=0.2225, simple_loss=0.3, pruned_loss=0.0725, over 3712876.74 frames. ], batch size: 49, lr: 1.45e-02, grad_scale: 16.0
2022-12-22 21:25:51,223 WARNING [train.py:1060] (2/4) Exclude cut with ID 7255-291500-0005-134304-0_sp0.9 from training. Duration: 27.8166875
2022-12-22 21:25:54,627 INFO [zipformer.py:660] (2/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=26250.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 21:26:14,834 WARNING [train.py:1060] (2/4) Exclude cut with ID 7357-94126-0021-15852-0_sp1.1 from training. Duration: 22.5090625
2022-12-22 21:26:20,997 WARNING [train.py:1060] (2/4) Exclude cut with ID 7255-291500-0005-134304-0 from training. Duration: 25.035
2022-12-22 21:26:26,091 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.7883, 1.7530, 1.3564, 1.8350, 1.6625, 1.6712, 1.5800, 1.8770],
       device='cuda:2'), covar=tensor([0.1545, 0.2107, 0.1446, 0.1817, 0.2293, 0.0783, 0.1918, 0.0583],
       device='cuda:2'), in_proj_covar=tensor([0.0264, 0.0253, 0.0217, 0.0326, 0.0240, 0.0205, 0.0251, 0.0177],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0003, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-22 21:26:41,504 WARNING [train.py:1060] (2/4) Exclude cut with ID 774-127930-0014-10412-0_sp1.1 from training. Duration: 0.95
2022-12-22 21:26:58,085 WARNING [train.py:1060] (2/4) Exclude cut with ID 3033-130750-0096-55598-0_sp0.9 from training. Duration: 0.92225
2022-12-22 21:27:02,525 INFO [train.py:894] (2/4) Epoch 8, batch 1750, loss[loss=0.2231, simple_loss=0.3055, pruned_loss=0.07037, over 18577.00 frames. ], tot_loss[loss=0.2259, simple_loss=0.3015, pruned_loss=0.07514, over 3713357.22 frames. ], batch size: 51, lr: 1.45e-02, grad_scale: 16.0
2022-12-22 21:27:03,220 INFO [scaling.py:679] (2/4) Whitening: num_groups=1, num_channels=384, metric=2.96 vs. limit=5.0
2022-12-22 21:27:23,908 WARNING [train.py:1060] (2/4) Exclude cut with ID 4511-76322-0006-80011-0 from training. Duration: 21.97
2022-12-22 21:27:28,828 INFO [zipformer.py:660] (2/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=26311.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 21:27:42,793 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([2.5908, 1.9271, 0.5738, 2.1096, 2.4710, 1.6694, 2.1400, 2.6163],
       device='cuda:2'), covar=tensor([0.1440, 0.1880, 0.2820, 0.1411, 0.1661, 0.1603, 0.1390, 0.1462],
       device='cuda:2'), in_proj_covar=tensor([0.0090, 0.0102, 0.0122, 0.0097, 0.0111, 0.0092, 0.0097, 0.0097],
       device='cuda:2'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003],
       device='cuda:2')
2022-12-22 21:27:43,954 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 3.748e+02 6.270e+02 7.569e+02 9.269e+02 2.899e+03, threshold=1.514e+03, percent-clipped=6.0
2022-12-22 21:27:43,997 WARNING [train.py:1060] (2/4) Exclude cut with ID 7492-105653-0055-62765-0_sp0.9 from training. Duration: 21.97225
2022-12-22 21:27:44,031 WARNING [train.py:1060] (2/4) Exclude cut with ID 453-131332-0000-47844-0_sp0.9 from training. Duration: 25.3333125
2022-12-22 21:27:55,241 WARNING [train.py:1060] (2/4) Exclude cut with ID 5172-29468-0015-19128-0_sp0.9 from training. Duration: 21.5055625
2022-12-22 21:28:06,078 WARNING [train.py:1060] (2/4) Exclude cut with ID 453-131332-0000-47844-0_sp1.1 from training. Duration: 20.72725
2022-12-22 21:28:18,994 INFO [train.py:894] (2/4) Epoch 8, batch 1800, loss[loss=0.2238, simple_loss=0.3036, pruned_loss=0.07196, over 18540.00 frames. ], tot_loss[loss=0.2307, simple_loss=0.3048, pruned_loss=0.07829, over 3713806.03 frames. ], batch size: 55, lr: 1.44e-02, grad_scale: 16.0
2022-12-22 21:28:37,637 WARNING [train.py:1060] (2/4) Exclude cut with ID 8631-249866-0030-130156-0_sp0.9 from training. Duration: 26.32775
2022-12-22 21:28:37,913 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.3926, 1.4197, 1.0109, 1.5908, 1.6212, 2.9340, 1.1712, 1.4952],
       device='cuda:2'), covar=tensor([0.0904, 0.1789, 0.1223, 0.0917, 0.1366, 0.0291, 0.1466, 0.1466],
       device='cuda:2'), in_proj_covar=tensor([0.0078, 0.0087, 0.0078, 0.0080, 0.0095, 0.0074, 0.0087, 0.0081],
       device='cuda:2'), out_proj_covar=tensor([0.0003, 0.0004, 0.0003, 0.0003, 0.0004, 0.0003, 0.0004, 0.0003],
       device='cuda:2')
2022-12-22 21:29:10,993 WARNING [train.py:1060] (2/4) Exclude cut with ID 3867-173237-0077-144769-0 from training. Duration: 20.025
2022-12-22 21:29:16,151 WARNING [train.py:1060] (2/4) Exclude cut with ID 6709-74022-0004-86860-0_sp1.1 from training. Duration: 0.9409375
2022-12-22 21:29:17,728 WARNING [train.py:1060] (2/4) Exclude cut with ID 4757-1811-0023-62229-0_sp0.9 from training. Duration: 21.37775
2022-12-22 21:29:34,924 INFO [train.py:894] (2/4) Epoch 8, batch 1850, loss[loss=0.2585, simple_loss=0.3319, pruned_loss=0.09252, over 18581.00 frames. ], tot_loss[loss=0.2347, simple_loss=0.307, pruned_loss=0.08115, over 3713912.20 frames. ], batch size: 78, lr: 1.44e-02, grad_scale: 16.0
2022-12-22 21:29:40,021 WARNING [train.py:1060] (2/4) Exclude cut with ID 1250-135782-0004-25974-0_sp0.9 from training. Duration: 21.17225
2022-12-22 21:29:40,033 WARNING [train.py:1060] (2/4) Exclude cut with ID 7357-94126-0021-15852-0_sp0.9 from training. Duration: 27.511125
2022-12-22 21:30:05,990 INFO [zipformer.py:660] (2/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=26414.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 21:30:08,039 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.4809, 1.0833, 1.5273, 2.5659, 1.9031, 2.0696, 0.8193, 1.6792],
       device='cuda:2'), covar=tensor([0.1757, 0.1967, 0.1568, 0.0609, 0.1190, 0.1177, 0.2328, 0.1508],
       device='cuda:2'), in_proj_covar=tensor([0.0103, 0.0112, 0.0126, 0.0115, 0.0103, 0.0128, 0.0128, 0.0107],
       device='cuda:2'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003],
       device='cuda:2')
2022-12-22 21:30:12,249 WARNING [train.py:1060] (2/4) Exclude cut with ID 453-131332-0000-47844-0 from training. Duration: 22.8
2022-12-22 21:30:17,098 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 4.061e+02 5.870e+02 7.240e+02 8.563e+02 1.417e+03, threshold=1.448e+03, percent-clipped=0.0
2022-12-22 21:30:17,118 WARNING [train.py:1060] (2/4) Exclude cut with ID 4964-30587-0040-44509-0 from training. Duration: 22.585
2022-12-22 21:30:47,648 WARNING [train.py:1060] (2/4) Exclude cut with ID 6978-92210-0001-146967-0_sp0.9 from training. Duration: 22.0166875
2022-12-22 21:30:53,355 INFO [train.py:894] (2/4) Epoch 8, batch 1900, loss[loss=0.2182, simple_loss=0.3, pruned_loss=0.06818, over 18729.00 frames. ], tot_loss[loss=0.2361, simple_loss=0.3075, pruned_loss=0.08229, over 3714544.00 frames. ], batch size: 52, lr: 1.44e-02, grad_scale: 16.0
2022-12-22 21:31:04,364 WARNING [train.py:1060] (2/4) Exclude cut with ID 7255-291500-0003-134302-0_sp1.1 from training. Duration: 24.395375
2022-12-22 21:31:11,951 WARNING [train.py:1060] (2/4) Exclude cut with ID 3488-85273-0017-41203-0_sp0.9 from training. Duration: 27.47775
2022-12-22 21:31:14,975 INFO [zipformer.py:660] (2/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=26458.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 21:31:16,234 WARNING [train.py:1060] (2/4) Exclude cut with ID 432-122774-0017-62487-0_sp0.9 from training. Duration: 24.8833125
2022-12-22 21:31:18,899 WARNING [train.py:1060] (2/4) Exclude cut with ID 6758-72288-0033-108368-0 from training. Duration: 23.39
2022-12-22 21:31:25,709 WARNING [train.py:1060] (2/4) Exclude cut with ID 8291-282929-0007-12994-0_sp0.9 from training. Duration: 28.72225
2022-12-22 21:31:36,482 WARNING [train.py:1060] (2/4) Exclude cut with ID 585-294811-0110-133686-0_sp0.9 from training. Duration: 20.8944375
2022-12-22 21:31:41,215 INFO [zipformer.py:660] (2/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=26475.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 21:31:52,726 WARNING [train.py:1060] (2/4) Exclude cut with ID 5796-66357-0007-116447-0_sp0.9 from training. Duration: 23.8444375
2022-12-22 21:32:09,416 INFO [train.py:894] (2/4) Epoch 8, batch 1950, loss[loss=0.261, simple_loss=0.3273, pruned_loss=0.09733, over 18630.00 frames. ], tot_loss[loss=0.2384, simple_loss=0.3092, pruned_loss=0.08382, over 3714918.75 frames. ], batch size: 53, lr: 1.44e-02, grad_scale: 16.0
2022-12-22 21:32:15,721 WARNING [train.py:1060] (2/4) Exclude cut with ID 8291-282929-0007-12994-0 from training. Duration: 25.85
2022-12-22 21:32:15,730 WARNING [train.py:1060] (2/4) Exclude cut with ID 8291-282929-0023-13010-0 from training. Duration: 21.39
2022-12-22 21:32:17,416 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([0.5230, 1.5921, 2.0361, 0.8273, 1.4431, 2.3650, 1.8510, 1.8148],
       device='cuda:2'), covar=tensor([0.0560, 0.0314, 0.0289, 0.0361, 0.0321, 0.0309, 0.0214, 0.0550],
       device='cuda:2'), in_proj_covar=tensor([0.0127, 0.0148, 0.0103, 0.0125, 0.0131, 0.0112, 0.0133, 0.0131],
       device='cuda:2'), out_proj_covar=tensor([1.1128e-04, 1.3229e-04, 9.0857e-05, 1.0920e-04, 1.1526e-04, 9.9820e-05,
        1.2008e-04, 1.1735e-04], device='cuda:2')
2022-12-22 21:32:27,293 WARNING [train.py:1060] (2/4) Exclude cut with ID 7357-94126-0014-15845-0 from training. Duration: 27.92
2022-12-22 21:32:39,782 INFO [zipformer.py:660] (2/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=26514.0, num_to_drop=1, layers_to_drop={2}
2022-12-22 21:32:50,234 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 3.849e+02 6.560e+02 7.949e+02 9.276e+02 2.232e+03, threshold=1.590e+03, percent-clipped=3.0
2022-12-22 21:32:54,894 WARNING [train.py:1060] (2/4) Exclude cut with ID 8631-249866-0039-130165-0_sp0.9 from training. Duration: 20.661125
2022-12-22 21:32:55,006 INFO [zipformer.py:660] (2/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=26524.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 21:33:17,710 WARNING [train.py:1060] (2/4) Exclude cut with ID 7357-94126-0043-15874-0_sp0.9 from training. Duration: 20.07225
2022-12-22 21:33:25,146 INFO [train.py:894] (2/4) Epoch 8, batch 2000, loss[loss=0.2503, simple_loss=0.3218, pruned_loss=0.08943, over 18672.00 frames. ], tot_loss[loss=0.2413, simple_loss=0.3105, pruned_loss=0.08608, over 3714771.65 frames. ], batch size: 60, lr: 1.44e-02, grad_scale: 16.0
2022-12-22 21:33:25,216 WARNING [train.py:1060] (2/4) Exclude cut with ID 1085-156170-0017-128270-0 from training. Duration: 21.01
2022-12-22 21:33:27,120 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([2.7573, 2.3335, 1.8654, 0.9172, 1.9478, 1.8995, 1.3666, 2.0774],
       device='cuda:2'), covar=tensor([0.0541, 0.0565, 0.1256, 0.1729, 0.1356, 0.1522, 0.1719, 0.0836],
       device='cuda:2'), in_proj_covar=tensor([0.0159, 0.0176, 0.0203, 0.0195, 0.0205, 0.0187, 0.0201, 0.0190],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-22 21:33:52,774 INFO [zipformer.py:660] (2/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=26562.0, num_to_drop=1, layers_to_drop={0}
2022-12-22 21:34:36,231 WARNING [train.py:1060] (2/4) Exclude cut with ID 2195-150901-0045-59933-0 from training. Duration: 20.65
2022-12-22 21:34:40,618 INFO [train.py:894] (2/4) Epoch 8, batch 2050, loss[loss=0.2234, simple_loss=0.2892, pruned_loss=0.0788, over 18663.00 frames. ], tot_loss[loss=0.2412, simple_loss=0.3098, pruned_loss=0.08631, over 3714492.31 frames. ], batch size: 48, lr: 1.44e-02, grad_scale: 8.0
2022-12-22 21:34:41,909 WARNING [train.py:1060] (2/4) Exclude cut with ID 5796-66357-0007-116447-0 from training. Duration: 21.46
2022-12-22 21:34:58,856 INFO [zipformer.py:660] (2/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=26606.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 21:35:08,623 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.4423, 0.9867, 1.9055, 3.0561, 2.2172, 2.2380, 0.5638, 2.0201],
       device='cuda:2'), covar=tensor([0.1846, 0.1974, 0.1461, 0.0675, 0.1205, 0.1393, 0.2556, 0.1282],
       device='cuda:2'), in_proj_covar=tensor([0.0104, 0.0115, 0.0126, 0.0118, 0.0105, 0.0131, 0.0130, 0.0108],
       device='cuda:2'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003],
       device='cuda:2')
2022-12-22 21:35:23,326 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 4.180e+02 6.124e+02 7.302e+02 9.759e+02 1.688e+03, threshold=1.460e+03, percent-clipped=2.0
2022-12-22 21:35:28,006 WARNING [train.py:1060] (2/4) Exclude cut with ID 3557-8342-0013-54691-0 from training. Duration: 0.92
2022-12-22 21:35:36,117 WARNING [train.py:1060] (2/4) Exclude cut with ID 8291-282929-0023-13010-0_sp0.9 from training. Duration: 23.7666875
2022-12-22 21:35:56,666 INFO [train.py:894] (2/4) Epoch 8, batch 2100, loss[loss=0.2498, simple_loss=0.3049, pruned_loss=0.09736, over 18493.00 frames. ], tot_loss[loss=0.2412, simple_loss=0.3097, pruned_loss=0.08633, over 3714381.81 frames. ], batch size: 43, lr: 1.44e-02, grad_scale: 8.0
2022-12-22 21:36:13,181 WARNING [train.py:1060] (2/4) Exclude cut with ID 8544-281189-0060-101339-0_sp0.9 from training. Duration: 20.861125
2022-12-22 21:36:23,248 WARNING [train.py:1060] (2/4) Exclude cut with ID 3488-65654-0031-41259-0_sp0.9 from training. Duration: 22.711125
2022-12-22 21:37:05,840 WARNING [train.py:1060] (2/4) Exclude cut with ID 6951-79737-0043-132310-0_sp1.1 from training. Duration: 22.986375
2022-12-22 21:37:11,693 INFO [train.py:894] (2/4) Epoch 8, batch 2150, loss[loss=0.2189, simple_loss=0.2881, pruned_loss=0.07481, over 18385.00 frames. ], tot_loss[loss=0.2423, simple_loss=0.3102, pruned_loss=0.08718, over 3714946.47 frames. ], batch size: 46, lr: 1.43e-02, grad_scale: 8.0
2022-12-22 21:37:22,930 WARNING [train.py:1060] (2/4) Exclude cut with ID 8040-260924-0003-80960-0_sp0.9 from training. Duration: 22.07225
2022-12-22 21:37:27,556 WARNING [train.py:1060] (2/4) Exclude cut with ID 7699-105389-0045-26330-0_sp0.9 from training. Duration: 20.3055625
2022-12-22 21:37:30,329 WARNING [train.py:1060] (2/4) Exclude cut with ID 6356-271890-0060-94317-0_sp0.9 from training. Duration: 20.72225
2022-12-22 21:37:48,913 WARNING [train.py:1060] (2/4) Exclude cut with ID 3488-85273-0017-41203-0_sp1.1 from training. Duration: 22.4818125
2022-12-22 21:37:54,909 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 2.474e+02 5.655e+02 7.658e+02 9.829e+02 2.063e+03, threshold=1.532e+03, percent-clipped=2.0
2022-12-22 21:38:13,631 WARNING [train.py:1060] (2/4) Exclude cut with ID 4964-30587-0040-44509-0_sp0.9 from training. Duration: 25.0944375
2022-12-22 21:38:15,538 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([2.3886, 2.4296, 1.6317, 1.0103, 3.1925, 2.7818, 2.0711, 1.6570],
       device='cuda:2'), covar=tensor([0.0360, 0.0313, 0.0587, 0.0784, 0.0114, 0.0280, 0.0505, 0.0788],
       device='cuda:2'), in_proj_covar=tensor([0.0123, 0.0117, 0.0132, 0.0122, 0.0084, 0.0116, 0.0138, 0.0151],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0001, 0.0002, 0.0002, 0.0001, 0.0001, 0.0002, 0.0002],
       device='cuda:2')
2022-12-22 21:38:18,038 WARNING [train.py:1060] (2/4) Exclude cut with ID 6533-399-0047-104881-0 from training. Duration: 21.515
2022-12-22 21:38:23,869 WARNING [train.py:1060] (2/4) Exclude cut with ID 7357-94126-0009-15840-0_sp0.9 from training. Duration: 27.02225
2022-12-22 21:38:26,593 INFO [train.py:894] (2/4) Epoch 8, batch 2200, loss[loss=0.237, simple_loss=0.3104, pruned_loss=0.08177, over 18533.00 frames. ], tot_loss[loss=0.242, simple_loss=0.3096, pruned_loss=0.08715, over 3715201.14 frames. ], batch size: 77, lr: 1.43e-02, grad_scale: 8.0
2022-12-22 21:38:29,776 WARNING [train.py:1060] (2/4) Exclude cut with ID 432-122774-0010-62480-0_sp0.9 from training. Duration: 22.22225
2022-12-22 21:38:36,290 WARNING [train.py:1060] (2/4) Exclude cut with ID 4964-30587-0085-44554-0_sp0.9 from training. Duration: 20.85
2022-12-22 21:38:49,272 INFO [zipformer.py:660] (2/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=26758.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 21:38:57,612 INFO [scaling.py:679] (2/4) Whitening: num_groups=8, num_channels=192, metric=1.93 vs. limit=2.0
2022-12-22 21:39:07,834 INFO [zipformer.py:660] (2/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=26770.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 21:39:11,979 WARNING [train.py:1060] (2/4) Exclude cut with ID 4295-39940-0007-92567-0 from training. Duration: 21.54
2022-12-22 21:39:16,142 WARNING [train.py:1060] (2/4) Exclude cut with ID 4964-30587-0040-44509-0_sp1.1 from training. Duration: 20.5318125
2022-12-22 21:39:25,887 WARNING [train.py:1060] (2/4) Exclude cut with ID 7255-291500-0012-134311-0_sp0.9 from training. Duration: 21.9333125
2022-12-22 21:39:42,227 INFO [train.py:894] (2/4) Epoch 8, batch 2250, loss[loss=0.2466, simple_loss=0.3201, pruned_loss=0.08655, over 18404.00 frames. ], tot_loss[loss=0.2413, simple_loss=0.3091, pruned_loss=0.08673, over 3715218.98 frames. ], batch size: 53, lr: 1.43e-02, grad_scale: 8.0
2022-12-22 21:39:57,606 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([2.3938, 3.4389, 3.4952, 1.6276, 3.5581, 2.6625, 0.5820, 2.4208],
       device='cuda:2'), covar=tensor([0.2227, 0.1069, 0.1418, 0.3738, 0.1000, 0.1166, 0.5876, 0.1703],
       device='cuda:2'), in_proj_covar=tensor([0.0131, 0.0114, 0.0148, 0.0120, 0.0119, 0.0103, 0.0141, 0.0109],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-22 21:40:02,305 INFO [zipformer.py:660] (2/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=26806.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 21:40:14,692 WARNING [train.py:1060] (2/4) Exclude cut with ID 8631-249866-0025-130151-0_sp0.9 from training. Duration: 21.7944375
2022-12-22 21:40:27,435 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 3.517e+02 5.710e+02 7.345e+02 8.844e+02 3.050e+03, threshold=1.469e+03, percent-clipped=4.0
2022-12-22 21:40:27,468 WARNING [train.py:1060] (2/4) Exclude cut with ID 8291-282929-0002-12989-0_sp0.9 from training. Duration: 22.4666875
2022-12-22 21:40:30,730 INFO [zipformer.py:660] (2/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=26824.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 21:40:35,141 WARNING [train.py:1060] (2/4) Exclude cut with ID 6121-9014-0076-24124-0 from training. Duration: 21.635
2022-12-22 21:40:41,034 WARNING [train.py:1060] (2/4) Exclude cut with ID 6121-9014-0076-24124-0_sp0.9 from training. Duration: 24.038875
2022-12-22 21:41:01,011 INFO [train.py:894] (2/4) Epoch 8, batch 2300, loss[loss=0.2896, simple_loss=0.3486, pruned_loss=0.1152, over 18707.00 frames. ], tot_loss[loss=0.2408, simple_loss=0.3087, pruned_loss=0.0865, over 3715031.26 frames. ], batch size: 60, lr: 1.43e-02, grad_scale: 8.0
2022-12-22 21:41:24,837 WARNING [train.py:1060] (2/4) Exclude cut with ID 7255-291500-0009-134308-0_sp1.1 from training. Duration: 21.786375
2022-12-22 21:41:38,869 WARNING [train.py:1060] (2/4) Exclude cut with ID 8291-282929-0002-12989-0 from training. Duration: 20.22
2022-12-22 21:41:44,735 INFO [zipformer.py:660] (2/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=26872.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 21:42:18,278 INFO [train.py:894] (2/4) Epoch 8, batch 2350, loss[loss=0.2199, simple_loss=0.306, pruned_loss=0.06686, over 18648.00 frames. ], tot_loss[loss=0.2399, simple_loss=0.3083, pruned_loss=0.08573, over 3714209.63 frames. ], batch size: 53, lr: 1.43e-02, grad_scale: 8.0
2022-12-22 21:42:34,261 INFO [zipformer.py:660] (2/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=26904.0, num_to_drop=1, layers_to_drop={0}
2022-12-22 21:42:37,259 INFO [zipformer.py:660] (2/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=26906.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 21:42:38,890 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([2.7177, 2.1346, 2.2094, 1.8603, 2.3772, 3.2481, 2.9548, 2.1806],
       device='cuda:2'), covar=tensor([0.0390, 0.0285, 0.0294, 0.0267, 0.0221, 0.0196, 0.0239, 0.0237],
       device='cuda:2'), in_proj_covar=tensor([0.0079, 0.0114, 0.0135, 0.0121, 0.0105, 0.0102, 0.0087, 0.0112],
       device='cuda:2'), out_proj_covar=tensor([7.0361e-05, 1.0025e-04, 1.2367e-04, 1.0723e-04, 9.5022e-05, 8.6767e-05,
        7.6685e-05, 9.7338e-05], device='cuda:2')
2022-12-22 21:42:51,246 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([0.1278, 1.1932, 1.6838, 0.4447, 0.8103, 1.8239, 1.6001, 1.4040],
       device='cuda:2'), covar=tensor([0.0558, 0.0267, 0.0204, 0.0332, 0.0378, 0.0283, 0.0199, 0.0482],
       device='cuda:2'), in_proj_covar=tensor([0.0131, 0.0151, 0.0106, 0.0127, 0.0135, 0.0114, 0.0138, 0.0137],
       device='cuda:2'), out_proj_covar=tensor([1.1471e-04, 1.3479e-04, 9.3075e-05, 1.1069e-04, 1.1842e-04, 1.0154e-04,
        1.2458e-04, 1.2244e-04], device='cuda:2')
2022-12-22 21:43:00,410 INFO [scaling.py:679] (2/4) Whitening: num_groups=8, num_channels=96, metric=1.43 vs. limit=2.0
2022-12-22 21:43:00,839 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 3.511e+02 5.486e+02 6.533e+02 8.294e+02 1.980e+03, threshold=1.307e+03, percent-clipped=2.0
2022-12-22 21:43:34,064 INFO [train.py:894] (2/4) Epoch 8, batch 2400, loss[loss=0.2297, simple_loss=0.3001, pruned_loss=0.07968, over 18458.00 frames. ], tot_loss[loss=0.239, simple_loss=0.3077, pruned_loss=0.08514, over 3713418.96 frames. ], batch size: 50, lr: 1.43e-02, grad_scale: 8.0
2022-12-22 21:43:39,088 WARNING [train.py:1060] (2/4) Exclude cut with ID 6951-79737-0043-132310-0 from training. Duration: 25.285
2022-12-22 21:43:43,460 INFO [zipformer.py:660] (2/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=26950.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 21:43:43,831 INFO [scaling.py:679] (2/4) Whitening: num_groups=8, num_channels=192, metric=1.78 vs. limit=2.0
2022-12-22 21:43:49,119 INFO [zipformer.py:660] (2/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=26954.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 21:44:06,519 INFO [zipformer.py:660] (2/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=26965.0, num_to_drop=1, layers_to_drop={2}
2022-12-22 21:44:41,625 WARNING [train.py:1060] (2/4) Exclude cut with ID 811-130148-0001-63453-0_sp0.9 from training. Duration: 20.861125
2022-12-22 21:44:50,476 INFO [train.py:894] (2/4) Epoch 8, batch 2450, loss[loss=0.2136, simple_loss=0.2925, pruned_loss=0.06731, over 18574.00 frames. ], tot_loss[loss=0.2389, simple_loss=0.3073, pruned_loss=0.08527, over 3713107.23 frames. ], batch size: 51, lr: 1.43e-02, grad_scale: 8.0
2022-12-22 21:44:56,873 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([2.2807, 1.3070, 2.0406, 4.1800, 2.9672, 2.6688, 0.6399, 2.8733],
       device='cuda:2'), covar=tensor([0.1666, 0.2038, 0.1839, 0.0454, 0.1099, 0.1340, 0.2815, 0.1040],
       device='cuda:2'), in_proj_covar=tensor([0.0104, 0.0114, 0.0126, 0.0118, 0.0104, 0.0128, 0.0130, 0.0106],
       device='cuda:2'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003],
       device='cuda:2')
2022-12-22 21:45:04,475 WARNING [train.py:1060] (2/4) Exclude cut with ID 6010-56788-0055-90261-0 from training. Duration: 20.88
2022-12-22 21:45:17,507 INFO [zipformer.py:660] (2/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=27011.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 21:45:27,102 INFO [scaling.py:679] (2/4) Whitening: num_groups=8, num_channels=192, metric=1.92 vs. limit=2.0
2022-12-22 21:45:33,581 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 3.757e+02 5.947e+02 7.436e+02 9.112e+02 3.064e+03, threshold=1.487e+03, percent-clipped=6.0
2022-12-22 21:45:35,127 WARNING [train.py:1060] (2/4) Exclude cut with ID 7357-94126-0045-15876-0_sp0.9 from training. Duration: 23.4166875
2022-12-22 21:45:39,945 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.9699, 1.4176, 1.5051, 1.6487, 2.0281, 2.0250, 2.1099, 1.3408],
       device='cuda:2'), covar=tensor([0.0238, 0.0238, 0.0382, 0.0207, 0.0152, 0.0246, 0.0208, 0.0252],
       device='cuda:2'), in_proj_covar=tensor([0.0080, 0.0114, 0.0138, 0.0121, 0.0104, 0.0103, 0.0088, 0.0113],
       device='cuda:2'), out_proj_covar=tensor([7.0701e-05, 1.0019e-04, 1.2664e-04, 1.0717e-04, 9.4977e-05, 8.7950e-05,
        7.7335e-05, 9.8031e-05], device='cuda:2')
2022-12-22 21:45:41,355 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.5204, 1.6479, 1.6732, 1.7042, 1.2641, 3.4935, 1.7936, 2.0716],
       device='cuda:2'), covar=tensor([0.3269, 0.1913, 0.1811, 0.1813, 0.1408, 0.0186, 0.1342, 0.0902],
       device='cuda:2'), in_proj_covar=tensor([0.0143, 0.0121, 0.0135, 0.0124, 0.0107, 0.0102, 0.0102, 0.0099],
       device='cuda:2'), out_proj_covar=tensor([0.0005, 0.0004, 0.0004, 0.0004, 0.0003, 0.0003, 0.0004, 0.0004],
       device='cuda:2')
2022-12-22 21:45:43,152 INFO [scaling.py:679] (2/4) Whitening: num_groups=8, num_channels=96, metric=1.43 vs. limit=2.0
2022-12-22 21:45:57,341 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.5634, 1.3821, 1.1970, 1.7412, 1.5531, 3.2464, 1.2749, 1.5512],
       device='cuda:2'), covar=tensor([0.0891, 0.1784, 0.1195, 0.0996, 0.1534, 0.0231, 0.1382, 0.1492],
       device='cuda:2'), in_proj_covar=tensor([0.0077, 0.0086, 0.0078, 0.0080, 0.0096, 0.0072, 0.0087, 0.0081],
       device='cuda:2'), out_proj_covar=tensor([0.0003, 0.0004, 0.0003, 0.0003, 0.0004, 0.0003, 0.0004, 0.0003],
       device='cuda:2')
2022-12-22 21:46:07,398 INFO [train.py:894] (2/4) Epoch 8, batch 2500, loss[loss=0.2335, simple_loss=0.3109, pruned_loss=0.07804, over 18391.00 frames. ], tot_loss[loss=0.24, simple_loss=0.3078, pruned_loss=0.08607, over 3713220.66 frames. ], batch size: 53, lr: 1.43e-02, grad_scale: 8.0
2022-12-22 21:46:46,340 INFO [zipformer.py:660] (2/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=27070.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 21:46:52,951 WARNING [train.py:1060] (2/4) Exclude cut with ID 7255-291500-0006-134305-0 from training. Duration: 21.24
2022-12-22 21:46:52,965 WARNING [train.py:1060] (2/4) Exclude cut with ID 6533-399-0047-104881-0_sp0.9 from training. Duration: 23.9055625
2022-12-22 21:47:23,371 INFO [train.py:894] (2/4) Epoch 8, batch 2550, loss[loss=0.2198, simple_loss=0.2949, pruned_loss=0.07232, over 18596.00 frames. ], tot_loss[loss=0.2395, simple_loss=0.3076, pruned_loss=0.08569, over 3713671.32 frames. ], batch size: 51, lr: 1.42e-02, grad_scale: 8.0
2022-12-22 21:47:27,661 WARNING [train.py:1060] (2/4) Exclude cut with ID 6758-72288-0033-108368-0_sp0.9 from training. Duration: 25.988875
2022-12-22 21:47:37,380 WARNING [train.py:1060] (2/4) Exclude cut with ID 7255-291500-0001-134300-0_sp0.9 from training. Duration: 20.67225
2022-12-22 21:47:49,091 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([2.5480, 2.2378, 2.2220, 2.0870, 2.2557, 2.0642, 2.3772, 1.9694],
       device='cuda:2'), covar=tensor([0.1021, 0.1554, 0.1156, 0.1630, 0.0999, 0.0576, 0.1420, 0.0640],
       device='cuda:2'), in_proj_covar=tensor([0.0238, 0.0270, 0.0245, 0.0273, 0.0258, 0.0223, 0.0276, 0.0213],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-22 21:48:00,713 INFO [zipformer.py:660] (2/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=27118.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 21:48:06,476 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 3.238e+02 5.522e+02 7.040e+02 8.791e+02 1.415e+03, threshold=1.408e+03, percent-clipped=0.0
2022-12-22 21:48:22,454 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([2.1270, 2.6128, 2.5842, 1.2907, 3.0226, 2.7866, 2.0482, 3.5936],
       device='cuda:2'), covar=tensor([0.1329, 0.1688, 0.1848, 0.2563, 0.0865, 0.1386, 0.2232, 0.0554],
       device='cuda:2'), in_proj_covar=tensor([0.0205, 0.0203, 0.0205, 0.0194, 0.0188, 0.0212, 0.0210, 0.0183],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-22 21:48:25,568 WARNING [train.py:1060] (2/4) Exclude cut with ID 3488-85273-0038-41224-0 from training. Duration: 20.34
2022-12-22 21:48:28,839 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([2.5871, 1.7337, 0.7173, 2.2838, 2.7014, 1.8746, 2.5585, 2.6054],
       device='cuda:2'), covar=tensor([0.1944, 0.2819, 0.3159, 0.1720, 0.1771, 0.1882, 0.1656, 0.2098],
       device='cuda:2'), in_proj_covar=tensor([0.0091, 0.0103, 0.0121, 0.0096, 0.0113, 0.0092, 0.0097, 0.0097],
       device='cuda:2'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003],
       device='cuda:2')
2022-12-22 21:48:40,630 INFO [train.py:894] (2/4) Epoch 8, batch 2600, loss[loss=0.2427, simple_loss=0.3099, pruned_loss=0.08771, over 18500.00 frames. ], tot_loss[loss=0.2374, simple_loss=0.3054, pruned_loss=0.0847, over 3712093.14 frames. ], batch size: 52, lr: 1.42e-02, grad_scale: 8.0
2022-12-22 21:48:47,688 INFO [zipformer.py:660] (2/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=27148.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 21:49:19,518 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([2.0475, 1.3539, 2.0815, 4.2097, 3.0497, 2.4434, 0.7838, 2.7058],
       device='cuda:2'), covar=tensor([0.1786, 0.2006, 0.1754, 0.0508, 0.1080, 0.1493, 0.2690, 0.1134],
       device='cuda:2'), in_proj_covar=tensor([0.0104, 0.0115, 0.0127, 0.0118, 0.0104, 0.0131, 0.0131, 0.0107],
       device='cuda:2'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003],
       device='cuda:2')
2022-12-22 21:49:39,806 WARNING [train.py:1060] (2/4) Exclude cut with ID 7357-94126-0026-15857-0_sp0.9 from training. Duration: 25.061125
2022-12-22 21:49:50,507 WARNING [train.py:1060] (2/4) Exclude cut with ID 3033-130750-0096-55598-0 from training. Duration: 0.83
2022-12-22 21:49:57,938 INFO [train.py:894] (2/4) Epoch 8, batch 2650, loss[loss=0.2241, simple_loss=0.3052, pruned_loss=0.07143, over 18628.00 frames. ], tot_loss[loss=0.238, simple_loss=0.3061, pruned_loss=0.08498, over 3713211.57 frames. ], batch size: 53, lr: 1.42e-02, grad_scale: 8.0
2022-12-22 21:50:16,658 WARNING [train.py:1060] (2/4) Exclude cut with ID 3488-85273-0017-41203-0 from training. Duration: 24.73
2022-12-22 21:50:18,601 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([2.0865, 2.4502, 1.4176, 2.9033, 2.8507, 2.2228, 3.7565, 2.2576],
       device='cuda:2'), covar=tensor([0.1034, 0.1654, 0.2708, 0.2033, 0.1553, 0.1049, 0.0809, 0.1175],
       device='cuda:2'), in_proj_covar=tensor([0.0173, 0.0189, 0.0229, 0.0281, 0.0226, 0.0180, 0.0201, 0.0190],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-22 21:50:21,936 INFO [zipformer.py:660] (2/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=27209.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 21:50:31,773 WARNING [train.py:1060] (2/4) Exclude cut with ID 7255-291500-0009-134308-0 from training. Duration: 23.965
2022-12-22 21:50:41,439 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 3.270e+02 5.404e+02 7.023e+02 8.570e+02 1.922e+03, threshold=1.405e+03, percent-clipped=4.0
2022-12-22 21:50:41,486 WARNING [train.py:1060] (2/4) Exclude cut with ID 6978-92210-0030-146996-0_sp0.9 from training. Duration: 22.088875
2022-12-22 21:50:57,203 WARNING [train.py:1060] (2/4) Exclude cut with ID 7255-291500-0006-134305-0_sp0.9 from training. Duration: 23.6
2022-12-22 21:51:11,994 INFO [scaling.py:679] (2/4) Whitening: num_groups=8, num_channels=96, metric=1.22 vs. limit=2.0
2022-12-22 21:51:15,439 INFO [train.py:894] (2/4) Epoch 8, batch 2700, loss[loss=0.2857, simple_loss=0.3366, pruned_loss=0.1174, over 18688.00 frames. ], tot_loss[loss=0.2373, simple_loss=0.3055, pruned_loss=0.08454, over 3712857.20 frames. ], batch size: 180, lr: 1.42e-02, grad_scale: 8.0
2022-12-22 21:51:39,782 INFO [zipformer.py:660] (2/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=27260.0, num_to_drop=1, layers_to_drop={0}
2022-12-22 21:52:32,323 INFO [train.py:894] (2/4) Epoch 8, batch 2750, loss[loss=0.2832, simple_loss=0.3401, pruned_loss=0.1131, over 18579.00 frames. ], tot_loss[loss=0.2375, simple_loss=0.3057, pruned_loss=0.08467, over 3713160.65 frames. ], batch size: 174, lr: 1.42e-02, grad_scale: 8.0
2022-12-22 21:52:36,730 WARNING [train.py:1060] (2/4) Exclude cut with ID 8291-282929-0024-13011-0 from training. Duration: 23.795
2022-12-22 21:52:50,750 INFO [zipformer.py:660] (2/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=27306.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 21:52:53,720 WARNING [train.py:1060] (2/4) Exclude cut with ID 8631-249866-0030-130156-0_sp1.1 from training. Duration: 21.5409375
2022-12-22 21:52:55,170 WARNING [train.py:1060] (2/4) Exclude cut with ID 6978-92210-0019-146985-0_sp0.9 from training. Duration: 24.97775
2022-12-22 21:53:06,766 WARNING [train.py:1060] (2/4) Exclude cut with ID 1085-156170-0017-128270-0_sp0.9 from training. Duration: 23.3444375
2022-12-22 21:53:15,432 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 4.310e+02 5.953e+02 7.157e+02 9.092e+02 1.661e+03, threshold=1.431e+03, percent-clipped=6.0
2022-12-22 21:53:33,703 WARNING [train.py:1060] (2/4) Exclude cut with ID 6010-56788-0055-90261-0_sp0.9 from training. Duration: 23.2
2022-12-22 21:53:39,356 WARNING [train.py:1060] (2/4) Exclude cut with ID 5653-46179-0060-117930-0_sp0.9 from training. Duration: 21.17225
2022-12-22 21:53:47,791 INFO [train.py:894] (2/4) Epoch 8, batch 2800, loss[loss=0.2591, simple_loss=0.3257, pruned_loss=0.0963, over 18508.00 frames. ], tot_loss[loss=0.2377, simple_loss=0.3057, pruned_loss=0.08486, over 3712559.96 frames. ], batch size: 52, lr: 1.42e-02, grad_scale: 8.0
2022-12-22 21:54:00,152 WARNING [train.py:1060] (2/4) Exclude cut with ID 7255-291500-0008-134307-0_sp0.9 from training. Duration: 24.6555625
2022-12-22 21:54:27,255 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([2.7757, 2.7878, 2.2029, 2.0210, 3.2440, 3.1645, 2.6163, 2.1486],
       device='cuda:2'), covar=tensor([0.0270, 0.0234, 0.0444, 0.0500, 0.0135, 0.0184, 0.0342, 0.0591],
       device='cuda:2'), in_proj_covar=tensor([0.0122, 0.0118, 0.0132, 0.0121, 0.0084, 0.0117, 0.0138, 0.0150],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0001, 0.0002, 0.0001, 0.0001, 0.0001, 0.0002, 0.0002],
       device='cuda:2')
2022-12-22 21:54:53,161 WARNING [train.py:1060] (2/4) Exclude cut with ID 3488-65654-0031-41259-0 from training. Duration: 20.44
2022-12-22 21:55:04,965 INFO [train.py:894] (2/4) Epoch 8, batch 2850, loss[loss=0.2395, simple_loss=0.314, pruned_loss=0.08253, over 18631.00 frames. ], tot_loss[loss=0.238, simple_loss=0.3061, pruned_loss=0.08502, over 3712584.75 frames. ], batch size: 62, lr: 1.42e-02, grad_scale: 8.0
2022-12-22 21:55:07,946 WARNING [train.py:1060] (2/4) Exclude cut with ID 6951-79737-0018-132285-0_sp0.9 from training. Duration: 23.45
2022-12-22 21:55:38,532 WARNING [train.py:1060] (2/4) Exclude cut with ID 6945-60535-0076-12784-0_sp0.9 from training. Duration: 20.52225
2022-12-22 21:55:47,152 WARNING [train.py:1060] (2/4) Exclude cut with ID 7255-291500-0008-134307-0 from training. Duration: 22.19
2022-12-22 21:55:48,489 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 3.802e+02 6.087e+02 7.386e+02 9.087e+02 2.096e+03, threshold=1.477e+03, percent-clipped=5.0
2022-12-22 21:55:55,752 WARNING [train.py:1060] (2/4) Exclude cut with ID 7357-94126-0014-15845-0_sp1.1 from training. Duration: 25.3818125
2022-12-22 21:56:13,627 WARNING [train.py:1060] (2/4) Exclude cut with ID 6951-79737-0043-132310-0_sp0.9 from training. Duration: 28.0944375
2022-12-22 21:56:21,331 INFO [train.py:894] (2/4) Epoch 8, batch 2900, loss[loss=0.2441, simple_loss=0.3174, pruned_loss=0.08537, over 18386.00 frames. ], tot_loss[loss=0.2368, simple_loss=0.305, pruned_loss=0.08426, over 3712590.38 frames. ], batch size: 53, lr: 1.42e-02, grad_scale: 8.0
2022-12-22 21:56:21,385 WARNING [train.py:1060] (2/4) Exclude cut with ID 2195-150901-0045-59933-0_sp0.9 from training. Duration: 22.9444375
2022-12-22 21:56:28,886 WARNING [train.py:1060] (2/4) Exclude cut with ID 8291-282929-0024-13011-0_sp1.1 from training. Duration: 21.6318125
2022-12-22 21:56:48,885 WARNING [train.py:1060] (2/4) Exclude cut with ID 8631-249866-0030-130156-0 from training. Duration: 23.695
2022-12-22 21:57:14,389 WARNING [train.py:1060] (2/4) Exclude cut with ID 7699-105389-0094-26379-0 from training. Duration: 23.955
2022-12-22 21:57:32,522 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([0.1892, 1.2220, 1.6514, 0.5159, 0.8255, 1.8336, 1.5778, 1.4521],
       device='cuda:2'), covar=tensor([0.0603, 0.0303, 0.0278, 0.0343, 0.0430, 0.0314, 0.0223, 0.0511],
       device='cuda:2'), in_proj_covar=tensor([0.0132, 0.0152, 0.0107, 0.0129, 0.0137, 0.0117, 0.0140, 0.0138],
       device='cuda:2'), out_proj_covar=tensor([1.1493e-04, 1.3414e-04, 9.3535e-05, 1.1160e-04, 1.1923e-04, 1.0387e-04,
        1.2558e-04, 1.2206e-04], device='cuda:2')
2022-12-22 21:57:36,463 INFO [train.py:894] (2/4) Epoch 8, batch 2950, loss[loss=0.2819, simple_loss=0.3407, pruned_loss=0.1115, over 18657.00 frames. ], tot_loss[loss=0.2364, simple_loss=0.3051, pruned_loss=0.08382, over 3713057.73 frames. ], batch size: 60, lr: 1.41e-02, grad_scale: 8.0
2022-12-22 21:57:46,460 WARNING [train.py:1060] (2/4) Exclude cut with ID 8291-282929-0024-13011-0_sp0.9 from training. Duration: 26.438875
2022-12-22 21:57:53,233 INFO [zipformer.py:660] (2/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=27504.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 21:58:19,628 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 2.835e+02 5.842e+02 7.166e+02 9.533e+02 2.194e+03, threshold=1.433e+03, percent-clipped=8.0
2022-12-22 21:58:27,508 WARNING [train.py:1060] (2/4) Exclude cut with ID 7699-105389-0021-26306-0_sp0.9 from training. Duration: 21.2444375
2022-12-22 21:58:29,105 WARNING [train.py:1060] (2/4) Exclude cut with ID 7357-94126-0014-15845-0_sp0.9 from training. Duration: 31.02225
2022-12-22 21:58:34,744 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([0.5709, 1.9293, 2.1999, 0.7955, 1.4208, 2.5757, 1.5996, 1.8306],
       device='cuda:2'), covar=tensor([0.0779, 0.0354, 0.0272, 0.0418, 0.0377, 0.0312, 0.0335, 0.0658],
       device='cuda:2'), in_proj_covar=tensor([0.0130, 0.0150, 0.0105, 0.0127, 0.0136, 0.0115, 0.0138, 0.0136],
       device='cuda:2'), out_proj_covar=tensor([1.1309e-04, 1.3256e-04, 9.1620e-05, 1.0984e-04, 1.1831e-04, 1.0193e-04,
        1.2360e-04, 1.2031e-04], device='cuda:2')
2022-12-22 21:58:38,816 WARNING [train.py:1060] (2/4) Exclude cut with ID 432-122774-0017-62487-0 from training. Duration: 22.395
2022-12-22 21:58:51,736 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([2.3381, 1.9026, 1.6745, 0.5272, 1.5999, 1.8500, 1.4717, 1.7382],
       device='cuda:2'), covar=tensor([0.0522, 0.0406, 0.0868, 0.1350, 0.0806, 0.1207, 0.1357, 0.0551],
       device='cuda:2'), in_proj_covar=tensor([0.0161, 0.0177, 0.0201, 0.0195, 0.0206, 0.0188, 0.0200, 0.0191],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-22 21:58:54,001 INFO [train.py:894] (2/4) Epoch 8, batch 3000, loss[loss=0.2813, simple_loss=0.3493, pruned_loss=0.1066, over 18669.00 frames. ], tot_loss[loss=0.2368, simple_loss=0.3056, pruned_loss=0.08404, over 3713138.05 frames. ], batch size: 99, lr: 1.41e-02, grad_scale: 8.0
2022-12-22 21:58:54,001 INFO [train.py:919] (2/4) Computing validation loss
2022-12-22 21:59:05,189 INFO [train.py:928] (2/4) Epoch 8, validation: loss=0.1857, simple_loss=0.2837, pruned_loss=0.04382, over 944034.00 frames. 
2022-12-22 21:59:05,190 INFO [train.py:929] (2/4) Maximum memory allocated so far is 24680MB
2022-12-22 21:59:08,107 WARNING [train.py:1060] (2/4) Exclude cut with ID 7357-94126-0045-15876-0 from training. Duration: 21.075
2022-12-22 21:59:14,122 WARNING [train.py:1060] (2/4) Exclude cut with ID 6482-98857-0025-147532-0_sp0.9 from training. Duration: 20.0055625
2022-12-22 21:59:14,133 WARNING [train.py:1060] (2/4) Exclude cut with ID 6951-79737-0037-132304-0_sp0.9 from training. Duration: 22.05
2022-12-22 21:59:14,144 WARNING [train.py:1060] (2/4) Exclude cut with ID 7255-291500-0003-134302-0 from training. Duration: 26.8349375
2022-12-22 21:59:17,067 WARNING [train.py:1060] (2/4) Exclude cut with ID 7357-94126-0009-15840-0_sp1.1 from training. Duration: 22.1090625
2022-12-22 21:59:24,293 WARNING [train.py:1060] (2/4) Exclude cut with ID 7699-105389-0094-26379-0_sp0.9 from training. Duration: 26.6166875
2022-12-22 21:59:29,530 INFO [zipformer.py:660] (2/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=27560.0, num_to_drop=1, layers_to_drop={2}
2022-12-22 21:59:42,692 WARNING [train.py:1060] (2/4) Exclude cut with ID 2046-178027-0000-53705-0_sp0.9 from training. Duration: 20.3055625
2022-12-22 22:00:05,927 WARNING [train.py:1060] (2/4) Exclude cut with ID 7205-50138-0008-5373-0_sp0.9 from training. Duration: 20.7
2022-12-22 22:00:21,696 INFO [train.py:894] (2/4) Epoch 8, batch 3050, loss[loss=0.232, simple_loss=0.3069, pruned_loss=0.07853, over 18532.00 frames. ], tot_loss[loss=0.2357, simple_loss=0.3046, pruned_loss=0.08336, over 3712950.27 frames. ], batch size: 55, lr: 1.41e-02, grad_scale: 8.0
2022-12-22 22:00:39,882 INFO [zipformer.py:660] (2/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=27606.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 22:00:41,901 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.7268, 1.1168, 1.8153, 3.0666, 2.0213, 2.3961, 0.8884, 2.1980],
       device='cuda:2'), covar=tensor([0.1776, 0.1938, 0.1692, 0.0646, 0.1311, 0.1233, 0.2520, 0.1169],
       device='cuda:2'), in_proj_covar=tensor([0.0103, 0.0112, 0.0125, 0.0117, 0.0103, 0.0129, 0.0128, 0.0106],
       device='cuda:2'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003],
       device='cuda:2')
2022-12-22 22:00:43,299 INFO [zipformer.py:660] (2/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=27608.0, num_to_drop=1, layers_to_drop={1}
2022-12-22 22:00:51,723 WARNING [train.py:1060] (2/4) Exclude cut with ID 6978-92210-0019-146985-0 from training. Duration: 22.48
2022-12-22 22:01:03,205 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 3.695e+02 5.726e+02 6.987e+02 9.097e+02 1.949e+03, threshold=1.397e+03, percent-clipped=1.0
2022-12-22 22:01:06,738 WARNING [train.py:1060] (2/4) Exclude cut with ID 7255-291500-0003-134302-0_sp0.9 from training. Duration: 29.816625
2022-12-22 22:01:27,371 WARNING [train.py:1060] (2/4) Exclude cut with ID 7255-291500-0005-134304-0_sp1.1 from training. Duration: 22.7590625
2022-12-22 22:01:31,857 WARNING [train.py:1060] (2/4) Exclude cut with ID 7357-94126-0026-15857-0 from training. Duration: 22.555
2022-12-22 22:01:37,865 INFO [train.py:894] (2/4) Epoch 8, batch 3100, loss[loss=0.2326, simple_loss=0.3082, pruned_loss=0.07848, over 18579.00 frames. ], tot_loss[loss=0.2375, simple_loss=0.3063, pruned_loss=0.08434, over 3713927.59 frames. ], batch size: 77, lr: 1.41e-02, grad_scale: 8.0
2022-12-22 22:01:53,085 WARNING [train.py:1060] (2/4) Exclude cut with ID 1250-135782-0005-25975-0_sp0.9 from training. Duration: 21.688875
2022-12-22 22:01:53,201 INFO [zipformer.py:660] (2/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=27654.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 22:01:59,421 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([2.0058, 2.1383, 1.3166, 2.2316, 2.2547, 1.9693, 2.9400, 2.0830],
       device='cuda:2'), covar=tensor([0.0844, 0.1458, 0.2580, 0.1859, 0.1621, 0.0812, 0.0906, 0.1031],
       device='cuda:2'), in_proj_covar=tensor([0.0171, 0.0189, 0.0228, 0.0278, 0.0223, 0.0179, 0.0200, 0.0190],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-22 22:02:27,540 WARNING [train.py:1060] (2/4) Exclude cut with ID 3488-85273-0038-41224-0_sp0.9 from training. Duration: 22.6
2022-12-22 22:02:54,057 INFO [train.py:894] (2/4) Epoch 8, batch 3150, loss[loss=0.2216, simple_loss=0.2914, pruned_loss=0.07588, over 18546.00 frames. ], tot_loss[loss=0.2369, simple_loss=0.3056, pruned_loss=0.0841, over 3714362.69 frames. ], batch size: 47, lr: 1.41e-02, grad_scale: 8.0
2022-12-22 22:03:04,766 WARNING [train.py:1060] (2/4) Exclude cut with ID 7357-94126-0009-15840-0 from training. Duration: 24.32
2022-12-22 22:03:25,301 INFO [zipformer.py:660] (2/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=27714.0, num_to_drop=1, layers_to_drop={1}
2022-12-22 22:03:36,990 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 3.801e+02 5.956e+02 7.094e+02 8.866e+02 2.469e+03, threshold=1.419e+03, percent-clipped=7.0
2022-12-22 22:04:05,087 WARNING [train.py:1060] (2/4) Exclude cut with ID 8291-276745-0093-13116-0_sp0.9 from training. Duration: 21.061125
2022-12-22 22:04:10,928 INFO [train.py:894] (2/4) Epoch 8, batch 3200, loss[loss=0.228, simple_loss=0.3066, pruned_loss=0.07472, over 18575.00 frames. ], tot_loss[loss=0.2341, simple_loss=0.3031, pruned_loss=0.08256, over 3714425.94 frames. ], batch size: 51, lr: 1.41e-02, grad_scale: 8.0
2022-12-22 22:04:19,030 WARNING [train.py:1060] (2/4) Exclude cut with ID 7357-94126-0024-15855-0_sp0.9 from training. Duration: 20.32225
2022-12-22 22:04:31,514 WARNING [train.py:1060] (2/4) Exclude cut with ID 3033-130750-0096-55598-0_sp1.1 from training. Duration: 0.7545625
2022-12-22 22:04:31,952 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([2.0110, 2.1388, 1.1415, 2.3512, 2.2196, 1.9744, 3.1263, 2.0746],
       device='cuda:2'), covar=tensor([0.0841, 0.1550, 0.2739, 0.1965, 0.1685, 0.0793, 0.0839, 0.1098],
       device='cuda:2'), in_proj_covar=tensor([0.0171, 0.0188, 0.0227, 0.0279, 0.0223, 0.0179, 0.0201, 0.0191],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-22 22:04:41,569 INFO [scaling.py:679] (2/4) Whitening: num_groups=8, num_channels=192, metric=1.87 vs. limit=2.0
2022-12-22 22:04:46,397 WARNING [train.py:1060] (2/4) Exclude cut with ID 4295-39940-0007-92567-0_sp0.9 from training. Duration: 23.9333125
2022-12-22 22:04:59,747 INFO [zipformer.py:660] (2/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=27775.0, num_to_drop=1, layers_to_drop={0}
2022-12-22 22:05:08,848 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([3.3428, 2.3516, 1.6342, 2.8750, 2.1628, 1.9427, 2.3003, 3.6734],
       device='cuda:2'), covar=tensor([0.1386, 0.2559, 0.1469, 0.2486, 0.2879, 0.0891, 0.2450, 0.0405],
       device='cuda:2'), in_proj_covar=tensor([0.0268, 0.0256, 0.0218, 0.0335, 0.0242, 0.0208, 0.0253, 0.0179],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0003, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-22 22:05:17,524 WARNING [train.py:1060] (2/4) Exclude cut with ID 7255-291500-0008-134307-0_sp1.1 from training. Duration: 20.17275
2022-12-22 22:05:23,995 WARNING [train.py:1060] (2/4) Exclude cut with ID 6978-92210-0019-146985-0_sp1.1 from training. Duration: 20.436375
2022-12-22 22:05:28,395 INFO [train.py:894] (2/4) Epoch 8, batch 3250, loss[loss=0.2266, simple_loss=0.3028, pruned_loss=0.07521, over 18696.00 frames. ], tot_loss[loss=0.2345, simple_loss=0.3039, pruned_loss=0.08261, over 3714269.77 frames. ], batch size: 62, lr: 1.41e-02, grad_scale: 8.0
2022-12-22 22:05:44,361 INFO [zipformer.py:660] (2/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=27804.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 22:06:11,162 INFO [scaling.py:679] (2/4) Whitening: num_groups=8, num_channels=96, metric=1.17 vs. limit=2.0
2022-12-22 22:06:11,577 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 3.550e+02 6.210e+02 7.285e+02 9.598e+02 1.821e+03, threshold=1.457e+03, percent-clipped=2.0
2022-12-22 22:06:43,553 WARNING [train.py:1060] (2/4) Exclude cut with ID 4234-40345-0022-142709-0_sp0.9 from training. Duration: 23.1055625
2022-12-22 22:06:44,990 INFO [train.py:894] (2/4) Epoch 8, batch 3300, loss[loss=0.2007, simple_loss=0.2712, pruned_loss=0.06514, over 18596.00 frames. ], tot_loss[loss=0.2336, simple_loss=0.3031, pruned_loss=0.08205, over 3713435.62 frames. ], batch size: 45, lr: 1.41e-02, grad_scale: 8.0
2022-12-22 22:06:45,026 WARNING [train.py:1060] (2/4) Exclude cut with ID 8291-282929-0007-12994-0_sp1.1 from training. Duration: 23.5
2022-12-22 22:06:56,588 WARNING [train.py:1060] (2/4) Exclude cut with ID 7255-291500-0009-134308-0_sp0.9 from training. Duration: 26.62775
2022-12-22 22:06:58,372 INFO [zipformer.py:660] (2/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=27852.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 22:07:10,151 WARNING [train.py:1060] (2/4) Exclude cut with ID 6951-79737-0018-132285-0 from training. Duration: 21.105
2022-12-22 22:07:15,233 WARNING [train.py:1060] (2/4) Exclude cut with ID 4511-76322-0006-80011-0_sp0.9 from training. Duration: 24.411125
2022-12-22 22:07:42,975 WARNING [train.py:1060] (2/4) Exclude cut with ID 6758-72288-0033-108368-0_sp1.1 from training. Duration: 21.263625
2022-12-22 22:08:02,676 INFO [train.py:894] (2/4) Epoch 8, batch 3350, loss[loss=0.2305, simple_loss=0.3111, pruned_loss=0.07494, over 18715.00 frames. ], tot_loss[loss=0.233, simple_loss=0.3029, pruned_loss=0.08152, over 3714167.95 frames. ], batch size: 62, lr: 1.40e-02, grad_scale: 8.0
2022-12-22 22:08:16,329 WARNING [train.py:1060] (2/4) Exclude cut with ID 4234-40345-0022-142709-0 from training. Duration: 20.795
2022-12-22 22:08:27,167 WARNING [train.py:1060] (2/4) Exclude cut with ID 7357-94126-0021-15852-0 from training. Duration: 24.76
2022-12-22 22:08:27,188 WARNING [train.py:1060] (2/4) Exclude cut with ID 3867-173237-0077-144769-0_sp0.9 from training. Duration: 22.25
2022-12-22 22:08:30,339 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([2.2941, 2.7505, 2.6829, 1.3484, 2.7458, 1.9745, 0.4935, 1.8554],
       device='cuda:2'), covar=tensor([0.2111, 0.1150, 0.1701, 0.3624, 0.1274, 0.1369, 0.5369, 0.1779],
       device='cuda:2'), in_proj_covar=tensor([0.0133, 0.0119, 0.0154, 0.0119, 0.0126, 0.0104, 0.0145, 0.0111],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0002, 0.0003, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-22 22:08:45,535 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 3.385e+02 6.137e+02 7.435e+02 8.806e+02 1.482e+03, threshold=1.487e+03, percent-clipped=1.0
2022-12-22 22:08:52,490 WARNING [train.py:1060] (2/4) Exclude cut with ID 7357-94126-0026-15857-0_sp1.1 from training. Duration: 20.5045625
2022-12-22 22:09:19,703 INFO [scaling.py:679] (2/4) Whitening: num_groups=1, num_channels=384, metric=4.81 vs. limit=5.0
2022-12-22 22:09:20,022 INFO [train.py:894] (2/4) Epoch 8, batch 3400, loss[loss=0.2318, simple_loss=0.3111, pruned_loss=0.0763, over 18659.00 frames. ], tot_loss[loss=0.2318, simple_loss=0.3015, pruned_loss=0.08103, over 3714481.86 frames. ], batch size: 78, lr: 1.40e-02, grad_scale: 8.0
2022-12-22 22:09:32,102 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([2.6483, 2.4008, 2.6965, 1.4906, 3.0090, 2.9232, 2.0944, 3.7569],
       device='cuda:2'), covar=tensor([0.1031, 0.1465, 0.1337, 0.2026, 0.0767, 0.1128, 0.1847, 0.0407],
       device='cuda:2'), in_proj_covar=tensor([0.0196, 0.0192, 0.0195, 0.0185, 0.0180, 0.0206, 0.0205, 0.0176],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-22 22:10:32,703 INFO [train.py:894] (2/4) Epoch 8, batch 3450, loss[loss=0.2334, simple_loss=0.2989, pruned_loss=0.08399, over 18703.00 frames. ], tot_loss[loss=0.2321, simple_loss=0.3016, pruned_loss=0.0813, over 3714737.29 frames. ], batch size: 50, lr: 1.40e-02, grad_scale: 8.0
2022-12-22 22:11:05,052 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.8259, 1.4339, 0.5902, 1.3896, 2.0311, 1.3003, 1.7055, 1.8091],
       device='cuda:2'), covar=tensor([0.1624, 0.2200, 0.2970, 0.1619, 0.1874, 0.1790, 0.1500, 0.1693],
       device='cuda:2'), in_proj_covar=tensor([0.0091, 0.0104, 0.0123, 0.0098, 0.0114, 0.0093, 0.0098, 0.0097],
       device='cuda:2'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003],
       device='cuda:2')
2022-12-22 22:11:16,876 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 3.454e+02 5.907e+02 7.440e+02 9.598e+02 2.157e+03, threshold=1.488e+03, percent-clipped=3.0
2022-12-22 22:11:31,375 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([2.7633, 2.1919, 1.6370, 2.5325, 1.9701, 1.9226, 2.0554, 2.8507],
       device='cuda:2'), covar=tensor([0.1627, 0.2730, 0.1646, 0.2748, 0.2917, 0.0966, 0.2539, 0.0568],
       device='cuda:2'), in_proj_covar=tensor([0.0273, 0.0260, 0.0221, 0.0340, 0.0244, 0.0212, 0.0256, 0.0182],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0003, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-22 22:11:49,480 INFO [train.py:894] (2/4) Epoch 8, batch 3500, loss[loss=0.2547, simple_loss=0.3207, pruned_loss=0.09432, over 18586.00 frames. ], tot_loss[loss=0.2331, simple_loss=0.3024, pruned_loss=0.08184, over 3714923.73 frames. ], batch size: 98, lr: 1.40e-02, grad_scale: 8.0
2022-12-22 22:12:10,227 WARNING [train.py:1060] (2/4) Exclude cut with ID 7859-102521-0017-7548-0_sp1.1 from training. Duration: 22.2954375
2022-12-22 22:12:22,581 INFO [train.py:894] (2/4) Epoch 9, batch 0, loss[loss=0.2247, simple_loss=0.31, pruned_loss=0.06966, over 18722.00 frames. ], tot_loss[loss=0.2247, simple_loss=0.31, pruned_loss=0.06966, over 18722.00 frames. ], batch size: 54, lr: 1.33e-02, grad_scale: 8.0
2022-12-22 22:12:22,581 INFO [train.py:919] (2/4) Computing validation loss
2022-12-22 22:12:33,731 INFO [train.py:928] (2/4) Epoch 9, validation: loss=0.1816, simple_loss=0.2807, pruned_loss=0.04127, over 944034.00 frames. 
2022-12-22 22:12:33,732 INFO [train.py:929] (2/4) Maximum memory allocated so far is 24680MB
2022-12-22 22:12:56,682 INFO [scaling.py:679] (2/4) Whitening: num_groups=8, num_channels=96, metric=1.57 vs. limit=2.0
2022-12-22 22:13:06,128 INFO [zipformer.py:660] (2/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=28070.0, num_to_drop=1, layers_to_drop={2}
2022-12-22 22:13:26,222 WARNING [train.py:1060] (2/4) Exclude cut with ID 298-126791-0067-24026-0_sp0.9 from training. Duration: 21.438875
2022-12-22 22:13:32,547 WARNING [train.py:1060] (2/4) Exclude cut with ID 5652-39938-0025-23684-0_sp0.9 from training. Duration: 22.2055625
2022-12-22 22:13:52,625 INFO [train.py:894] (2/4) Epoch 9, batch 50, loss[loss=0.2194, simple_loss=0.3013, pruned_loss=0.0687, over 18527.00 frames. ], tot_loss[loss=0.2172, simple_loss=0.2988, pruned_loss=0.06781, over 838974.32 frames. ], batch size: 55, lr: 1.32e-02, grad_scale: 8.0
2022-12-22 22:13:59,892 INFO [scaling.py:679] (2/4) Whitening: num_groups=8, num_channels=96, metric=1.13 vs. limit=2.0
2022-12-22 22:14:26,944 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 2.901e+02 4.305e+02 5.351e+02 6.877e+02 1.696e+03, threshold=1.070e+03, percent-clipped=2.0
2022-12-22 22:14:46,559 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([3.3720, 3.5308, 1.4544, 1.0089, 3.9314, 3.8617, 2.5404, 2.3703],
       device='cuda:2'), covar=tensor([0.0295, 0.0251, 0.0750, 0.0868, 0.0080, 0.0275, 0.0539, 0.0685],
       device='cuda:2'), in_proj_covar=tensor([0.0117, 0.0116, 0.0130, 0.0119, 0.0082, 0.0116, 0.0135, 0.0150],
       device='cuda:2'), out_proj_covar=tensor([0.0001, 0.0001, 0.0002, 0.0001, 0.0001, 0.0001, 0.0002, 0.0002],
       device='cuda:2')
2022-12-22 22:15:02,485 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.6049, 1.4094, 1.1685, 0.6052, 1.8587, 1.5306, 1.3440, 1.1448],
       device='cuda:2'), covar=tensor([0.0375, 0.0464, 0.0573, 0.0814, 0.0277, 0.0403, 0.0526, 0.0953],
       device='cuda:2'), in_proj_covar=tensor([0.0117, 0.0115, 0.0129, 0.0119, 0.0082, 0.0116, 0.0134, 0.0150],
       device='cuda:2'), out_proj_covar=tensor([0.0001, 0.0001, 0.0002, 0.0001, 0.0001, 0.0001, 0.0002, 0.0002],
       device='cuda:2')
2022-12-22 22:15:07,904 INFO [train.py:894] (2/4) Epoch 9, batch 100, loss[loss=0.1821, simple_loss=0.2712, pruned_loss=0.04643, over 18381.00 frames. ], tot_loss[loss=0.2117, simple_loss=0.2931, pruned_loss=0.06514, over 1475494.24 frames. ], batch size: 46, lr: 1.32e-02, grad_scale: 8.0
2022-12-22 22:16:22,823 INFO [train.py:894] (2/4) Epoch 9, batch 150, loss[loss=0.1885, simple_loss=0.2675, pruned_loss=0.05473, over 18412.00 frames. ], tot_loss[loss=0.2099, simple_loss=0.2914, pruned_loss=0.06423, over 1971380.58 frames. ], batch size: 46, lr: 1.32e-02, grad_scale: 8.0
2022-12-22 22:16:42,103 WARNING [train.py:1060] (2/4) Exclude cut with ID 7859-102521-0017-7548-0 from training. Duration: 24.525
2022-12-22 22:16:57,976 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 2.622e+02 4.291e+02 5.287e+02 6.560e+02 1.269e+03, threshold=1.057e+03, percent-clipped=2.0
2022-12-22 22:17:09,705 INFO [zipformer.py:660] (2/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=28229.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 22:17:13,903 WARNING [train.py:1060] (2/4) Exclude cut with ID 3699-47246-0007-3408-0_sp0.9 from training. Duration: 20.26675
2022-12-22 22:17:27,230 WARNING [train.py:1060] (2/4) Exclude cut with ID 7859-102521-0017-7548-0_sp0.9 from training. Duration: 27.25
2022-12-22 22:17:41,453 INFO [train.py:894] (2/4) Epoch 9, batch 200, loss[loss=0.2019, simple_loss=0.2846, pruned_loss=0.05958, over 18549.00 frames. ], tot_loss[loss=0.208, simple_loss=0.2897, pruned_loss=0.06321, over 2358092.91 frames. ], batch size: 57, lr: 1.32e-02, grad_scale: 8.0
2022-12-22 22:17:48,926 INFO [zipformer.py:660] (2/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=28255.0, num_to_drop=1, layers_to_drop={1}
2022-12-22 22:18:08,273 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([3.4014, 2.4219, 1.9710, 3.1658, 2.3538, 2.3768, 2.5702, 3.9030],
       device='cuda:2'), covar=tensor([0.1442, 0.2711, 0.1412, 0.2557, 0.3197, 0.0805, 0.2670, 0.0404],
       device='cuda:2'), in_proj_covar=tensor([0.0274, 0.0262, 0.0223, 0.0341, 0.0247, 0.0212, 0.0258, 0.0182],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0003, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-22 22:18:38,287 INFO [zipformer.py:660] (2/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=28287.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 22:18:42,373 INFO [zipformer.py:660] (2/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=28290.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 22:18:44,734 WARNING [train.py:1060] (2/4) Exclude cut with ID 6426-64292-0017-15984-0 from training. Duration: 21.68
2022-12-22 22:18:55,136 WARNING [train.py:1060] (2/4) Exclude cut with ID 4278-13270-0007-59342-0 from training. Duration: 21.6300625
2022-12-22 22:18:57,279 INFO [train.py:894] (2/4) Epoch 9, batch 250, loss[loss=0.1987, simple_loss=0.2763, pruned_loss=0.06054, over 18723.00 frames. ], tot_loss[loss=0.2058, simple_loss=0.2877, pruned_loss=0.062, over 2658755.97 frames. ], batch size: 52, lr: 1.32e-02, grad_scale: 8.0
2022-12-22 22:19:20,901 WARNING [train.py:1060] (2/4) Exclude cut with ID 4278-13270-0007-59342-0_sp0.9 from training. Duration: 24.033375
2022-12-22 22:19:22,871 INFO [zipformer.py:660] (2/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=28316.0, num_to_drop=1, layers_to_drop={0}
2022-12-22 22:19:31,399 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 2.623e+02 3.974e+02 4.954e+02 5.974e+02 1.449e+03, threshold=9.907e+02, percent-clipped=4.0
2022-12-22 22:20:10,967 INFO [zipformer.py:660] (2/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=28348.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 22:20:13,526 INFO [train.py:894] (2/4) Epoch 9, batch 300, loss[loss=0.2349, simple_loss=0.3024, pruned_loss=0.0837, over 18609.00 frames. ], tot_loss[loss=0.2055, simple_loss=0.2865, pruned_loss=0.06219, over 2891392.98 frames. ], batch size: 177, lr: 1.32e-02, grad_scale: 8.0
2022-12-22 22:20:15,317 WARNING [train.py:1060] (2/4) Exclude cut with ID 4278-13270-0009-59344-0 from training. Duration: 22.905
2022-12-22 22:20:16,777 WARNING [train.py:1060] (2/4) Exclude cut with ID 5622-44585-0006-90525-0_sp1.1 from training. Duration: 23.4318125
2022-12-22 22:20:45,294 INFO [zipformer.py:660] (2/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=28370.0, num_to_drop=1, layers_to_drop={0}
2022-12-22 22:21:29,908 INFO [train.py:894] (2/4) Epoch 9, batch 350, loss[loss=0.1913, simple_loss=0.2767, pruned_loss=0.05295, over 18686.00 frames. ], tot_loss[loss=0.2057, simple_loss=0.2868, pruned_loss=0.06223, over 3073679.77 frames. ], batch size: 46, lr: 1.32e-02, grad_scale: 8.0
2022-12-22 22:21:58,742 INFO [zipformer.py:660] (2/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=28418.0, num_to_drop=1, layers_to_drop={1}
2022-12-22 22:22:04,378 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 3.510e+02 4.571e+02 5.481e+02 6.652e+02 2.001e+03, threshold=1.096e+03, percent-clipped=7.0
2022-12-22 22:22:16,074 WARNING [train.py:1060] (2/4) Exclude cut with ID 4278-13270-0009-59344-0_sp1.1 from training. Duration: 20.82275
2022-12-22 22:22:17,510 WARNING [train.py:1060] (2/4) Exclude cut with ID 4278-13270-0009-59344-0_sp0.9 from training. Duration: 25.45
2022-12-22 22:22:46,140 INFO [train.py:894] (2/4) Epoch 9, batch 400, loss[loss=0.2096, simple_loss=0.2796, pruned_loss=0.06983, over 18487.00 frames. ], tot_loss[loss=0.2089, simple_loss=0.2896, pruned_loss=0.06407, over 3214671.27 frames. ], batch size: 43, lr: 1.32e-02, grad_scale: 8.0
2022-12-22 22:23:14,958 WARNING [train.py:1060] (2/4) Exclude cut with ID 5622-44585-0006-90525-0 from training. Duration: 25.775
2022-12-22 22:23:21,559 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([2.1765, 2.5348, 1.4970, 2.8482, 2.9836, 2.4240, 3.9033, 2.4970],
       device='cuda:2'), covar=tensor([0.0762, 0.1526, 0.2377, 0.1827, 0.1310, 0.0758, 0.0732, 0.0990],
       device='cuda:2'), in_proj_covar=tensor([0.0167, 0.0188, 0.0225, 0.0272, 0.0221, 0.0177, 0.0197, 0.0189],
       device='cuda:2'), out_proj_covar=tensor([0.0001, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-22 22:23:35,607 WARNING [train.py:1060] (2/4) Exclude cut with ID 2929-85685-0071-62375-0_sp0.9 from training. Duration: 22.25
2022-12-22 22:23:41,880 INFO [zipformer.py:660] (2/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=28486.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 22:23:51,251 INFO [zipformer.py:660] (2/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=28492.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 22:24:03,935 INFO [train.py:894] (2/4) Epoch 9, batch 450, loss[loss=0.2433, simple_loss=0.3233, pruned_loss=0.08166, over 18643.00 frames. ], tot_loss[loss=0.2117, simple_loss=0.2922, pruned_loss=0.06561, over 3326317.73 frames. ], batch size: 60, lr: 1.32e-02, grad_scale: 8.0
2022-12-22 22:24:03,963 WARNING [train.py:1060] (2/4) Exclude cut with ID 3972-170212-0014-23379-0 from training. Duration: 26.205
2022-12-22 22:24:20,879 WARNING [train.py:1060] (2/4) Exclude cut with ID 5239-32139-0047-9341-0_sp0.9 from training. Duration: 30.1555625
2022-12-22 22:24:22,910 INFO [scaling.py:679] (2/4) Whitening: num_groups=8, num_channels=96, metric=1.39 vs. limit=2.0
2022-12-22 22:24:26,245 WARNING [train.py:1060] (2/4) Exclude cut with ID 1265-135635-0050-6781-0_sp0.9 from training. Duration: 21.8333125
2022-12-22 22:24:35,502 WARNING [train.py:1060] (2/4) Exclude cut with ID 1914-133440-0024-94914-0_sp1.1 from training. Duration: 20.6545625
2022-12-22 22:24:36,879 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 2.819e+02 4.572e+02 5.901e+02 7.346e+02 1.900e+03, threshold=1.180e+03, percent-clipped=4.0
2022-12-22 22:25:08,503 INFO [scaling.py:679] (2/4) Whitening: num_groups=8, num_channels=96, metric=1.91 vs. limit=2.0
2022-12-22 22:25:14,232 INFO [zipformer.py:660] (2/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=28547.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 22:25:18,610 INFO [train.py:894] (2/4) Epoch 9, batch 500, loss[loss=0.2362, simple_loss=0.3188, pruned_loss=0.07677, over 18526.00 frames. ], tot_loss[loss=0.2127, simple_loss=0.293, pruned_loss=0.06616, over 3412768.65 frames. ], batch size: 52, lr: 1.31e-02, grad_scale: 8.0
2022-12-22 22:25:18,656 WARNING [train.py:1060] (2/4) Exclude cut with ID 7395-89880-0045-39920-0_sp0.9 from training. Duration: 20.52225
2022-12-22 22:25:23,647 INFO [zipformer.py:660] (2/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=28553.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 22:25:31,512 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.6547, 1.1369, 1.6400, 3.1110, 2.2187, 2.3154, 0.8428, 2.0856],
       device='cuda:2'), covar=tensor([0.1846, 0.2016, 0.1731, 0.0565, 0.1260, 0.1265, 0.2472, 0.1247],
       device='cuda:2'), in_proj_covar=tensor([0.0103, 0.0113, 0.0125, 0.0119, 0.0104, 0.0131, 0.0129, 0.0108],
       device='cuda:2'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003],
       device='cuda:2')
2022-12-22 22:25:38,809 WARNING [train.py:1060] (2/4) Exclude cut with ID 3972-170212-0014-23379-0_sp0.9 from training. Duration: 29.1166875
2022-12-22 22:26:12,602 INFO [zipformer.py:660] (2/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=28585.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 22:26:36,465 INFO [train.py:894] (2/4) Epoch 9, batch 550, loss[loss=0.211, simple_loss=0.2902, pruned_loss=0.06587, over 18536.00 frames. ], tot_loss[loss=0.2126, simple_loss=0.2932, pruned_loss=0.06599, over 3479265.71 frames. ], batch size: 47, lr: 1.31e-02, grad_scale: 16.0
2022-12-22 22:26:41,202 WARNING [train.py:1060] (2/4) Exclude cut with ID 543-133211-0007-59831-0_sp0.9 from training. Duration: 21.388875
2022-12-22 22:26:53,789 INFO [zipformer.py:660] (2/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=28611.0, num_to_drop=1, layers_to_drop={1}
2022-12-22 22:27:10,199 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 3.016e+02 4.589e+02 5.717e+02 7.953e+02 1.813e+03, threshold=1.143e+03, percent-clipped=5.0
2022-12-22 22:27:17,431 WARNING [train.py:1060] (2/4) Exclude cut with ID 1914-133440-0024-94914-0 from training. Duration: 22.72
2022-12-22 22:27:18,876 WARNING [train.py:1060] (2/4) Exclude cut with ID 1914-133440-0031-94921-0_sp0.9 from training. Duration: 22.7444375
2022-12-22 22:27:42,201 INFO [zipformer.py:660] (2/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=28643.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 22:27:53,947 INFO [train.py:894] (2/4) Epoch 9, batch 600, loss[loss=0.2403, simple_loss=0.3189, pruned_loss=0.08084, over 18660.00 frames. ], tot_loss[loss=0.2125, simple_loss=0.293, pruned_loss=0.06599, over 3530907.52 frames. ], batch size: 62, lr: 1.31e-02, grad_scale: 16.0
2022-12-22 22:28:00,853 WARNING [train.py:1060] (2/4) Exclude cut with ID 4133-6541-0027-40495-0_sp1.1 from training. Duration: 0.9681875
2022-12-22 22:28:05,033 WARNING [train.py:1060] (2/4) Exclude cut with ID 6330-62851-0022-91297-0_sp0.9 from training. Duration: 22.3166875
2022-12-22 22:28:09,616 WARNING [train.py:1060] (2/4) Exclude cut with ID 543-133212-0015-59917-0_sp0.9 from training. Duration: 21.8166875
2022-12-22 22:28:54,413 INFO [scaling.py:679] (2/4) Whitening: num_groups=8, num_channels=96, metric=1.25 vs. limit=2.0
2022-12-22 22:29:09,504 INFO [train.py:894] (2/4) Epoch 9, batch 650, loss[loss=0.2021, simple_loss=0.2886, pruned_loss=0.05778, over 18391.00 frames. ], tot_loss[loss=0.2125, simple_loss=0.2931, pruned_loss=0.06591, over 3571456.05 frames. ], batch size: 53, lr: 1.31e-02, grad_scale: 16.0
2022-12-22 22:29:46,461 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 3.349e+02 4.652e+02 5.570e+02 7.061e+02 1.295e+03, threshold=1.114e+03, percent-clipped=4.0
2022-12-22 22:29:56,834 WARNING [train.py:1060] (2/4) Exclude cut with ID 4957-30119-0041-23990-0_sp0.9 from training. Duration: 20.22775
2022-12-22 22:30:29,677 INFO [train.py:894] (2/4) Epoch 9, batch 700, loss[loss=0.2214, simple_loss=0.2963, pruned_loss=0.07325, over 18543.00 frames. ], tot_loss[loss=0.213, simple_loss=0.2935, pruned_loss=0.06629, over 3603408.87 frames. ], batch size: 44, lr: 1.31e-02, grad_scale: 16.0
2022-12-22 22:30:41,318 WARNING [train.py:1060] (2/4) Exclude cut with ID 5239-32139-0047-9341-0_sp1.1 from training. Duration: 24.67275
2022-12-22 22:31:09,780 WARNING [train.py:1060] (2/4) Exclude cut with ID 3082-165428-0081-50734-0_sp0.9 from training. Duration: 21.8055625
2022-12-22 22:31:45,293 INFO [train.py:894] (2/4) Epoch 9, batch 750, loss[loss=0.2441, simple_loss=0.3167, pruned_loss=0.08571, over 18471.00 frames. ], tot_loss[loss=0.2141, simple_loss=0.2946, pruned_loss=0.0668, over 3627407.43 frames. ], batch size: 50, lr: 1.31e-02, grad_scale: 16.0
2022-12-22 22:31:45,400 WARNING [train.py:1060] (2/4) Exclude cut with ID 3340-169293-0054-76830-0_sp0.9 from training. Duration: 22.6666875
2022-12-22 22:32:08,402 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.6371, 1.7962, 1.8539, 1.8155, 1.5485, 3.8112, 1.7758, 2.3191],
       device='cuda:2'), covar=tensor([0.3111, 0.1941, 0.1702, 0.1773, 0.1228, 0.0134, 0.1386, 0.0762],
       device='cuda:2'), in_proj_covar=tensor([0.0142, 0.0121, 0.0132, 0.0123, 0.0107, 0.0100, 0.0101, 0.0099],
       device='cuda:2'), out_proj_covar=tensor([0.0005, 0.0004, 0.0004, 0.0004, 0.0004, 0.0003, 0.0004, 0.0004],
       device='cuda:2')
2022-12-22 22:32:18,225 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 2.703e+02 4.340e+02 5.308e+02 6.932e+02 2.206e+03, threshold=1.062e+03, percent-clipped=3.0
2022-12-22 22:32:48,891 INFO [zipformer.py:660] (2/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=28842.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 22:32:50,210 WARNING [train.py:1060] (2/4) Exclude cut with ID 2411-132532-0017-82279-0_sp1.1 from training. Duration: 0.9681875
2022-12-22 22:32:51,059 INFO [scaling.py:679] (2/4) Whitening: num_groups=1, num_channels=384, metric=3.88 vs. limit=5.0
2022-12-22 22:32:57,680 INFO [zipformer.py:660] (2/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=28848.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 22:33:01,176 INFO [train.py:894] (2/4) Epoch 9, batch 800, loss[loss=0.1957, simple_loss=0.2697, pruned_loss=0.06085, over 18474.00 frames. ], tot_loss[loss=0.2138, simple_loss=0.2943, pruned_loss=0.0666, over 3645991.72 frames. ], batch size: 43, lr: 1.31e-02, grad_scale: 16.0
2022-12-22 22:33:16,191 WARNING [train.py:1060] (2/4) Exclude cut with ID 6330-62850-0007-91323-0 from training. Duration: 22.485
2022-12-22 22:33:18,059 INFO [zipformer.py:660] (2/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=28861.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 22:33:54,500 INFO [zipformer.py:660] (2/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=28885.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 22:33:55,761 WARNING [train.py:1060] (2/4) Exclude cut with ID 3972-170212-0014-23379-0_sp1.1 from training. Duration: 23.82275
2022-12-22 22:34:09,443 WARNING [train.py:1060] (2/4) Exclude cut with ID 4860-13185-0032-76709-0 from training. Duration: 20.77
2022-12-22 22:34:17,218 INFO [train.py:894] (2/4) Epoch 9, batch 850, loss[loss=0.2034, simple_loss=0.2935, pruned_loss=0.05663, over 18497.00 frames. ], tot_loss[loss=0.214, simple_loss=0.2947, pruned_loss=0.06667, over 3661433.54 frames. ], batch size: 52, lr: 1.31e-02, grad_scale: 16.0
2022-12-22 22:34:17,290 WARNING [train.py:1060] (2/4) Exclude cut with ID 6426-64292-0017-15984-0_sp0.9 from training. Duration: 24.088875
2022-12-22 22:34:33,698 INFO [zipformer.py:660] (2/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=28911.0, num_to_drop=1, layers_to_drop={0}
2022-12-22 22:34:46,551 WARNING [train.py:1060] (2/4) Exclude cut with ID 6330-62850-0007-91323-0_sp1.1 from training. Duration: 20.4409375
2022-12-22 22:34:49,300 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 2.759e+02 4.128e+02 5.297e+02 6.595e+02 1.319e+03, threshold=1.059e+03, percent-clipped=1.0
2022-12-22 22:34:50,503 INFO [zipformer.py:660] (2/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=28922.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 22:35:01,967 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.5444, 1.1378, 2.0399, 2.9663, 1.9993, 2.2356, 0.6927, 1.9485],
       device='cuda:2'), covar=tensor([0.1756, 0.1809, 0.1265, 0.0522, 0.1110, 0.1246, 0.2375, 0.1238],
       device='cuda:2'), in_proj_covar=tensor([0.0103, 0.0113, 0.0123, 0.0118, 0.0101, 0.0130, 0.0127, 0.0107],
       device='cuda:2'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003],
       device='cuda:2')
2022-12-22 22:35:05,789 INFO [zipformer.py:660] (2/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=28933.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 22:35:21,073 INFO [zipformer.py:660] (2/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=28943.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 22:35:26,936 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.3258, 1.7642, 1.4192, 2.3193, 2.1496, 1.4249, 1.4368, 1.1908],
       device='cuda:2'), covar=tensor([0.2031, 0.1841, 0.1506, 0.0832, 0.1479, 0.1268, 0.1913, 0.1585],
       device='cuda:2'), in_proj_covar=tensor([0.0230, 0.0207, 0.0196, 0.0182, 0.0246, 0.0183, 0.0205, 0.0184],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-22 22:35:31,621 INFO [train.py:894] (2/4) Epoch 9, batch 900, loss[loss=0.2497, simple_loss=0.3224, pruned_loss=0.08853, over 18616.00 frames. ], tot_loss[loss=0.214, simple_loss=0.2945, pruned_loss=0.06678, over 3673052.54 frames. ], batch size: 176, lr: 1.31e-02, grad_scale: 16.0
2022-12-22 22:35:44,984 INFO [zipformer.py:660] (2/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=28959.0, num_to_drop=1, layers_to_drop={1}
2022-12-22 22:36:04,297 WARNING [train.py:1060] (2/4) Exclude cut with ID 2929-85685-0062-62366-0_sp0.9 from training. Duration: 22.511125
2022-12-22 22:36:04,319 WARNING [train.py:1060] (2/4) Exclude cut with ID 7395-89880-0031-39906-0 from training. Duration: 20.675
2022-12-22 22:36:23,711 INFO [scaling.py:679] (2/4) Whitening: num_groups=1, num_channels=384, metric=3.35 vs. limit=5.0
2022-12-22 22:36:33,552 INFO [zipformer.py:660] (2/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=28991.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 22:36:36,920 INFO [scaling.py:679] (2/4) Whitening: num_groups=8, num_channels=96, metric=1.22 vs. limit=2.0
2022-12-22 22:36:47,765 INFO [train.py:894] (2/4) Epoch 9, batch 950, loss[loss=0.1848, simple_loss=0.262, pruned_loss=0.05382, over 18436.00 frames. ], tot_loss[loss=0.213, simple_loss=0.2939, pruned_loss=0.06604, over 3681424.87 frames. ], batch size: 42, lr: 1.30e-02, grad_scale: 16.0
2022-12-22 22:36:51,425 INFO [zipformer.py:660] (2/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=29002.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 22:37:16,863 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([2.2211, 1.4571, 1.7767, 1.9404, 2.1141, 2.0015, 1.9852, 1.4473],
       device='cuda:2'), covar=tensor([0.1471, 0.2311, 0.1824, 0.1948, 0.1361, 0.0704, 0.2014, 0.0888],
       device='cuda:2'), in_proj_covar=tensor([0.0243, 0.0277, 0.0250, 0.0284, 0.0264, 0.0226, 0.0285, 0.0216],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-22 22:37:21,573 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 3.190e+02 4.506e+02 5.344e+02 6.616e+02 1.256e+03, threshold=1.069e+03, percent-clipped=2.0
2022-12-22 22:37:43,321 WARNING [train.py:1060] (2/4) Exclude cut with ID 6330-62850-0007-91323-0_sp0.9 from training. Duration: 24.9833125
2022-12-22 22:38:04,948 INFO [train.py:894] (2/4) Epoch 9, batch 1000, loss[loss=0.2484, simple_loss=0.3299, pruned_loss=0.08343, over 18677.00 frames. ], tot_loss[loss=0.2128, simple_loss=0.2941, pruned_loss=0.06577, over 3688210.98 frames. ], batch size: 78, lr: 1.30e-02, grad_scale: 16.0
2022-12-22 22:38:08,990 INFO [scaling.py:679] (2/4) Whitening: num_groups=8, num_channels=96, metric=1.22 vs. limit=2.0
2022-12-22 22:38:15,773 WARNING [train.py:1060] (2/4) Exclude cut with ID 5239-32139-0047-9341-0 from training. Duration: 27.14
2022-12-22 22:38:25,252 INFO [zipformer.py:660] (2/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=29063.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 22:38:30,551 WARNING [train.py:1060] (2/4) Exclude cut with ID 2929-85685-0044-62348-0 from training. Duration: 22.44
2022-12-22 22:39:15,606 INFO [scaling.py:679] (2/4) Whitening: num_groups=8, num_channels=96, metric=1.22 vs. limit=2.0
2022-12-22 22:39:22,240 INFO [train.py:894] (2/4) Epoch 9, batch 1050, loss[loss=0.1804, simple_loss=0.2557, pruned_loss=0.05251, over 18685.00 frames. ], tot_loss[loss=0.2124, simple_loss=0.2938, pruned_loss=0.06548, over 3695269.95 frames. ], batch size: 46, lr: 1.30e-02, grad_scale: 16.0
2022-12-22 22:39:51,620 WARNING [train.py:1060] (2/4) Exclude cut with ID 2929-85685-0060-62364-0_sp0.9 from training. Duration: 21.361125
2022-12-22 22:39:55,942 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 3.054e+02 4.752e+02 5.475e+02 6.695e+02 1.514e+03, threshold=1.095e+03, percent-clipped=3.0
2022-12-22 22:39:57,361 WARNING [train.py:1060] (2/4) Exclude cut with ID 2929-85685-0079-62383-0_sp1.1 from training. Duration: 27.0318125
2022-12-22 22:40:06,903 WARNING [train.py:1060] (2/4) Exclude cut with ID 5622-44585-0006-90525-0_sp0.9 from training. Duration: 28.638875
2022-12-22 22:40:23,246 WARNING [train.py:1060] (2/4) Exclude cut with ID 3340-169293-0054-76830-0 from training. Duration: 20.4
2022-12-22 22:40:27,086 INFO [zipformer.py:660] (2/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=29142.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 22:40:35,803 INFO [zipformer.py:660] (2/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=29148.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 22:40:37,604 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([3.0806, 2.3869, 1.6821, 3.0017, 2.1584, 2.2164, 2.3980, 3.6179],
       device='cuda:2'), covar=tensor([0.1494, 0.2519, 0.1497, 0.2463, 0.3080, 0.0849, 0.2517, 0.0452],
       device='cuda:2'), in_proj_covar=tensor([0.0266, 0.0257, 0.0219, 0.0331, 0.0240, 0.0206, 0.0255, 0.0179],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0003, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-22 22:40:38,490 INFO [train.py:894] (2/4) Epoch 9, batch 1100, loss[loss=0.2132, simple_loss=0.2941, pruned_loss=0.06608, over 18387.00 frames. ], tot_loss[loss=0.2116, simple_loss=0.2933, pruned_loss=0.06493, over 3699498.81 frames. ], batch size: 53, lr: 1.30e-02, grad_scale: 16.0
2022-12-22 22:40:55,876 WARNING [train.py:1060] (2/4) Exclude cut with ID 2929-85685-0071-62375-0 from training. Duration: 20.025
2022-12-22 22:40:55,892 WARNING [train.py:1060] (2/4) Exclude cut with ID 2364-131735-0112-64612-0_sp0.9 from training. Duration: 20.488875
2022-12-22 22:41:02,685 WARNING [train.py:1060] (2/4) Exclude cut with ID 2929-85685-0079-62383-0 from training. Duration: 29.735
2022-12-22 22:41:40,813 INFO [zipformer.py:660] (2/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=29190.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 22:41:49,303 INFO [zipformer.py:660] (2/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=29196.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 22:41:54,971 INFO [train.py:894] (2/4) Epoch 9, batch 1150, loss[loss=0.1727, simple_loss=0.2524, pruned_loss=0.04652, over 18533.00 frames. ], tot_loss[loss=0.2113, simple_loss=0.2924, pruned_loss=0.0651, over 3703426.08 frames. ], batch size: 44, lr: 1.30e-02, grad_scale: 16.0
2022-12-22 22:42:04,347 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([2.2387, 1.6075, 1.6814, 1.9735, 2.0517, 2.1046, 1.9924, 1.4559],
       device='cuda:2'), covar=tensor([0.1675, 0.2830, 0.2101, 0.2505, 0.1735, 0.0883, 0.2660, 0.1145],
       device='cuda:2'), in_proj_covar=tensor([0.0243, 0.0278, 0.0249, 0.0285, 0.0265, 0.0227, 0.0287, 0.0216],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-22 22:42:21,039 INFO [zipformer.py:660] (2/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=29217.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 22:42:25,344 WARNING [train.py:1060] (2/4) Exclude cut with ID 7276-92427-0014-12983-0_sp0.9 from training. Duration: 21.3055625
2022-12-22 22:42:26,826 WARNING [train.py:1060] (2/4) Exclude cut with ID 1025-75365-0008-79168-0_sp0.9 from training. Duration: 22.0666875
2022-12-22 22:42:28,686 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 2.563e+02 4.647e+02 6.106e+02 7.932e+02 1.840e+03, threshold=1.221e+03, percent-clipped=6.0
2022-12-22 22:42:50,149 INFO [zipformer.py:660] (2/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=29236.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 22:43:11,134 INFO [train.py:894] (2/4) Epoch 9, batch 1200, loss[loss=0.1905, simple_loss=0.278, pruned_loss=0.05152, over 18548.00 frames. ], tot_loss[loss=0.2106, simple_loss=0.2917, pruned_loss=0.06476, over 3705226.80 frames. ], batch size: 55, lr: 1.30e-02, grad_scale: 16.0
2022-12-22 22:44:15,704 WARNING [train.py:1060] (2/4) Exclude cut with ID 2929-85685-0062-62366-0 from training. Duration: 20.26
2022-12-22 22:44:23,450 INFO [zipformer.py:660] (2/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=29297.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 22:44:27,739 INFO [train.py:894] (2/4) Epoch 9, batch 1250, loss[loss=0.1841, simple_loss=0.2639, pruned_loss=0.05212, over 18680.00 frames. ], tot_loss[loss=0.2106, simple_loss=0.2918, pruned_loss=0.06472, over 3707523.05 frames. ], batch size: 46, lr: 1.30e-02, grad_scale: 8.0
2022-12-22 22:44:29,241 WARNING [train.py:1060] (2/4) Exclude cut with ID 5239-32139-0030-9324-0_sp0.9 from training. Duration: 21.3444375
2022-12-22 22:45:03,216 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 2.891e+02 4.534e+02 5.383e+02 6.900e+02 1.545e+03, threshold=1.077e+03, percent-clipped=4.0
2022-12-22 22:45:25,741 WARNING [train.py:1060] (2/4) Exclude cut with ID 497-129325-0061-62254-0_sp1.1 from training. Duration: 0.97725
2022-12-22 22:45:43,222 INFO [train.py:894] (2/4) Epoch 9, batch 1300, loss[loss=0.2018, simple_loss=0.2898, pruned_loss=0.05691, over 18565.00 frames. ], tot_loss[loss=0.211, simple_loss=0.2919, pruned_loss=0.06511, over 3709407.93 frames. ], batch size: 49, lr: 1.30e-02, grad_scale: 8.0
2022-12-22 22:45:55,773 INFO [zipformer.py:660] (2/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=29358.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 22:46:08,658 WARNING [train.py:1060] (2/4) Exclude cut with ID 7395-89880-0031-39906-0_sp0.9 from training. Duration: 22.97225
2022-12-22 22:46:39,059 WARNING [train.py:1060] (2/4) Exclude cut with ID 7395-89880-0047-39922-0_sp0.9 from training. Duration: 21.97775
2022-12-22 22:46:51,186 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.9275, 0.9380, 1.6661, 1.5577, 1.9615, 1.8227, 1.6047, 1.3761],
       device='cuda:2'), covar=tensor([0.1561, 0.2324, 0.1843, 0.1910, 0.1309, 0.0705, 0.1966, 0.0935],
       device='cuda:2'), in_proj_covar=tensor([0.0241, 0.0275, 0.0246, 0.0282, 0.0264, 0.0225, 0.0284, 0.0214],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-22 22:46:53,509 WARNING [train.py:1060] (2/4) Exclude cut with ID 1112-1043-0006-89194-0_sp0.9 from training. Duration: 21.8333125
2022-12-22 22:46:59,057 INFO [train.py:894] (2/4) Epoch 9, batch 1350, loss[loss=0.214, simple_loss=0.3032, pruned_loss=0.06243, over 18549.00 frames. ], tot_loss[loss=0.2106, simple_loss=0.2913, pruned_loss=0.0649, over 3710541.89 frames. ], batch size: 55, lr: 1.30e-02, grad_scale: 8.0
2022-12-22 22:47:03,872 WARNING [train.py:1060] (2/4) Exclude cut with ID 1914-133440-0031-94921-0 from training. Duration: 20.47
2022-12-22 22:47:19,807 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([0.4161, 1.6579, 2.0341, 0.6107, 1.1216, 2.1913, 1.8912, 1.7845],
       device='cuda:2'), covar=tensor([0.0587, 0.0270, 0.0284, 0.0343, 0.0360, 0.0332, 0.0203, 0.0510],
       device='cuda:2'), in_proj_covar=tensor([0.0129, 0.0148, 0.0105, 0.0125, 0.0133, 0.0116, 0.0135, 0.0136],
       device='cuda:2'), out_proj_covar=tensor([1.1032e-04, 1.2908e-04, 8.9930e-05, 1.0565e-04, 1.1396e-04, 1.0128e-04,
        1.1947e-04, 1.1831e-04], device='cuda:2')
2022-12-22 22:47:28,238 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([2.4792, 4.0763, 3.8094, 1.8738, 3.9500, 2.9461, 0.8518, 2.7983],
       device='cuda:2'), covar=tensor([0.2268, 0.0886, 0.1475, 0.3707, 0.0916, 0.1016, 0.5643, 0.1500],
       device='cuda:2'), in_proj_covar=tensor([0.0131, 0.0115, 0.0146, 0.0116, 0.0121, 0.0101, 0.0140, 0.0106],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-22 22:47:35,241 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 2.581e+02 4.720e+02 6.038e+02 7.473e+02 1.219e+03, threshold=1.208e+03, percent-clipped=1.0
2022-12-22 22:48:11,273 WARNING [train.py:1060] (2/4) Exclude cut with ID 7395-89880-0037-39912-0_sp0.9 from training. Duration: 20.67225
2022-12-22 22:48:15,741 INFO [train.py:894] (2/4) Epoch 9, batch 1400, loss[loss=0.216, simple_loss=0.277, pruned_loss=0.07752, over 18489.00 frames. ], tot_loss[loss=0.2102, simple_loss=0.2911, pruned_loss=0.06461, over 3710668.71 frames. ], batch size: 43, lr: 1.30e-02, grad_scale: 8.0
2022-12-22 22:48:31,010 WARNING [train.py:1060] (2/4) Exclude cut with ID 1914-133440-0024-94914-0_sp0.9 from training. Duration: 25.2444375
2022-12-22 22:48:46,980 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([2.5148, 2.1953, 1.6763, 2.3230, 1.8117, 1.9051, 2.0317, 2.6573],
       device='cuda:2'), covar=tensor([0.1452, 0.2411, 0.1301, 0.2279, 0.2680, 0.0819, 0.2171, 0.0526],
       device='cuda:2'), in_proj_covar=tensor([0.0269, 0.0261, 0.0221, 0.0334, 0.0244, 0.0209, 0.0258, 0.0181],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0003, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-22 22:48:50,631 INFO [scaling.py:679] (2/4) Whitening: num_groups=8, num_channels=96, metric=1.35 vs. limit=2.0
2022-12-22 22:48:54,648 INFO [scaling.py:679] (2/4) Whitening: num_groups=8, num_channels=192, metric=1.91 vs. limit=2.0
2022-12-22 22:48:55,302 WARNING [train.py:1060] (2/4) Exclude cut with ID 3340-169293-0021-76797-0_sp0.9 from training. Duration: 21.1445
2022-12-22 22:49:21,486 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.9885, 1.6644, 1.6332, 1.1031, 2.3687, 2.0602, 1.7881, 1.3898],
       device='cuda:2'), covar=tensor([0.0313, 0.0427, 0.0474, 0.0687, 0.0186, 0.0300, 0.0452, 0.0880],
       device='cuda:2'), in_proj_covar=tensor([0.0121, 0.0117, 0.0128, 0.0122, 0.0083, 0.0115, 0.0136, 0.0152],
       device='cuda:2'), out_proj_covar=tensor([0.0001, 0.0001, 0.0002, 0.0002, 0.0001, 0.0001, 0.0002, 0.0002],
       device='cuda:2')
2022-12-22 22:49:31,472 INFO [train.py:894] (2/4) Epoch 9, batch 1450, loss[loss=0.2297, simple_loss=0.3178, pruned_loss=0.07077, over 18659.00 frames. ], tot_loss[loss=0.2107, simple_loss=0.2916, pruned_loss=0.06492, over 3711546.74 frames. ], batch size: 60, lr: 1.29e-02, grad_scale: 8.0
2022-12-22 22:49:57,128 INFO [zipformer.py:660] (2/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=29517.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 22:50:05,922 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 2.999e+02 4.337e+02 5.237e+02 6.782e+02 1.195e+03, threshold=1.047e+03, percent-clipped=0.0
2022-12-22 22:50:10,113 WARNING [train.py:1060] (2/4) Exclude cut with ID 2929-85685-0079-62383-0_sp0.9 from training. Duration: 33.038875
2022-12-22 22:50:46,417 INFO [train.py:894] (2/4) Epoch 9, batch 1500, loss[loss=0.1936, simple_loss=0.2698, pruned_loss=0.05869, over 18490.00 frames. ], tot_loss[loss=0.2095, simple_loss=0.2906, pruned_loss=0.06418, over 3711676.66 frames. ], batch size: 43, lr: 1.29e-02, grad_scale: 8.0
2022-12-22 22:50:47,859 WARNING [train.py:1060] (2/4) Exclude cut with ID 6426-64291-0000-16059-0_sp0.9 from training. Duration: 20.0944375
2022-12-22 22:50:48,175 INFO [zipformer.py:660] (2/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=29551.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 22:51:02,523 WARNING [train.py:1060] (2/4) Exclude cut with ID 2929-85685-0044-62348-0_sp1.1 from training. Duration: 20.4
2022-12-22 22:51:09,784 INFO [zipformer.py:660] (2/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=29565.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 22:51:11,061 WARNING [train.py:1060] (2/4) Exclude cut with ID 6330-62851-0022-91297-0 from training. Duration: 20.085
2022-12-22 22:51:21,758 WARNING [train.py:1060] (2/4) Exclude cut with ID 4860-13185-0032-76709-0_sp0.9 from training. Duration: 23.07775
2022-12-22 22:51:49,986 INFO [zipformer.py:660] (2/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=29592.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 22:52:01,851 INFO [train.py:894] (2/4) Epoch 9, batch 1550, loss[loss=0.1765, simple_loss=0.2557, pruned_loss=0.04867, over 18521.00 frames. ], tot_loss[loss=0.2081, simple_loss=0.2896, pruned_loss=0.06332, over 3711638.01 frames. ], batch size: 44, lr: 1.29e-02, grad_scale: 8.0
2022-12-22 22:52:08,055 WARNING [train.py:1060] (2/4) Exclude cut with ID 2929-85685-0044-62348-0_sp0.9 from training. Duration: 24.9333125
2022-12-22 22:52:20,492 INFO [zipformer.py:660] (2/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=29612.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 22:52:37,791 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 2.815e+02 4.732e+02 5.774e+02 7.173e+02 1.545e+03, threshold=1.155e+03, percent-clipped=6.0
2022-12-22 22:52:51,426 WARNING [train.py:1060] (2/4) Exclude cut with ID 5118-111612-0016-124680-0_sp0.9 from training. Duration: 20.388875
2022-12-22 22:52:57,581 WARNING [train.py:1060] (2/4) Exclude cut with ID 432-122774-0017-62487-0_sp1.1 from training. Duration: 20.3590625
2022-12-22 22:53:19,293 INFO [train.py:894] (2/4) Epoch 9, batch 1600, loss[loss=0.1821, simple_loss=0.2588, pruned_loss=0.05272, over 18493.00 frames. ], tot_loss[loss=0.2084, simple_loss=0.2898, pruned_loss=0.06357, over 3711950.65 frames. ], batch size: 43, lr: 1.29e-02, grad_scale: 8.0
2022-12-22 22:53:31,312 INFO [zipformer.py:660] (2/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=29658.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 22:53:37,795 INFO [zipformer.py:660] (2/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=29662.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 22:54:07,611 WARNING [train.py:1060] (2/4) Exclude cut with ID 3557-8342-0013-54691-0_sp1.1 from training. Duration: 0.836375
2022-12-22 22:54:35,185 INFO [train.py:894] (2/4) Epoch 9, batch 1650, loss[loss=0.1876, simple_loss=0.2693, pruned_loss=0.05297, over 18685.00 frames. ], tot_loss[loss=0.2103, simple_loss=0.2906, pruned_loss=0.06499, over 3711729.01 frames. ], batch size: 48, lr: 1.29e-02, grad_scale: 8.0
2022-12-22 22:54:43,979 INFO [zipformer.py:660] (2/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=29706.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 22:54:53,377 WARNING [train.py:1060] (2/4) Exclude cut with ID 8565-290391-0049-67394-0_sp0.9 from training. Duration: 21.3166875
2022-12-22 22:55:05,127 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([2.6524, 3.6603, 3.7132, 1.4345, 3.5396, 2.8658, 0.7841, 2.5712],
       device='cuda:2'), covar=tensor([0.2291, 0.1200, 0.1523, 0.4329, 0.1231, 0.1070, 0.5809, 0.1733],
       device='cuda:2'), in_proj_covar=tensor([0.0134, 0.0118, 0.0150, 0.0118, 0.0123, 0.0102, 0.0141, 0.0108],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-22 22:55:09,546 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 2.853e+02 4.793e+02 6.223e+02 8.275e+02 2.950e+03, threshold=1.245e+03, percent-clipped=9.0
2022-12-22 22:55:09,958 INFO [zipformer.py:660] (2/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=29723.0, num_to_drop=1, layers_to_drop={2}
2022-12-22 22:55:24,125 WARNING [train.py:1060] (2/4) Exclude cut with ID 6533-399-0029-104863-0_sp0.9 from training. Duration: 22.1055625
2022-12-22 22:55:34,723 WARNING [train.py:1060] (2/4) Exclude cut with ID 7699-105389-0094-26379-0_sp1.1 from training. Duration: 21.77725
2022-12-22 22:55:35,505 INFO [scaling.py:679] (2/4) Whitening: num_groups=8, num_channels=96, metric=1.90 vs. limit=2.0
2022-12-22 22:55:43,327 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.4826, 1.1157, 1.5953, 2.8039, 1.9009, 2.1086, 0.6627, 1.8934],
       device='cuda:2'), covar=tensor([0.1778, 0.1739, 0.1534, 0.0607, 0.1245, 0.1336, 0.2338, 0.1279],
       device='cuda:2'), in_proj_covar=tensor([0.0101, 0.0111, 0.0125, 0.0119, 0.0101, 0.0128, 0.0127, 0.0106],
       device='cuda:2'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003],
       device='cuda:2')
2022-12-22 22:55:50,804 INFO [train.py:894] (2/4) Epoch 9, batch 1700, loss[loss=0.2644, simple_loss=0.322, pruned_loss=0.1034, over 18614.00 frames. ], tot_loss[loss=0.216, simple_loss=0.2947, pruned_loss=0.06861, over 3713045.11 frames. ], batch size: 179, lr: 1.29e-02, grad_scale: 8.0
2022-12-22 22:55:53,878 WARNING [train.py:1060] (2/4) Exclude cut with ID 7255-291500-0005-134304-0_sp0.9 from training. Duration: 27.8166875
2022-12-22 22:56:19,187 WARNING [train.py:1060] (2/4) Exclude cut with ID 7357-94126-0021-15852-0_sp1.1 from training. Duration: 22.5090625
2022-12-22 22:56:25,260 WARNING [train.py:1060] (2/4) Exclude cut with ID 7255-291500-0005-134304-0 from training. Duration: 25.035
2022-12-22 22:56:30,116 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.6414, 1.5283, 1.6766, 1.7231, 1.3013, 3.8300, 1.7626, 2.3614],
       device='cuda:2'), covar=tensor([0.3221, 0.1996, 0.1844, 0.1850, 0.1363, 0.0158, 0.1461, 0.0783],
       device='cuda:2'), in_proj_covar=tensor([0.0141, 0.0121, 0.0132, 0.0123, 0.0105, 0.0100, 0.0102, 0.0098],
       device='cuda:2'), out_proj_covar=tensor([0.0005, 0.0004, 0.0004, 0.0004, 0.0004, 0.0003, 0.0004, 0.0004],
       device='cuda:2')
2022-12-22 22:56:40,721 INFO [scaling.py:679] (2/4) Whitening: num_groups=8, num_channels=192, metric=2.00 vs. limit=2.0
2022-12-22 22:56:43,149 WARNING [train.py:1060] (2/4) Exclude cut with ID 774-127930-0014-10412-0_sp1.1 from training. Duration: 0.95
2022-12-22 22:57:01,217 WARNING [train.py:1060] (2/4) Exclude cut with ID 3033-130750-0096-55598-0_sp0.9 from training. Duration: 0.92225
2022-12-22 22:57:07,347 INFO [train.py:894] (2/4) Epoch 9, batch 1750, loss[loss=0.2628, simple_loss=0.3227, pruned_loss=0.1015, over 18626.00 frames. ], tot_loss[loss=0.2185, simple_loss=0.2956, pruned_loss=0.0707, over 3714375.40 frames. ], batch size: 182, lr: 1.29e-02, grad_scale: 8.0
2022-12-22 22:57:11,807 INFO [zipformer.py:660] (2/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=29802.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 22:57:28,258 WARNING [train.py:1060] (2/4) Exclude cut with ID 4511-76322-0006-80011-0 from training. Duration: 21.97
2022-12-22 22:57:42,903 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 3.618e+02 5.406e+02 6.621e+02 8.145e+02 1.955e+03, threshold=1.324e+03, percent-clipped=2.0
2022-12-22 22:57:48,786 WARNING [train.py:1060] (2/4) Exclude cut with ID 7492-105653-0055-62765-0_sp0.9 from training. Duration: 21.97225
2022-12-22 22:57:50,047 WARNING [train.py:1060] (2/4) Exclude cut with ID 453-131332-0000-47844-0_sp0.9 from training. Duration: 25.3333125
2022-12-22 22:57:58,166 INFO [scaling.py:679] (2/4) Whitening: num_groups=8, num_channels=96, metric=1.42 vs. limit=2.0
2022-12-22 22:58:00,093 WARNING [train.py:1060] (2/4) Exclude cut with ID 5172-29468-0015-19128-0_sp0.9 from training. Duration: 21.5055625
2022-12-22 22:58:10,950 WARNING [train.py:1060] (2/4) Exclude cut with ID 453-131332-0000-47844-0_sp1.1 from training. Duration: 20.72725
2022-12-22 22:58:23,412 INFO [train.py:894] (2/4) Epoch 9, batch 1800, loss[loss=0.2122, simple_loss=0.2953, pruned_loss=0.06454, over 18393.00 frames. ], tot_loss[loss=0.2221, simple_loss=0.2977, pruned_loss=0.0732, over 3713976.30 frames. ], batch size: 53, lr: 1.29e-02, grad_scale: 8.0
2022-12-22 22:58:23,721 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([4.6022, 3.8539, 3.9477, 4.4734, 4.1273, 4.0936, 4.7351, 1.3039],
       device='cuda:2'), covar=tensor([0.0764, 0.0643, 0.0530, 0.0776, 0.1414, 0.1056, 0.0480, 0.4629],
       device='cuda:2'), in_proj_covar=tensor([0.0283, 0.0192, 0.0196, 0.0193, 0.0268, 0.0226, 0.0223, 0.0246],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0001, 0.0001, 0.0001, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-22 22:58:43,666 INFO [zipformer.py:660] (2/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=29863.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 22:58:45,290 WARNING [train.py:1060] (2/4) Exclude cut with ID 8631-249866-0030-130156-0_sp0.9 from training. Duration: 26.32775
2022-12-22 22:59:17,148 WARNING [train.py:1060] (2/4) Exclude cut with ID 3867-173237-0077-144769-0 from training. Duration: 20.025
2022-12-22 22:59:22,073 WARNING [train.py:1060] (2/4) Exclude cut with ID 6709-74022-0004-86860-0_sp1.1 from training. Duration: 0.9409375
2022-12-22 22:59:22,084 WARNING [train.py:1060] (2/4) Exclude cut with ID 4757-1811-0023-62229-0_sp0.9 from training. Duration: 21.37775
2022-12-22 22:59:26,759 INFO [zipformer.py:660] (2/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=29892.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 22:59:39,024 INFO [train.py:894] (2/4) Epoch 9, batch 1850, loss[loss=0.2344, simple_loss=0.2969, pruned_loss=0.08591, over 18680.00 frames. ], tot_loss[loss=0.2262, simple_loss=0.3, pruned_loss=0.07617, over 3714064.72 frames. ], batch size: 48, lr: 1.29e-02, grad_scale: 8.0
2022-12-22 22:59:42,544 WARNING [train.py:1060] (2/4) Exclude cut with ID 1250-135782-0004-25974-0_sp0.9 from training. Duration: 21.17225
2022-12-22 22:59:42,554 WARNING [train.py:1060] (2/4) Exclude cut with ID 7357-94126-0021-15852-0_sp0.9 from training. Duration: 27.511125
2022-12-22 22:59:50,536 INFO [zipformer.py:660] (2/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=29907.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 23:00:13,996 WARNING [train.py:1060] (2/4) Exclude cut with ID 453-131332-0000-47844-0 from training. Duration: 22.8
2022-12-22 23:00:15,336 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 3.806e+02 5.568e+02 6.649e+02 8.317e+02 1.675e+03, threshold=1.330e+03, percent-clipped=4.0
2022-12-22 23:00:18,511 WARNING [train.py:1060] (2/4) Exclude cut with ID 4964-30587-0040-44509-0 from training. Duration: 22.585
2022-12-22 23:00:42,386 INFO [zipformer.py:660] (2/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=29940.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 23:00:51,746 WARNING [train.py:1060] (2/4) Exclude cut with ID 6978-92210-0001-146967-0_sp0.9 from training. Duration: 22.0166875
2022-12-22 23:00:58,027 INFO [train.py:894] (2/4) Epoch 9, batch 1900, loss[loss=0.2003, simple_loss=0.2742, pruned_loss=0.06316, over 18529.00 frames. ], tot_loss[loss=0.2284, simple_loss=0.301, pruned_loss=0.0779, over 3715218.37 frames. ], batch size: 44, lr: 1.28e-02, grad_scale: 8.0
2022-12-22 23:01:10,431 WARNING [train.py:1060] (2/4) Exclude cut with ID 7255-291500-0003-134302-0_sp1.1 from training. Duration: 24.395375
2022-12-22 23:01:17,949 WARNING [train.py:1060] (2/4) Exclude cut with ID 3488-85273-0017-41203-0_sp0.9 from training. Duration: 27.47775
2022-12-22 23:01:21,036 WARNING [train.py:1060] (2/4) Exclude cut with ID 432-122774-0017-62487-0_sp0.9 from training. Duration: 24.8833125
2022-12-22 23:01:24,106 WARNING [train.py:1060] (2/4) Exclude cut with ID 6758-72288-0033-108368-0 from training. Duration: 23.39
2022-12-22 23:01:30,093 WARNING [train.py:1060] (2/4) Exclude cut with ID 8291-282929-0007-12994-0_sp0.9 from training. Duration: 28.72225
2022-12-22 23:01:36,189 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.6575, 1.7731, 1.9019, 1.2002, 2.0811, 1.9086, 1.4448, 2.6343],
       device='cuda:2'), covar=tensor([0.1030, 0.1465, 0.1349, 0.1711, 0.0695, 0.1153, 0.2017, 0.0407],
       device='cuda:2'), in_proj_covar=tensor([0.0195, 0.0191, 0.0196, 0.0186, 0.0181, 0.0204, 0.0204, 0.0177],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-22 23:01:40,300 WARNING [train.py:1060] (2/4) Exclude cut with ID 585-294811-0110-133686-0_sp0.9 from training. Duration: 20.8944375
2022-12-22 23:01:49,826 INFO [scaling.py:679] (2/4) Whitening: num_groups=8, num_channels=192, metric=1.89 vs. limit=2.0
2022-12-22 23:01:53,171 WARNING [train.py:1060] (2/4) Exclude cut with ID 5796-66357-0007-116447-0_sp0.9 from training. Duration: 23.8444375
2022-12-22 23:02:15,625 INFO [train.py:894] (2/4) Epoch 9, batch 1950, loss[loss=0.2704, simple_loss=0.3363, pruned_loss=0.1022, over 18558.00 frames. ], tot_loss[loss=0.2287, simple_loss=0.3006, pruned_loss=0.07844, over 3715332.94 frames. ], batch size: 57, lr: 1.28e-02, grad_scale: 8.0
2022-12-22 23:02:22,798 WARNING [train.py:1060] (2/4) Exclude cut with ID 8291-282929-0007-12994-0 from training. Duration: 25.85
2022-12-22 23:02:22,805 WARNING [train.py:1060] (2/4) Exclude cut with ID 8291-282929-0023-13010-0 from training. Duration: 21.39
2022-12-22 23:02:34,551 WARNING [train.py:1060] (2/4) Exclude cut with ID 7357-94126-0014-15845-0 from training. Duration: 27.92
2022-12-22 23:02:46,598 INFO [zipformer.py:660] (2/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=30018.0, num_to_drop=1, layers_to_drop={0}
2022-12-22 23:02:49,907 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([2.2606, 1.4704, 1.8301, 1.9007, 2.0440, 1.9017, 2.0802, 1.4578],
       device='cuda:2'), covar=tensor([0.1431, 0.2453, 0.1709, 0.1914, 0.1305, 0.0731, 0.1990, 0.0897],
       device='cuda:2'), in_proj_covar=tensor([0.0244, 0.0279, 0.0247, 0.0280, 0.0264, 0.0226, 0.0284, 0.0215],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-22 23:02:53,459 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 3.837e+02 5.184e+02 6.235e+02 8.086e+02 1.649e+03, threshold=1.247e+03, percent-clipped=1.0
2022-12-22 23:03:04,148 WARNING [train.py:1060] (2/4) Exclude cut with ID 8631-249866-0039-130165-0_sp0.9 from training. Duration: 20.661125
2022-12-22 23:03:27,767 WARNING [train.py:1060] (2/4) Exclude cut with ID 7357-94126-0043-15874-0_sp0.9 from training. Duration: 20.07225
2022-12-22 23:03:35,303 INFO [train.py:894] (2/4) Epoch 9, batch 2000, loss[loss=0.2166, simple_loss=0.2776, pruned_loss=0.07777, over 18367.00 frames. ], tot_loss[loss=0.2305, simple_loss=0.3016, pruned_loss=0.07968, over 3714621.64 frames. ], batch size: 46, lr: 1.28e-02, grad_scale: 8.0
2022-12-22 23:03:35,390 WARNING [train.py:1060] (2/4) Exclude cut with ID 1085-156170-0017-128270-0 from training. Duration: 21.01
2022-12-22 23:04:35,045 INFO [scaling.py:679] (2/4) Whitening: num_groups=8, num_channels=96, metric=1.92 vs. limit=2.0
2022-12-22 23:04:43,468 WARNING [train.py:1060] (2/4) Exclude cut with ID 2195-150901-0045-59933-0 from training. Duration: 20.65
2022-12-22 23:04:51,205 WARNING [train.py:1060] (2/4) Exclude cut with ID 5796-66357-0007-116447-0 from training. Duration: 21.46
2022-12-22 23:04:52,759 INFO [train.py:894] (2/4) Epoch 9, batch 2050, loss[loss=0.1977, simple_loss=0.2654, pruned_loss=0.06498, over 18480.00 frames. ], tot_loss[loss=0.2318, simple_loss=0.3025, pruned_loss=0.08056, over 3714561.65 frames. ], batch size: 43, lr: 1.28e-02, grad_scale: 8.0
2022-12-22 23:05:10,797 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.6768, 1.6933, 1.8473, 2.1740, 2.1356, 4.5351, 1.4273, 1.8622],
       device='cuda:2'), covar=tensor([0.0945, 0.1687, 0.1032, 0.0975, 0.1339, 0.0194, 0.1430, 0.1466],
       device='cuda:2'), in_proj_covar=tensor([0.0075, 0.0085, 0.0077, 0.0076, 0.0094, 0.0073, 0.0087, 0.0078],
       device='cuda:2'), out_proj_covar=tensor([0.0003, 0.0004, 0.0003, 0.0003, 0.0004, 0.0003, 0.0004, 0.0003],
       device='cuda:2')
2022-12-22 23:05:27,811 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 3.734e+02 6.048e+02 7.232e+02 9.558e+02 2.262e+03, threshold=1.446e+03, percent-clipped=10.0
2022-12-22 23:05:31,165 INFO [zipformer.py:660] (2/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=30125.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 23:05:36,338 WARNING [train.py:1060] (2/4) Exclude cut with ID 3557-8342-0013-54691-0 from training. Duration: 0.92
2022-12-22 23:05:42,563 WARNING [train.py:1060] (2/4) Exclude cut with ID 8291-282929-0023-13010-0_sp0.9 from training. Duration: 23.7666875
2022-12-22 23:05:49,698 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.4728, 1.8566, 1.3049, 2.2165, 2.5838, 1.4143, 1.5240, 1.1625],
       device='cuda:2'), covar=tensor([0.2000, 0.1698, 0.1584, 0.0884, 0.1257, 0.1247, 0.1867, 0.1585],
       device='cuda:2'), in_proj_covar=tensor([0.0235, 0.0207, 0.0198, 0.0184, 0.0251, 0.0184, 0.0207, 0.0186],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-22 23:06:09,349 INFO [train.py:894] (2/4) Epoch 9, batch 2100, loss[loss=0.2477, simple_loss=0.3188, pruned_loss=0.08832, over 18616.00 frames. ], tot_loss[loss=0.2324, simple_loss=0.3029, pruned_loss=0.08096, over 3714909.31 frames. ], batch size: 53, lr: 1.28e-02, grad_scale: 8.0
2022-12-22 23:06:17,343 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.5900, 1.4854, 1.3249, 2.0124, 1.5263, 3.1368, 1.1590, 1.3483],
       device='cuda:2'), covar=tensor([0.1195, 0.2408, 0.1527, 0.1103, 0.1933, 0.0327, 0.2060, 0.2266],
       device='cuda:2'), in_proj_covar=tensor([0.0075, 0.0085, 0.0077, 0.0076, 0.0094, 0.0073, 0.0086, 0.0078],
       device='cuda:2'), out_proj_covar=tensor([0.0003, 0.0004, 0.0003, 0.0003, 0.0004, 0.0003, 0.0004, 0.0003],
       device='cuda:2')
2022-12-22 23:06:19,809 WARNING [train.py:1060] (2/4) Exclude cut with ID 8544-281189-0060-101339-0_sp0.9 from training. Duration: 20.861125
2022-12-22 23:06:21,290 INFO [zipformer.py:660] (2/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=30158.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 23:06:29,781 WARNING [train.py:1060] (2/4) Exclude cut with ID 3488-65654-0031-41259-0_sp0.9 from training. Duration: 22.711125
2022-12-22 23:07:04,566 INFO [zipformer.py:660] (2/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=30186.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 23:07:13,123 WARNING [train.py:1060] (2/4) Exclude cut with ID 6951-79737-0043-132310-0_sp1.1 from training. Duration: 22.986375
2022-12-22 23:07:24,961 INFO [train.py:894] (2/4) Epoch 9, batch 2150, loss[loss=0.2983, simple_loss=0.3488, pruned_loss=0.1239, over 18610.00 frames. ], tot_loss[loss=0.2334, simple_loss=0.3035, pruned_loss=0.08163, over 3715287.55 frames. ], batch size: 178, lr: 1.28e-02, grad_scale: 8.0
2022-12-22 23:07:29,452 WARNING [train.py:1060] (2/4) Exclude cut with ID 8040-260924-0003-80960-0_sp0.9 from training. Duration: 22.07225
2022-12-22 23:07:33,739 WARNING [train.py:1060] (2/4) Exclude cut with ID 7699-105389-0045-26330-0_sp0.9 from training. Duration: 20.3055625
2022-12-22 23:07:35,568 INFO [zipformer.py:660] (2/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=30207.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 23:07:36,605 WARNING [train.py:1060] (2/4) Exclude cut with ID 6356-271890-0060-94317-0_sp0.9 from training. Duration: 20.72225
2022-12-22 23:07:38,554 INFO [zipformer.py:660] (2/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=30209.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 23:07:55,147 WARNING [train.py:1060] (2/4) Exclude cut with ID 3488-85273-0017-41203-0_sp1.1 from training. Duration: 22.4818125
2022-12-22 23:07:59,366 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 2.960e+02 5.109e+02 5.943e+02 7.734e+02 1.669e+03, threshold=1.189e+03, percent-clipped=3.0
2022-12-22 23:08:20,635 WARNING [train.py:1060] (2/4) Exclude cut with ID 4964-30587-0040-44509-0_sp0.9 from training. Duration: 25.0944375
2022-12-22 23:08:25,297 WARNING [train.py:1060] (2/4) Exclude cut with ID 6533-399-0047-104881-0 from training. Duration: 21.515
2022-12-22 23:08:27,085 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.8505, 1.6007, 1.8885, 1.8709, 1.6744, 4.6424, 1.9671, 2.4608],
       device='cuda:2'), covar=tensor([0.3252, 0.2097, 0.1901, 0.1890, 0.1337, 0.0130, 0.1469, 0.0922],
       device='cuda:2'), in_proj_covar=tensor([0.0142, 0.0121, 0.0132, 0.0122, 0.0106, 0.0101, 0.0100, 0.0097],
       device='cuda:2'), out_proj_covar=tensor([0.0005, 0.0004, 0.0005, 0.0004, 0.0004, 0.0003, 0.0004, 0.0004],
       device='cuda:2')
2022-12-22 23:08:32,203 WARNING [train.py:1060] (2/4) Exclude cut with ID 7357-94126-0009-15840-0_sp0.9 from training. Duration: 27.02225
2022-12-22 23:08:36,572 WARNING [train.py:1060] (2/4) Exclude cut with ID 432-122774-0010-62480-0_sp0.9 from training. Duration: 22.22225
2022-12-22 23:08:40,701 INFO [train.py:894] (2/4) Epoch 9, batch 2200, loss[loss=0.2345, simple_loss=0.3076, pruned_loss=0.08069, over 18653.00 frames. ], tot_loss[loss=0.2352, simple_loss=0.3044, pruned_loss=0.08299, over 3714854.87 frames. ], batch size: 97, lr: 1.28e-02, grad_scale: 8.0
2022-12-22 23:08:42,114 WARNING [train.py:1060] (2/4) Exclude cut with ID 4964-30587-0085-44554-0_sp0.9 from training. Duration: 20.85
2022-12-22 23:08:48,429 INFO [zipformer.py:660] (2/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=30255.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 23:09:11,314 INFO [zipformer.py:660] (2/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=30270.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 23:09:17,715 WARNING [train.py:1060] (2/4) Exclude cut with ID 4295-39940-0007-92567-0 from training. Duration: 21.54
2022-12-22 23:09:19,546 INFO [scaling.py:679] (2/4) Whitening: num_groups=8, num_channels=192, metric=1.80 vs. limit=2.0
2022-12-22 23:09:21,719 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([2.2519, 2.0082, 1.6619, 0.9889, 2.7450, 2.2414, 2.0120, 1.4462],
       device='cuda:2'), covar=tensor([0.0292, 0.0329, 0.0499, 0.0720, 0.0153, 0.0287, 0.0423, 0.0821],
       device='cuda:2'), in_proj_covar=tensor([0.0115, 0.0113, 0.0124, 0.0117, 0.0079, 0.0112, 0.0131, 0.0146],
       device='cuda:2'), out_proj_covar=tensor([1.4216e-04, 1.3911e-04, 1.5181e-04, 1.4391e-04, 9.9723e-05, 1.3494e-04,
        1.6058e-04, 1.7920e-04], device='cuda:2')
2022-12-22 23:09:22,840 WARNING [train.py:1060] (2/4) Exclude cut with ID 4964-30587-0040-44509-0_sp1.1 from training. Duration: 20.5318125
2022-12-22 23:09:32,495 WARNING [train.py:1060] (2/4) Exclude cut with ID 7255-291500-0012-134311-0_sp0.9 from training. Duration: 21.9333125
2022-12-22 23:09:57,724 INFO [train.py:894] (2/4) Epoch 9, batch 2250, loss[loss=0.2175, simple_loss=0.287, pruned_loss=0.07399, over 18433.00 frames. ], tot_loss[loss=0.2343, simple_loss=0.3035, pruned_loss=0.08259, over 3713798.99 frames. ], batch size: 48, lr: 1.28e-02, grad_scale: 8.0
2022-12-22 23:10:00,518 INFO [scaling.py:679] (2/4) Whitening: num_groups=8, num_channels=96, metric=1.58 vs. limit=2.0
2022-12-22 23:10:20,947 WARNING [train.py:1060] (2/4) Exclude cut with ID 8631-249866-0025-130151-0_sp0.9 from training. Duration: 21.7944375
2022-12-22 23:10:25,703 INFO [zipformer.py:660] (2/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=30318.0, num_to_drop=1, layers_to_drop={2}
2022-12-22 23:10:33,675 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 3.779e+02 5.688e+02 7.340e+02 9.327e+02 2.181e+03, threshold=1.468e+03, percent-clipped=8.0
2022-12-22 23:10:33,733 WARNING [train.py:1060] (2/4) Exclude cut with ID 8291-282929-0002-12989-0_sp0.9 from training. Duration: 22.4666875
2022-12-22 23:10:41,185 WARNING [train.py:1060] (2/4) Exclude cut with ID 6121-9014-0076-24124-0 from training. Duration: 21.635
2022-12-22 23:10:47,541 WARNING [train.py:1060] (2/4) Exclude cut with ID 6121-9014-0076-24124-0_sp0.9 from training. Duration: 24.038875
2022-12-22 23:11:06,423 INFO [scaling.py:679] (2/4) Whitening: num_groups=1, num_channels=384, metric=4.23 vs. limit=5.0
2022-12-22 23:11:14,217 INFO [train.py:894] (2/4) Epoch 9, batch 2300, loss[loss=0.2052, simple_loss=0.269, pruned_loss=0.07071, over 18561.00 frames. ], tot_loss[loss=0.2334, simple_loss=0.3027, pruned_loss=0.08209, over 3714390.64 frames. ], batch size: 44, lr: 1.28e-02, grad_scale: 8.0
2022-12-22 23:11:28,645 WARNING [train.py:1060] (2/4) Exclude cut with ID 7255-291500-0009-134308-0_sp1.1 from training. Duration: 21.786375
2022-12-22 23:11:30,414 INFO [zipformer.py:660] (2/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=30360.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 23:11:39,399 INFO [zipformer.py:660] (2/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=30366.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 23:11:42,260 WARNING [train.py:1060] (2/4) Exclude cut with ID 8291-282929-0002-12989-0 from training. Duration: 20.22
2022-12-22 23:12:29,575 INFO [train.py:894] (2/4) Epoch 9, batch 2350, loss[loss=0.2188, simple_loss=0.2875, pruned_loss=0.07502, over 18531.00 frames. ], tot_loss[loss=0.2332, simple_loss=0.3025, pruned_loss=0.08197, over 3714466.43 frames. ], batch size: 41, lr: 1.28e-02, grad_scale: 8.0
2022-12-22 23:13:03,132 INFO [zipformer.py:660] (2/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=30421.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 23:13:06,427 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 3.215e+02 5.140e+02 6.090e+02 7.966e+02 3.738e+03, threshold=1.218e+03, percent-clipped=4.0
2022-12-22 23:13:40,548 WARNING [train.py:1060] (2/4) Exclude cut with ID 6951-79737-0043-132310-0 from training. Duration: 25.285
2022-12-22 23:13:46,558 INFO [train.py:894] (2/4) Epoch 9, batch 2400, loss[loss=0.2825, simple_loss=0.3327, pruned_loss=0.1161, over 18605.00 frames. ], tot_loss[loss=0.2327, simple_loss=0.3021, pruned_loss=0.08166, over 3715243.62 frames. ], batch size: 179, lr: 1.27e-02, grad_scale: 8.0
2022-12-22 23:13:59,164 INFO [zipformer.py:660] (2/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=30458.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 23:14:04,935 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([2.1724, 2.0732, 1.7178, 1.2876, 2.9224, 2.5832, 2.1196, 1.4026],
       device='cuda:2'), covar=tensor([0.0356, 0.0356, 0.0510, 0.0684, 0.0132, 0.0230, 0.0426, 0.0938],
       device='cuda:2'), in_proj_covar=tensor([0.0114, 0.0113, 0.0124, 0.0117, 0.0080, 0.0112, 0.0131, 0.0146],
       device='cuda:2'), out_proj_covar=tensor([0.0001, 0.0001, 0.0002, 0.0001, 0.0001, 0.0001, 0.0002, 0.0002],
       device='cuda:2')
2022-12-22 23:14:34,764 INFO [zipformer.py:660] (2/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=30481.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 23:14:46,294 WARNING [train.py:1060] (2/4) Exclude cut with ID 811-130148-0001-63453-0_sp0.9 from training. Duration: 20.861125
2022-12-22 23:14:52,964 INFO [scaling.py:679] (2/4) Whitening: num_groups=8, num_channels=96, metric=1.37 vs. limit=2.0
2022-12-22 23:15:03,529 INFO [train.py:894] (2/4) Epoch 9, batch 2450, loss[loss=0.2208, simple_loss=0.3006, pruned_loss=0.0705, over 18562.00 frames. ], tot_loss[loss=0.2309, simple_loss=0.301, pruned_loss=0.08036, over 3714753.54 frames. ], batch size: 57, lr: 1.27e-02, grad_scale: 8.0
2022-12-22 23:15:11,081 WARNING [train.py:1060] (2/4) Exclude cut with ID 6010-56788-0055-90261-0 from training. Duration: 20.88
2022-12-22 23:15:12,704 INFO [zipformer.py:660] (2/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=30506.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 23:15:14,296 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([0.3931, 1.6754, 1.9886, 0.5952, 1.1408, 2.2331, 1.8915, 1.6255],
       device='cuda:2'), covar=tensor([0.0685, 0.0345, 0.0293, 0.0408, 0.0387, 0.0361, 0.0237, 0.0600],
       device='cuda:2'), in_proj_covar=tensor([0.0132, 0.0153, 0.0106, 0.0128, 0.0136, 0.0119, 0.0139, 0.0141],
       device='cuda:2'), out_proj_covar=tensor([1.1240e-04, 1.3187e-04, 8.9533e-05, 1.0709e-04, 1.1574e-04, 1.0294e-04,
        1.2203e-04, 1.2152e-04], device='cuda:2')
2022-12-22 23:15:24,848 INFO [scaling.py:679] (2/4) Whitening: num_groups=1, num_channels=384, metric=3.27 vs. limit=5.0
2022-12-22 23:15:39,226 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 3.497e+02 5.250e+02 6.536e+02 8.030e+02 2.143e+03, threshold=1.307e+03, percent-clipped=5.0
2022-12-22 23:15:44,719 WARNING [train.py:1060] (2/4) Exclude cut with ID 7357-94126-0045-15876-0_sp0.9 from training. Duration: 23.4166875
2022-12-22 23:16:20,845 INFO [train.py:894] (2/4) Epoch 9, batch 2500, loss[loss=0.2102, simple_loss=0.293, pruned_loss=0.06375, over 18573.00 frames. ], tot_loss[loss=0.2305, simple_loss=0.3008, pruned_loss=0.08008, over 3714279.62 frames. ], batch size: 51, lr: 1.27e-02, grad_scale: 8.0
2022-12-22 23:16:44,682 INFO [zipformer.py:660] (2/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=30565.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 23:16:57,301 WARNING [train.py:1060] (2/4) Exclude cut with ID 7255-291500-0006-134305-0 from training. Duration: 21.24
2022-12-22 23:16:58,679 WARNING [train.py:1060] (2/4) Exclude cut with ID 6533-399-0047-104881-0_sp0.9 from training. Duration: 23.9055625
2022-12-22 23:17:33,688 WARNING [train.py:1060] (2/4) Exclude cut with ID 6758-72288-0033-108368-0_sp0.9 from training. Duration: 25.988875
2022-12-22 23:17:35,717 INFO [train.py:894] (2/4) Epoch 9, batch 2550, loss[loss=0.2464, simple_loss=0.3067, pruned_loss=0.09299, over 18462.00 frames. ], tot_loss[loss=0.2291, simple_loss=0.2991, pruned_loss=0.07954, over 3713110.23 frames. ], batch size: 50, lr: 1.27e-02, grad_scale: 8.0
2022-12-22 23:17:40,987 WARNING [train.py:1060] (2/4) Exclude cut with ID 7255-291500-0001-134300-0_sp0.9 from training. Duration: 20.67225
2022-12-22 23:17:54,682 INFO [scaling.py:679] (2/4) Whitening: num_groups=8, num_channels=96, metric=1.20 vs. limit=2.0
2022-12-22 23:18:10,866 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 3.281e+02 5.961e+02 7.202e+02 9.301e+02 2.076e+03, threshold=1.440e+03, percent-clipped=8.0
2022-12-22 23:18:27,874 WARNING [train.py:1060] (2/4) Exclude cut with ID 3488-85273-0038-41224-0 from training. Duration: 20.34
2022-12-22 23:18:41,686 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([0.0490, 1.0282, 1.2096, 0.2411, 0.5195, 1.2766, 1.2085, 1.1569],
       device='cuda:2'), covar=tensor([0.0617, 0.0293, 0.0286, 0.0368, 0.0445, 0.0405, 0.0232, 0.0523],
       device='cuda:2'), in_proj_covar=tensor([0.0133, 0.0153, 0.0105, 0.0128, 0.0136, 0.0120, 0.0139, 0.0141],
       device='cuda:2'), out_proj_covar=tensor([1.1286e-04, 1.3236e-04, 8.9299e-05, 1.0701e-04, 1.1577e-04, 1.0316e-04,
        1.2203e-04, 1.2087e-04], device='cuda:2')
2022-12-22 23:18:53,713 INFO [train.py:894] (2/4) Epoch 9, batch 2600, loss[loss=0.2232, simple_loss=0.3023, pruned_loss=0.0721, over 18543.00 frames. ], tot_loss[loss=0.2296, simple_loss=0.2998, pruned_loss=0.07966, over 3714121.12 frames. ], batch size: 55, lr: 1.27e-02, grad_scale: 8.0
2022-12-22 23:19:01,943 INFO [zipformer.py:660] (2/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=30655.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 23:19:40,495 WARNING [train.py:1060] (2/4) Exclude cut with ID 7357-94126-0026-15857-0_sp0.9 from training. Duration: 25.061125
2022-12-22 23:19:53,541 WARNING [train.py:1060] (2/4) Exclude cut with ID 3033-130750-0096-55598-0 from training. Duration: 0.83
2022-12-22 23:20:11,067 INFO [train.py:894] (2/4) Epoch 9, batch 2650, loss[loss=0.2355, simple_loss=0.3076, pruned_loss=0.08173, over 18524.00 frames. ], tot_loss[loss=0.229, simple_loss=0.2996, pruned_loss=0.07916, over 3714938.33 frames. ], batch size: 69, lr: 1.27e-02, grad_scale: 8.0
2022-12-22 23:20:18,702 WARNING [train.py:1060] (2/4) Exclude cut with ID 3488-85273-0017-41203-0 from training. Duration: 24.73
2022-12-22 23:20:31,038 WARNING [train.py:1060] (2/4) Exclude cut with ID 7255-291500-0009-134308-0 from training. Duration: 23.965
2022-12-22 23:20:35,386 INFO [zipformer.py:660] (2/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=30716.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 23:20:35,684 INFO [zipformer.py:660] (2/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=30716.0, num_to_drop=1, layers_to_drop={3}
2022-12-22 23:20:41,061 WARNING [train.py:1060] (2/4) Exclude cut with ID 6978-92210-0030-146996-0_sp0.9 from training. Duration: 22.088875
2022-12-22 23:20:45,423 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 3.321e+02 5.727e+02 6.778e+02 8.704e+02 1.736e+03, threshold=1.356e+03, percent-clipped=1.0
2022-12-22 23:20:54,557 WARNING [train.py:1060] (2/4) Exclude cut with ID 7255-291500-0006-134305-0_sp0.9 from training. Duration: 23.6
2022-12-22 23:21:27,279 INFO [train.py:894] (2/4) Epoch 9, batch 2700, loss[loss=0.2453, simple_loss=0.3121, pruned_loss=0.08926, over 18551.00 frames. ], tot_loss[loss=0.2297, simple_loss=0.3001, pruned_loss=0.07967, over 3713939.24 frames. ], batch size: 69, lr: 1.27e-02, grad_scale: 8.0
2022-12-22 23:21:34,006 INFO [scaling.py:679] (2/4) Whitening: num_groups=8, num_channels=192, metric=1.86 vs. limit=2.0
2022-12-22 23:22:14,267 INFO [zipformer.py:660] (2/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=30781.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 23:22:40,573 WARNING [train.py:1060] (2/4) Exclude cut with ID 8291-282929-0024-13011-0 from training. Duration: 23.795
2022-12-22 23:22:43,530 INFO [train.py:894] (2/4) Epoch 9, batch 2750, loss[loss=0.189, simple_loss=0.265, pruned_loss=0.05643, over 18588.00 frames. ], tot_loss[loss=0.2298, simple_loss=0.2996, pruned_loss=0.07995, over 3713669.40 frames. ], batch size: 49, lr: 1.27e-02, grad_scale: 8.0
2022-12-22 23:22:56,853 WARNING [train.py:1060] (2/4) Exclude cut with ID 8631-249866-0030-130156-0_sp1.1 from training. Duration: 21.5409375
2022-12-22 23:22:59,734 WARNING [train.py:1060] (2/4) Exclude cut with ID 6978-92210-0019-146985-0_sp0.9 from training. Duration: 24.97775
2022-12-22 23:23:11,899 WARNING [train.py:1060] (2/4) Exclude cut with ID 1085-156170-0017-128270-0_sp0.9 from training. Duration: 23.3444375
2022-12-22 23:23:17,511 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 4.042e+02 6.530e+02 8.093e+02 1.038e+03 2.043e+03, threshold=1.619e+03, percent-clipped=4.0
2022-12-22 23:23:24,807 INFO [scaling.py:679] (2/4) Whitening: num_groups=1, num_channels=384, metric=3.23 vs. limit=5.0
2022-12-22 23:23:27,009 INFO [zipformer.py:660] (2/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=30829.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 23:23:33,539 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.3736, 1.7352, 1.3358, 2.0144, 2.2489, 1.4984, 1.2820, 1.1425],
       device='cuda:2'), covar=tensor([0.1881, 0.1587, 0.1483, 0.0936, 0.1052, 0.1073, 0.1739, 0.1428],
       device='cuda:2'), in_proj_covar=tensor([0.0236, 0.0210, 0.0201, 0.0187, 0.0252, 0.0186, 0.0211, 0.0188],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-22 23:23:36,504 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.4177, 1.4604, 1.3991, 1.4102, 1.3018, 3.2322, 1.5824, 2.0983],
       device='cuda:2'), covar=tensor([0.4744, 0.3051, 0.2712, 0.2886, 0.1527, 0.0352, 0.1745, 0.1003],
       device='cuda:2'), in_proj_covar=tensor([0.0141, 0.0121, 0.0132, 0.0122, 0.0105, 0.0101, 0.0100, 0.0097],
       device='cuda:2'), out_proj_covar=tensor([0.0005, 0.0004, 0.0005, 0.0004, 0.0004, 0.0003, 0.0004, 0.0004],
       device='cuda:2')
2022-12-22 23:23:40,656 WARNING [train.py:1060] (2/4) Exclude cut with ID 6010-56788-0055-90261-0_sp0.9 from training. Duration: 23.2
2022-12-22 23:23:47,423 WARNING [train.py:1060] (2/4) Exclude cut with ID 5653-46179-0060-117930-0_sp0.9 from training. Duration: 21.17225
2022-12-22 23:23:59,697 INFO [train.py:894] (2/4) Epoch 9, batch 2800, loss[loss=0.2071, simple_loss=0.2726, pruned_loss=0.07085, over 18427.00 frames. ], tot_loss[loss=0.2305, simple_loss=0.3004, pruned_loss=0.08032, over 3713657.07 frames. ], batch size: 42, lr: 1.27e-02, grad_scale: 8.0
2022-12-22 23:24:07,355 WARNING [train.py:1060] (2/4) Exclude cut with ID 7255-291500-0008-134307-0_sp0.9 from training. Duration: 24.6555625
2022-12-22 23:24:21,977 INFO [zipformer.py:660] (2/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=30865.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 23:24:40,671 INFO [scaling.py:679] (2/4) Whitening: num_groups=8, num_channels=96, metric=1.36 vs. limit=2.0
2022-12-22 23:24:42,764 INFO [scaling.py:679] (2/4) Whitening: num_groups=8, num_channels=192, metric=2.02 vs. limit=2.0
2022-12-22 23:25:06,020 WARNING [train.py:1060] (2/4) Exclude cut with ID 3488-65654-0031-41259-0 from training. Duration: 20.44
2022-12-22 23:25:16,618 INFO [train.py:894] (2/4) Epoch 9, batch 2850, loss[loss=0.2196, simple_loss=0.2818, pruned_loss=0.07877, over 18698.00 frames. ], tot_loss[loss=0.2282, simple_loss=0.2988, pruned_loss=0.07876, over 3713348.39 frames. ], batch size: 46, lr: 1.27e-02, grad_scale: 8.0
2022-12-22 23:25:18,142 WARNING [train.py:1060] (2/4) Exclude cut with ID 6951-79737-0018-132285-0_sp0.9 from training. Duration: 23.45
2022-12-22 23:25:21,365 INFO [zipformer.py:660] (2/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=30903.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 23:25:37,113 INFO [zipformer.py:660] (2/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=30913.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 23:25:40,760 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.9643, 1.7577, 2.1912, 2.3324, 1.9642, 4.9929, 2.0240, 2.6434],
       device='cuda:2'), covar=tensor([0.3193, 0.2207, 0.1786, 0.1729, 0.1159, 0.0108, 0.1354, 0.0828],
       device='cuda:2'), in_proj_covar=tensor([0.0141, 0.0122, 0.0132, 0.0124, 0.0106, 0.0101, 0.0101, 0.0097],
       device='cuda:2'), out_proj_covar=tensor([0.0005, 0.0004, 0.0005, 0.0004, 0.0004, 0.0003, 0.0004, 0.0004],
       device='cuda:2')
2022-12-22 23:25:47,873 WARNING [train.py:1060] (2/4) Exclude cut with ID 6945-60535-0076-12784-0_sp0.9 from training. Duration: 20.52225
2022-12-22 23:25:52,777 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 3.339e+02 5.387e+02 6.998e+02 8.710e+02 1.625e+03, threshold=1.400e+03, percent-clipped=1.0
2022-12-22 23:25:55,117 WARNING [train.py:1060] (2/4) Exclude cut with ID 7255-291500-0008-134307-0 from training. Duration: 22.19
2022-12-22 23:25:55,843 INFO [scaling.py:679] (2/4) Whitening: num_groups=1, num_channels=384, metric=3.08 vs. limit=5.0
2022-12-22 23:25:56,935 INFO [zipformer.py:660] (2/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=30925.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 23:26:06,249 WARNING [train.py:1060] (2/4) Exclude cut with ID 7357-94126-0014-15845-0_sp1.1 from training. Duration: 25.3818125
2022-12-22 23:26:10,613 INFO [zipformer.py:660] (2/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=30934.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 23:26:16,520 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.7236, 2.3951, 1.6391, 2.7327, 3.2592, 1.6587, 2.0169, 1.3845],
       device='cuda:2'), covar=tensor([0.1771, 0.1497, 0.1395, 0.0789, 0.1309, 0.1100, 0.1701, 0.1400],
       device='cuda:2'), in_proj_covar=tensor([0.0234, 0.0209, 0.0200, 0.0185, 0.0251, 0.0187, 0.0212, 0.0188],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-22 23:26:22,053 WARNING [train.py:1060] (2/4) Exclude cut with ID 6951-79737-0043-132310-0_sp0.9 from training. Duration: 28.0944375
2022-12-22 23:26:29,384 WARNING [train.py:1060] (2/4) Exclude cut with ID 2195-150901-0045-59933-0_sp0.9 from training. Duration: 22.9444375
2022-12-22 23:26:33,902 INFO [train.py:894] (2/4) Epoch 9, batch 2900, loss[loss=0.2661, simple_loss=0.3262, pruned_loss=0.103, over 18722.00 frames. ], tot_loss[loss=0.2284, simple_loss=0.2991, pruned_loss=0.07883, over 3713543.88 frames. ], batch size: 52, lr: 1.26e-02, grad_scale: 8.0
2022-12-22 23:26:36,709 WARNING [train.py:1060] (2/4) Exclude cut with ID 8291-282929-0024-13011-0_sp1.1 from training. Duration: 21.6318125
2022-12-22 23:26:55,080 WARNING [train.py:1060] (2/4) Exclude cut with ID 8631-249866-0030-130156-0 from training. Duration: 23.695
2022-12-22 23:26:55,421 INFO [zipformer.py:660] (2/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=30964.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 23:27:22,697 WARNING [train.py:1060] (2/4) Exclude cut with ID 7699-105389-0094-26379-0 from training. Duration: 23.955
2022-12-22 23:27:28,902 INFO [zipformer.py:660] (2/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=30986.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 23:27:41,679 INFO [zipformer.py:660] (2/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=30995.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 23:27:48,925 INFO [train.py:894] (2/4) Epoch 9, batch 2950, loss[loss=0.2235, simple_loss=0.2948, pruned_loss=0.07606, over 18603.00 frames. ], tot_loss[loss=0.2288, simple_loss=0.2997, pruned_loss=0.07891, over 3713486.38 frames. ], batch size: 78, lr: 1.26e-02, grad_scale: 8.0
2022-12-22 23:27:52,885 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([4.7630, 4.1558, 4.3334, 4.6315, 4.4366, 4.3201, 4.8289, 2.5891],
       device='cuda:2'), covar=tensor([0.0559, 0.0473, 0.0461, 0.0608, 0.1003, 0.0769, 0.0450, 0.3189],
       device='cuda:2'), in_proj_covar=tensor([0.0289, 0.0193, 0.0201, 0.0199, 0.0269, 0.0228, 0.0228, 0.0249],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0001, 0.0001, 0.0001, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-22 23:27:53,152 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.9910, 1.9590, 1.5905, 1.8833, 1.9395, 2.1123, 2.1962, 1.9000],
       device='cuda:2'), covar=tensor([0.0297, 0.0260, 0.0362, 0.0208, 0.0305, 0.0301, 0.0235, 0.0279],
       device='cuda:2'), in_proj_covar=tensor([0.0085, 0.0116, 0.0142, 0.0123, 0.0107, 0.0105, 0.0090, 0.0118],
       device='cuda:2'), out_proj_covar=tensor([7.4356e-05, 9.9855e-05, 1.2748e-04, 1.0603e-04, 9.4651e-05, 8.7629e-05,
        7.7475e-05, 1.0024e-04], device='cuda:2')
2022-12-22 23:27:54,396 WARNING [train.py:1060] (2/4) Exclude cut with ID 8291-282929-0024-13011-0_sp0.9 from training. Duration: 26.438875
2022-12-22 23:28:07,228 INFO [zipformer.py:660] (2/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=31011.0, num_to_drop=1, layers_to_drop={2}
2022-12-22 23:28:07,479 INFO [zipformer.py:660] (2/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=31011.0, num_to_drop=1, layers_to_drop={0}
2022-12-22 23:28:15,969 INFO [zipformer.py:660] (2/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=31016.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 23:28:27,057 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 3.802e+02 5.929e+02 7.348e+02 8.660e+02 3.219e+03, threshold=1.470e+03, percent-clipped=1.0
2022-12-22 23:28:29,084 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([2.5676, 2.1413, 1.5132, 1.1256, 3.1551, 2.8196, 2.0964, 1.5733],
       device='cuda:2'), covar=tensor([0.0332, 0.0362, 0.0635, 0.0789, 0.0132, 0.0275, 0.0458, 0.0870],
       device='cuda:2'), in_proj_covar=tensor([0.0117, 0.0116, 0.0126, 0.0119, 0.0083, 0.0116, 0.0135, 0.0148],
       device='cuda:2'), out_proj_covar=tensor([0.0001, 0.0001, 0.0002, 0.0001, 0.0001, 0.0001, 0.0002, 0.0002],
       device='cuda:2')
2022-12-22 23:28:39,387 WARNING [train.py:1060] (2/4) Exclude cut with ID 7699-105389-0021-26306-0_sp0.9 from training. Duration: 21.2444375
2022-12-22 23:28:41,051 WARNING [train.py:1060] (2/4) Exclude cut with ID 7357-94126-0014-15845-0_sp0.9 from training. Duration: 31.02225
2022-12-22 23:28:49,576 WARNING [train.py:1060] (2/4) Exclude cut with ID 432-122774-0017-62487-0 from training. Duration: 22.395
2022-12-22 23:29:06,540 INFO [train.py:894] (2/4) Epoch 9, batch 3000, loss[loss=0.1937, simple_loss=0.2758, pruned_loss=0.05583, over 18600.00 frames. ], tot_loss[loss=0.2296, simple_loss=0.3006, pruned_loss=0.07928, over 3713256.94 frames. ], batch size: 53, lr: 1.26e-02, grad_scale: 8.0
2022-12-22 23:29:06,540 INFO [train.py:919] (2/4) Computing validation loss
2022-12-22 23:29:17,603 INFO [train.py:928] (2/4) Epoch 9, validation: loss=0.1783, simple_loss=0.2774, pruned_loss=0.03963, over 944034.00 frames. 
2022-12-22 23:29:17,604 INFO [train.py:929] (2/4) Maximum memory allocated so far is 24680MB
2022-12-22 23:29:17,630 WARNING [train.py:1060] (2/4) Exclude cut with ID 7357-94126-0045-15876-0 from training. Duration: 21.075
2022-12-22 23:29:23,233 WARNING [train.py:1060] (2/4) Exclude cut with ID 6482-98857-0025-147532-0_sp0.9 from training. Duration: 20.0055625
2022-12-22 23:29:23,239 WARNING [train.py:1060] (2/4) Exclude cut with ID 6951-79737-0037-132304-0_sp0.9 from training. Duration: 22.05
2022-12-22 23:29:23,253 WARNING [train.py:1060] (2/4) Exclude cut with ID 7255-291500-0003-134302-0 from training. Duration: 26.8349375
2022-12-22 23:29:26,668 WARNING [train.py:1060] (2/4) Exclude cut with ID 7357-94126-0009-15840-0_sp1.1 from training. Duration: 22.1090625
2022-12-22 23:29:33,264 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([2.3005, 1.9714, 1.5221, 0.4987, 1.3505, 1.9253, 1.5564, 1.9207],
       device='cuda:2'), covar=tensor([0.0549, 0.0398, 0.0916, 0.1469, 0.1050, 0.1290, 0.1449, 0.0560],
       device='cuda:2'), in_proj_covar=tensor([0.0161, 0.0177, 0.0203, 0.0191, 0.0205, 0.0188, 0.0200, 0.0188],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-22 23:29:34,404 WARNING [train.py:1060] (2/4) Exclude cut with ID 7699-105389-0094-26379-0_sp0.9 from training. Duration: 26.6166875
2022-12-22 23:29:38,759 INFO [zipformer.py:660] (2/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=31064.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 23:29:44,109 INFO [scaling.py:679] (2/4) Whitening: num_groups=8, num_channels=96, metric=1.44 vs. limit=2.0
2022-12-22 23:29:50,693 WARNING [train.py:1060] (2/4) Exclude cut with ID 2046-178027-0000-53705-0_sp0.9 from training. Duration: 20.3055625
2022-12-22 23:29:51,100 INFO [zipformer.py:660] (2/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=31072.0, num_to_drop=1, layers_to_drop={0}
2022-12-22 23:29:57,433 INFO [scaling.py:679] (2/4) Whitening: num_groups=8, num_channels=192, metric=1.88 vs. limit=2.0
2022-12-22 23:30:00,153 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.9102, 2.0118, 1.2775, 2.2687, 2.2560, 1.8417, 2.9521, 2.0809],
       device='cuda:2'), covar=tensor([0.0973, 0.1578, 0.2842, 0.1886, 0.1623, 0.0941, 0.0948, 0.1142],
       device='cuda:2'), in_proj_covar=tensor([0.0172, 0.0191, 0.0232, 0.0280, 0.0220, 0.0181, 0.0204, 0.0191],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-22 23:30:13,566 WARNING [train.py:1060] (2/4) Exclude cut with ID 7205-50138-0008-5373-0_sp0.9 from training. Duration: 20.7
2022-12-22 23:30:34,677 INFO [train.py:894] (2/4) Epoch 9, batch 3050, loss[loss=0.214, simple_loss=0.2983, pruned_loss=0.06489, over 18389.00 frames. ], tot_loss[loss=0.2283, simple_loss=0.2995, pruned_loss=0.07857, over 3712743.27 frames. ], batch size: 53, lr: 1.26e-02, grad_scale: 8.0
2022-12-22 23:30:56,853 WARNING [train.py:1060] (2/4) Exclude cut with ID 6978-92210-0019-146985-0 from training. Duration: 22.48
2022-12-22 23:31:08,707 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 3.995e+02 5.870e+02 7.181e+02 8.516e+02 1.347e+03, threshold=1.436e+03, percent-clipped=0.0
2022-12-22 23:31:13,016 WARNING [train.py:1060] (2/4) Exclude cut with ID 7255-291500-0003-134302-0_sp0.9 from training. Duration: 29.816625
2022-12-22 23:31:19,944 INFO [scaling.py:679] (2/4) Whitening: num_groups=1, num_channels=384, metric=2.96 vs. limit=5.0
2022-12-22 23:31:33,847 WARNING [train.py:1060] (2/4) Exclude cut with ID 7255-291500-0005-134304-0_sp1.1 from training. Duration: 22.7590625
2022-12-22 23:31:38,511 WARNING [train.py:1060] (2/4) Exclude cut with ID 7357-94126-0026-15857-0 from training. Duration: 22.555
2022-12-22 23:31:51,838 INFO [train.py:894] (2/4) Epoch 9, batch 3100, loss[loss=0.2129, simple_loss=0.2803, pruned_loss=0.07277, over 18426.00 frames. ], tot_loss[loss=0.2286, simple_loss=0.2996, pruned_loss=0.07874, over 3712871.82 frames. ], batch size: 42, lr: 1.26e-02, grad_scale: 8.0
2022-12-22 23:32:00,443 WARNING [train.py:1060] (2/4) Exclude cut with ID 1250-135782-0005-25975-0_sp0.9 from training. Duration: 21.688875
2022-12-22 23:32:22,969 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([0.1422, 1.2817, 1.7071, 0.5531, 0.8790, 1.7915, 1.5640, 1.4933],
       device='cuda:2'), covar=tensor([0.0609, 0.0305, 0.0248, 0.0326, 0.0377, 0.0327, 0.0201, 0.0491],
       device='cuda:2'), in_proj_covar=tensor([0.0132, 0.0152, 0.0105, 0.0127, 0.0133, 0.0118, 0.0141, 0.0141],
       device='cuda:2'), out_proj_covar=tensor([1.1180e-04, 1.3055e-04, 8.8423e-05, 1.0619e-04, 1.1273e-04, 1.0119e-04,
        1.2310e-04, 1.2054e-04], device='cuda:2')
2022-12-22 23:32:36,507 WARNING [train.py:1060] (2/4) Exclude cut with ID 3488-85273-0038-41224-0_sp0.9 from training. Duration: 22.6
2022-12-22 23:32:53,289 INFO [scaling.py:679] (2/4) Whitening: num_groups=1, num_channels=384, metric=2.97 vs. limit=5.0
2022-12-22 23:33:08,576 INFO [train.py:894] (2/4) Epoch 9, batch 3150, loss[loss=0.2153, simple_loss=0.2877, pruned_loss=0.07147, over 18696.00 frames. ], tot_loss[loss=0.2284, simple_loss=0.2993, pruned_loss=0.07873, over 3713507.04 frames. ], batch size: 50, lr: 1.26e-02, grad_scale: 8.0
2022-12-22 23:33:13,524 WARNING [train.py:1060] (2/4) Exclude cut with ID 7357-94126-0009-15840-0 from training. Duration: 24.32
2022-12-22 23:33:44,242 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 3.945e+02 5.767e+02 6.869e+02 8.584e+02 1.481e+03, threshold=1.374e+03, percent-clipped=1.0
2022-12-22 23:34:14,808 WARNING [train.py:1060] (2/4) Exclude cut with ID 8291-276745-0093-13116-0_sp0.9 from training. Duration: 21.061125
2022-12-22 23:34:28,080 WARNING [train.py:1060] (2/4) Exclude cut with ID 7357-94126-0024-15855-0_sp0.9 from training. Duration: 20.32225
2022-12-22 23:34:29,492 INFO [train.py:894] (2/4) Epoch 9, batch 3200, loss[loss=0.2099, simple_loss=0.278, pruned_loss=0.07091, over 18393.00 frames. ], tot_loss[loss=0.2278, simple_loss=0.2991, pruned_loss=0.07822, over 3713128.57 frames. ], batch size: 46, lr: 1.26e-02, grad_scale: 8.0
2022-12-22 23:34:41,483 WARNING [train.py:1060] (2/4) Exclude cut with ID 3033-130750-0096-55598-0_sp1.1 from training. Duration: 0.7545625
2022-12-22 23:34:43,036 INFO [zipformer.py:660] (2/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=31259.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 23:34:55,826 WARNING [train.py:1060] (2/4) Exclude cut with ID 4295-39940-0007-92567-0_sp0.9 from training. Duration: 23.9333125
2022-12-22 23:35:17,396 INFO [zipformer.py:660] (2/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=31281.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 23:35:29,392 WARNING [train.py:1060] (2/4) Exclude cut with ID 7255-291500-0008-134307-0_sp1.1 from training. Duration: 20.17275
2022-12-22 23:35:30,950 INFO [zipformer.py:660] (2/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=31290.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 23:35:35,197 WARNING [train.py:1060] (2/4) Exclude cut with ID 6978-92210-0019-146985-0_sp1.1 from training. Duration: 20.436375
2022-12-22 23:35:45,272 INFO [train.py:894] (2/4) Epoch 9, batch 3250, loss[loss=0.2424, simple_loss=0.312, pruned_loss=0.08639, over 18583.00 frames. ], tot_loss[loss=0.2273, simple_loss=0.2986, pruned_loss=0.07803, over 3713158.00 frames. ], batch size: 69, lr: 1.26e-02, grad_scale: 16.0
2022-12-22 23:36:01,528 INFO [zipformer.py:660] (2/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=31311.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 23:36:15,449 INFO [scaling.py:679] (2/4) Whitening: num_groups=1, num_channels=384, metric=3.78 vs. limit=5.0
2022-12-22 23:36:19,275 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 3.404e+02 5.691e+02 7.000e+02 8.164e+02 1.618e+03, threshold=1.400e+03, percent-clipped=2.0
2022-12-22 23:36:21,721 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([4.7577, 4.1021, 4.0637, 4.6176, 4.3390, 4.1953, 4.7812, 1.3553],
       device='cuda:2'), covar=tensor([0.0590, 0.0561, 0.0567, 0.0537, 0.1197, 0.0910, 0.0501, 0.4620],
       device='cuda:2'), in_proj_covar=tensor([0.0288, 0.0195, 0.0200, 0.0201, 0.0273, 0.0226, 0.0231, 0.0252],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0001, 0.0001, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-22 23:36:28,012 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.7552, 1.4380, 1.0271, 1.4659, 1.9385, 1.4715, 1.7171, 2.0692],
       device='cuda:2'), covar=tensor([0.1648, 0.2094, 0.2577, 0.1505, 0.1950, 0.1647, 0.1452, 0.1498],
       device='cuda:2'), in_proj_covar=tensor([0.0092, 0.0103, 0.0122, 0.0099, 0.0117, 0.0092, 0.0098, 0.0097],
       device='cuda:2'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003],
       device='cuda:2')
2022-12-22 23:36:57,752 WARNING [train.py:1060] (2/4) Exclude cut with ID 4234-40345-0022-142709-0_sp0.9 from training. Duration: 23.1055625
2022-12-22 23:36:59,090 WARNING [train.py:1060] (2/4) Exclude cut with ID 8291-282929-0007-12994-0_sp1.1 from training. Duration: 23.5
2022-12-22 23:37:00,500 INFO [train.py:894] (2/4) Epoch 9, batch 3300, loss[loss=0.2428, simple_loss=0.3161, pruned_loss=0.08475, over 18720.00 frames. ], tot_loss[loss=0.2287, simple_loss=0.3002, pruned_loss=0.0786, over 3713416.83 frames. ], batch size: 52, lr: 1.26e-02, grad_scale: 16.0
2022-12-22 23:37:09,690 WARNING [train.py:1060] (2/4) Exclude cut with ID 7255-291500-0009-134308-0_sp0.9 from training. Duration: 26.62775
2022-12-22 23:37:14,148 INFO [zipformer.py:660] (2/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=31359.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 23:37:21,751 WARNING [train.py:1060] (2/4) Exclude cut with ID 6951-79737-0018-132285-0 from training. Duration: 21.105
2022-12-22 23:37:26,556 WARNING [train.py:1060] (2/4) Exclude cut with ID 4511-76322-0006-80011-0_sp0.9 from training. Duration: 24.411125
2022-12-22 23:37:26,711 INFO [zipformer.py:660] (2/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=31367.0, num_to_drop=1, layers_to_drop={3}
2022-12-22 23:37:54,876 WARNING [train.py:1060] (2/4) Exclude cut with ID 6758-72288-0033-108368-0_sp1.1 from training. Duration: 21.263625
2022-12-22 23:38:17,129 INFO [train.py:894] (2/4) Epoch 9, batch 3350, loss[loss=0.2391, simple_loss=0.313, pruned_loss=0.08257, over 18689.00 frames. ], tot_loss[loss=0.2289, simple_loss=0.3004, pruned_loss=0.07871, over 3713518.72 frames. ], batch size: 60, lr: 1.26e-02, grad_scale: 16.0
2022-12-22 23:38:27,892 WARNING [train.py:1060] (2/4) Exclude cut with ID 4234-40345-0022-142709-0 from training. Duration: 20.795
2022-12-22 23:38:31,930 INFO [zipformer.py:660] (2/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=31409.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 23:38:38,887 WARNING [train.py:1060] (2/4) Exclude cut with ID 7357-94126-0021-15852-0 from training. Duration: 24.76
2022-12-22 23:38:38,906 WARNING [train.py:1060] (2/4) Exclude cut with ID 3867-173237-0077-144769-0_sp0.9 from training. Duration: 22.25
2022-12-22 23:38:51,969 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([2.6800, 2.8434, 2.8596, 1.4625, 3.3922, 3.2688, 2.1198, 3.7652],
       device='cuda:2'), covar=tensor([0.1048, 0.1342, 0.1312, 0.2123, 0.0666, 0.1049, 0.1903, 0.0432],
       device='cuda:2'), in_proj_covar=tensor([0.0194, 0.0196, 0.0199, 0.0187, 0.0180, 0.0207, 0.0206, 0.0178],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-22 23:38:52,961 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 3.333e+02 5.442e+02 7.006e+02 9.337e+02 1.866e+03, threshold=1.401e+03, percent-clipped=5.0
2022-12-22 23:39:06,430 WARNING [train.py:1060] (2/4) Exclude cut with ID 7357-94126-0026-15857-0_sp1.1 from training. Duration: 20.5045625
2022-12-22 23:39:17,168 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.8027, 1.5700, 1.2585, 1.9838, 1.7908, 3.5045, 1.3893, 1.5271],
       device='cuda:2'), covar=tensor([0.0922, 0.1783, 0.1310, 0.0962, 0.1435, 0.0235, 0.1437, 0.1636],
       device='cuda:2'), in_proj_covar=tensor([0.0076, 0.0084, 0.0077, 0.0077, 0.0093, 0.0072, 0.0086, 0.0078],
       device='cuda:2'), out_proj_covar=tensor([0.0003, 0.0004, 0.0003, 0.0003, 0.0004, 0.0003, 0.0004, 0.0004],
       device='cuda:2')
2022-12-22 23:39:32,935 INFO [train.py:894] (2/4) Epoch 9, batch 3400, loss[loss=0.2576, simple_loss=0.3205, pruned_loss=0.09739, over 18582.00 frames. ], tot_loss[loss=0.2263, simple_loss=0.2981, pruned_loss=0.0773, over 3712648.98 frames. ], batch size: 182, lr: 1.25e-02, grad_scale: 16.0
2022-12-22 23:40:00,038 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.9926, 1.8853, 1.7765, 0.7458, 2.4498, 2.0827, 1.7815, 1.3266],
       device='cuda:2'), covar=tensor([0.0317, 0.0362, 0.0391, 0.0732, 0.0180, 0.0306, 0.0498, 0.0893],
       device='cuda:2'), in_proj_covar=tensor([0.0119, 0.0119, 0.0128, 0.0120, 0.0084, 0.0117, 0.0138, 0.0151],
       device='cuda:2'), out_proj_covar=tensor([0.0001, 0.0001, 0.0002, 0.0001, 0.0001, 0.0001, 0.0002, 0.0002],
       device='cuda:2')
2022-12-22 23:40:01,587 INFO [zipformer.py:660] (2/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=31470.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 23:40:32,674 INFO [zipformer.py:660] (2/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=31491.0, num_to_drop=1, layers_to_drop={0}
2022-12-22 23:40:45,260 INFO [train.py:894] (2/4) Epoch 9, batch 3450, loss[loss=0.2051, simple_loss=0.283, pruned_loss=0.06365, over 18676.00 frames. ], tot_loss[loss=0.2247, simple_loss=0.2971, pruned_loss=0.07612, over 3713305.39 frames. ], batch size: 48, lr: 1.25e-02, grad_scale: 16.0
2022-12-22 23:41:20,628 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 3.481e+02 5.213e+02 6.398e+02 7.592e+02 2.358e+03, threshold=1.280e+03, percent-clipped=3.0
2022-12-22 23:41:57,741 INFO [scaling.py:679] (2/4) Whitening: num_groups=8, num_channels=96, metric=1.20 vs. limit=2.0
2022-12-22 23:42:01,851 INFO [train.py:894] (2/4) Epoch 9, batch 3500, loss[loss=0.2639, simple_loss=0.328, pruned_loss=0.09983, over 18700.00 frames. ], tot_loss[loss=0.2263, simple_loss=0.2982, pruned_loss=0.07717, over 3714582.09 frames. ], batch size: 97, lr: 1.25e-02, grad_scale: 16.0
2022-12-22 23:42:05,258 INFO [zipformer.py:660] (2/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=31552.0, num_to_drop=1, layers_to_drop={3}
2022-12-22 23:42:22,134 WARNING [train.py:1060] (2/4) Exclude cut with ID 7859-102521-0017-7548-0_sp1.1 from training. Duration: 22.2954375
2022-12-22 23:42:34,706 INFO [train.py:894] (2/4) Epoch 10, batch 0, loss[loss=0.2013, simple_loss=0.2913, pruned_loss=0.05563, over 18480.00 frames. ], tot_loss[loss=0.2013, simple_loss=0.2913, pruned_loss=0.05563, over 18480.00 frames. ], batch size: 54, lr: 1.19e-02, grad_scale: 16.0
2022-12-22 23:42:34,706 INFO [train.py:919] (2/4) Computing validation loss
2022-12-22 23:42:45,648 INFO [train.py:928] (2/4) Epoch 10, validation: loss=0.1774, simple_loss=0.277, pruned_loss=0.03887, over 944034.00 frames. 
2022-12-22 23:42:45,648 INFO [train.py:929] (2/4) Maximum memory allocated so far is 24680MB
2022-12-22 23:42:50,462 INFO [zipformer.py:660] (2/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=31559.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 23:43:25,351 INFO [zipformer.py:660] (2/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=31581.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 23:43:36,764 WARNING [train.py:1060] (2/4) Exclude cut with ID 298-126791-0067-24026-0_sp0.9 from training. Duration: 21.438875
2022-12-22 23:43:38,504 INFO [zipformer.py:660] (2/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=31590.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 23:43:41,250 WARNING [train.py:1060] (2/4) Exclude cut with ID 5652-39938-0025-23684-0_sp0.9 from training. Duration: 22.2055625
2022-12-22 23:43:54,012 INFO [scaling.py:679] (2/4) Whitening: num_groups=1, num_channels=384, metric=4.19 vs. limit=5.0
2022-12-22 23:44:02,467 INFO [train.py:894] (2/4) Epoch 10, batch 50, loss[loss=0.2143, simple_loss=0.2999, pruned_loss=0.06434, over 18633.00 frames. ], tot_loss[loss=0.208, simple_loss=0.2881, pruned_loss=0.06395, over 837747.61 frames. ], batch size: 53, lr: 1.19e-02, grad_scale: 16.0
2022-12-22 23:44:03,919 INFO [zipformer.py:660] (2/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=31607.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 23:44:28,597 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 2.286e+02 4.881e+02 5.764e+02 7.674e+02 1.540e+03, threshold=1.153e+03, percent-clipped=4.0
2022-12-22 23:44:37,726 INFO [zipformer.py:660] (2/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=31629.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 23:44:50,677 INFO [zipformer.py:660] (2/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=31638.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 23:45:12,262 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.4371, 1.9857, 1.4442, 2.4651, 2.5290, 1.4289, 1.6008, 1.1633],
       device='cuda:2'), covar=tensor([0.2161, 0.1750, 0.1694, 0.0876, 0.1529, 0.1318, 0.2020, 0.1762],
       device='cuda:2'), in_proj_covar=tensor([0.0234, 0.0212, 0.0198, 0.0185, 0.0251, 0.0185, 0.0209, 0.0187],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-22 23:45:17,536 INFO [train.py:894] (2/4) Epoch 10, batch 100, loss[loss=0.2128, simple_loss=0.276, pruned_loss=0.0748, over 18490.00 frames. ], tot_loss[loss=0.2077, simple_loss=0.2879, pruned_loss=0.06368, over 1476297.58 frames. ], batch size: 43, lr: 1.19e-02, grad_scale: 16.0
2022-12-22 23:45:22,980 INFO [scaling.py:679] (2/4) Whitening: num_groups=8, num_channels=96, metric=1.28 vs. limit=2.0
2022-12-22 23:45:35,762 INFO [zipformer.py:660] (2/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=31667.0, num_to_drop=1, layers_to_drop={2}
2022-12-22 23:46:23,233 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.8044, 1.6841, 2.1548, 1.2252, 2.2456, 2.1037, 1.4413, 2.4808],
       device='cuda:2'), covar=tensor([0.1231, 0.1694, 0.1242, 0.1898, 0.0724, 0.1134, 0.2116, 0.0472],
       device='cuda:2'), in_proj_covar=tensor([0.0196, 0.0197, 0.0202, 0.0191, 0.0182, 0.0208, 0.0206, 0.0179],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-22 23:46:30,454 INFO [zipformer.py:660] (2/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=31704.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 23:46:33,008 INFO [train.py:894] (2/4) Epoch 10, batch 150, loss[loss=0.2115, simple_loss=0.3015, pruned_loss=0.06075, over 18553.00 frames. ], tot_loss[loss=0.2058, simple_loss=0.2878, pruned_loss=0.06195, over 1972178.95 frames. ], batch size: 69, lr: 1.19e-02, grad_scale: 16.0
2022-12-22 23:46:36,038 WARNING [train.py:1060] (2/4) Exclude cut with ID 7859-102521-0017-7548-0 from training. Duration: 24.525
2022-12-22 23:46:47,725 INFO [zipformer.py:660] (2/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=31715.0, num_to_drop=1, layers_to_drop={0}
2022-12-22 23:46:58,976 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 2.574e+02 4.062e+02 5.113e+02 6.573e+02 1.615e+03, threshold=1.023e+03, percent-clipped=2.0
2022-12-22 23:47:09,529 INFO [scaling.py:679] (2/4) Whitening: num_groups=8, num_channels=192, metric=1.94 vs. limit=2.0
2022-12-22 23:47:12,881 WARNING [train.py:1060] (2/4) Exclude cut with ID 3699-47246-0007-3408-0_sp0.9 from training. Duration: 20.26675
2022-12-22 23:47:25,947 WARNING [train.py:1060] (2/4) Exclude cut with ID 7859-102521-0017-7548-0_sp0.9 from training. Duration: 27.25
2022-12-22 23:47:48,632 INFO [train.py:894] (2/4) Epoch 10, batch 200, loss[loss=0.2299, simple_loss=0.3165, pruned_loss=0.07167, over 18517.00 frames. ], tot_loss[loss=0.2027, simple_loss=0.2844, pruned_loss=0.06055, over 2357469.08 frames. ], batch size: 55, lr: 1.19e-02, grad_scale: 16.0
2022-12-22 23:48:02,754 INFO [zipformer.py:660] (2/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=31765.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 23:48:03,009 INFO [zipformer.py:660] (2/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=31765.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 23:48:37,902 INFO [zipformer.py:660] (2/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=31788.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 23:48:40,568 WARNING [train.py:1060] (2/4) Exclude cut with ID 6426-64292-0017-15984-0 from training. Duration: 21.68
2022-12-22 23:48:52,742 WARNING [train.py:1060] (2/4) Exclude cut with ID 4278-13270-0007-59342-0 from training. Duration: 21.6300625
2022-12-22 23:49:04,718 INFO [train.py:894] (2/4) Epoch 10, batch 250, loss[loss=0.1921, simple_loss=0.2807, pruned_loss=0.05175, over 18514.00 frames. ], tot_loss[loss=0.1999, simple_loss=0.2822, pruned_loss=0.05883, over 2657365.08 frames. ], batch size: 77, lr: 1.19e-02, grad_scale: 16.0
2022-12-22 23:49:12,096 INFO [scaling.py:679] (2/4) Whitening: num_groups=8, num_channels=96, metric=1.13 vs. limit=2.0
2022-12-22 23:49:17,736 WARNING [train.py:1060] (2/4) Exclude cut with ID 4278-13270-0007-59342-0_sp0.9 from training. Duration: 24.033375
2022-12-22 23:49:30,996 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 2.762e+02 3.941e+02 4.740e+02 5.956e+02 1.524e+03, threshold=9.481e+02, percent-clipped=1.0
2022-12-22 23:49:31,257 INFO [zipformer.py:660] (2/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=31823.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 23:50:07,719 INFO [zipformer.py:660] (2/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=31847.0, num_to_drop=1, layers_to_drop={2}
2022-12-22 23:50:10,598 INFO [zipformer.py:660] (2/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=31849.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 23:50:12,673 INFO [scaling.py:679] (2/4) Whitening: num_groups=8, num_channels=96, metric=1.32 vs. limit=2.0
2022-12-22 23:50:16,251 WARNING [train.py:1060] (2/4) Exclude cut with ID 4278-13270-0009-59344-0 from training. Duration: 22.905
2022-12-22 23:50:17,671 WARNING [train.py:1060] (2/4) Exclude cut with ID 5622-44585-0006-90525-0_sp1.1 from training. Duration: 23.4318125
2022-12-22 23:50:21,129 INFO [train.py:894] (2/4) Epoch 10, batch 300, loss[loss=0.2116, simple_loss=0.3003, pruned_loss=0.06142, over 18705.00 frames. ], tot_loss[loss=0.2013, simple_loss=0.2837, pruned_loss=0.05945, over 2890651.87 frames. ], batch size: 62, lr: 1.19e-02, grad_scale: 16.0
2022-12-22 23:50:47,880 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([2.8735, 3.9464, 3.8221, 1.8756, 3.9527, 2.9364, 1.2797, 2.8672],
       device='cuda:2'), covar=tensor([0.1940, 0.0977, 0.1274, 0.3627, 0.0827, 0.1017, 0.4727, 0.1525],
       device='cuda:2'), in_proj_covar=tensor([0.0133, 0.0118, 0.0146, 0.0117, 0.0121, 0.0103, 0.0141, 0.0108],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-22 23:50:55,772 INFO [zipformer.py:660] (2/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=31878.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 23:51:04,450 INFO [zipformer.py:660] (2/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=31884.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 23:51:11,236 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.6450, 1.5791, 1.6697, 1.7013, 1.4763, 3.5824, 1.4503, 2.2085],
       device='cuda:2'), covar=tensor([0.3204, 0.1955, 0.1921, 0.1901, 0.1276, 0.0167, 0.1577, 0.0830],
       device='cuda:2'), in_proj_covar=tensor([0.0139, 0.0121, 0.0129, 0.0122, 0.0104, 0.0099, 0.0098, 0.0094],
       device='cuda:2'), out_proj_covar=tensor([0.0005, 0.0004, 0.0005, 0.0004, 0.0004, 0.0003, 0.0004, 0.0004],
       device='cuda:2')
2022-12-22 23:51:37,208 INFO [train.py:894] (2/4) Epoch 10, batch 350, loss[loss=0.2116, simple_loss=0.2935, pruned_loss=0.0649, over 18600.00 frames. ], tot_loss[loss=0.2045, simple_loss=0.2864, pruned_loss=0.06133, over 3073925.39 frames. ], batch size: 69, lr: 1.18e-02, grad_scale: 16.0
2022-12-22 23:51:48,719 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.3438, 1.0776, 1.5275, 2.7590, 2.0206, 2.1085, 0.8650, 1.9689],
       device='cuda:2'), covar=tensor([0.1939, 0.1996, 0.1702, 0.0668, 0.1254, 0.1322, 0.2334, 0.1300],
       device='cuda:2'), in_proj_covar=tensor([0.0103, 0.0116, 0.0129, 0.0123, 0.0103, 0.0133, 0.0130, 0.0110],
       device='cuda:2'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003],
       device='cuda:2')
2022-12-22 23:51:51,991 INFO [zipformer.py:660] (2/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=31915.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 23:52:03,206 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 3.016e+02 4.446e+02 5.363e+02 6.750e+02 1.599e+03, threshold=1.073e+03, percent-clipped=7.0
2022-12-22 23:52:13,342 WARNING [train.py:1060] (2/4) Exclude cut with ID 4278-13270-0009-59344-0_sp1.1 from training. Duration: 20.82275
2022-12-22 23:52:14,602 WARNING [train.py:1060] (2/4) Exclude cut with ID 4278-13270-0009-59344-0_sp0.9 from training. Duration: 25.45
2022-12-22 23:52:27,819 INFO [zipformer.py:660] (2/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=31939.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 23:52:37,136 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.8005, 1.3871, 1.0910, 0.2830, 1.2658, 1.5152, 1.0262, 1.5032],
       device='cuda:2'), covar=tensor([0.0612, 0.0638, 0.1061, 0.1583, 0.0991, 0.1355, 0.1757, 0.0619],
       device='cuda:2'), in_proj_covar=tensor([0.0164, 0.0181, 0.0204, 0.0195, 0.0207, 0.0190, 0.0203, 0.0189],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-22 23:52:44,507 INFO [zipformer.py:660] (2/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=31950.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 23:52:53,564 INFO [train.py:894] (2/4) Epoch 10, batch 400, loss[loss=0.2047, simple_loss=0.2813, pruned_loss=0.06406, over 18390.00 frames. ], tot_loss[loss=0.2039, simple_loss=0.2858, pruned_loss=0.06103, over 3214831.95 frames. ], batch size: 46, lr: 1.18e-02, grad_scale: 16.0
2022-12-22 23:53:13,506 WARNING [train.py:1060] (2/4) Exclude cut with ID 5622-44585-0006-90525-0 from training. Duration: 25.775
2022-12-22 23:53:24,091 INFO [zipformer.py:660] (2/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=31976.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 23:53:34,137 WARNING [train.py:1060] (2/4) Exclude cut with ID 2929-85685-0071-62375-0_sp0.9 from training. Duration: 22.25
2022-12-22 23:54:06,272 WARNING [train.py:1060] (2/4) Exclude cut with ID 3972-170212-0014-23379-0 from training. Duration: 26.205
2022-12-22 23:54:12,705 INFO [train.py:894] (2/4) Epoch 10, batch 450, loss[loss=0.2096, simple_loss=0.2919, pruned_loss=0.06362, over 18588.00 frames. ], tot_loss[loss=0.2058, simple_loss=0.2876, pruned_loss=0.06194, over 3325474.24 frames. ], batch size: 51, lr: 1.18e-02, grad_scale: 16.0
2022-12-22 23:54:21,090 INFO [zipformer.py:660] (2/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=32011.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 23:54:23,868 WARNING [train.py:1060] (2/4) Exclude cut with ID 5239-32139-0047-9341-0_sp0.9 from training. Duration: 30.1555625
2022-12-22 23:54:28,314 WARNING [train.py:1060] (2/4) Exclude cut with ID 1265-135635-0050-6781-0_sp0.9 from training. Duration: 21.8333125
2022-12-22 23:54:38,590 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 2.601e+02 4.140e+02 5.222e+02 6.476e+02 1.497e+03, threshold=1.044e+03, percent-clipped=4.0
2022-12-22 23:54:40,120 WARNING [train.py:1060] (2/4) Exclude cut with ID 1914-133440-0024-94914-0_sp1.1 from training. Duration: 20.6545625
2022-12-22 23:55:22,788 WARNING [train.py:1060] (2/4) Exclude cut with ID 7395-89880-0045-39920-0_sp0.9 from training. Duration: 20.52225
2022-12-22 23:55:28,524 INFO [train.py:894] (2/4) Epoch 10, batch 500, loss[loss=0.1654, simple_loss=0.2469, pruned_loss=0.04196, over 18439.00 frames. ], tot_loss[loss=0.2072, simple_loss=0.2882, pruned_loss=0.0631, over 3411794.69 frames. ], batch size: 42, lr: 1.18e-02, grad_scale: 16.0
2022-12-22 23:55:35,976 INFO [zipformer.py:660] (2/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=32060.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 23:55:43,640 INFO [zipformer.py:660] (2/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=32065.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 23:55:44,870 WARNING [train.py:1060] (2/4) Exclude cut with ID 3972-170212-0014-23379-0_sp0.9 from training. Duration: 29.1166875
2022-12-22 23:56:45,408 INFO [train.py:894] (2/4) Epoch 10, batch 550, loss[loss=0.2067, simple_loss=0.2879, pruned_loss=0.06279, over 18583.00 frames. ], tot_loss[loss=0.2071, simple_loss=0.288, pruned_loss=0.0631, over 3479355.91 frames. ], batch size: 51, lr: 1.18e-02, grad_scale: 16.0
2022-12-22 23:56:45,459 WARNING [train.py:1060] (2/4) Exclude cut with ID 543-133211-0007-59831-0_sp0.9 from training. Duration: 21.388875
2022-12-22 23:56:56,144 INFO [zipformer.py:660] (2/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=32113.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 23:57:10,608 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 3.007e+02 4.171e+02 5.313e+02 6.192e+02 1.458e+03, threshold=1.063e+03, percent-clipped=4.0
2022-12-22 23:57:23,903 WARNING [train.py:1060] (2/4) Exclude cut with ID 1914-133440-0024-94914-0 from training. Duration: 22.72
2022-12-22 23:57:25,357 WARNING [train.py:1060] (2/4) Exclude cut with ID 1914-133440-0031-94921-0_sp0.9 from training. Duration: 22.7444375
2022-12-22 23:57:42,786 INFO [zipformer.py:660] (2/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=32144.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 23:57:47,272 INFO [zipformer.py:660] (2/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=32147.0, num_to_drop=1, layers_to_drop={0}
2022-12-22 23:57:50,676 INFO [scaling.py:679] (2/4) Whitening: num_groups=1, num_channels=384, metric=3.34 vs. limit=5.0
2022-12-22 23:58:00,334 INFO [train.py:894] (2/4) Epoch 10, batch 600, loss[loss=0.1936, simple_loss=0.2759, pruned_loss=0.05565, over 18526.00 frames. ], tot_loss[loss=0.2069, simple_loss=0.2877, pruned_loss=0.06302, over 3530894.02 frames. ], batch size: 47, lr: 1.18e-02, grad_scale: 16.0
2022-12-22 23:58:10,333 WARNING [train.py:1060] (2/4) Exclude cut with ID 4133-6541-0027-40495-0_sp1.1 from training. Duration: 0.9681875
2022-12-22 23:58:11,673 WARNING [train.py:1060] (2/4) Exclude cut with ID 6330-62851-0022-91297-0_sp0.9 from training. Duration: 22.3166875
2022-12-22 23:58:14,150 INFO [scaling.py:679] (2/4) Whitening: num_groups=8, num_channels=192, metric=1.92 vs. limit=2.0
2022-12-22 23:58:17,794 WARNING [train.py:1060] (2/4) Exclude cut with ID 543-133212-0015-59917-0_sp0.9 from training. Duration: 21.8166875
2022-12-22 23:58:35,930 INFO [zipformer.py:660] (2/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=32179.0, num_to_drop=0, layers_to_drop=set()
2022-12-22 23:59:00,476 INFO [zipformer.py:660] (2/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=32195.0, num_to_drop=1, layers_to_drop={1}
2022-12-22 23:59:17,367 INFO [train.py:894] (2/4) Epoch 10, batch 650, loss[loss=0.1852, simple_loss=0.2585, pruned_loss=0.05595, over 18422.00 frames. ], tot_loss[loss=0.2071, simple_loss=0.2883, pruned_loss=0.06301, over 3571571.96 frames. ], batch size: 42, lr: 1.18e-02, grad_scale: 16.0
2022-12-22 23:59:43,717 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 2.998e+02 4.804e+02 5.575e+02 7.166e+02 1.378e+03, threshold=1.115e+03, percent-clipped=4.0
2022-12-22 23:59:49,340 INFO [scaling.py:679] (2/4) Whitening: num_groups=1, num_channels=384, metric=4.82 vs. limit=5.0
2022-12-23 00:00:00,259 INFO [zipformer.py:660] (2/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=32234.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 00:00:04,188 WARNING [train.py:1060] (2/4) Exclude cut with ID 4957-30119-0041-23990-0_sp0.9 from training. Duration: 20.22775
2022-12-23 00:00:35,984 INFO [train.py:894] (2/4) Epoch 10, batch 700, loss[loss=0.1996, simple_loss=0.2765, pruned_loss=0.06132, over 18677.00 frames. ], tot_loss[loss=0.2074, simple_loss=0.2888, pruned_loss=0.063, over 3603756.96 frames. ], batch size: 48, lr: 1.18e-02, grad_scale: 16.0
2022-12-23 00:00:39,188 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.9733, 1.8543, 2.1024, 1.3072, 2.4669, 2.1489, 1.5964, 2.7732],
       device='cuda:2'), covar=tensor([0.1015, 0.1556, 0.1312, 0.1787, 0.0666, 0.1157, 0.2073, 0.0461],
       device='cuda:2'), in_proj_covar=tensor([0.0193, 0.0193, 0.0200, 0.0189, 0.0178, 0.0208, 0.0206, 0.0179],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-23 00:00:50,408 WARNING [train.py:1060] (2/4) Exclude cut with ID 5239-32139-0047-9341-0_sp1.1 from training. Duration: 24.67275
2022-12-23 00:00:57,940 INFO [zipformer.py:660] (2/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=32271.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 00:01:08,824 INFO [scaling.py:679] (2/4) Whitening: num_groups=1, num_channels=384, metric=3.46 vs. limit=5.0
2022-12-23 00:01:18,623 WARNING [train.py:1060] (2/4) Exclude cut with ID 3082-165428-0081-50734-0_sp0.9 from training. Duration: 21.8055625
2022-12-23 00:01:31,771 INFO [scaling.py:679] (2/4) Whitening: num_groups=8, num_channels=192, metric=1.79 vs. limit=2.0
2022-12-23 00:01:50,521 INFO [train.py:894] (2/4) Epoch 10, batch 750, loss[loss=0.2127, simple_loss=0.2993, pruned_loss=0.06306, over 18468.00 frames. ], tot_loss[loss=0.2081, simple_loss=0.2899, pruned_loss=0.06316, over 3627692.59 frames. ], batch size: 54, lr: 1.18e-02, grad_scale: 16.0
2022-12-23 00:01:50,701 INFO [zipformer.py:660] (2/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=32306.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 00:01:57,739 WARNING [train.py:1060] (2/4) Exclude cut with ID 3340-169293-0054-76830-0_sp0.9 from training. Duration: 22.6666875
2022-12-23 00:02:15,109 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 2.578e+02 4.144e+02 4.884e+02 6.281e+02 1.414e+03, threshold=9.769e+02, percent-clipped=2.0
2022-12-23 00:02:59,126 WARNING [train.py:1060] (2/4) Exclude cut with ID 2411-132532-0017-82279-0_sp1.1 from training. Duration: 0.9681875
2022-12-23 00:03:05,508 INFO [train.py:894] (2/4) Epoch 10, batch 800, loss[loss=0.2004, simple_loss=0.2684, pruned_loss=0.06624, over 18694.00 frames. ], tot_loss[loss=0.2077, simple_loss=0.2894, pruned_loss=0.06304, over 3646192.74 frames. ], batch size: 41, lr: 1.18e-02, grad_scale: 16.0
2022-12-23 00:03:12,422 INFO [zipformer.py:660] (2/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=32360.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 00:03:25,620 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.7040, 1.6559, 1.4900, 2.0013, 1.6503, 3.6056, 1.3441, 1.6145],
       device='cuda:2'), covar=tensor([0.0915, 0.1731, 0.1177, 0.0944, 0.1492, 0.0200, 0.1419, 0.1561],
       device='cuda:2'), in_proj_covar=tensor([0.0075, 0.0082, 0.0077, 0.0076, 0.0093, 0.0072, 0.0086, 0.0078],
       device='cuda:2'), out_proj_covar=tensor([0.0003, 0.0004, 0.0003, 0.0003, 0.0004, 0.0003, 0.0004, 0.0004],
       device='cuda:2')
2022-12-23 00:03:26,856 WARNING [train.py:1060] (2/4) Exclude cut with ID 6330-62850-0007-91323-0 from training. Duration: 22.485
2022-12-23 00:04:03,819 WARNING [train.py:1060] (2/4) Exclude cut with ID 3972-170212-0014-23379-0_sp1.1 from training. Duration: 23.82275
2022-12-23 00:04:17,779 WARNING [train.py:1060] (2/4) Exclude cut with ID 4860-13185-0032-76709-0 from training. Duration: 20.77
2022-12-23 00:04:22,594 INFO [train.py:894] (2/4) Epoch 10, batch 850, loss[loss=0.2176, simple_loss=0.3013, pruned_loss=0.06698, over 18540.00 frames. ], tot_loss[loss=0.2078, simple_loss=0.2896, pruned_loss=0.06301, over 3660342.06 frames. ], batch size: 55, lr: 1.18e-02, grad_scale: 16.0
2022-12-23 00:04:25,597 INFO [zipformer.py:660] (2/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=32408.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 00:04:27,107 WARNING [train.py:1060] (2/4) Exclude cut with ID 6426-64292-0017-15984-0_sp0.9 from training. Duration: 24.088875
2022-12-23 00:04:48,421 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 2.428e+02 4.593e+02 5.340e+02 6.790e+02 1.639e+03, threshold=1.068e+03, percent-clipped=5.0
2022-12-23 00:04:55,591 WARNING [train.py:1060] (2/4) Exclude cut with ID 6330-62850-0007-91323-0_sp1.1 from training. Duration: 20.4409375
2022-12-23 00:05:20,418 INFO [zipformer.py:660] (2/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=32444.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 00:05:39,521 INFO [train.py:894] (2/4) Epoch 10, batch 900, loss[loss=0.2342, simple_loss=0.3183, pruned_loss=0.07506, over 18509.00 frames. ], tot_loss[loss=0.2075, simple_loss=0.2894, pruned_loss=0.06279, over 3671596.32 frames. ], batch size: 64, lr: 1.17e-02, grad_scale: 16.0
2022-12-23 00:05:59,634 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.9684, 2.0275, 1.3843, 2.2751, 2.2476, 1.8259, 2.9089, 2.0623],
       device='cuda:2'), covar=tensor([0.0803, 0.1514, 0.2495, 0.1571, 0.1479, 0.0849, 0.0847, 0.1115],
       device='cuda:2'), in_proj_covar=tensor([0.0171, 0.0191, 0.0229, 0.0274, 0.0216, 0.0178, 0.0202, 0.0191],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-23 00:06:12,506 WARNING [train.py:1060] (2/4) Exclude cut with ID 2929-85685-0062-62366-0_sp0.9 from training. Duration: 22.511125
2022-12-23 00:06:13,826 WARNING [train.py:1060] (2/4) Exclude cut with ID 7395-89880-0031-39906-0 from training. Duration: 20.675
2022-12-23 00:06:14,100 INFO [zipformer.py:660] (2/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=32479.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 00:06:33,717 INFO [zipformer.py:660] (2/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=32492.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 00:06:34,454 INFO [scaling.py:679] (2/4) Whitening: num_groups=8, num_channels=96, metric=1.38 vs. limit=2.0
2022-12-23 00:06:37,335 INFO [scaling.py:679] (2/4) Whitening: num_groups=1, num_channels=384, metric=3.24 vs. limit=5.0
2022-12-23 00:06:55,146 INFO [train.py:894] (2/4) Epoch 10, batch 950, loss[loss=0.1899, simple_loss=0.2613, pruned_loss=0.05927, over 18483.00 frames. ], tot_loss[loss=0.208, simple_loss=0.2899, pruned_loss=0.06305, over 3680852.56 frames. ], batch size: 41, lr: 1.17e-02, grad_scale: 16.0
2022-12-23 00:07:19,172 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 3.034e+02 4.294e+02 5.171e+02 6.457e+02 1.266e+03, threshold=1.034e+03, percent-clipped=2.0
2022-12-23 00:07:25,136 INFO [zipformer.py:660] (2/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=32527.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 00:07:36,178 INFO [zipformer.py:660] (2/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=32534.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 00:07:50,225 WARNING [train.py:1060] (2/4) Exclude cut with ID 6330-62850-0007-91323-0_sp0.9 from training. Duration: 24.9833125
2022-12-23 00:08:10,204 INFO [train.py:894] (2/4) Epoch 10, batch 1000, loss[loss=0.2128, simple_loss=0.2905, pruned_loss=0.06759, over 18433.00 frames. ], tot_loss[loss=0.2083, simple_loss=0.2902, pruned_loss=0.06322, over 3687268.72 frames. ], batch size: 48, lr: 1.17e-02, grad_scale: 16.0
2022-12-23 00:08:20,630 WARNING [train.py:1060] (2/4) Exclude cut with ID 5239-32139-0047-9341-0 from training. Duration: 27.14
2022-12-23 00:08:28,054 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([2.6632, 4.0108, 3.8696, 1.7337, 3.9854, 2.8509, 0.5501, 2.4780],
       device='cuda:2'), covar=tensor([0.2067, 0.0872, 0.1208, 0.3884, 0.0810, 0.1107, 0.5639, 0.1874],
       device='cuda:2'), in_proj_covar=tensor([0.0134, 0.0118, 0.0147, 0.0119, 0.0122, 0.0103, 0.0142, 0.0109],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-23 00:08:32,425 INFO [zipformer.py:660] (2/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=32571.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 00:08:35,413 WARNING [train.py:1060] (2/4) Exclude cut with ID 2929-85685-0044-62348-0 from training. Duration: 22.44
2022-12-23 00:08:50,737 INFO [zipformer.py:660] (2/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=32582.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 00:09:27,549 INFO [train.py:894] (2/4) Epoch 10, batch 1050, loss[loss=0.1916, simple_loss=0.2779, pruned_loss=0.05265, over 18694.00 frames. ], tot_loss[loss=0.2092, simple_loss=0.2914, pruned_loss=0.06346, over 3693197.95 frames. ], batch size: 50, lr: 1.17e-02, grad_scale: 8.0
2022-12-23 00:09:27,958 INFO [zipformer.py:660] (2/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=32606.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 00:09:47,492 INFO [zipformer.py:660] (2/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=32619.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 00:09:51,891 WARNING [train.py:1060] (2/4) Exclude cut with ID 2929-85685-0060-62364-0_sp0.9 from training. Duration: 21.361125
2022-12-23 00:09:54,779 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 2.674e+02 4.245e+02 5.244e+02 7.030e+02 1.488e+03, threshold=1.049e+03, percent-clipped=6.0
2022-12-23 00:09:57,825 WARNING [train.py:1060] (2/4) Exclude cut with ID 2929-85685-0079-62383-0_sp1.1 from training. Duration: 27.0318125
2022-12-23 00:10:08,590 WARNING [train.py:1060] (2/4) Exclude cut with ID 5622-44585-0006-90525-0_sp0.9 from training. Duration: 28.638875
2022-12-23 00:10:24,103 WARNING [train.py:1060] (2/4) Exclude cut with ID 3340-169293-0054-76830-0 from training. Duration: 20.4
2022-12-23 00:10:24,803 INFO [scaling.py:679] (2/4) Whitening: num_groups=1, num_channels=384, metric=3.10 vs. limit=5.0
2022-12-23 00:10:40,848 INFO [zipformer.py:660] (2/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=32654.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 00:10:43,585 INFO [train.py:894] (2/4) Epoch 10, batch 1100, loss[loss=0.1824, simple_loss=0.2636, pruned_loss=0.05064, over 18536.00 frames. ], tot_loss[loss=0.2086, simple_loss=0.2907, pruned_loss=0.06329, over 3698655.97 frames. ], batch size: 47, lr: 1.17e-02, grad_scale: 8.0
2022-12-23 00:10:59,667 WARNING [train.py:1060] (2/4) Exclude cut with ID 2929-85685-0071-62375-0 from training. Duration: 20.025
2022-12-23 00:10:59,680 WARNING [train.py:1060] (2/4) Exclude cut with ID 2364-131735-0112-64612-0_sp0.9 from training. Duration: 20.488875
2022-12-23 00:11:04,313 WARNING [train.py:1060] (2/4) Exclude cut with ID 2929-85685-0079-62383-0 from training. Duration: 29.735
2022-12-23 00:11:44,757 INFO [scaling.py:679] (2/4) Whitening: num_groups=1, num_channels=384, metric=4.58 vs. limit=5.0
2022-12-23 00:11:45,327 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.5169, 1.0315, 1.9424, 3.1938, 2.1028, 2.4049, 0.9041, 2.0275],
       device='cuda:2'), covar=tensor([0.1831, 0.1921, 0.1446, 0.0483, 0.1169, 0.1185, 0.2252, 0.1189],
       device='cuda:2'), in_proj_covar=tensor([0.0101, 0.0113, 0.0127, 0.0122, 0.0102, 0.0131, 0.0127, 0.0108],
       device='cuda:2'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003],
       device='cuda:2')
2022-12-23 00:11:52,848 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.6977, 1.6447, 1.2940, 1.6082, 1.8428, 1.5390, 2.1468, 1.8414],
       device='cuda:2'), covar=tensor([0.0867, 0.1631, 0.2614, 0.1644, 0.1668, 0.0901, 0.1080, 0.1154],
       device='cuda:2'), in_proj_covar=tensor([0.0174, 0.0194, 0.0233, 0.0279, 0.0221, 0.0181, 0.0206, 0.0194],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-23 00:12:00,066 INFO [train.py:894] (2/4) Epoch 10, batch 1150, loss[loss=0.1828, simple_loss=0.258, pruned_loss=0.05378, over 18608.00 frames. ], tot_loss[loss=0.2066, simple_loss=0.2886, pruned_loss=0.06226, over 3702568.32 frames. ], batch size: 45, lr: 1.17e-02, grad_scale: 8.0
2022-12-23 00:12:19,697 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([2.1678, 2.3202, 1.3818, 2.5813, 2.4134, 2.0497, 3.2310, 2.2436],
       device='cuda:2'), covar=tensor([0.0835, 0.1538, 0.2683, 0.1722, 0.1463, 0.0860, 0.0857, 0.1117],
       device='cuda:2'), in_proj_covar=tensor([0.0172, 0.0192, 0.0230, 0.0275, 0.0218, 0.0179, 0.0204, 0.0192],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-23 00:12:27,863 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 2.652e+02 4.092e+02 4.850e+02 6.073e+02 1.087e+03, threshold=9.700e+02, percent-clipped=2.0
2022-12-23 00:12:27,974 WARNING [train.py:1060] (2/4) Exclude cut with ID 7276-92427-0014-12983-0_sp0.9 from training. Duration: 21.3055625
2022-12-23 00:12:29,242 WARNING [train.py:1060] (2/4) Exclude cut with ID 1025-75365-0008-79168-0_sp0.9 from training. Duration: 22.0666875
2022-12-23 00:13:16,600 INFO [train.py:894] (2/4) Epoch 10, batch 1200, loss[loss=0.201, simple_loss=0.2777, pruned_loss=0.06221, over 18550.00 frames. ], tot_loss[loss=0.2067, simple_loss=0.2888, pruned_loss=0.06234, over 3705791.33 frames. ], batch size: 47, lr: 1.17e-02, grad_scale: 8.0
2022-12-23 00:13:44,038 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([2.5415, 1.3214, 0.9283, 1.9794, 2.4650, 1.7303, 2.2963, 2.3016],
       device='cuda:2'), covar=tensor([0.1455, 0.2558, 0.2596, 0.1595, 0.1531, 0.1502, 0.1535, 0.1576],
       device='cuda:2'), in_proj_covar=tensor([0.0089, 0.0099, 0.0119, 0.0096, 0.0113, 0.0090, 0.0097, 0.0095],
       device='cuda:2'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003],
       device='cuda:2')
2022-12-23 00:14:05,487 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([2.5069, 3.8552, 3.6860, 1.6460, 3.8189, 3.0766, 0.6192, 2.8530],
       device='cuda:2'), covar=tensor([0.2213, 0.0942, 0.1551, 0.3910, 0.0915, 0.0893, 0.6032, 0.1452],
       device='cuda:2'), in_proj_covar=tensor([0.0131, 0.0118, 0.0144, 0.0118, 0.0121, 0.0102, 0.0142, 0.0108],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-23 00:14:22,913 WARNING [train.py:1060] (2/4) Exclude cut with ID 2929-85685-0062-62366-0 from training. Duration: 20.26
2022-12-23 00:14:33,118 INFO [train.py:894] (2/4) Epoch 10, batch 1250, loss[loss=0.2093, simple_loss=0.3025, pruned_loss=0.05809, over 18731.00 frames. ], tot_loss[loss=0.2064, simple_loss=0.2883, pruned_loss=0.06226, over 3706891.78 frames. ], batch size: 54, lr: 1.17e-02, grad_scale: 8.0
2022-12-23 00:14:36,116 WARNING [train.py:1060] (2/4) Exclude cut with ID 5239-32139-0030-9324-0_sp0.9 from training. Duration: 21.3444375
2022-12-23 00:14:36,382 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.3897, 1.0979, 1.4774, 2.6055, 1.7907, 2.2672, 0.9129, 1.7686],
       device='cuda:2'), covar=tensor([0.1879, 0.1727, 0.1477, 0.0588, 0.1203, 0.1083, 0.2061, 0.1277],
       device='cuda:2'), in_proj_covar=tensor([0.0103, 0.0113, 0.0129, 0.0123, 0.0102, 0.0133, 0.0129, 0.0109],
       device='cuda:2'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003],
       device='cuda:2')
2022-12-23 00:15:00,106 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 2.696e+02 4.479e+02 5.169e+02 6.782e+02 1.244e+03, threshold=1.034e+03, percent-clipped=4.0
2022-12-23 00:15:24,692 INFO [scaling.py:679] (2/4) Whitening: num_groups=8, num_channels=96, metric=1.33 vs. limit=2.0
2022-12-23 00:15:36,291 WARNING [train.py:1060] (2/4) Exclude cut with ID 497-129325-0061-62254-0_sp1.1 from training. Duration: 0.97725
2022-12-23 00:15:49,582 INFO [train.py:894] (2/4) Epoch 10, batch 1300, loss[loss=0.1957, simple_loss=0.2734, pruned_loss=0.05896, over 18621.00 frames. ], tot_loss[loss=0.2066, simple_loss=0.2886, pruned_loss=0.06228, over 3708675.12 frames. ], batch size: 45, lr: 1.17e-02, grad_scale: 8.0
2022-12-23 00:16:17,490 WARNING [train.py:1060] (2/4) Exclude cut with ID 7395-89880-0031-39906-0_sp0.9 from training. Duration: 22.97225
2022-12-23 00:16:48,466 WARNING [train.py:1060] (2/4) Exclude cut with ID 7395-89880-0047-39922-0_sp0.9 from training. Duration: 21.97775
2022-12-23 00:16:54,735 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.4970, 1.5933, 1.4975, 1.6181, 1.3837, 3.7000, 1.8041, 2.2211],
       device='cuda:2'), covar=tensor([0.3987, 0.2441, 0.2253, 0.2284, 0.1418, 0.0178, 0.1418, 0.0921],
       device='cuda:2'), in_proj_covar=tensor([0.0139, 0.0121, 0.0130, 0.0121, 0.0104, 0.0098, 0.0098, 0.0095],
       device='cuda:2'), out_proj_covar=tensor([0.0005, 0.0005, 0.0005, 0.0004, 0.0004, 0.0003, 0.0004, 0.0004],
       device='cuda:2')
2022-12-23 00:17:02,926 WARNING [train.py:1060] (2/4) Exclude cut with ID 1112-1043-0006-89194-0_sp0.9 from training. Duration: 21.8333125
2022-12-23 00:17:05,852 INFO [train.py:894] (2/4) Epoch 10, batch 1350, loss[loss=0.2102, simple_loss=0.3014, pruned_loss=0.05946, over 18649.00 frames. ], tot_loss[loss=0.207, simple_loss=0.2894, pruned_loss=0.06227, over 3710513.73 frames. ], batch size: 78, lr: 1.17e-02, grad_scale: 8.0
2022-12-23 00:17:12,927 WARNING [train.py:1060] (2/4) Exclude cut with ID 1914-133440-0031-94921-0 from training. Duration: 20.47
2022-12-23 00:17:33,576 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 3.207e+02 4.434e+02 5.433e+02 6.652e+02 1.215e+03, threshold=1.087e+03, percent-clipped=3.0
2022-12-23 00:18:19,064 WARNING [train.py:1060] (2/4) Exclude cut with ID 7395-89880-0037-39912-0_sp0.9 from training. Duration: 20.67225
2022-12-23 00:18:22,022 INFO [train.py:894] (2/4) Epoch 10, batch 1400, loss[loss=0.1793, simple_loss=0.2645, pruned_loss=0.04706, over 18678.00 frames. ], tot_loss[loss=0.2052, simple_loss=0.2877, pruned_loss=0.06135, over 3710796.11 frames. ], batch size: 46, lr: 1.17e-02, grad_scale: 8.0
2022-12-23 00:18:38,003 INFO [zipformer.py:660] (2/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=32966.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 00:18:39,200 WARNING [train.py:1060] (2/4) Exclude cut with ID 1914-133440-0024-94914-0_sp0.9 from training. Duration: 25.2444375
2022-12-23 00:18:52,145 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.6187, 1.6193, 1.1506, 1.5779, 1.7667, 1.5351, 2.1641, 1.7350],
       device='cuda:2'), covar=tensor([0.0891, 0.1532, 0.2499, 0.1658, 0.1627, 0.0870, 0.0986, 0.1118],
       device='cuda:2'), in_proj_covar=tensor([0.0172, 0.0192, 0.0231, 0.0279, 0.0221, 0.0181, 0.0205, 0.0193],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-23 00:19:03,584 WARNING [train.py:1060] (2/4) Exclude cut with ID 3340-169293-0021-76797-0_sp0.9 from training. Duration: 21.1445
2022-12-23 00:19:37,772 INFO [train.py:894] (2/4) Epoch 10, batch 1450, loss[loss=0.1962, simple_loss=0.2657, pruned_loss=0.06331, over 18397.00 frames. ], tot_loss[loss=0.2044, simple_loss=0.2868, pruned_loss=0.06102, over 3710362.02 frames. ], batch size: 46, lr: 1.16e-02, grad_scale: 8.0
2022-12-23 00:20:05,660 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 2.915e+02 4.368e+02 5.270e+02 6.859e+02 1.163e+03, threshold=1.054e+03, percent-clipped=2.0
2022-12-23 00:20:10,968 INFO [zipformer.py:660] (2/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=33027.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 00:20:16,464 WARNING [train.py:1060] (2/4) Exclude cut with ID 2929-85685-0079-62383-0_sp0.9 from training. Duration: 33.038875
2022-12-23 00:20:53,388 INFO [train.py:894] (2/4) Epoch 10, batch 1500, loss[loss=0.1997, simple_loss=0.2864, pruned_loss=0.05645, over 18588.00 frames. ], tot_loss[loss=0.2049, simple_loss=0.2876, pruned_loss=0.06113, over 3711794.24 frames. ], batch size: 51, lr: 1.16e-02, grad_scale: 8.0
2022-12-23 00:20:53,432 WARNING [train.py:1060] (2/4) Exclude cut with ID 6426-64291-0000-16059-0_sp0.9 from training. Duration: 20.0944375
2022-12-23 00:21:09,561 WARNING [train.py:1060] (2/4) Exclude cut with ID 2929-85685-0044-62348-0_sp1.1 from training. Duration: 20.4
2022-12-23 00:21:14,081 INFO [zipformer.py:660] (2/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=33069.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 00:21:17,299 WARNING [train.py:1060] (2/4) Exclude cut with ID 6330-62851-0022-91297-0 from training. Duration: 20.085
2022-12-23 00:21:27,373 WARNING [train.py:1060] (2/4) Exclude cut with ID 4860-13185-0032-76709-0_sp0.9 from training. Duration: 23.07775
2022-12-23 00:22:08,911 INFO [train.py:894] (2/4) Epoch 10, batch 1550, loss[loss=0.2102, simple_loss=0.3005, pruned_loss=0.05997, over 18516.00 frames. ], tot_loss[loss=0.2053, simple_loss=0.2878, pruned_loss=0.06142, over 3712365.76 frames. ], batch size: 52, lr: 1.16e-02, grad_scale: 8.0
2022-12-23 00:22:12,909 WARNING [train.py:1060] (2/4) Exclude cut with ID 2929-85685-0044-62348-0_sp0.9 from training. Duration: 24.9333125
2022-12-23 00:22:37,936 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 2.803e+02 4.567e+02 5.574e+02 6.708e+02 1.321e+03, threshold=1.115e+03, percent-clipped=4.0
2022-12-23 00:22:48,019 INFO [zipformer.py:660] (2/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=33130.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 00:22:57,516 WARNING [train.py:1060] (2/4) Exclude cut with ID 5118-111612-0016-124680-0_sp0.9 from training. Duration: 20.388875
2022-12-23 00:23:03,644 WARNING [train.py:1060] (2/4) Exclude cut with ID 432-122774-0017-62487-0_sp1.1 from training. Duration: 20.3590625
2022-12-23 00:23:26,068 INFO [zipformer.py:660] (2/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=33155.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 00:23:27,240 INFO [train.py:894] (2/4) Epoch 10, batch 1600, loss[loss=0.201, simple_loss=0.2657, pruned_loss=0.06811, over 18575.00 frames. ], tot_loss[loss=0.2046, simple_loss=0.2869, pruned_loss=0.06114, over 3712055.26 frames. ], batch size: 41, lr: 1.16e-02, grad_scale: 8.0
2022-12-23 00:24:10,545 WARNING [train.py:1060] (2/4) Exclude cut with ID 3557-8342-0013-54691-0_sp1.1 from training. Duration: 0.836375
2022-12-23 00:24:16,017 INFO [scaling.py:679] (2/4) Whitening: num_groups=8, num_channels=96, metric=1.36 vs. limit=2.0
2022-12-23 00:24:43,958 INFO [train.py:894] (2/4) Epoch 10, batch 1650, loss[loss=0.2396, simple_loss=0.3159, pruned_loss=0.08165, over 18681.00 frames. ], tot_loss[loss=0.2067, simple_loss=0.2886, pruned_loss=0.06245, over 3713208.81 frames. ], batch size: 69, lr: 1.16e-02, grad_scale: 8.0
2022-12-23 00:24:56,496 WARNING [train.py:1060] (2/4) Exclude cut with ID 8565-290391-0049-67394-0_sp0.9 from training. Duration: 21.3166875
2022-12-23 00:24:59,869 INFO [zipformer.py:660] (2/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=33216.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 00:25:08,721 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.5586, 1.8930, 1.1954, 2.2104, 2.4983, 1.4290, 1.5504, 1.1587],
       device='cuda:2'), covar=tensor([0.1780, 0.1585, 0.1590, 0.0902, 0.1233, 0.1179, 0.1757, 0.1444],
       device='cuda:2'), in_proj_covar=tensor([0.0233, 0.0207, 0.0198, 0.0185, 0.0248, 0.0186, 0.0205, 0.0185],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-23 00:25:11,643 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 2.064e+02 4.439e+02 5.632e+02 7.567e+02 2.415e+03, threshold=1.126e+03, percent-clipped=5.0
2022-12-23 00:25:25,813 WARNING [train.py:1060] (2/4) Exclude cut with ID 6533-399-0029-104863-0_sp0.9 from training. Duration: 22.1055625
2022-12-23 00:25:36,009 WARNING [train.py:1060] (2/4) Exclude cut with ID 7699-105389-0094-26379-0_sp1.1 from training. Duration: 21.77725
2022-12-23 00:25:58,161 WARNING [train.py:1060] (2/4) Exclude cut with ID 7255-291500-0005-134304-0_sp0.9 from training. Duration: 27.8166875
2022-12-23 00:26:00,239 INFO [train.py:894] (2/4) Epoch 10, batch 1700, loss[loss=0.2041, simple_loss=0.2777, pruned_loss=0.0652, over 18415.00 frames. ], tot_loss[loss=0.2104, simple_loss=0.2908, pruned_loss=0.06498, over 3712960.77 frames. ], batch size: 48, lr: 1.16e-02, grad_scale: 8.0
2022-12-23 00:26:21,873 WARNING [train.py:1060] (2/4) Exclude cut with ID 7357-94126-0021-15852-0_sp1.1 from training. Duration: 22.5090625
2022-12-23 00:26:30,218 WARNING [train.py:1060] (2/4) Exclude cut with ID 7255-291500-0005-134304-0 from training. Duration: 25.035
2022-12-23 00:26:47,905 WARNING [train.py:1060] (2/4) Exclude cut with ID 774-127930-0014-10412-0_sp1.1 from training. Duration: 0.95
2022-12-23 00:27:07,655 WARNING [train.py:1060] (2/4) Exclude cut with ID 3033-130750-0096-55598-0_sp0.9 from training. Duration: 0.92225
2022-12-23 00:27:16,768 INFO [train.py:894] (2/4) Epoch 10, batch 1750, loss[loss=0.2765, simple_loss=0.3285, pruned_loss=0.1122, over 18455.00 frames. ], tot_loss[loss=0.2149, simple_loss=0.2936, pruned_loss=0.06811, over 3713994.33 frames. ], batch size: 50, lr: 1.16e-02, grad_scale: 8.0
2022-12-23 00:27:34,197 WARNING [train.py:1060] (2/4) Exclude cut with ID 4511-76322-0006-80011-0 from training. Duration: 21.97
2022-12-23 00:27:42,354 INFO [zipformer.py:660] (2/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=33322.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 00:27:45,019 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 3.376e+02 5.253e+02 6.174e+02 7.896e+02 1.469e+03, threshold=1.235e+03, percent-clipped=6.0
2022-12-23 00:27:53,781 WARNING [train.py:1060] (2/4) Exclude cut with ID 7492-105653-0055-62765-0_sp0.9 from training. Duration: 21.97225
2022-12-23 00:27:55,141 WARNING [train.py:1060] (2/4) Exclude cut with ID 453-131332-0000-47844-0_sp0.9 from training. Duration: 25.3333125
2022-12-23 00:28:07,221 WARNING [train.py:1060] (2/4) Exclude cut with ID 5172-29468-0015-19128-0_sp0.9 from training. Duration: 21.5055625
2022-12-23 00:28:15,118 WARNING [train.py:1060] (2/4) Exclude cut with ID 453-131332-0000-47844-0_sp1.1 from training. Duration: 20.72725
2022-12-23 00:28:33,082 INFO [train.py:894] (2/4) Epoch 10, batch 1800, loss[loss=0.2379, simple_loss=0.3088, pruned_loss=0.08346, over 18721.00 frames. ], tot_loss[loss=0.2168, simple_loss=0.2939, pruned_loss=0.06985, over 3713808.55 frames. ], batch size: 52, lr: 1.16e-02, grad_scale: 8.0
2022-12-23 00:28:47,398 INFO [scaling.py:679] (2/4) Whitening: num_groups=8, num_channels=192, metric=1.85 vs. limit=2.0
2022-12-23 00:28:49,009 WARNING [train.py:1060] (2/4) Exclude cut with ID 8631-249866-0030-130156-0_sp0.9 from training. Duration: 26.32775
2022-12-23 00:28:57,294 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([3.1756, 2.4135, 1.9777, 2.9306, 2.2304, 2.2440, 2.3073, 3.4914],
       device='cuda:2'), covar=tensor([0.1363, 0.2371, 0.1367, 0.2306, 0.2740, 0.0801, 0.2362, 0.0411],
       device='cuda:2'), in_proj_covar=tensor([0.0272, 0.0263, 0.0220, 0.0334, 0.0245, 0.0208, 0.0257, 0.0181],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0003, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-23 00:29:20,151 WARNING [train.py:1060] (2/4) Exclude cut with ID 3867-173237-0077-144769-0 from training. Duration: 20.025
2022-12-23 00:29:24,474 WARNING [train.py:1060] (2/4) Exclude cut with ID 6709-74022-0004-86860-0_sp1.1 from training. Duration: 0.9409375
2022-12-23 00:29:24,489 WARNING [train.py:1060] (2/4) Exclude cut with ID 4757-1811-0023-62229-0_sp0.9 from training. Duration: 21.37775
2022-12-23 00:29:48,010 WARNING [train.py:1060] (2/4) Exclude cut with ID 1250-135782-0004-25974-0_sp0.9 from training. Duration: 21.17225
2022-12-23 00:29:48,021 WARNING [train.py:1060] (2/4) Exclude cut with ID 7357-94126-0021-15852-0_sp0.9 from training. Duration: 27.511125
2022-12-23 00:29:51,057 INFO [train.py:894] (2/4) Epoch 10, batch 1850, loss[loss=0.2051, simple_loss=0.2838, pruned_loss=0.06316, over 18602.00 frames. ], tot_loss[loss=0.2205, simple_loss=0.2958, pruned_loss=0.07256, over 3713990.38 frames. ], batch size: 51, lr: 1.16e-02, grad_scale: 4.0
2022-12-23 00:30:19,986 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 3.458e+02 5.783e+02 6.744e+02 8.262e+02 1.612e+03, threshold=1.349e+03, percent-clipped=7.0
2022-12-23 00:30:20,196 INFO [zipformer.py:660] (2/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=33425.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 00:30:21,548 WARNING [train.py:1060] (2/4) Exclude cut with ID 453-131332-0000-47844-0 from training. Duration: 22.8
2022-12-23 00:30:24,354 WARNING [train.py:1060] (2/4) Exclude cut with ID 4964-30587-0040-44509-0 from training. Duration: 22.585
2022-12-23 00:30:55,948 WARNING [train.py:1060] (2/4) Exclude cut with ID 6978-92210-0001-146967-0_sp0.9 from training. Duration: 22.0166875
2022-12-23 00:31:02,519 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.4163, 1.9254, 1.4322, 2.4325, 2.4636, 1.4986, 1.5824, 1.2643],
       device='cuda:2'), covar=tensor([0.1950, 0.1594, 0.1488, 0.0823, 0.1387, 0.1162, 0.1964, 0.1418],
       device='cuda:2'), in_proj_covar=tensor([0.0234, 0.0206, 0.0198, 0.0184, 0.0249, 0.0185, 0.0206, 0.0185],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-23 00:31:07,984 INFO [train.py:894] (2/4) Epoch 10, batch 1900, loss[loss=0.1886, simple_loss=0.2743, pruned_loss=0.05144, over 18550.00 frames. ], tot_loss[loss=0.223, simple_loss=0.2971, pruned_loss=0.07448, over 3713626.63 frames. ], batch size: 49, lr: 1.16e-02, grad_scale: 4.0
2022-12-23 00:31:11,682 WARNING [train.py:1060] (2/4) Exclude cut with ID 7255-291500-0003-134302-0_sp1.1 from training. Duration: 24.395375
2022-12-23 00:31:19,001 WARNING [train.py:1060] (2/4) Exclude cut with ID 3488-85273-0017-41203-0_sp0.9 from training. Duration: 27.47775
2022-12-23 00:31:23,408 WARNING [train.py:1060] (2/4) Exclude cut with ID 432-122774-0017-62487-0_sp0.9 from training. Duration: 24.8833125
2022-12-23 00:31:26,432 WARNING [train.py:1060] (2/4) Exclude cut with ID 6758-72288-0033-108368-0 from training. Duration: 23.39
2022-12-23 00:31:31,939 WARNING [train.py:1060] (2/4) Exclude cut with ID 8291-282929-0007-12994-0_sp0.9 from training. Duration: 28.72225
2022-12-23 00:31:32,183 INFO [zipformer.py:660] (2/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=33472.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 00:31:40,728 WARNING [train.py:1060] (2/4) Exclude cut with ID 585-294811-0110-133686-0_sp0.9 from training. Duration: 20.8944375
2022-12-23 00:31:58,317 WARNING [train.py:1060] (2/4) Exclude cut with ID 5796-66357-0007-116447-0_sp0.9 from training. Duration: 23.8444375
2022-12-23 00:32:24,037 INFO [train.py:894] (2/4) Epoch 10, batch 1950, loss[loss=0.2415, simple_loss=0.3077, pruned_loss=0.08763, over 18499.00 frames. ], tot_loss[loss=0.224, simple_loss=0.2973, pruned_loss=0.07531, over 3713834.33 frames. ], batch size: 76, lr: 1.16e-02, grad_scale: 4.0
2022-12-23 00:32:25,446 WARNING [train.py:1060] (2/4) Exclude cut with ID 8291-282929-0007-12994-0 from training. Duration: 25.85
2022-12-23 00:32:25,457 WARNING [train.py:1060] (2/4) Exclude cut with ID 8291-282929-0023-13010-0 from training. Duration: 21.39
2022-12-23 00:32:31,460 INFO [zipformer.py:660] (2/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=33511.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 00:32:35,916 WARNING [train.py:1060] (2/4) Exclude cut with ID 7357-94126-0014-15845-0 from training. Duration: 27.92
2022-12-23 00:32:51,914 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 3.820e+02 5.421e+02 6.446e+02 7.879e+02 3.029e+03, threshold=1.289e+03, percent-clipped=5.0
2022-12-23 00:33:02,303 WARNING [train.py:1060] (2/4) Exclude cut with ID 8631-249866-0039-130165-0_sp0.9 from training. Duration: 20.661125
2022-12-23 00:33:04,249 INFO [zipformer.py:660] (2/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=33533.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 00:33:29,354 WARNING [train.py:1060] (2/4) Exclude cut with ID 7357-94126-0043-15874-0_sp0.9 from training. Duration: 20.07225
2022-12-23 00:33:36,900 WARNING [train.py:1060] (2/4) Exclude cut with ID 1085-156170-0017-128270-0 from training. Duration: 21.01
2022-12-23 00:33:40,395 INFO [train.py:894] (2/4) Epoch 10, batch 2000, loss[loss=0.1978, simple_loss=0.286, pruned_loss=0.05486, over 18617.00 frames. ], tot_loss[loss=0.2237, simple_loss=0.2967, pruned_loss=0.07538, over 3714011.74 frames. ], batch size: 53, lr: 1.16e-02, grad_scale: 8.0
2022-12-23 00:33:48,609 INFO [scaling.py:679] (2/4) Whitening: num_groups=1, num_channels=384, metric=3.17 vs. limit=5.0
2022-12-23 00:34:49,397 INFO [scaling.py:679] (2/4) Whitening: num_groups=8, num_channels=96, metric=1.36 vs. limit=2.0
2022-12-23 00:34:51,301 WARNING [train.py:1060] (2/4) Exclude cut with ID 2195-150901-0045-59933-0 from training. Duration: 20.65
2022-12-23 00:34:56,525 INFO [train.py:894] (2/4) Epoch 10, batch 2050, loss[loss=0.2053, simple_loss=0.2857, pruned_loss=0.06245, over 18599.00 frames. ], tot_loss[loss=0.2232, simple_loss=0.2961, pruned_loss=0.07516, over 3714357.65 frames. ], batch size: 51, lr: 1.15e-02, grad_scale: 8.0
2022-12-23 00:34:58,002 WARNING [train.py:1060] (2/4) Exclude cut with ID 5796-66357-0007-116447-0 from training. Duration: 21.46
2022-12-23 00:35:21,233 INFO [zipformer.py:660] (2/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=33622.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 00:35:25,501 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 3.406e+02 5.626e+02 7.229e+02 8.971e+02 2.818e+03, threshold=1.446e+03, percent-clipped=6.0
2022-12-23 00:35:42,082 WARNING [train.py:1060] (2/4) Exclude cut with ID 3557-8342-0013-54691-0 from training. Duration: 0.92
2022-12-23 00:35:48,299 WARNING [train.py:1060] (2/4) Exclude cut with ID 8291-282929-0023-13010-0_sp0.9 from training. Duration: 23.7666875
2022-12-23 00:36:12,680 INFO [train.py:894] (2/4) Epoch 10, batch 2100, loss[loss=0.2501, simple_loss=0.3237, pruned_loss=0.0883, over 18564.00 frames. ], tot_loss[loss=0.2242, simple_loss=0.2971, pruned_loss=0.07559, over 3714817.89 frames. ], batch size: 57, lr: 1.15e-02, grad_scale: 8.0
2022-12-23 00:36:25,021 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.7656, 1.7615, 1.3893, 1.7208, 1.6470, 1.6108, 1.5658, 1.8177],
       device='cuda:2'), covar=tensor([0.1964, 0.2483, 0.1562, 0.2071, 0.2633, 0.0921, 0.2233, 0.0707],
       device='cuda:2'), in_proj_covar=tensor([0.0278, 0.0267, 0.0225, 0.0344, 0.0250, 0.0213, 0.0262, 0.0186],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0003, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-23 00:36:25,877 WARNING [train.py:1060] (2/4) Exclude cut with ID 8544-281189-0060-101339-0_sp0.9 from training. Duration: 20.861125
2022-12-23 00:36:33,425 INFO [zipformer.py:660] (2/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=33670.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 00:36:37,497 WARNING [train.py:1060] (2/4) Exclude cut with ID 3488-65654-0031-41259-0_sp0.9 from training. Duration: 22.711125
2022-12-23 00:36:43,681 INFO [zipformer.py:660] (2/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=33677.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 00:37:22,058 WARNING [train.py:1060] (2/4) Exclude cut with ID 6951-79737-0043-132310-0_sp1.1 from training. Duration: 22.986375
2022-12-23 00:37:27,737 INFO [train.py:894] (2/4) Epoch 10, batch 2150, loss[loss=0.1955, simple_loss=0.2778, pruned_loss=0.0566, over 18543.00 frames. ], tot_loss[loss=0.2244, simple_loss=0.2968, pruned_loss=0.07606, over 3714790.26 frames. ], batch size: 55, lr: 1.15e-02, grad_scale: 8.0
2022-12-23 00:37:37,922 WARNING [train.py:1060] (2/4) Exclude cut with ID 8040-260924-0003-80960-0_sp0.9 from training. Duration: 22.07225
2022-12-23 00:37:42,271 WARNING [train.py:1060] (2/4) Exclude cut with ID 7699-105389-0045-26330-0_sp0.9 from training. Duration: 20.3055625
2022-12-23 00:37:45,587 WARNING [train.py:1060] (2/4) Exclude cut with ID 6356-271890-0060-94317-0_sp0.9 from training. Duration: 20.72225
2022-12-23 00:37:55,475 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 3.380e+02 5.930e+02 7.248e+02 8.853e+02 1.852e+03, threshold=1.450e+03, percent-clipped=4.0
2022-12-23 00:37:55,820 INFO [zipformer.py:660] (2/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=33725.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 00:38:03,017 WARNING [train.py:1060] (2/4) Exclude cut with ID 3488-85273-0017-41203-0_sp1.1 from training. Duration: 22.4818125
2022-12-23 00:38:17,539 INFO [zipformer.py:660] (2/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=33738.0, num_to_drop=1, layers_to_drop={3}
2022-12-23 00:38:29,809 WARNING [train.py:1060] (2/4) Exclude cut with ID 4964-30587-0040-44509-0_sp0.9 from training. Duration: 25.0944375
2022-12-23 00:38:32,657 WARNING [train.py:1060] (2/4) Exclude cut with ID 6533-399-0047-104881-0 from training. Duration: 21.515
2022-12-23 00:38:40,057 WARNING [train.py:1060] (2/4) Exclude cut with ID 7357-94126-0009-15840-0_sp0.9 from training. Duration: 27.02225
2022-12-23 00:38:44,312 INFO [train.py:894] (2/4) Epoch 10, batch 2200, loss[loss=0.2192, simple_loss=0.2961, pruned_loss=0.07121, over 18670.00 frames. ], tot_loss[loss=0.2255, simple_loss=0.2977, pruned_loss=0.07664, over 3715083.16 frames. ], batch size: 62, lr: 1.15e-02, grad_scale: 8.0
2022-12-23 00:38:45,868 WARNING [train.py:1060] (2/4) Exclude cut with ID 432-122774-0010-62480-0_sp0.9 from training. Duration: 22.22225
2022-12-23 00:38:53,446 WARNING [train.py:1060] (2/4) Exclude cut with ID 4964-30587-0085-44554-0_sp0.9 from training. Duration: 20.85
2022-12-23 00:38:55,563 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([2.3010, 1.7981, 1.7910, 1.7422, 2.2124, 2.6452, 2.5573, 2.0724],
       device='cuda:2'), covar=tensor([0.0376, 0.0341, 0.0417, 0.0286, 0.0266, 0.0373, 0.0473, 0.0289],
       device='cuda:2'), in_proj_covar=tensor([0.0083, 0.0117, 0.0142, 0.0121, 0.0107, 0.0106, 0.0089, 0.0117],
       device='cuda:2'), out_proj_covar=tensor([7.0632e-05, 9.8422e-05, 1.2546e-04, 1.0300e-04, 9.4131e-05, 8.7130e-05,
        7.4858e-05, 9.8413e-05], device='cuda:2')
2022-12-23 00:38:59,458 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([2.7223, 3.6359, 3.6635, 1.6438, 3.5573, 2.7484, 0.9775, 2.5299],
       device='cuda:2'), covar=tensor([0.2044, 0.1014, 0.1333, 0.3678, 0.1110, 0.1032, 0.5065, 0.1568],
       device='cuda:2'), in_proj_covar=tensor([0.0136, 0.0122, 0.0150, 0.0121, 0.0126, 0.0106, 0.0144, 0.0109],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-23 00:39:09,566 INFO [zipformer.py:660] (2/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=33773.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 00:39:26,331 WARNING [train.py:1060] (2/4) Exclude cut with ID 4295-39940-0007-92567-0 from training. Duration: 21.54
2022-12-23 00:39:30,665 WARNING [train.py:1060] (2/4) Exclude cut with ID 4964-30587-0040-44509-0_sp1.1 from training. Duration: 20.5318125
2022-12-23 00:39:41,703 WARNING [train.py:1060] (2/4) Exclude cut with ID 7255-291500-0012-134311-0_sp0.9 from training. Duration: 21.9333125
2022-12-23 00:40:00,937 INFO [train.py:894] (2/4) Epoch 10, batch 2250, loss[loss=0.2228, simple_loss=0.3041, pruned_loss=0.07077, over 18592.00 frames. ], tot_loss[loss=0.2243, simple_loss=0.2967, pruned_loss=0.0759, over 3714674.98 frames. ], batch size: 56, lr: 1.15e-02, grad_scale: 8.0
2022-12-23 00:40:08,722 INFO [zipformer.py:660] (2/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=33811.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 00:40:29,047 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 3.109e+02 5.120e+02 6.230e+02 8.011e+02 1.544e+03, threshold=1.246e+03, percent-clipped=1.0
2022-12-23 00:40:29,085 WARNING [train.py:1060] (2/4) Exclude cut with ID 8631-249866-0025-130151-0_sp0.9 from training. Duration: 21.7944375
2022-12-23 00:40:34,265 INFO [zipformer.py:660] (2/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=33828.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 00:40:36,689 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.3832, 1.1379, 1.2451, 1.3947, 1.6409, 1.4828, 1.5776, 1.0909],
       device='cuda:2'), covar=tensor([0.0279, 0.0231, 0.0425, 0.0191, 0.0179, 0.0332, 0.0213, 0.0266],
       device='cuda:2'), in_proj_covar=tensor([0.0084, 0.0118, 0.0144, 0.0122, 0.0108, 0.0106, 0.0089, 0.0118],
       device='cuda:2'), out_proj_covar=tensor([7.1441e-05, 9.9327e-05, 1.2698e-04, 1.0352e-04, 9.4660e-05, 8.6856e-05,
        7.5226e-05, 9.8975e-05], device='cuda:2')
2022-12-23 00:40:42,360 WARNING [train.py:1060] (2/4) Exclude cut with ID 8291-282929-0002-12989-0_sp0.9 from training. Duration: 22.4666875
2022-12-23 00:40:50,355 WARNING [train.py:1060] (2/4) Exclude cut with ID 6121-9014-0076-24124-0 from training. Duration: 21.635
2022-12-23 00:40:54,664 WARNING [train.py:1060] (2/4) Exclude cut with ID 6121-9014-0076-24124-0_sp0.9 from training. Duration: 24.038875
2022-12-23 00:41:12,551 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([5.9236, 4.9857, 5.1491, 5.8531, 5.3255, 5.2505, 5.8287, 1.8225],
       device='cuda:2'), covar=tensor([0.0501, 0.0497, 0.0463, 0.0615, 0.1152, 0.0842, 0.0404, 0.4477],
       device='cuda:2'), in_proj_covar=tensor([0.0300, 0.0200, 0.0211, 0.0215, 0.0287, 0.0241, 0.0243, 0.0260],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0001, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-23 00:41:16,739 INFO [train.py:894] (2/4) Epoch 10, batch 2300, loss[loss=0.1862, simple_loss=0.2568, pruned_loss=0.05784, over 18491.00 frames. ], tot_loss[loss=0.2245, simple_loss=0.2966, pruned_loss=0.07616, over 3714827.84 frames. ], batch size: 43, lr: 1.15e-02, grad_scale: 8.0
2022-12-23 00:41:21,675 INFO [zipformer.py:660] (2/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=33859.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 00:41:37,415 WARNING [train.py:1060] (2/4) Exclude cut with ID 7255-291500-0009-134308-0_sp1.1 from training. Duration: 21.786375
2022-12-23 00:41:50,706 WARNING [train.py:1060] (2/4) Exclude cut with ID 8291-282929-0002-12989-0 from training. Duration: 20.22
2022-12-23 00:42:33,581 INFO [train.py:894] (2/4) Epoch 10, batch 2350, loss[loss=0.2169, simple_loss=0.2971, pruned_loss=0.06839, over 18393.00 frames. ], tot_loss[loss=0.2246, simple_loss=0.2967, pruned_loss=0.07628, over 3714851.47 frames. ], batch size: 53, lr: 1.15e-02, grad_scale: 8.0
2022-12-23 00:42:44,917 INFO [scaling.py:679] (2/4) Whitening: num_groups=8, num_channels=96, metric=1.13 vs. limit=2.0
2022-12-23 00:43:02,606 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 3.812e+02 5.316e+02 6.848e+02 8.119e+02 2.151e+03, threshold=1.370e+03, percent-clipped=4.0
2022-12-23 00:43:13,876 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.7992, 1.0720, 0.8677, 1.2332, 2.0501, 1.2007, 1.6446, 1.7643],
       device='cuda:2'), covar=tensor([0.1652, 0.2363, 0.2484, 0.1721, 0.1876, 0.1678, 0.1530, 0.1694],
       device='cuda:2'), in_proj_covar=tensor([0.0091, 0.0101, 0.0121, 0.0097, 0.0114, 0.0090, 0.0097, 0.0094],
       device='cuda:2'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003],
       device='cuda:2')
2022-12-23 00:43:24,614 INFO [zipformer.py:660] (2/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=33939.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 00:43:48,627 INFO [train.py:894] (2/4) Epoch 10, batch 2400, loss[loss=0.2143, simple_loss=0.279, pruned_loss=0.07483, over 18551.00 frames. ], tot_loss[loss=0.2247, simple_loss=0.2965, pruned_loss=0.07642, over 3715481.75 frames. ], batch size: 44, lr: 1.15e-02, grad_scale: 8.0
2022-12-23 00:43:51,579 WARNING [train.py:1060] (2/4) Exclude cut with ID 6951-79737-0043-132310-0 from training. Duration: 25.285
2022-12-23 00:44:04,227 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([3.7687, 3.1886, 2.8110, 1.3368, 2.6182, 2.5336, 2.0664, 2.3589],
       device='cuda:2'), covar=tensor([0.0516, 0.0476, 0.1291, 0.1656, 0.1495, 0.1165, 0.1439, 0.0969],
       device='cuda:2'), in_proj_covar=tensor([0.0164, 0.0180, 0.0201, 0.0196, 0.0206, 0.0190, 0.0205, 0.0193],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-23 00:44:58,832 INFO [zipformer.py:660] (2/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=34000.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 00:44:59,800 WARNING [train.py:1060] (2/4) Exclude cut with ID 811-130148-0001-63453-0_sp0.9 from training. Duration: 20.861125
2022-12-23 00:45:07,229 INFO [train.py:894] (2/4) Epoch 10, batch 2450, loss[loss=0.2217, simple_loss=0.3013, pruned_loss=0.07102, over 18584.00 frames. ], tot_loss[loss=0.2249, simple_loss=0.2966, pruned_loss=0.07659, over 3715476.36 frames. ], batch size: 56, lr: 1.15e-02, grad_scale: 8.0
2022-12-23 00:45:21,398 WARNING [train.py:1060] (2/4) Exclude cut with ID 6010-56788-0055-90261-0 from training. Duration: 20.88
2022-12-23 00:45:38,421 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 3.233e+02 5.336e+02 6.652e+02 9.061e+02 1.953e+03, threshold=1.330e+03, percent-clipped=5.0
2022-12-23 00:45:50,191 INFO [zipformer.py:660] (2/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=34033.0, num_to_drop=1, layers_to_drop={2}
2022-12-23 00:45:50,860 INFO [scaling.py:679] (2/4) Whitening: num_groups=8, num_channels=96, metric=1.28 vs. limit=2.0
2022-12-23 00:45:54,409 WARNING [train.py:1060] (2/4) Exclude cut with ID 7357-94126-0045-15876-0_sp0.9 from training. Duration: 23.4166875
2022-12-23 00:46:07,184 INFO [zipformer.py:660] (2/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=34044.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 00:46:24,161 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([2.5831, 1.7234, 2.0206, 2.2302, 2.3215, 2.1514, 2.2823, 1.5460],
       device='cuda:2'), covar=tensor([0.1589, 0.2632, 0.1942, 0.2217, 0.1303, 0.0726, 0.2252, 0.0977],
       device='cuda:2'), in_proj_covar=tensor([0.0250, 0.0286, 0.0258, 0.0291, 0.0274, 0.0234, 0.0300, 0.0223],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-23 00:46:24,995 INFO [train.py:894] (2/4) Epoch 10, batch 2500, loss[loss=0.2203, simple_loss=0.2958, pruned_loss=0.07243, over 18472.00 frames. ], tot_loss[loss=0.2243, simple_loss=0.2963, pruned_loss=0.07616, over 3714649.26 frames. ], batch size: 54, lr: 1.15e-02, grad_scale: 8.0
2022-12-23 00:47:10,370 WARNING [train.py:1060] (2/4) Exclude cut with ID 7255-291500-0006-134305-0 from training. Duration: 21.24
2022-12-23 00:47:10,385 WARNING [train.py:1060] (2/4) Exclude cut with ID 6533-399-0047-104881-0_sp0.9 from training. Duration: 23.9055625
2022-12-23 00:47:29,024 INFO [scaling.py:679] (2/4) Whitening: num_groups=8, num_channels=96, metric=1.46 vs. limit=2.0
2022-12-23 00:47:40,807 INFO [zipformer.py:660] (2/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=34105.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 00:47:41,792 INFO [train.py:894] (2/4) Epoch 10, batch 2550, loss[loss=0.2364, simple_loss=0.3135, pruned_loss=0.07962, over 18532.00 frames. ], tot_loss[loss=0.2249, simple_loss=0.2967, pruned_loss=0.07657, over 3714864.52 frames. ], batch size: 55, lr: 1.15e-02, grad_scale: 8.0
2022-12-23 00:47:43,600 WARNING [train.py:1060] (2/4) Exclude cut with ID 6758-72288-0033-108368-0_sp0.9 from training. Duration: 25.988875
2022-12-23 00:47:53,005 WARNING [train.py:1060] (2/4) Exclude cut with ID 7255-291500-0001-134300-0_sp0.9 from training. Duration: 20.67225
2022-12-23 00:48:04,737 INFO [scaling.py:679] (2/4) Whitening: num_groups=8, num_channels=96, metric=1.25 vs. limit=2.0
2022-12-23 00:48:10,955 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 3.099e+02 5.662e+02 6.760e+02 8.493e+02 2.029e+03, threshold=1.352e+03, percent-clipped=2.0
2022-12-23 00:48:15,780 INFO [zipformer.py:660] (2/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=34128.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 00:48:39,445 WARNING [train.py:1060] (2/4) Exclude cut with ID 3488-85273-0038-41224-0 from training. Duration: 20.34
2022-12-23 00:48:41,302 INFO [zipformer.py:660] (2/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=34145.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 00:48:59,218 INFO [train.py:894] (2/4) Epoch 10, batch 2600, loss[loss=0.2055, simple_loss=0.2942, pruned_loss=0.05841, over 18515.00 frames. ], tot_loss[loss=0.2248, simple_loss=0.2966, pruned_loss=0.07648, over 3713924.63 frames. ], batch size: 77, lr: 1.15e-02, grad_scale: 8.0
2022-12-23 00:49:03,475 INFO [scaling.py:679] (2/4) Whitening: num_groups=8, num_channels=96, metric=1.19 vs. limit=2.0
2022-12-23 00:49:30,098 INFO [zipformer.py:660] (2/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=34176.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 00:49:36,168 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.5399, 1.1211, 1.4473, 2.5704, 1.7520, 2.2434, 0.7569, 1.8148],
       device='cuda:2'), covar=tensor([0.1812, 0.1904, 0.1697, 0.0729, 0.1340, 0.1145, 0.2399, 0.1278],
       device='cuda:2'), in_proj_covar=tensor([0.0105, 0.0116, 0.0132, 0.0127, 0.0105, 0.0134, 0.0132, 0.0111],
       device='cuda:2'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0004, 0.0003, 0.0003],
       device='cuda:2')
2022-12-23 00:49:54,766 WARNING [train.py:1060] (2/4) Exclude cut with ID 7357-94126-0026-15857-0_sp0.9 from training. Duration: 25.061125
2022-12-23 00:50:06,422 WARNING [train.py:1060] (2/4) Exclude cut with ID 3033-130750-0096-55598-0 from training. Duration: 0.83
2022-12-23 00:50:17,347 INFO [train.py:894] (2/4) Epoch 10, batch 2650, loss[loss=0.1842, simple_loss=0.2599, pruned_loss=0.0543, over 18613.00 frames. ], tot_loss[loss=0.224, simple_loss=0.296, pruned_loss=0.076, over 3713426.08 frames. ], batch size: 45, lr: 1.14e-02, grad_scale: 8.0
2022-12-23 00:50:17,745 INFO [zipformer.py:660] (2/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=34206.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 00:50:34,675 WARNING [train.py:1060] (2/4) Exclude cut with ID 3488-85273-0017-41203-0 from training. Duration: 24.73
2022-12-23 00:50:44,953 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 3.713e+02 5.212e+02 6.485e+02 7.863e+02 1.303e+03, threshold=1.297e+03, percent-clipped=0.0
2022-12-23 00:50:46,289 WARNING [train.py:1060] (2/4) Exclude cut with ID 7255-291500-0009-134308-0 from training. Duration: 23.965
2022-12-23 00:50:54,353 WARNING [train.py:1060] (2/4) Exclude cut with ID 6978-92210-0030-146996-0_sp0.9 from training. Duration: 22.088875
2022-12-23 00:51:10,943 WARNING [train.py:1060] (2/4) Exclude cut with ID 7255-291500-0006-134305-0_sp0.9 from training. Duration: 23.6
2022-12-23 00:51:33,674 INFO [train.py:894] (2/4) Epoch 10, batch 2700, loss[loss=0.3605, simple_loss=0.3869, pruned_loss=0.1671, over 18552.00 frames. ], tot_loss[loss=0.2242, simple_loss=0.2958, pruned_loss=0.07627, over 3713831.83 frames. ], batch size: 179, lr: 1.14e-02, grad_scale: 8.0
2022-12-23 00:51:37,956 INFO [scaling.py:679] (2/4) Whitening: num_groups=8, num_channels=192, metric=2.00 vs. limit=2.0
2022-12-23 00:51:41,960 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.6802, 1.2707, 1.8342, 2.9145, 2.0262, 2.4466, 1.1835, 1.9888],
       device='cuda:2'), covar=tensor([0.1813, 0.1829, 0.1569, 0.0653, 0.1229, 0.1598, 0.2129, 0.1267],
       device='cuda:2'), in_proj_covar=tensor([0.0105, 0.0116, 0.0132, 0.0127, 0.0105, 0.0134, 0.0131, 0.0111],
       device='cuda:2'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0004, 0.0003, 0.0003],
       device='cuda:2')
2022-12-23 00:52:32,397 INFO [zipformer.py:660] (2/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=34295.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 00:52:50,115 INFO [train.py:894] (2/4) Epoch 10, batch 2750, loss[loss=0.2367, simple_loss=0.3125, pruned_loss=0.08039, over 18477.00 frames. ], tot_loss[loss=0.2241, simple_loss=0.2962, pruned_loss=0.07604, over 3714770.31 frames. ], batch size: 54, lr: 1.14e-02, grad_scale: 8.0
2022-12-23 00:52:54,411 WARNING [train.py:1060] (2/4) Exclude cut with ID 8291-282929-0024-13011-0 from training. Duration: 23.795
2022-12-23 00:53:11,516 WARNING [train.py:1060] (2/4) Exclude cut with ID 8631-249866-0030-130156-0_sp1.1 from training. Duration: 21.5409375
2022-12-23 00:53:14,416 WARNING [train.py:1060] (2/4) Exclude cut with ID 6978-92210-0019-146985-0_sp0.9 from training. Duration: 24.97775
2022-12-23 00:53:17,278 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 3.657e+02 5.849e+02 7.144e+02 9.388e+02 2.029e+03, threshold=1.429e+03, percent-clipped=11.0
2022-12-23 00:53:21,200 INFO [scaling.py:679] (2/4) Whitening: num_groups=1, num_channels=384, metric=4.19 vs. limit=5.0
2022-12-23 00:53:26,025 WARNING [train.py:1060] (2/4) Exclude cut with ID 1085-156170-0017-128270-0_sp0.9 from training. Duration: 23.3444375
2022-12-23 00:53:29,251 INFO [zipformer.py:660] (2/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=34333.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 00:53:54,452 WARNING [train.py:1060] (2/4) Exclude cut with ID 6010-56788-0055-90261-0_sp0.9 from training. Duration: 23.2
2022-12-23 00:53:59,373 WARNING [train.py:1060] (2/4) Exclude cut with ID 5653-46179-0060-117930-0_sp0.9 from training. Duration: 21.17225
2022-12-23 00:54:05,528 INFO [train.py:894] (2/4) Epoch 10, batch 2800, loss[loss=0.2216, simple_loss=0.301, pruned_loss=0.0711, over 18659.00 frames. ], tot_loss[loss=0.2241, simple_loss=0.2962, pruned_loss=0.07601, over 3715327.42 frames. ], batch size: 60, lr: 1.14e-02, grad_scale: 8.0
2022-12-23 00:54:16,476 INFO [scaling.py:679] (2/4) Whitening: num_groups=1, num_channels=384, metric=3.17 vs. limit=5.0
2022-12-23 00:54:20,294 WARNING [train.py:1060] (2/4) Exclude cut with ID 7255-291500-0008-134307-0_sp0.9 from training. Duration: 24.6555625
2022-12-23 00:54:37,778 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.8697, 0.6491, 1.6279, 1.5527, 1.8142, 1.7095, 1.5054, 1.4156],
       device='cuda:2'), covar=tensor([0.1335, 0.2173, 0.1672, 0.1581, 0.1294, 0.0758, 0.1722, 0.0876],
       device='cuda:2'), in_proj_covar=tensor([0.0251, 0.0286, 0.0257, 0.0288, 0.0274, 0.0234, 0.0299, 0.0223],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-23 00:54:42,540 INFO [zipformer.py:660] (2/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=34381.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 00:55:12,425 INFO [zipformer.py:660] (2/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=34400.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 00:55:13,598 WARNING [train.py:1060] (2/4) Exclude cut with ID 3488-65654-0031-41259-0 from training. Duration: 20.44
2022-12-23 00:55:20,880 INFO [train.py:894] (2/4) Epoch 10, batch 2850, loss[loss=0.223, simple_loss=0.2984, pruned_loss=0.07381, over 18468.00 frames. ], tot_loss[loss=0.224, simple_loss=0.2959, pruned_loss=0.07606, over 3714168.40 frames. ], batch size: 50, lr: 1.14e-02, grad_scale: 8.0
2022-12-23 00:55:21,328 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([2.5851, 2.0234, 1.7935, 0.6601, 1.6122, 1.9679, 1.5504, 1.9354],
       device='cuda:2'), covar=tensor([0.0518, 0.0540, 0.1096, 0.1510, 0.1136, 0.1345, 0.1533, 0.0638],
       device='cuda:2'), in_proj_covar=tensor([0.0165, 0.0182, 0.0202, 0.0197, 0.0205, 0.0189, 0.0204, 0.0192],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-23 00:55:29,936 WARNING [train.py:1060] (2/4) Exclude cut with ID 6951-79737-0018-132285-0_sp0.9 from training. Duration: 23.45
2022-12-23 00:55:48,954 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.5982, 2.0052, 1.3729, 2.3825, 2.8383, 1.5308, 1.5892, 1.3392],
       device='cuda:2'), covar=tensor([0.1832, 0.1615, 0.1581, 0.0813, 0.1146, 0.1194, 0.1876, 0.1421],
       device='cuda:2'), in_proj_covar=tensor([0.0238, 0.0213, 0.0203, 0.0185, 0.0254, 0.0189, 0.0212, 0.0189],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-23 00:55:49,871 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 3.394e+02 5.367e+02 6.476e+02 7.793e+02 1.701e+03, threshold=1.295e+03, percent-clipped=3.0
2022-12-23 00:55:56,713 WARNING [train.py:1060] (2/4) Exclude cut with ID 6945-60535-0076-12784-0_sp0.9 from training. Duration: 20.52225
2022-12-23 00:56:04,352 WARNING [train.py:1060] (2/4) Exclude cut with ID 7255-291500-0008-134307-0 from training. Duration: 22.19
2022-12-23 00:56:15,082 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([2.2159, 2.0241, 1.8669, 1.5457, 2.1438, 2.7351, 2.4483, 1.9380],
       device='cuda:2'), covar=tensor([0.0373, 0.0246, 0.0383, 0.0265, 0.0287, 0.0281, 0.0333, 0.0270],
       device='cuda:2'), in_proj_covar=tensor([0.0085, 0.0116, 0.0144, 0.0123, 0.0109, 0.0108, 0.0091, 0.0118],
       device='cuda:2'), out_proj_covar=tensor([7.2411e-05, 9.7535e-05, 1.2673e-04, 1.0399e-04, 9.4945e-05, 8.8162e-05,
        7.6152e-05, 9.8944e-05], device='cuda:2')
2022-12-23 00:56:16,114 WARNING [train.py:1060] (2/4) Exclude cut with ID 7357-94126-0014-15845-0_sp1.1 from training. Duration: 25.3818125
2022-12-23 00:56:32,896 WARNING [train.py:1060] (2/4) Exclude cut with ID 6951-79737-0043-132310-0_sp0.9 from training. Duration: 28.0944375
2022-12-23 00:56:38,513 INFO [train.py:894] (2/4) Epoch 10, batch 2900, loss[loss=0.2282, simple_loss=0.2978, pruned_loss=0.07936, over 18599.00 frames. ], tot_loss[loss=0.2235, simple_loss=0.2955, pruned_loss=0.07574, over 3713954.47 frames. ], batch size: 51, lr: 1.14e-02, grad_scale: 8.0
2022-12-23 00:56:38,621 WARNING [train.py:1060] (2/4) Exclude cut with ID 2195-150901-0045-59933-0_sp0.9 from training. Duration: 22.9444375
2022-12-23 00:56:45,364 INFO [scaling.py:679] (2/4) Whitening: num_groups=8, num_channels=192, metric=1.91 vs. limit=2.0
2022-12-23 00:56:45,765 WARNING [train.py:1060] (2/4) Exclude cut with ID 8291-282929-0024-13011-0_sp1.1 from training. Duration: 21.6318125
2022-12-23 00:57:02,545 WARNING [train.py:1060] (2/4) Exclude cut with ID 8631-249866-0030-130156-0 from training. Duration: 23.695
2022-12-23 00:57:07,833 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([2.0556, 2.0109, 1.3895, 2.0741, 1.8105, 1.7515, 1.8209, 2.1089],
       device='cuda:2'), covar=tensor([0.1909, 0.2377, 0.1738, 0.2240, 0.2522, 0.0986, 0.2233, 0.0744],
       device='cuda:2'), in_proj_covar=tensor([0.0281, 0.0270, 0.0227, 0.0344, 0.0252, 0.0215, 0.0267, 0.0190],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0003, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-23 00:57:29,855 WARNING [train.py:1060] (2/4) Exclude cut with ID 7699-105389-0094-26379-0 from training. Duration: 23.955
2022-12-23 00:57:32,326 INFO [scaling.py:679] (2/4) Whitening: num_groups=8, num_channels=96, metric=1.33 vs. limit=2.0
2022-12-23 00:57:49,523 INFO [zipformer.py:660] (2/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=34501.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 00:57:56,468 INFO [train.py:894] (2/4) Epoch 10, batch 2950, loss[loss=0.2233, simple_loss=0.3004, pruned_loss=0.07307, over 18528.00 frames. ], tot_loss[loss=0.2234, simple_loss=0.2956, pruned_loss=0.07558, over 3714407.18 frames. ], batch size: 69, lr: 1.14e-02, grad_scale: 8.0
2022-12-23 00:58:02,566 WARNING [train.py:1060] (2/4) Exclude cut with ID 8291-282929-0024-13011-0_sp0.9 from training. Duration: 26.438875
2022-12-23 00:58:21,883 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.5881, 1.3906, 1.3644, 0.7567, 1.8640, 1.5775, 1.4781, 1.1570],
       device='cuda:2'), covar=tensor([0.0367, 0.0438, 0.0454, 0.0670, 0.0258, 0.0302, 0.0422, 0.0895],
       device='cuda:2'), in_proj_covar=tensor([0.0119, 0.0118, 0.0126, 0.0118, 0.0085, 0.0117, 0.0136, 0.0149],
       device='cuda:2'), out_proj_covar=tensor([0.0001, 0.0001, 0.0002, 0.0001, 0.0001, 0.0001, 0.0002, 0.0002],
       device='cuda:2')
2022-12-23 00:58:25,800 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 3.157e+02 5.308e+02 6.490e+02 8.090e+02 1.978e+03, threshold=1.298e+03, percent-clipped=5.0
2022-12-23 00:58:43,288 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([0.1629, 1.0687, 1.1585, 0.3896, 0.6594, 1.2470, 1.2208, 1.1444],
       device='cuda:2'), covar=tensor([0.0623, 0.0266, 0.0291, 0.0349, 0.0395, 0.0396, 0.0233, 0.0493],
       device='cuda:2'), in_proj_covar=tensor([0.0135, 0.0152, 0.0107, 0.0128, 0.0138, 0.0122, 0.0141, 0.0144],
       device='cuda:2'), out_proj_covar=tensor([1.1181e-04, 1.2732e-04, 8.7864e-05, 1.0404e-04, 1.1411e-04, 1.0345e-04,
        1.2032e-04, 1.2089e-04], device='cuda:2')
2022-12-23 00:58:45,971 WARNING [train.py:1060] (2/4) Exclude cut with ID 7699-105389-0021-26306-0_sp0.9 from training. Duration: 21.2444375
2022-12-23 00:58:45,992 WARNING [train.py:1060] (2/4) Exclude cut with ID 7357-94126-0014-15845-0_sp0.9 from training. Duration: 31.02225
2022-12-23 00:58:56,991 WARNING [train.py:1060] (2/4) Exclude cut with ID 432-122774-0017-62487-0 from training. Duration: 22.395
2022-12-23 00:59:13,105 INFO [train.py:894] (2/4) Epoch 10, batch 3000, loss[loss=0.2583, simple_loss=0.3232, pruned_loss=0.09667, over 18671.00 frames. ], tot_loss[loss=0.2241, simple_loss=0.2965, pruned_loss=0.0759, over 3713540.19 frames. ], batch size: 62, lr: 1.14e-02, grad_scale: 8.0
2022-12-23 00:59:13,106 INFO [train.py:919] (2/4) Computing validation loss
2022-12-23 00:59:16,943 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.7261, 2.0662, 1.6792, 2.2446, 2.4755, 1.5917, 1.9713, 1.5584],
       device='cuda:2'), covar=tensor([0.1648, 0.1364, 0.1289, 0.0826, 0.1193, 0.1107, 0.1640, 0.1225],
       device='cuda:2'), in_proj_covar=tensor([0.0240, 0.0214, 0.0204, 0.0187, 0.0257, 0.0191, 0.0213, 0.0191],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-23 00:59:18,970 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([0.2326, 1.4063, 1.5936, 0.7748, 0.9056, 1.6997, 1.6000, 1.4281],
       device='cuda:2'), covar=tensor([0.0719, 0.0273, 0.0344, 0.0322, 0.0405, 0.0433, 0.0223, 0.0571],
       device='cuda:2'), in_proj_covar=tensor([0.0135, 0.0152, 0.0107, 0.0128, 0.0137, 0.0123, 0.0141, 0.0145],
       device='cuda:2'), out_proj_covar=tensor([1.1215e-04, 1.2730e-04, 8.8226e-05, 1.0423e-04, 1.1381e-04, 1.0389e-04,
        1.2038e-04, 1.2112e-04], device='cuda:2')
2022-12-23 00:59:24,294 INFO [train.py:928] (2/4) Epoch 10, validation: loss=0.175, simple_loss=0.2744, pruned_loss=0.03777, over 944034.00 frames. 
2022-12-23 00:59:24,295 INFO [train.py:929] (2/4) Maximum memory allocated so far is 24680MB
2022-12-23 00:59:27,393 WARNING [train.py:1060] (2/4) Exclude cut with ID 7357-94126-0045-15876-0 from training. Duration: 21.075
2022-12-23 00:59:28,040 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([2.5590, 1.7902, 2.0574, 2.2807, 2.3000, 2.1583, 2.2306, 1.6930],
       device='cuda:2'), covar=tensor([0.1584, 0.2606, 0.1927, 0.2117, 0.1316, 0.0719, 0.2310, 0.0942],
       device='cuda:2'), in_proj_covar=tensor([0.0250, 0.0284, 0.0258, 0.0289, 0.0274, 0.0235, 0.0299, 0.0223],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-23 00:59:32,102 WARNING [train.py:1060] (2/4) Exclude cut with ID 6482-98857-0025-147532-0_sp0.9 from training. Duration: 20.0055625
2022-12-23 00:59:32,109 WARNING [train.py:1060] (2/4) Exclude cut with ID 6951-79737-0037-132304-0_sp0.9 from training. Duration: 22.05
2022-12-23 00:59:32,120 WARNING [train.py:1060] (2/4) Exclude cut with ID 7255-291500-0003-134302-0 from training. Duration: 26.8349375
2022-12-23 00:59:35,228 WARNING [train.py:1060] (2/4) Exclude cut with ID 7357-94126-0009-15840-0_sp1.1 from training. Duration: 22.1090625
2022-12-23 00:59:43,945 WARNING [train.py:1060] (2/4) Exclude cut with ID 7699-105389-0094-26379-0_sp0.9 from training. Duration: 26.6166875
2022-12-23 01:00:01,139 WARNING [train.py:1060] (2/4) Exclude cut with ID 2046-178027-0000-53705-0_sp0.9 from training. Duration: 20.3055625
2022-12-23 01:00:21,773 INFO [zipformer.py:660] (2/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=34593.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 01:00:24,334 WARNING [train.py:1060] (2/4) Exclude cut with ID 7205-50138-0008-5373-0_sp0.9 from training. Duration: 20.7
2022-12-23 01:00:24,669 INFO [zipformer.py:660] (2/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=34595.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 01:00:41,531 INFO [train.py:894] (2/4) Epoch 10, batch 3050, loss[loss=0.196, simple_loss=0.2631, pruned_loss=0.06444, over 18688.00 frames. ], tot_loss[loss=0.2238, simple_loss=0.296, pruned_loss=0.07576, over 3713601.08 frames. ], batch size: 46, lr: 1.14e-02, grad_scale: 8.0
2022-12-23 01:00:56,761 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.1547, 1.4899, 1.1501, 1.7378, 1.7811, 1.3510, 0.9970, 1.1408],
       device='cuda:2'), covar=tensor([0.2113, 0.1764, 0.1608, 0.1035, 0.1268, 0.1261, 0.2018, 0.1601],
       device='cuda:2'), in_proj_covar=tensor([0.0236, 0.0211, 0.0201, 0.0184, 0.0254, 0.0188, 0.0211, 0.0188],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-23 01:01:05,388 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.5821, 1.3840, 1.2865, 0.7862, 1.8299, 1.5271, 1.4770, 1.1949],
       device='cuda:2'), covar=tensor([0.0352, 0.0428, 0.0480, 0.0717, 0.0272, 0.0328, 0.0450, 0.0885],
       device='cuda:2'), in_proj_covar=tensor([0.0121, 0.0119, 0.0128, 0.0119, 0.0086, 0.0119, 0.0138, 0.0152],
       device='cuda:2'), out_proj_covar=tensor([0.0001, 0.0001, 0.0002, 0.0001, 0.0001, 0.0001, 0.0002, 0.0002],
       device='cuda:2')
2022-12-23 01:01:06,492 WARNING [train.py:1060] (2/4) Exclude cut with ID 6978-92210-0019-146985-0 from training. Duration: 22.48
2022-12-23 01:01:06,779 INFO [zipformer.py:660] (2/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=34622.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 01:01:11,055 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 3.837e+02 5.772e+02 6.600e+02 8.445e+02 2.116e+03, threshold=1.320e+03, percent-clipped=3.0
2022-12-23 01:01:12,874 INFO [zipformer.py:660] (2/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=34626.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 01:01:21,856 WARNING [train.py:1060] (2/4) Exclude cut with ID 7255-291500-0003-134302-0_sp0.9 from training. Duration: 29.816625
2022-12-23 01:01:39,164 INFO [zipformer.py:660] (2/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=34643.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 01:01:42,054 WARNING [train.py:1060] (2/4) Exclude cut with ID 7255-291500-0005-134304-0_sp1.1 from training. Duration: 22.7590625
2022-12-23 01:01:47,796 WARNING [train.py:1060] (2/4) Exclude cut with ID 7357-94126-0026-15857-0 from training. Duration: 22.555
2022-12-23 01:01:56,388 INFO [zipformer.py:660] (2/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=34654.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 01:01:59,069 INFO [train.py:894] (2/4) Epoch 10, batch 3100, loss[loss=0.1827, simple_loss=0.2538, pruned_loss=0.05578, over 18547.00 frames. ], tot_loss[loss=0.2248, simple_loss=0.2972, pruned_loss=0.0762, over 3715259.28 frames. ], batch size: 44, lr: 1.14e-02, grad_scale: 8.0
2022-12-23 01:02:07,528 WARNING [train.py:1060] (2/4) Exclude cut with ID 1250-135782-0005-25975-0_sp0.9 from training. Duration: 21.688875
2022-12-23 01:02:08,033 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.3047, 2.0181, 1.3241, 2.1563, 2.3757, 1.4231, 1.5256, 1.1618],
       device='cuda:2'), covar=tensor([0.2106, 0.1461, 0.1547, 0.0912, 0.1360, 0.1220, 0.1893, 0.1500],
       device='cuda:2'), in_proj_covar=tensor([0.0237, 0.0212, 0.0202, 0.0185, 0.0253, 0.0189, 0.0212, 0.0189],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-23 01:02:41,049 INFO [zipformer.py:660] (2/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=34683.0, num_to_drop=1, layers_to_drop={0}
2022-12-23 01:02:43,636 WARNING [train.py:1060] (2/4) Exclude cut with ID 3488-85273-0038-41224-0_sp0.9 from training. Duration: 22.6
2022-12-23 01:02:47,296 INFO [zipformer.py:660] (2/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=34687.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 01:03:00,108 INFO [scaling.py:679] (2/4) Whitening: num_groups=8, num_channels=192, metric=1.75 vs. limit=2.0
2022-12-23 01:03:08,244 INFO [zipformer.py:660] (2/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=34700.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 01:03:17,605 INFO [train.py:894] (2/4) Epoch 10, batch 3150, loss[loss=0.1751, simple_loss=0.2569, pruned_loss=0.04662, over 18424.00 frames. ], tot_loss[loss=0.2238, simple_loss=0.2964, pruned_loss=0.07566, over 3715198.80 frames. ], batch size: 48, lr: 1.14e-02, grad_scale: 8.0
2022-12-23 01:03:18,014 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.4110, 1.0249, 1.2940, 1.3740, 1.7147, 1.5432, 1.5154, 1.1361],
       device='cuda:2'), covar=tensor([0.0234, 0.0219, 0.0414, 0.0183, 0.0161, 0.0282, 0.0220, 0.0242],
       device='cuda:2'), in_proj_covar=tensor([0.0085, 0.0116, 0.0143, 0.0124, 0.0108, 0.0109, 0.0091, 0.0119],
       device='cuda:2'), out_proj_covar=tensor([7.2214e-05, 9.7728e-05, 1.2608e-04, 1.0477e-04, 9.4283e-05, 8.8827e-05,
        7.6422e-05, 9.9016e-05], device='cuda:2')
2022-12-23 01:03:23,854 INFO [zipformer.py:660] (2/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=34710.0, num_to_drop=1, layers_to_drop={1}
2022-12-23 01:03:24,905 WARNING [train.py:1060] (2/4) Exclude cut with ID 7357-94126-0009-15840-0 from training. Duration: 24.32
2022-12-23 01:03:46,377 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 3.556e+02 5.245e+02 6.711e+02 8.370e+02 2.399e+03, threshold=1.342e+03, percent-clipped=5.0
2022-12-23 01:04:21,948 WARNING [train.py:1060] (2/4) Exclude cut with ID 8291-276745-0093-13116-0_sp0.9 from training. Duration: 21.061125
2022-12-23 01:04:22,031 INFO [zipformer.py:660] (2/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=34748.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 01:04:34,344 INFO [train.py:894] (2/4) Epoch 10, batch 3200, loss[loss=0.231, simple_loss=0.3077, pruned_loss=0.07716, over 18662.00 frames. ], tot_loss[loss=0.2226, simple_loss=0.2952, pruned_loss=0.07496, over 3714329.13 frames. ], batch size: 62, lr: 1.14e-02, grad_scale: 8.0
2022-12-23 01:04:35,819 WARNING [train.py:1060] (2/4) Exclude cut with ID 7357-94126-0024-15855-0_sp0.9 from training. Duration: 20.32225
2022-12-23 01:04:36,201 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.5091, 1.5728, 1.5141, 1.6517, 1.1209, 3.7316, 1.6513, 2.1031],
       device='cuda:2'), covar=tensor([0.3319, 0.2016, 0.1978, 0.1941, 0.1409, 0.0167, 0.1504, 0.0883],
       device='cuda:2'), in_proj_covar=tensor([0.0139, 0.0118, 0.0128, 0.0120, 0.0102, 0.0101, 0.0098, 0.0093],
       device='cuda:2'), out_proj_covar=tensor([0.0005, 0.0004, 0.0005, 0.0004, 0.0004, 0.0003, 0.0004, 0.0004],
       device='cuda:2')
2022-12-23 01:04:50,757 WARNING [train.py:1060] (2/4) Exclude cut with ID 3033-130750-0096-55598-0_sp1.1 from training. Duration: 0.7545625
2022-12-23 01:04:56,747 INFO [zipformer.py:660] (2/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=34771.0, num_to_drop=1, layers_to_drop={1}
2022-12-23 01:05:05,747 WARNING [train.py:1060] (2/4) Exclude cut with ID 4295-39940-0007-92567-0_sp0.9 from training. Duration: 23.9333125
2022-12-23 01:05:40,002 WARNING [train.py:1060] (2/4) Exclude cut with ID 7255-291500-0008-134307-0_sp1.1 from training. Duration: 20.17275
2022-12-23 01:05:43,545 INFO [zipformer.py:660] (2/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=34801.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 01:05:46,247 WARNING [train.py:1060] (2/4) Exclude cut with ID 6978-92210-0019-146985-0_sp1.1 from training. Duration: 20.436375
2022-12-23 01:05:50,487 INFO [train.py:894] (2/4) Epoch 10, batch 3250, loss[loss=0.1858, simple_loss=0.2626, pruned_loss=0.05448, over 18543.00 frames. ], tot_loss[loss=0.2222, simple_loss=0.295, pruned_loss=0.0747, over 3714172.96 frames. ], batch size: 47, lr: 1.14e-02, grad_scale: 8.0
2022-12-23 01:06:19,409 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 3.549e+02 5.831e+02 6.841e+02 8.207e+02 1.292e+03, threshold=1.368e+03, percent-clipped=0.0
2022-12-23 01:06:36,887 INFO [scaling.py:679] (2/4) Whitening: num_groups=8, num_channels=96, metric=1.85 vs. limit=2.0
2022-12-23 01:06:56,429 INFO [zipformer.py:660] (2/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=34849.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 01:07:06,011 INFO [train.py:894] (2/4) Epoch 10, batch 3300, loss[loss=0.2282, simple_loss=0.305, pruned_loss=0.07564, over 18527.00 frames. ], tot_loss[loss=0.2219, simple_loss=0.2944, pruned_loss=0.07473, over 3714354.10 frames. ], batch size: 78, lr: 1.13e-02, grad_scale: 8.0
2022-12-23 01:07:07,692 WARNING [train.py:1060] (2/4) Exclude cut with ID 4234-40345-0022-142709-0_sp0.9 from training. Duration: 23.1055625
2022-12-23 01:07:09,425 WARNING [train.py:1060] (2/4) Exclude cut with ID 8291-282929-0007-12994-0_sp1.1 from training. Duration: 23.5
2022-12-23 01:07:21,385 WARNING [train.py:1060] (2/4) Exclude cut with ID 7255-291500-0009-134308-0_sp0.9 from training. Duration: 26.62775
2022-12-23 01:07:35,173 WARNING [train.py:1060] (2/4) Exclude cut with ID 6951-79737-0018-132285-0 from training. Duration: 21.105
2022-12-23 01:07:39,433 WARNING [train.py:1060] (2/4) Exclude cut with ID 4511-76322-0006-80011-0_sp0.9 from training. Duration: 24.411125
2022-12-23 01:07:47,795 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([3.6033, 3.2369, 3.3013, 3.4036, 2.9519, 2.9652, 3.7912, 1.3011],
       device='cuda:2'), covar=tensor([0.1434, 0.1297, 0.1303, 0.1823, 0.2742, 0.2209, 0.1210, 0.6274],
       device='cuda:2'), in_proj_covar=tensor([0.0305, 0.0202, 0.0212, 0.0220, 0.0290, 0.0242, 0.0248, 0.0261],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0001, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-23 01:08:08,269 WARNING [train.py:1060] (2/4) Exclude cut with ID 6758-72288-0033-108368-0_sp1.1 from training. Duration: 21.263625
2022-12-23 01:08:23,700 INFO [train.py:894] (2/4) Epoch 10, batch 3350, loss[loss=0.2084, simple_loss=0.2713, pruned_loss=0.07269, over 18511.00 frames. ], tot_loss[loss=0.2217, simple_loss=0.2941, pruned_loss=0.07466, over 3714137.16 frames. ], batch size: 44, lr: 1.13e-02, grad_scale: 8.0
2022-12-23 01:08:30,309 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([4.5723, 3.8831, 3.8736, 4.4731, 4.1352, 4.0591, 4.7119, 1.5507],
       device='cuda:2'), covar=tensor([0.0570, 0.0645, 0.0658, 0.0653, 0.1263, 0.1029, 0.0467, 0.4560],
       device='cuda:2'), in_proj_covar=tensor([0.0304, 0.0202, 0.0211, 0.0218, 0.0288, 0.0242, 0.0247, 0.0260],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0001, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-23 01:08:41,579 WARNING [train.py:1060] (2/4) Exclude cut with ID 4234-40345-0022-142709-0 from training. Duration: 20.795
2022-12-23 01:08:51,655 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 3.438e+02 5.551e+02 6.581e+02 7.829e+02 1.593e+03, threshold=1.316e+03, percent-clipped=1.0
2022-12-23 01:08:51,701 WARNING [train.py:1060] (2/4) Exclude cut with ID 7357-94126-0021-15852-0 from training. Duration: 24.76
2022-12-23 01:08:52,896 WARNING [train.py:1060] (2/4) Exclude cut with ID 3867-173237-0077-144769-0_sp0.9 from training. Duration: 22.25
2022-12-23 01:09:05,688 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([2.3698, 1.5594, 0.8245, 1.5245, 2.2695, 1.5736, 2.1131, 2.6004],
       device='cuda:2'), covar=tensor([0.1482, 0.2116, 0.2632, 0.1630, 0.1850, 0.1571, 0.1414, 0.1304],
       device='cuda:2'), in_proj_covar=tensor([0.0091, 0.0100, 0.0119, 0.0095, 0.0113, 0.0090, 0.0097, 0.0094],
       device='cuda:2'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003],
       device='cuda:2')
2022-12-23 01:09:06,107 INFO [scaling.py:679] (2/4) Whitening: num_groups=8, num_channels=96, metric=1.14 vs. limit=2.0
2022-12-23 01:09:17,927 WARNING [train.py:1060] (2/4) Exclude cut with ID 7357-94126-0026-15857-0_sp1.1 from training. Duration: 20.5045625
2022-12-23 01:09:28,659 INFO [zipformer.py:660] (2/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=34949.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 01:09:38,882 INFO [train.py:894] (2/4) Epoch 10, batch 3400, loss[loss=0.252, simple_loss=0.3214, pruned_loss=0.09131, over 18682.00 frames. ], tot_loss[loss=0.2227, simple_loss=0.2948, pruned_loss=0.0753, over 3714859.13 frames. ], batch size: 60, lr: 1.13e-02, grad_scale: 8.0
2022-12-23 01:10:11,156 INFO [zipformer.py:660] (2/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=34978.0, num_to_drop=1, layers_to_drop={2}
2022-12-23 01:10:16,886 INFO [zipformer.py:660] (2/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=34982.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 01:10:52,552 INFO [train.py:894] (2/4) Epoch 10, batch 3450, loss[loss=0.2216, simple_loss=0.3004, pruned_loss=0.07142, over 18527.00 frames. ], tot_loss[loss=0.2231, simple_loss=0.2953, pruned_loss=0.07544, over 3714844.70 frames. ], batch size: 77, lr: 1.13e-02, grad_scale: 8.0
2022-12-23 01:11:20,657 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 3.418e+02 5.044e+02 6.339e+02 8.246e+02 2.038e+03, threshold=1.268e+03, percent-clipped=3.0
2022-12-23 01:11:47,046 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([0.9880, 0.9504, 0.9916, 0.9239, 0.6498, 1.5021, 0.7393, 1.0569],
       device='cuda:2'), covar=tensor([0.2966, 0.1851, 0.1632, 0.1717, 0.1196, 0.0397, 0.1421, 0.0781],
       device='cuda:2'), in_proj_covar=tensor([0.0139, 0.0119, 0.0129, 0.0120, 0.0103, 0.0102, 0.0098, 0.0094],
       device='cuda:2'), out_proj_covar=tensor([0.0005, 0.0005, 0.0005, 0.0004, 0.0004, 0.0003, 0.0004, 0.0004],
       device='cuda:2')
2022-12-23 01:12:06,957 INFO [train.py:894] (2/4) Epoch 10, batch 3500, loss[loss=0.2477, simple_loss=0.3127, pruned_loss=0.09137, over 18611.00 frames. ], tot_loss[loss=0.2227, simple_loss=0.2949, pruned_loss=0.07521, over 3714861.77 frames. ], batch size: 176, lr: 1.13e-02, grad_scale: 8.0
2022-12-23 01:12:30,139 WARNING [train.py:1060] (2/4) Exclude cut with ID 7859-102521-0017-7548-0_sp1.1 from training. Duration: 22.2954375
2022-12-23 01:12:41,276 INFO [train.py:894] (2/4) Epoch 11, batch 0, loss[loss=0.1895, simple_loss=0.2839, pruned_loss=0.04757, over 18603.00 frames. ], tot_loss[loss=0.1895, simple_loss=0.2839, pruned_loss=0.04757, over 18603.00 frames. ], batch size: 53, lr: 1.08e-02, grad_scale: 8.0
2022-12-23 01:12:41,276 INFO [train.py:919] (2/4) Computing validation loss
2022-12-23 01:12:51,411 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([2.5797, 3.6660, 3.5329, 1.4650, 3.6051, 2.6148, 0.6842, 2.1826],
       device='cuda:2'), covar=tensor([0.2110, 0.0667, 0.1211, 0.4846, 0.1097, 0.1233, 0.6036, 0.2289],
       device='cuda:2'), in_proj_covar=tensor([0.0135, 0.0125, 0.0152, 0.0121, 0.0129, 0.0105, 0.0142, 0.0111],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0002, 0.0003, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-23 01:12:52,251 INFO [train.py:928] (2/4) Epoch 11, validation: loss=0.1739, simple_loss=0.2739, pruned_loss=0.03697, over 944034.00 frames. 
2022-12-23 01:12:52,252 INFO [train.py:929] (2/4) Maximum memory allocated so far is 24680MB
2022-12-23 01:12:56,511 INFO [zipformer.py:660] (2/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=35066.0, num_to_drop=1, layers_to_drop={1}
2022-12-23 01:13:24,563 INFO [scaling.py:679] (2/4) Whitening: num_groups=8, num_channels=192, metric=1.93 vs. limit=2.0
2022-12-23 01:13:25,402 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.7911, 1.7866, 1.7242, 1.9463, 1.5735, 5.0931, 2.3179, 2.5638],
       device='cuda:2'), covar=tensor([0.3346, 0.2026, 0.2009, 0.1923, 0.1359, 0.0083, 0.1344, 0.0845],
       device='cuda:2'), in_proj_covar=tensor([0.0140, 0.0119, 0.0129, 0.0120, 0.0103, 0.0102, 0.0099, 0.0094],
       device='cuda:2'), out_proj_covar=tensor([0.0005, 0.0005, 0.0005, 0.0004, 0.0004, 0.0003, 0.0004, 0.0004],
       device='cuda:2')
2022-12-23 01:13:35,546 INFO [zipformer.py:660] (2/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=35092.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 01:13:36,929 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([2.3022, 1.9702, 1.4526, 0.5586, 1.3772, 1.8508, 1.5572, 1.8002],
       device='cuda:2'), covar=tensor([0.0536, 0.0380, 0.0997, 0.1463, 0.1086, 0.1247, 0.1357, 0.0555],
       device='cuda:2'), in_proj_covar=tensor([0.0164, 0.0181, 0.0200, 0.0192, 0.0205, 0.0188, 0.0201, 0.0190],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-23 01:13:44,322 WARNING [train.py:1060] (2/4) Exclude cut with ID 298-126791-0067-24026-0_sp0.9 from training. Duration: 21.438875
2022-12-23 01:13:48,860 WARNING [train.py:1060] (2/4) Exclude cut with ID 5652-39938-0025-23684-0_sp0.9 from training. Duration: 22.2055625
2022-12-23 01:14:06,705 INFO [train.py:894] (2/4) Epoch 11, batch 50, loss[loss=0.1766, simple_loss=0.2595, pruned_loss=0.04684, over 18432.00 frames. ], tot_loss[loss=0.2075, simple_loss=0.2907, pruned_loss=0.0621, over 838133.02 frames. ], batch size: 42, lr: 1.08e-02, grad_scale: 8.0
2022-12-23 01:14:17,357 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([2.4309, 2.3652, 1.5753, 3.1722, 2.8009, 2.3637, 4.2274, 2.3814],
       device='cuda:2'), covar=tensor([0.0758, 0.1807, 0.2628, 0.1727, 0.1465, 0.0803, 0.0588, 0.1046],
       device='cuda:2'), in_proj_covar=tensor([0.0173, 0.0195, 0.0236, 0.0279, 0.0222, 0.0181, 0.0205, 0.0193],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-23 01:14:25,778 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 2.877e+02 4.401e+02 5.411e+02 7.454e+02 1.490e+03, threshold=1.082e+03, percent-clipped=5.0
2022-12-23 01:14:44,503 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([2.5284, 1.9840, 0.8151, 1.7291, 2.8120, 1.7951, 2.6780, 2.8058],
       device='cuda:2'), covar=tensor([0.1484, 0.1821, 0.2874, 0.1525, 0.1502, 0.1513, 0.1229, 0.1342],
       device='cuda:2'), in_proj_covar=tensor([0.0091, 0.0101, 0.0120, 0.0096, 0.0114, 0.0091, 0.0098, 0.0096],
       device='cuda:2'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003],
       device='cuda:2')
2022-12-23 01:14:44,867 INFO [scaling.py:679] (2/4) Whitening: num_groups=8, num_channels=96, metric=1.22 vs. limit=2.0
2022-12-23 01:14:59,112 INFO [scaling.py:679] (2/4) Whitening: num_groups=1, num_channels=384, metric=4.74 vs. limit=5.0
2022-12-23 01:15:07,642 INFO [zipformer.py:660] (2/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=35153.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 01:15:21,342 INFO [train.py:894] (2/4) Epoch 11, batch 100, loss[loss=0.1992, simple_loss=0.2847, pruned_loss=0.05682, over 18567.00 frames. ], tot_loss[loss=0.201, simple_loss=0.2845, pruned_loss=0.05879, over 1474303.62 frames. ], batch size: 49, lr: 1.08e-02, grad_scale: 8.0
2022-12-23 01:16:30,981 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.5441, 1.4143, 1.3233, 1.3287, 1.5923, 1.5988, 1.5608, 1.2323],
       device='cuda:2'), covar=tensor([0.0251, 0.0220, 0.0324, 0.0227, 0.0191, 0.0290, 0.0208, 0.0267],
       device='cuda:2'), in_proj_covar=tensor([0.0086, 0.0117, 0.0146, 0.0126, 0.0111, 0.0110, 0.0094, 0.0119],
       device='cuda:2'), out_proj_covar=tensor([7.3162e-05, 9.8195e-05, 1.2806e-04, 1.0645e-04, 9.6798e-05, 8.9827e-05,
        7.8955e-05, 9.8938e-05], device='cuda:2')
2022-12-23 01:16:33,966 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.1592, 1.1038, 0.6781, 1.3443, 1.4480, 2.5211, 1.1077, 1.2364],
       device='cuda:2'), covar=tensor([0.0918, 0.1799, 0.1243, 0.0963, 0.1482, 0.0279, 0.1413, 0.1652],
       device='cuda:2'), in_proj_covar=tensor([0.0074, 0.0082, 0.0075, 0.0074, 0.0092, 0.0072, 0.0085, 0.0077],
       device='cuda:2'), out_proj_covar=tensor([0.0003, 0.0004, 0.0003, 0.0003, 0.0004, 0.0003, 0.0004, 0.0004],
       device='cuda:2')
2022-12-23 01:16:38,007 INFO [train.py:894] (2/4) Epoch 11, batch 150, loss[loss=0.2297, simple_loss=0.3175, pruned_loss=0.0709, over 18489.00 frames. ], tot_loss[loss=0.1997, simple_loss=0.2826, pruned_loss=0.05838, over 1970374.34 frames. ], batch size: 64, lr: 1.08e-02, grad_scale: 8.0
2022-12-23 01:16:53,147 WARNING [train.py:1060] (2/4) Exclude cut with ID 7859-102521-0017-7548-0 from training. Duration: 24.525
2022-12-23 01:16:57,294 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 2.549e+02 4.088e+02 5.112e+02 6.380e+02 1.930e+03, threshold=1.022e+03, percent-clipped=3.0
2022-12-23 01:17:25,996 WARNING [train.py:1060] (2/4) Exclude cut with ID 3699-47246-0007-3408-0_sp0.9 from training. Duration: 20.26675
2022-12-23 01:17:33,874 INFO [zipformer.py:660] (2/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=35249.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 01:17:39,328 WARNING [train.py:1060] (2/4) Exclude cut with ID 7859-102521-0017-7548-0_sp0.9 from training. Duration: 27.25
2022-12-23 01:17:54,749 INFO [train.py:894] (2/4) Epoch 11, batch 200, loss[loss=0.2463, simple_loss=0.3225, pruned_loss=0.08505, over 18682.00 frames. ], tot_loss[loss=0.1995, simple_loss=0.283, pruned_loss=0.05802, over 2357627.83 frames. ], batch size: 60, lr: 1.08e-02, grad_scale: 8.0
2022-12-23 01:17:58,046 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([2.3382, 2.8355, 2.7668, 1.3691, 2.6971, 2.4297, 1.8454, 3.5861],
       device='cuda:2'), covar=tensor([0.1365, 0.1432, 0.1492, 0.2341, 0.0931, 0.1477, 0.2227, 0.0578],
       device='cuda:2'), in_proj_covar=tensor([0.0199, 0.0198, 0.0203, 0.0192, 0.0180, 0.0211, 0.0210, 0.0179],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-23 01:18:09,712 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([0.1564, 1.4375, 1.7374, 0.5835, 0.8023, 1.8523, 1.7587, 1.5855],
       device='cuda:2'), covar=tensor([0.0592, 0.0281, 0.0290, 0.0333, 0.0389, 0.0364, 0.0194, 0.0450],
       device='cuda:2'), in_proj_covar=tensor([0.0137, 0.0157, 0.0109, 0.0132, 0.0141, 0.0126, 0.0144, 0.0146],
       device='cuda:2'), out_proj_covar=tensor([1.1285e-04, 1.3203e-04, 8.9251e-05, 1.0686e-04, 1.1671e-04, 1.0561e-04,
        1.2264e-04, 1.2158e-04], device='cuda:2')
2022-12-23 01:18:16,513 INFO [zipformer.py:660] (2/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=35278.0, num_to_drop=1, layers_to_drop={1}
2022-12-23 01:18:22,935 INFO [zipformer.py:660] (2/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=35282.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 01:18:30,703 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.5812, 1.2907, 1.8929, 3.1961, 2.4274, 2.4183, 0.8177, 2.0212],
       device='cuda:2'), covar=tensor([0.1789, 0.1725, 0.1495, 0.0503, 0.0986, 0.1209, 0.2420, 0.1237],
       device='cuda:2'), in_proj_covar=tensor([0.0100, 0.0114, 0.0127, 0.0124, 0.0101, 0.0130, 0.0128, 0.0109],
       device='cuda:2'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003],
       device='cuda:2')
2022-12-23 01:18:45,686 INFO [zipformer.py:660] (2/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=35297.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 01:18:53,641 WARNING [train.py:1060] (2/4) Exclude cut with ID 6426-64292-0017-15984-0 from training. Duration: 21.68
2022-12-23 01:19:04,836 WARNING [train.py:1060] (2/4) Exclude cut with ID 4278-13270-0007-59342-0 from training. Duration: 21.6300625
2022-12-23 01:19:10,610 INFO [train.py:894] (2/4) Epoch 11, batch 250, loss[loss=0.1836, simple_loss=0.2581, pruned_loss=0.05459, over 18569.00 frames. ], tot_loss[loss=0.1972, simple_loss=0.2804, pruned_loss=0.05701, over 2658038.57 frames. ], batch size: 45, lr: 1.08e-02, grad_scale: 8.0
2022-12-23 01:19:29,239 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 2.669e+02 4.290e+02 5.310e+02 7.018e+02 1.757e+03, threshold=1.062e+03, percent-clipped=3.0
2022-12-23 01:19:29,305 WARNING [train.py:1060] (2/4) Exclude cut with ID 4278-13270-0007-59342-0_sp0.9 from training. Duration: 24.033375
2022-12-23 01:19:30,838 INFO [zipformer.py:660] (2/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=35326.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 01:19:37,104 INFO [zipformer.py:660] (2/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=35330.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 01:19:59,275 INFO [zipformer.py:660] (2/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=35345.0, num_to_drop=1, layers_to_drop={1}
2022-12-23 01:20:05,749 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([2.8845, 2.4937, 2.1032, 0.8441, 1.9995, 2.2213, 1.9675, 2.1515],
       device='cuda:2'), covar=tensor([0.0550, 0.0481, 0.1077, 0.1735, 0.1272, 0.1290, 0.1337, 0.0769],
       device='cuda:2'), in_proj_covar=tensor([0.0160, 0.0177, 0.0195, 0.0190, 0.0202, 0.0186, 0.0199, 0.0187],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-23 01:20:26,271 INFO [train.py:894] (2/4) Epoch 11, batch 300, loss[loss=0.1961, simple_loss=0.2909, pruned_loss=0.0507, over 18578.00 frames. ], tot_loss[loss=0.196, simple_loss=0.2796, pruned_loss=0.05621, over 2892007.90 frames. ], batch size: 56, lr: 1.08e-02, grad_scale: 8.0
2022-12-23 01:20:31,309 WARNING [train.py:1060] (2/4) Exclude cut with ID 4278-13270-0009-59344-0 from training. Duration: 22.905
2022-12-23 01:20:31,602 INFO [zipformer.py:660] (2/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=35366.0, num_to_drop=1, layers_to_drop={2}
2022-12-23 01:20:32,825 WARNING [train.py:1060] (2/4) Exclude cut with ID 5622-44585-0006-90525-0_sp1.1 from training. Duration: 23.4318125
2022-12-23 01:20:43,064 INFO [zipformer.py:660] (2/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=35374.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 01:21:30,402 INFO [zipformer.py:660] (2/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=35406.0, num_to_drop=1, layers_to_drop={1}
2022-12-23 01:21:41,813 INFO [train.py:894] (2/4) Epoch 11, batch 350, loss[loss=0.2212, simple_loss=0.3039, pruned_loss=0.06926, over 18532.00 frames. ], tot_loss[loss=0.1981, simple_loss=0.2811, pruned_loss=0.05755, over 3074562.24 frames. ], batch size: 55, lr: 1.07e-02, grad_scale: 16.0
2022-12-23 01:21:42,228 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.6453, 1.5286, 1.7363, 1.7776, 1.3413, 3.7844, 1.7194, 2.3952],
       device='cuda:2'), covar=tensor([0.3272, 0.2042, 0.1936, 0.1863, 0.1364, 0.0152, 0.1551, 0.0797],
       device='cuda:2'), in_proj_covar=tensor([0.0139, 0.0119, 0.0129, 0.0119, 0.0104, 0.0101, 0.0098, 0.0094],
       device='cuda:2'), out_proj_covar=tensor([0.0005, 0.0005, 0.0005, 0.0004, 0.0004, 0.0003, 0.0004, 0.0004],
       device='cuda:2')
2022-12-23 01:21:43,380 INFO [zipformer.py:660] (2/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=35414.0, num_to_drop=1, layers_to_drop={1}
2022-12-23 01:22:00,005 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 2.294e+02 3.714e+02 4.522e+02 5.692e+02 1.135e+03, threshold=9.043e+02, percent-clipped=2.0
2022-12-23 01:22:15,537 INFO [zipformer.py:660] (2/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=35435.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 01:22:26,086 WARNING [train.py:1060] (2/4) Exclude cut with ID 4278-13270-0009-59344-0_sp1.1 from training. Duration: 20.82275
2022-12-23 01:22:26,124 WARNING [train.py:1060] (2/4) Exclude cut with ID 4278-13270-0009-59344-0_sp0.9 from training. Duration: 25.45
2022-12-23 01:22:34,075 INFO [zipformer.py:660] (2/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=35448.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 01:22:56,852 INFO [train.py:894] (2/4) Epoch 11, batch 400, loss[loss=0.1723, simple_loss=0.254, pruned_loss=0.04528, over 18384.00 frames. ], tot_loss[loss=0.2012, simple_loss=0.2842, pruned_loss=0.05914, over 3216184.86 frames. ], batch size: 46, lr: 1.07e-02, grad_scale: 8.0
2022-12-23 01:23:22,900 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.9605, 1.9212, 2.1704, 1.2059, 2.3179, 2.0688, 1.5763, 2.5974],
       device='cuda:2'), covar=tensor([0.1123, 0.1670, 0.1185, 0.1980, 0.0711, 0.1161, 0.2143, 0.0498],
       device='cuda:2'), in_proj_covar=tensor([0.0199, 0.0200, 0.0205, 0.0193, 0.0182, 0.0212, 0.0210, 0.0181],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-23 01:23:25,329 WARNING [train.py:1060] (2/4) Exclude cut with ID 5622-44585-0006-90525-0 from training. Duration: 25.775
2022-12-23 01:23:48,179 WARNING [train.py:1060] (2/4) Exclude cut with ID 2929-85685-0071-62375-0_sp0.9 from training. Duration: 22.25
2022-12-23 01:24:10,524 INFO [train.py:894] (2/4) Epoch 11, batch 450, loss[loss=0.2229, simple_loss=0.3005, pruned_loss=0.07271, over 18533.00 frames. ], tot_loss[loss=0.2023, simple_loss=0.2849, pruned_loss=0.05989, over 3326261.27 frames. ], batch size: 55, lr: 1.07e-02, grad_scale: 8.0
2022-12-23 01:24:15,181 WARNING [train.py:1060] (2/4) Exclude cut with ID 3972-170212-0014-23379-0 from training. Duration: 26.205
2022-12-23 01:24:29,643 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 2.813e+02 4.201e+02 5.115e+02 6.796e+02 1.250e+03, threshold=1.023e+03, percent-clipped=10.0
2022-12-23 01:24:31,130 WARNING [train.py:1060] (2/4) Exclude cut with ID 5239-32139-0047-9341-0_sp0.9 from training. Duration: 30.1555625
2022-12-23 01:24:35,980 INFO [zipformer.py:660] (2/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=35530.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 01:24:37,067 WARNING [train.py:1060] (2/4) Exclude cut with ID 1265-135635-0050-6781-0_sp0.9 from training. Duration: 21.8333125
2022-12-23 01:24:47,363 WARNING [train.py:1060] (2/4) Exclude cut with ID 1914-133440-0024-94914-0_sp1.1 from training. Duration: 20.6545625
2022-12-23 01:25:25,205 INFO [train.py:894] (2/4) Epoch 11, batch 500, loss[loss=0.1784, simple_loss=0.2508, pruned_loss=0.05295, over 18412.00 frames. ], tot_loss[loss=0.2038, simple_loss=0.2864, pruned_loss=0.06066, over 3412121.79 frames. ], batch size: 42, lr: 1.07e-02, grad_scale: 8.0
2022-12-23 01:25:28,161 WARNING [train.py:1060] (2/4) Exclude cut with ID 7395-89880-0045-39920-0_sp0.9 from training. Duration: 20.52225
2022-12-23 01:25:49,190 WARNING [train.py:1060] (2/4) Exclude cut with ID 3972-170212-0014-23379-0_sp0.9 from training. Duration: 29.1166875
2022-12-23 01:26:07,856 INFO [zipformer.py:660] (2/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=35591.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 01:26:40,521 INFO [train.py:894] (2/4) Epoch 11, batch 550, loss[loss=0.1729, simple_loss=0.2581, pruned_loss=0.04381, over 18683.00 frames. ], tot_loss[loss=0.2033, simple_loss=0.2861, pruned_loss=0.06025, over 3479743.13 frames. ], batch size: 46, lr: 1.07e-02, grad_scale: 8.0
2022-12-23 01:26:47,620 WARNING [train.py:1060] (2/4) Exclude cut with ID 543-133211-0007-59831-0_sp0.9 from training. Duration: 21.388875
2022-12-23 01:27:00,169 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 2.945e+02 4.347e+02 5.387e+02 6.927e+02 1.155e+03, threshold=1.077e+03, percent-clipped=5.0
2022-12-23 01:27:26,096 WARNING [train.py:1060] (2/4) Exclude cut with ID 1914-133440-0024-94914-0 from training. Duration: 22.72
2022-12-23 01:27:27,547 WARNING [train.py:1060] (2/4) Exclude cut with ID 1914-133440-0031-94921-0_sp0.9 from training. Duration: 22.7444375
2022-12-23 01:27:56,855 INFO [train.py:894] (2/4) Epoch 11, batch 600, loss[loss=0.2221, simple_loss=0.3029, pruned_loss=0.07064, over 18640.00 frames. ], tot_loss[loss=0.2046, simple_loss=0.2877, pruned_loss=0.06076, over 3533534.18 frames. ], batch size: 69, lr: 1.07e-02, grad_scale: 8.0
2022-12-23 01:28:08,704 INFO [scaling.py:679] (2/4) Whitening: num_groups=8, num_channels=192, metric=1.98 vs. limit=2.0
2022-12-23 01:28:12,154 WARNING [train.py:1060] (2/4) Exclude cut with ID 4133-6541-0027-40495-0_sp1.1 from training. Duration: 0.9681875
2022-12-23 01:28:15,333 WARNING [train.py:1060] (2/4) Exclude cut with ID 6330-62851-0022-91297-0_sp0.9 from training. Duration: 22.3166875
2022-12-23 01:28:16,117 INFO [scaling.py:679] (2/4) Whitening: num_groups=8, num_channels=192, metric=1.76 vs. limit=2.0
2022-12-23 01:28:21,029 WARNING [train.py:1060] (2/4) Exclude cut with ID 543-133212-0015-59917-0_sp0.9 from training. Duration: 21.8166875
2022-12-23 01:28:55,301 INFO [zipformer.py:660] (2/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=35701.0, num_to_drop=1, layers_to_drop={3}
2022-12-23 01:28:57,815 INFO [scaling.py:679] (2/4) Whitening: num_groups=8, num_channels=96, metric=1.39 vs. limit=2.0
2022-12-23 01:29:13,358 INFO [train.py:894] (2/4) Epoch 11, batch 650, loss[loss=0.1945, simple_loss=0.2721, pruned_loss=0.05851, over 18541.00 frames. ], tot_loss[loss=0.2033, simple_loss=0.286, pruned_loss=0.06033, over 3571994.78 frames. ], batch size: 47, lr: 1.07e-02, grad_scale: 8.0
2022-12-23 01:29:32,665 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 2.598e+02 4.065e+02 4.860e+02 5.934e+02 1.772e+03, threshold=9.721e+02, percent-clipped=1.0
2022-12-23 01:29:39,247 INFO [zipformer.py:660] (2/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=35730.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 01:30:03,832 WARNING [train.py:1060] (2/4) Exclude cut with ID 4957-30119-0041-23990-0_sp0.9 from training. Duration: 20.22775
2022-12-23 01:30:06,420 INFO [zipformer.py:660] (2/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=35748.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 01:30:28,938 INFO [train.py:894] (2/4) Epoch 11, batch 700, loss[loss=0.2076, simple_loss=0.2972, pruned_loss=0.05903, over 18581.00 frames. ], tot_loss[loss=0.2035, simple_loss=0.2863, pruned_loss=0.06037, over 3604089.50 frames. ], batch size: 56, lr: 1.07e-02, grad_scale: 8.0
2022-12-23 01:30:41,257 INFO [zipformer.py:660] (2/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=35771.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 01:30:49,617 WARNING [train.py:1060] (2/4) Exclude cut with ID 5239-32139-0047-9341-0_sp1.1 from training. Duration: 24.67275
2022-12-23 01:31:09,751 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([0.6356, 2.0503, 2.2257, 0.9148, 1.4489, 2.5666, 2.0459, 1.8969],
       device='cuda:2'), covar=tensor([0.0641, 0.0241, 0.0237, 0.0355, 0.0336, 0.0260, 0.0206, 0.0546],
       device='cuda:2'), in_proj_covar=tensor([0.0137, 0.0156, 0.0108, 0.0130, 0.0139, 0.0125, 0.0143, 0.0145],
       device='cuda:2'), out_proj_covar=tensor([1.1284e-04, 1.2995e-04, 8.8189e-05, 1.0551e-04, 1.1481e-04, 1.0501e-04,
        1.2082e-04, 1.2031e-04], device='cuda:2')
2022-12-23 01:31:17,925 INFO [zipformer.py:660] (2/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=35796.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 01:31:19,317 WARNING [train.py:1060] (2/4) Exclude cut with ID 3082-165428-0081-50734-0_sp0.9 from training. Duration: 21.8055625
2022-12-23 01:31:42,729 INFO [zipformer.py:660] (2/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=35811.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 01:31:45,336 INFO [train.py:894] (2/4) Epoch 11, batch 750, loss[loss=0.222, simple_loss=0.3042, pruned_loss=0.0699, over 18694.00 frames. ], tot_loss[loss=0.204, simple_loss=0.2869, pruned_loss=0.0605, over 3629388.72 frames. ], batch size: 62, lr: 1.07e-02, grad_scale: 8.0
2022-12-23 01:31:54,715 WARNING [train.py:1060] (2/4) Exclude cut with ID 3340-169293-0054-76830-0_sp0.9 from training. Duration: 22.6666875
2022-12-23 01:32:05,504 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 2.908e+02 4.276e+02 5.128e+02 6.060e+02 1.074e+03, threshold=1.026e+03, percent-clipped=3.0
2022-12-23 01:32:14,955 INFO [zipformer.py:660] (2/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=35832.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 01:32:18,028 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.7154, 2.2159, 1.4166, 2.2558, 3.3294, 1.5099, 1.9404, 1.2545],
       device='cuda:2'), covar=tensor([0.1902, 0.1661, 0.1569, 0.0982, 0.1139, 0.1204, 0.1842, 0.1561],
       device='cuda:2'), in_proj_covar=tensor([0.0233, 0.0210, 0.0200, 0.0185, 0.0247, 0.0187, 0.0210, 0.0187],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-23 01:32:57,731 WARNING [train.py:1060] (2/4) Exclude cut with ID 2411-132532-0017-82279-0_sp1.1 from training. Duration: 0.9681875
2022-12-23 01:33:02,620 INFO [train.py:894] (2/4) Epoch 11, batch 800, loss[loss=0.2168, simple_loss=0.3001, pruned_loss=0.06676, over 18727.00 frames. ], tot_loss[loss=0.2038, simple_loss=0.2868, pruned_loss=0.06045, over 3648092.91 frames. ], batch size: 52, lr: 1.07e-02, grad_scale: 8.0
2022-12-23 01:33:16,402 INFO [zipformer.py:660] (2/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=35872.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 01:33:23,213 WARNING [train.py:1060] (2/4) Exclude cut with ID 6330-62850-0007-91323-0 from training. Duration: 22.485
2022-12-23 01:33:37,226 INFO [zipformer.py:660] (2/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=35886.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 01:34:01,409 WARNING [train.py:1060] (2/4) Exclude cut with ID 3972-170212-0014-23379-0_sp1.1 from training. Duration: 23.82275
2022-12-23 01:34:14,535 WARNING [train.py:1060] (2/4) Exclude cut with ID 4860-13185-0032-76709-0 from training. Duration: 20.77
2022-12-23 01:34:17,528 INFO [train.py:894] (2/4) Epoch 11, batch 850, loss[loss=0.2187, simple_loss=0.3044, pruned_loss=0.06653, over 18547.00 frames. ], tot_loss[loss=0.2032, simple_loss=0.2857, pruned_loss=0.06029, over 3661449.50 frames. ], batch size: 55, lr: 1.07e-02, grad_scale: 8.0
2022-12-23 01:34:23,685 WARNING [train.py:1060] (2/4) Exclude cut with ID 6426-64292-0017-15984-0_sp0.9 from training. Duration: 24.088875
2022-12-23 01:34:32,439 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.7344, 1.7329, 1.6429, 1.9763, 1.5189, 5.0139, 2.2982, 2.6908],
       device='cuda:2'), covar=tensor([0.3544, 0.2311, 0.2133, 0.2081, 0.1446, 0.0120, 0.1271, 0.0901],
       device='cuda:2'), in_proj_covar=tensor([0.0141, 0.0120, 0.0130, 0.0122, 0.0106, 0.0101, 0.0099, 0.0094],
       device='cuda:2'), out_proj_covar=tensor([0.0005, 0.0005, 0.0005, 0.0004, 0.0004, 0.0003, 0.0004, 0.0004],
       device='cuda:2')
2022-12-23 01:34:36,373 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 3.023e+02 4.512e+02 5.363e+02 6.717e+02 1.218e+03, threshold=1.073e+03, percent-clipped=5.0
2022-12-23 01:34:53,411 INFO [scaling.py:679] (2/4) Whitening: num_groups=1, num_channels=384, metric=4.54 vs. limit=5.0
2022-12-23 01:34:53,763 WARNING [train.py:1060] (2/4) Exclude cut with ID 6330-62850-0007-91323-0_sp1.1 from training. Duration: 20.4409375
2022-12-23 01:35:34,147 INFO [train.py:894] (2/4) Epoch 11, batch 900, loss[loss=0.18, simple_loss=0.2565, pruned_loss=0.05177, over 18524.00 frames. ], tot_loss[loss=0.2013, simple_loss=0.2841, pruned_loss=0.05925, over 3673539.79 frames. ], batch size: 44, lr: 1.07e-02, grad_scale: 8.0
2022-12-23 01:36:08,618 WARNING [train.py:1060] (2/4) Exclude cut with ID 2929-85685-0062-62366-0_sp0.9 from training. Duration: 22.511125
2022-12-23 01:36:10,170 WARNING [train.py:1060] (2/4) Exclude cut with ID 7395-89880-0031-39906-0 from training. Duration: 20.675
2022-12-23 01:36:36,714 INFO [zipformer.py:660] (2/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=36001.0, num_to_drop=1, layers_to_drop={2}
2022-12-23 01:36:53,986 INFO [train.py:894] (2/4) Epoch 11, batch 950, loss[loss=0.2171, simple_loss=0.3102, pruned_loss=0.06201, over 18555.00 frames. ], tot_loss[loss=0.2017, simple_loss=0.2849, pruned_loss=0.05922, over 3683266.48 frames. ], batch size: 77, lr: 1.07e-02, grad_scale: 8.0
2022-12-23 01:37:11,939 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 2.706e+02 4.318e+02 5.198e+02 6.420e+02 1.014e+03, threshold=1.040e+03, percent-clipped=0.0
2022-12-23 01:37:14,401 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([0.1824, 1.4762, 1.6140, 0.8552, 0.9176, 1.7446, 1.6938, 1.5043],
       device='cuda:2'), covar=tensor([0.0629, 0.0242, 0.0278, 0.0288, 0.0396, 0.0378, 0.0193, 0.0458],
       device='cuda:2'), in_proj_covar=tensor([0.0138, 0.0159, 0.0110, 0.0131, 0.0141, 0.0129, 0.0146, 0.0147],
       device='cuda:2'), out_proj_covar=tensor([1.1374e-04, 1.3258e-04, 9.0064e-05, 1.0565e-04, 1.1673e-04, 1.0803e-04,
        1.2337e-04, 1.2203e-04], device='cuda:2')
2022-12-23 01:37:18,491 INFO [zipformer.py:660] (2/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=36030.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 01:37:47,114 INFO [zipformer.py:660] (2/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=36049.0, num_to_drop=1, layers_to_drop={0}
2022-12-23 01:37:49,889 WARNING [train.py:1060] (2/4) Exclude cut with ID 6330-62850-0007-91323-0_sp0.9 from training. Duration: 24.9833125
2022-12-23 01:38:08,626 INFO [train.py:894] (2/4) Epoch 11, batch 1000, loss[loss=0.2255, simple_loss=0.2997, pruned_loss=0.07563, over 18524.00 frames. ], tot_loss[loss=0.2015, simple_loss=0.2849, pruned_loss=0.05908, over 3689471.93 frames. ], batch size: 58, lr: 1.06e-02, grad_scale: 8.0
2022-12-23 01:38:21,525 WARNING [train.py:1060] (2/4) Exclude cut with ID 5239-32139-0047-9341-0 from training. Duration: 27.14
2022-12-23 01:38:31,041 INFO [zipformer.py:660] (2/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=36078.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 01:38:38,183 WARNING [train.py:1060] (2/4) Exclude cut with ID 2929-85685-0044-62348-0 from training. Duration: 22.44
2022-12-23 01:39:18,681 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.5705, 1.0860, 1.8056, 2.9330, 2.2737, 2.3460, 0.8314, 2.0643],
       device='cuda:2'), covar=tensor([0.1860, 0.1903, 0.1525, 0.0601, 0.1036, 0.1264, 0.2454, 0.1140],
       device='cuda:2'), in_proj_covar=tensor([0.0103, 0.0114, 0.0129, 0.0127, 0.0102, 0.0133, 0.0132, 0.0109],
       device='cuda:2'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0004, 0.0003, 0.0003],
       device='cuda:2')
2022-12-23 01:39:24,161 INFO [train.py:894] (2/4) Epoch 11, batch 1050, loss[loss=0.2087, simple_loss=0.3031, pruned_loss=0.05717, over 18568.00 frames. ], tot_loss[loss=0.2006, simple_loss=0.2841, pruned_loss=0.05858, over 3695116.60 frames. ], batch size: 56, lr: 1.06e-02, grad_scale: 8.0
2022-12-23 01:39:42,564 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 2.372e+02 4.570e+02 5.343e+02 6.692e+02 1.425e+03, threshold=1.069e+03, percent-clipped=5.0
2022-12-23 01:39:44,840 INFO [zipformer.py:660] (2/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=36127.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 01:39:58,736 WARNING [train.py:1060] (2/4) Exclude cut with ID 2929-85685-0060-62364-0_sp0.9 from training. Duration: 21.361125
2022-12-23 01:40:05,646 WARNING [train.py:1060] (2/4) Exclude cut with ID 2929-85685-0079-62383-0_sp1.1 from training. Duration: 27.0318125
2022-12-23 01:40:14,444 WARNING [train.py:1060] (2/4) Exclude cut with ID 5622-44585-0006-90525-0_sp0.9 from training. Duration: 28.638875
2022-12-23 01:40:17,639 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([4.2553, 3.6785, 3.6641, 4.1863, 3.7982, 3.8189, 4.4435, 1.4314],
       device='cuda:2'), covar=tensor([0.0824, 0.0674, 0.0676, 0.0786, 0.1532, 0.1046, 0.0589, 0.4993],
       device='cuda:2'), in_proj_covar=tensor([0.0296, 0.0192, 0.0201, 0.0208, 0.0276, 0.0229, 0.0240, 0.0254],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0001, 0.0001, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-23 01:40:32,443 WARNING [train.py:1060] (2/4) Exclude cut with ID 3340-169293-0054-76830-0 from training. Duration: 20.4
2022-12-23 01:40:39,902 INFO [train.py:894] (2/4) Epoch 11, batch 1100, loss[loss=0.1603, simple_loss=0.2422, pruned_loss=0.03918, over 18465.00 frames. ], tot_loss[loss=0.2011, simple_loss=0.2844, pruned_loss=0.05887, over 3699416.48 frames. ], batch size: 43, lr: 1.06e-02, grad_scale: 8.0
2022-12-23 01:40:40,281 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.4329, 2.8755, 3.2740, 0.8760, 2.6318, 3.4696, 2.4205, 2.9386],
       device='cuda:2'), covar=tensor([0.0812, 0.0312, 0.0214, 0.0473, 0.0344, 0.0275, 0.0319, 0.0545],
       device='cuda:2'), in_proj_covar=tensor([0.0136, 0.0157, 0.0109, 0.0129, 0.0139, 0.0127, 0.0143, 0.0146],
       device='cuda:2'), out_proj_covar=tensor([1.1210e-04, 1.3093e-04, 8.9098e-05, 1.0428e-04, 1.1411e-04, 1.0590e-04,
        1.2054e-04, 1.2063e-04], device='cuda:2')
2022-12-23 01:40:45,782 INFO [zipformer.py:660] (2/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=36167.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 01:41:04,442 WARNING [train.py:1060] (2/4) Exclude cut with ID 2929-85685-0071-62375-0 from training. Duration: 20.025
2022-12-23 01:41:04,450 WARNING [train.py:1060] (2/4) Exclude cut with ID 2364-131735-0112-64612-0_sp0.9 from training. Duration: 20.488875
2022-12-23 01:41:07,677 INFO [zipformer.py:660] (2/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=36181.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 01:41:11,051 WARNING [train.py:1060] (2/4) Exclude cut with ID 2929-85685-0079-62383-0 from training. Duration: 29.735
2022-12-23 01:41:15,788 INFO [zipformer.py:660] (2/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=36186.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 01:41:33,775 INFO [scaling.py:679] (2/4) Whitening: num_groups=8, num_channels=96, metric=1.48 vs. limit=2.0
2022-12-23 01:41:55,937 INFO [train.py:894] (2/4) Epoch 11, batch 1150, loss[loss=0.1983, simple_loss=0.2911, pruned_loss=0.05278, over 18712.00 frames. ], tot_loss[loss=0.2006, simple_loss=0.2841, pruned_loss=0.05858, over 3702252.46 frames. ], batch size: 54, lr: 1.06e-02, grad_scale: 8.0
2022-12-23 01:42:15,997 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 2.182e+02 4.452e+02 5.516e+02 6.913e+02 1.165e+03, threshold=1.103e+03, percent-clipped=1.0
2022-12-23 01:42:29,111 INFO [zipformer.py:660] (2/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=36234.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 01:42:33,998 WARNING [train.py:1060] (2/4) Exclude cut with ID 7276-92427-0014-12983-0_sp0.9 from training. Duration: 21.3055625
2022-12-23 01:42:35,479 WARNING [train.py:1060] (2/4) Exclude cut with ID 1025-75365-0008-79168-0_sp0.9 from training. Duration: 22.0666875
2022-12-23 01:42:41,807 INFO [zipformer.py:660] (2/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=36242.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 01:43:13,121 INFO [train.py:894] (2/4) Epoch 11, batch 1200, loss[loss=0.1786, simple_loss=0.2585, pruned_loss=0.04931, over 18536.00 frames. ], tot_loss[loss=0.2001, simple_loss=0.2835, pruned_loss=0.0583, over 3704320.76 frames. ], batch size: 44, lr: 1.06e-02, grad_scale: 8.0
2022-12-23 01:43:20,991 INFO [scaling.py:679] (2/4) Whitening: num_groups=1, num_channels=384, metric=3.19 vs. limit=5.0
2022-12-23 01:44:25,701 WARNING [train.py:1060] (2/4) Exclude cut with ID 2929-85685-0062-62366-0 from training. Duration: 20.26
2022-12-23 01:44:28,604 INFO [train.py:894] (2/4) Epoch 11, batch 1250, loss[loss=0.1939, simple_loss=0.285, pruned_loss=0.0514, over 18507.00 frames. ], tot_loss[loss=0.2002, simple_loss=0.2835, pruned_loss=0.05846, over 3706168.12 frames. ], batch size: 52, lr: 1.06e-02, grad_scale: 8.0
2022-12-23 01:44:38,819 WARNING [train.py:1060] (2/4) Exclude cut with ID 5239-32139-0030-9324-0_sp0.9 from training. Duration: 21.3444375
2022-12-23 01:44:48,462 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 2.493e+02 3.840e+02 4.722e+02 6.009e+02 1.141e+03, threshold=9.444e+02, percent-clipped=3.0
2022-12-23 01:45:34,677 WARNING [train.py:1060] (2/4) Exclude cut with ID 497-129325-0061-62254-0_sp1.1 from training. Duration: 0.97725
2022-12-23 01:45:44,710 INFO [train.py:894] (2/4) Epoch 11, batch 1300, loss[loss=0.1596, simple_loss=0.2408, pruned_loss=0.03918, over 18520.00 frames. ], tot_loss[loss=0.2009, simple_loss=0.2846, pruned_loss=0.05864, over 3707471.81 frames. ], batch size: 44, lr: 1.06e-02, grad_scale: 8.0
2022-12-23 01:45:59,139 INFO [scaling.py:679] (2/4) Whitening: num_groups=8, num_channels=96, metric=1.30 vs. limit=2.0
2022-12-23 01:46:17,888 WARNING [train.py:1060] (2/4) Exclude cut with ID 7395-89880-0031-39906-0_sp0.9 from training. Duration: 22.97225
2022-12-23 01:46:50,865 WARNING [train.py:1060] (2/4) Exclude cut with ID 7395-89880-0047-39922-0_sp0.9 from training. Duration: 21.97775
2022-12-23 01:46:59,613 INFO [train.py:894] (2/4) Epoch 11, batch 1350, loss[loss=0.164, simple_loss=0.2548, pruned_loss=0.03665, over 18675.00 frames. ], tot_loss[loss=0.2001, simple_loss=0.284, pruned_loss=0.05808, over 3709552.72 frames. ], batch size: 48, lr: 1.06e-02, grad_scale: 8.0
2022-12-23 01:47:05,266 WARNING [train.py:1060] (2/4) Exclude cut with ID 1112-1043-0006-89194-0_sp0.9 from training. Duration: 21.8333125
2022-12-23 01:47:14,732 WARNING [train.py:1060] (2/4) Exclude cut with ID 1914-133440-0031-94921-0 from training. Duration: 20.47
2022-12-23 01:47:19,943 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 2.513e+02 4.062e+02 4.949e+02 6.479e+02 1.390e+03, threshold=9.899e+02, percent-clipped=5.0
2022-12-23 01:47:21,735 INFO [zipformer.py:660] (2/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=36427.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 01:47:39,030 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([2.0879, 1.3788, 1.6794, 1.7502, 2.0643, 1.9628, 1.8603, 1.5133],
       device='cuda:2'), covar=tensor([0.1721, 0.2606, 0.2085, 0.2199, 0.1508, 0.0765, 0.2376, 0.0996],
       device='cuda:2'), in_proj_covar=tensor([0.0250, 0.0287, 0.0259, 0.0297, 0.0279, 0.0236, 0.0303, 0.0223],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-23 01:47:40,434 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([3.3422, 2.4167, 1.7168, 3.0366, 2.1965, 2.1654, 2.4070, 3.5583],
       device='cuda:2'), covar=tensor([0.1543, 0.2894, 0.1630, 0.2644, 0.3441, 0.0919, 0.2833, 0.0554],
       device='cuda:2'), in_proj_covar=tensor([0.0279, 0.0274, 0.0229, 0.0344, 0.0255, 0.0214, 0.0269, 0.0194],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0003, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-23 01:48:12,756 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.6484, 1.0812, 2.0052, 3.2865, 2.3087, 2.7410, 0.4948, 2.1369],
       device='cuda:2'), covar=tensor([0.1785, 0.1785, 0.1444, 0.0526, 0.1004, 0.1024, 0.2621, 0.1075],
       device='cuda:2'), in_proj_covar=tensor([0.0102, 0.0113, 0.0129, 0.0126, 0.0102, 0.0134, 0.0131, 0.0109],
       device='cuda:2'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0004, 0.0003, 0.0003],
       device='cuda:2')
2022-12-23 01:48:15,436 INFO [train.py:894] (2/4) Epoch 11, batch 1400, loss[loss=0.2246, simple_loss=0.3033, pruned_loss=0.07296, over 18585.00 frames. ], tot_loss[loss=0.2001, simple_loss=0.2839, pruned_loss=0.05817, over 3710395.45 frames. ], batch size: 57, lr: 1.06e-02, grad_scale: 8.0
2022-12-23 01:48:20,058 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.8453, 1.0834, 0.7446, 1.4129, 2.2125, 1.3460, 1.6548, 2.0953],
       device='cuda:2'), covar=tensor([0.1459, 0.2183, 0.2818, 0.1517, 0.1545, 0.1648, 0.1435, 0.1305],
       device='cuda:2'), in_proj_covar=tensor([0.0089, 0.0099, 0.0119, 0.0094, 0.0111, 0.0088, 0.0097, 0.0094],
       device='cuda:2'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003],
       device='cuda:2')
2022-12-23 01:48:21,477 INFO [zipformer.py:660] (2/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=36467.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 01:48:24,601 WARNING [train.py:1060] (2/4) Exclude cut with ID 7395-89880-0037-39912-0_sp0.9 from training. Duration: 20.67225
2022-12-23 01:48:34,503 INFO [zipformer.py:660] (2/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=36475.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 01:48:43,105 WARNING [train.py:1060] (2/4) Exclude cut with ID 1914-133440-0024-94914-0_sp0.9 from training. Duration: 25.2444375
2022-12-23 01:49:05,144 WARNING [train.py:1060] (2/4) Exclude cut with ID 3340-169293-0021-76797-0_sp0.9 from training. Duration: 21.1445
2022-12-23 01:49:33,104 INFO [train.py:894] (2/4) Epoch 11, batch 1450, loss[loss=0.2066, simple_loss=0.2915, pruned_loss=0.06083, over 18611.00 frames. ], tot_loss[loss=0.2003, simple_loss=0.2837, pruned_loss=0.05843, over 3710737.49 frames. ], batch size: 77, lr: 1.06e-02, grad_scale: 8.0
2022-12-23 01:49:36,318 INFO [zipformer.py:660] (2/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=36515.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 01:49:52,529 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 2.495e+02 4.266e+02 5.326e+02 7.004e+02 1.890e+03, threshold=1.065e+03, percent-clipped=3.0
2022-12-23 01:50:10,239 INFO [zipformer.py:660] (2/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=36537.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 01:50:15,000 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.4120, 1.0034, 1.0742, 1.0220, 1.4317, 1.3843, 1.5063, 1.0336],
       device='cuda:2'), covar=tensor([0.0223, 0.0274, 0.0415, 0.0227, 0.0196, 0.0311, 0.0196, 0.0285],
       device='cuda:2'), in_proj_covar=tensor([0.0088, 0.0118, 0.0145, 0.0125, 0.0111, 0.0110, 0.0093, 0.0120],
       device='cuda:2'), out_proj_covar=tensor([7.4041e-05, 9.8655e-05, 1.2706e-04, 1.0487e-04, 9.5633e-05, 8.9089e-05,
        7.6855e-05, 9.9918e-05], device='cuda:2')
2022-12-23 01:50:17,689 WARNING [train.py:1060] (2/4) Exclude cut with ID 2929-85685-0079-62383-0_sp0.9 from training. Duration: 33.038875
2022-12-23 01:50:49,217 INFO [train.py:894] (2/4) Epoch 11, batch 1500, loss[loss=0.2072, simple_loss=0.2952, pruned_loss=0.05956, over 18632.00 frames. ], tot_loss[loss=0.201, simple_loss=0.2843, pruned_loss=0.05889, over 3711037.95 frames. ], batch size: 53, lr: 1.06e-02, grad_scale: 8.0
2022-12-23 01:50:54,367 WARNING [train.py:1060] (2/4) Exclude cut with ID 6426-64291-0000-16059-0_sp0.9 from training. Duration: 20.0944375
2022-12-23 01:51:09,648 WARNING [train.py:1060] (2/4) Exclude cut with ID 2929-85685-0044-62348-0_sp1.1 from training. Duration: 20.4
2022-12-23 01:51:17,501 WARNING [train.py:1060] (2/4) Exclude cut with ID 6330-62851-0022-91297-0 from training. Duration: 20.085
2022-12-23 01:51:29,632 WARNING [train.py:1060] (2/4) Exclude cut with ID 4860-13185-0032-76709-0_sp0.9 from training. Duration: 23.07775
2022-12-23 01:51:53,805 INFO [zipformer.py:660] (2/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=36605.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 01:51:58,291 INFO [zipformer.py:660] (2/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=36608.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 01:52:04,773 INFO [zipformer.py:660] (2/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=36612.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 01:52:05,725 INFO [train.py:894] (2/4) Epoch 11, batch 1550, loss[loss=0.2015, simple_loss=0.2952, pruned_loss=0.05386, over 18658.00 frames. ], tot_loss[loss=0.1994, simple_loss=0.283, pruned_loss=0.05792, over 3711650.25 frames. ], batch size: 60, lr: 1.06e-02, grad_scale: 8.0
2022-12-23 01:52:17,985 WARNING [train.py:1060] (2/4) Exclude cut with ID 2929-85685-0044-62348-0_sp0.9 from training. Duration: 24.9333125
2022-12-23 01:52:25,465 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 2.783e+02 4.206e+02 5.136e+02 5.999e+02 2.048e+03, threshold=1.027e+03, percent-clipped=3.0
2022-12-23 01:52:55,562 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([2.6161, 2.3990, 1.7255, 1.5769, 3.1636, 2.8505, 2.4270, 1.9727],
       device='cuda:2'), covar=tensor([0.0316, 0.0328, 0.0561, 0.0618, 0.0130, 0.0253, 0.0374, 0.0663],
       device='cuda:2'), in_proj_covar=tensor([0.0120, 0.0119, 0.0129, 0.0120, 0.0087, 0.0118, 0.0135, 0.0148],
       device='cuda:2'), out_proj_covar=tensor([0.0001, 0.0001, 0.0002, 0.0001, 0.0001, 0.0001, 0.0002, 0.0002],
       device='cuda:2')
2022-12-23 01:53:01,349 WARNING [train.py:1060] (2/4) Exclude cut with ID 5118-111612-0016-124680-0_sp0.9 from training. Duration: 20.388875
2022-12-23 01:53:06,984 WARNING [train.py:1060] (2/4) Exclude cut with ID 432-122774-0017-62487-0_sp1.1 from training. Duration: 20.3590625
2022-12-23 01:53:22,337 INFO [train.py:894] (2/4) Epoch 11, batch 1600, loss[loss=0.1787, simple_loss=0.257, pruned_loss=0.05021, over 18601.00 frames. ], tot_loss[loss=0.1996, simple_loss=0.2829, pruned_loss=0.05812, over 3712907.18 frames. ], batch size: 45, lr: 1.06e-02, grad_scale: 8.0
2022-12-23 01:53:27,102 INFO [zipformer.py:660] (2/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=36666.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 01:53:31,381 INFO [zipformer.py:660] (2/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=36669.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 01:53:36,874 INFO [zipformer.py:660] (2/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=36673.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 01:54:15,012 WARNING [train.py:1060] (2/4) Exclude cut with ID 3557-8342-0013-54691-0_sp1.1 from training. Duration: 0.836375
2022-12-23 01:54:16,642 INFO [zipformer.py:660] (2/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=36700.0, num_to_drop=1, layers_to_drop={0}
2022-12-23 01:54:36,549 INFO [train.py:894] (2/4) Epoch 11, batch 1650, loss[loss=0.1987, simple_loss=0.2663, pruned_loss=0.06557, over 18525.00 frames. ], tot_loss[loss=0.2006, simple_loss=0.2832, pruned_loss=0.059, over 3712737.56 frames. ], batch size: 44, lr: 1.06e-02, grad_scale: 8.0
2022-12-23 01:54:55,462 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 2.822e+02 4.941e+02 6.129e+02 7.618e+02 1.418e+03, threshold=1.226e+03, percent-clipped=7.0
2022-12-23 01:54:59,909 WARNING [train.py:1060] (2/4) Exclude cut with ID 8565-290391-0049-67394-0_sp0.9 from training. Duration: 21.3166875
2022-12-23 01:55:29,853 WARNING [train.py:1060] (2/4) Exclude cut with ID 6533-399-0029-104863-0_sp0.9 from training. Duration: 22.1055625
2022-12-23 01:55:42,712 WARNING [train.py:1060] (2/4) Exclude cut with ID 7699-105389-0094-26379-0_sp1.1 from training. Duration: 21.77725
2022-12-23 01:55:49,097 INFO [zipformer.py:660] (2/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=36761.0, num_to_drop=1, layers_to_drop={2}
2022-12-23 01:55:51,584 INFO [train.py:894] (2/4) Epoch 11, batch 1700, loss[loss=0.2375, simple_loss=0.2999, pruned_loss=0.08753, over 18567.00 frames. ], tot_loss[loss=0.2038, simple_loss=0.2849, pruned_loss=0.06131, over 3712508.89 frames. ], batch size: 49, lr: 1.05e-02, grad_scale: 8.0
2022-12-23 01:55:53,811 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.5598, 2.1483, 1.3740, 2.3190, 2.4823, 1.4503, 1.5003, 1.2566],
       device='cuda:2'), covar=tensor([0.1783, 0.1504, 0.1520, 0.0881, 0.1274, 0.1179, 0.1816, 0.1404],
       device='cuda:2'), in_proj_covar=tensor([0.0236, 0.0213, 0.0202, 0.0187, 0.0250, 0.0188, 0.0208, 0.0189],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-23 01:55:59,831 WARNING [train.py:1060] (2/4) Exclude cut with ID 7255-291500-0005-134304-0_sp0.9 from training. Duration: 27.8166875
2022-12-23 01:56:06,278 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.6895, 1.2035, 0.8613, 1.2345, 1.9180, 0.9069, 1.3902, 1.5221],
       device='cuda:2'), covar=tensor([0.1575, 0.2122, 0.2136, 0.1500, 0.1725, 0.1640, 0.1365, 0.1612],
       device='cuda:2'), in_proj_covar=tensor([0.0089, 0.0099, 0.0117, 0.0094, 0.0112, 0.0089, 0.0097, 0.0094],
       device='cuda:2'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003],
       device='cuda:2')
2022-12-23 01:56:26,871 WARNING [train.py:1060] (2/4) Exclude cut with ID 7357-94126-0021-15852-0_sp1.1 from training. Duration: 22.5090625
2022-12-23 01:56:32,934 WARNING [train.py:1060] (2/4) Exclude cut with ID 7255-291500-0005-134304-0 from training. Duration: 25.035
2022-12-23 01:56:50,743 WARNING [train.py:1060] (2/4) Exclude cut with ID 774-127930-0014-10412-0_sp1.1 from training. Duration: 0.95
2022-12-23 01:56:58,223 INFO [zipformer.py:660] (2/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=36805.0, num_to_drop=1, layers_to_drop={0}
2022-12-23 01:57:09,349 INFO [train.py:894] (2/4) Epoch 11, batch 1750, loss[loss=0.2102, simple_loss=0.2784, pruned_loss=0.071, over 18385.00 frames. ], tot_loss[loss=0.2068, simple_loss=0.2866, pruned_loss=0.06354, over 3712580.04 frames. ], batch size: 46, lr: 1.05e-02, grad_scale: 8.0
2022-12-23 01:57:09,375 WARNING [train.py:1060] (2/4) Exclude cut with ID 3033-130750-0096-55598-0_sp0.9 from training. Duration: 0.92225
2022-12-23 01:57:30,107 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 2.720e+02 5.303e+02 6.412e+02 8.124e+02 1.465e+03, threshold=1.282e+03, percent-clipped=3.0
2022-12-23 01:57:37,407 WARNING [train.py:1060] (2/4) Exclude cut with ID 4511-76322-0006-80011-0 from training. Duration: 21.97
2022-12-23 01:57:46,814 INFO [zipformer.py:660] (2/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=36837.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 01:57:57,056 WARNING [train.py:1060] (2/4) Exclude cut with ID 7492-105653-0055-62765-0_sp0.9 from training. Duration: 21.97225
2022-12-23 01:57:58,465 WARNING [train.py:1060] (2/4) Exclude cut with ID 453-131332-0000-47844-0_sp0.9 from training. Duration: 25.3333125
2022-12-23 01:58:10,029 WARNING [train.py:1060] (2/4) Exclude cut with ID 5172-29468-0015-19128-0_sp0.9 from training. Duration: 21.5055625
2022-12-23 01:58:21,112 WARNING [train.py:1060] (2/4) Exclude cut with ID 453-131332-0000-47844-0_sp1.1 from training. Duration: 20.72725
2022-12-23 01:58:25,592 INFO [train.py:894] (2/4) Epoch 11, batch 1800, loss[loss=0.2636, simple_loss=0.3266, pruned_loss=0.1003, over 18719.00 frames. ], tot_loss[loss=0.2098, simple_loss=0.2881, pruned_loss=0.06571, over 3712557.01 frames. ], batch size: 54, lr: 1.05e-02, grad_scale: 8.0
2022-12-23 01:58:30,299 INFO [zipformer.py:660] (2/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=36866.0, num_to_drop=1, layers_to_drop={1}
2022-12-23 01:58:55,742 WARNING [train.py:1060] (2/4) Exclude cut with ID 8631-249866-0030-130156-0_sp0.9 from training. Duration: 26.32775
2022-12-23 01:58:58,764 INFO [zipformer.py:660] (2/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=36885.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 01:59:27,463 WARNING [train.py:1060] (2/4) Exclude cut with ID 3867-173237-0077-144769-0 from training. Duration: 20.025
2022-12-23 01:59:33,172 WARNING [train.py:1060] (2/4) Exclude cut with ID 6709-74022-0004-86860-0_sp1.1 from training. Duration: 0.9409375
2022-12-23 01:59:33,182 WARNING [train.py:1060] (2/4) Exclude cut with ID 4757-1811-0023-62229-0_sp0.9 from training. Duration: 21.37775
2022-12-23 01:59:39,330 INFO [zipformer.py:660] (2/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=36911.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 01:59:42,056 INFO [train.py:894] (2/4) Epoch 11, batch 1850, loss[loss=0.2064, simple_loss=0.2859, pruned_loss=0.06346, over 18515.00 frames. ], tot_loss[loss=0.2146, simple_loss=0.2913, pruned_loss=0.06897, over 3712997.94 frames. ], batch size: 52, lr: 1.05e-02, grad_scale: 8.0
2022-12-23 01:59:55,825 WARNING [train.py:1060] (2/4) Exclude cut with ID 1250-135782-0004-25974-0_sp0.9 from training. Duration: 21.17225
2022-12-23 01:59:55,834 WARNING [train.py:1060] (2/4) Exclude cut with ID 7357-94126-0021-15852-0_sp0.9 from training. Duration: 27.511125
2022-12-23 02:00:02,485 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 3.556e+02 5.200e+02 6.324e+02 7.983e+02 2.314e+03, threshold=1.265e+03, percent-clipped=3.0
2022-12-23 02:00:30,565 WARNING [train.py:1060] (2/4) Exclude cut with ID 453-131332-0000-47844-0 from training. Duration: 22.8
2022-12-23 02:00:35,064 WARNING [train.py:1060] (2/4) Exclude cut with ID 4964-30587-0040-44509-0 from training. Duration: 22.585
2022-12-23 02:00:52,193 INFO [zipformer.py:660] (2/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=36958.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 02:00:55,194 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([0.4705, 1.9108, 2.0770, 0.9472, 1.3579, 2.2098, 2.0088, 1.7973],
       device='cuda:2'), covar=tensor([0.0569, 0.0265, 0.0269, 0.0331, 0.0317, 0.0340, 0.0189, 0.0455],
       device='cuda:2'), in_proj_covar=tensor([0.0136, 0.0158, 0.0111, 0.0128, 0.0136, 0.0127, 0.0142, 0.0146],
       device='cuda:2'), out_proj_covar=tensor([1.1149e-04, 1.3134e-04, 9.0526e-05, 1.0289e-04, 1.1091e-04, 1.0635e-04,
        1.1913e-04, 1.2069e-04], device='cuda:2')
2022-12-23 02:00:56,521 INFO [zipformer.py:660] (2/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=36961.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 02:00:59,180 INFO [train.py:894] (2/4) Epoch 11, batch 1900, loss[loss=0.1962, simple_loss=0.2722, pruned_loss=0.06011, over 18395.00 frames. ], tot_loss[loss=0.2157, simple_loss=0.2915, pruned_loss=0.06996, over 3712540.16 frames. ], batch size: 46, lr: 1.05e-02, grad_scale: 8.0
2022-12-23 02:01:00,831 INFO [zipformer.py:660] (2/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=36964.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 02:01:06,906 WARNING [train.py:1060] (2/4) Exclude cut with ID 6978-92210-0001-146967-0_sp0.9 from training. Duration: 22.0166875
2022-12-23 02:01:07,047 INFO [zipformer.py:660] (2/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=36968.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 02:01:13,701 INFO [zipformer.py:660] (2/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=36972.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 02:01:24,029 WARNING [train.py:1060] (2/4) Exclude cut with ID 7255-291500-0003-134302-0_sp1.1 from training. Duration: 24.395375
2022-12-23 02:01:31,370 WARNING [train.py:1060] (2/4) Exclude cut with ID 3488-85273-0017-41203-0_sp0.9 from training. Duration: 27.47775
2022-12-23 02:01:35,823 WARNING [train.py:1060] (2/4) Exclude cut with ID 432-122774-0017-62487-0_sp0.9 from training. Duration: 24.8833125
2022-12-23 02:01:38,764 WARNING [train.py:1060] (2/4) Exclude cut with ID 6758-72288-0033-108368-0 from training. Duration: 23.39
2022-12-23 02:01:45,284 WARNING [train.py:1060] (2/4) Exclude cut with ID 8291-282929-0007-12994-0_sp0.9 from training. Duration: 28.72225
2022-12-23 02:01:52,337 WARNING [train.py:1060] (2/4) Exclude cut with ID 585-294811-0110-133686-0_sp0.9 from training. Duration: 20.8944375
2022-12-23 02:02:02,444 INFO [zipformer.py:660] (2/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=37004.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 02:02:09,301 WARNING [train.py:1060] (2/4) Exclude cut with ID 5796-66357-0007-116447-0_sp0.9 from training. Duration: 23.8444375
2022-12-23 02:02:15,054 INFO [train.py:894] (2/4) Epoch 11, batch 1950, loss[loss=0.2012, simple_loss=0.2756, pruned_loss=0.06336, over 18523.00 frames. ], tot_loss[loss=0.2166, simple_loss=0.2917, pruned_loss=0.07071, over 3713586.23 frames. ], batch size: 47, lr: 1.05e-02, grad_scale: 8.0
2022-12-23 02:02:24,880 INFO [zipformer.py:660] (2/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=37019.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 02:02:30,010 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.8813, 1.2878, 0.7178, 1.3198, 2.1888, 1.4364, 1.5617, 1.9050],
       device='cuda:2'), covar=tensor([0.1928, 0.2755, 0.3029, 0.2009, 0.1935, 0.1963, 0.2039, 0.2091],
       device='cuda:2'), in_proj_covar=tensor([0.0091, 0.0100, 0.0119, 0.0095, 0.0114, 0.0090, 0.0098, 0.0095],
       device='cuda:2'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003],
       device='cuda:2')
2022-12-23 02:02:34,187 WARNING [train.py:1060] (2/4) Exclude cut with ID 8291-282929-0007-12994-0 from training. Duration: 25.85
2022-12-23 02:02:35,548 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 3.431e+02 5.428e+02 6.678e+02 8.110e+02 1.320e+03, threshold=1.336e+03, percent-clipped=2.0
2022-12-23 02:02:35,572 WARNING [train.py:1060] (2/4) Exclude cut with ID 8291-282929-0023-13010-0 from training. Duration: 21.39
2022-12-23 02:02:43,464 WARNING [train.py:1060] (2/4) Exclude cut with ID 7357-94126-0014-15845-0 from training. Duration: 27.92
2022-12-23 02:02:57,942 INFO [scaling.py:679] (2/4) Whitening: num_groups=8, num_channels=96, metric=1.38 vs. limit=2.0
2022-12-23 02:03:12,193 WARNING [train.py:1060] (2/4) Exclude cut with ID 8631-249866-0039-130165-0_sp0.9 from training. Duration: 20.661125
2022-12-23 02:03:21,134 INFO [zipformer.py:660] (2/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=37056.0, num_to_drop=1, layers_to_drop={2}
2022-12-23 02:03:31,607 INFO [train.py:894] (2/4) Epoch 11, batch 2000, loss[loss=0.2563, simple_loss=0.3284, pruned_loss=0.0921, over 18559.00 frames. ], tot_loss[loss=0.2167, simple_loss=0.2912, pruned_loss=0.07107, over 3713781.74 frames. ], batch size: 57, lr: 1.05e-02, grad_scale: 8.0
2022-12-23 02:03:35,018 INFO [zipformer.py:660] (2/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=37065.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 02:03:36,132 WARNING [train.py:1060] (2/4) Exclude cut with ID 7357-94126-0043-15874-0_sp0.9 from training. Duration: 20.07225
2022-12-23 02:03:42,602 WARNING [train.py:1060] (2/4) Exclude cut with ID 1085-156170-0017-128270-0 from training. Duration: 21.01
2022-12-23 02:04:47,942 INFO [train.py:894] (2/4) Epoch 11, batch 2050, loss[loss=0.2095, simple_loss=0.2861, pruned_loss=0.06645, over 18413.00 frames. ], tot_loss[loss=0.2171, simple_loss=0.2916, pruned_loss=0.07132, over 3712642.80 frames. ], batch size: 53, lr: 1.05e-02, grad_scale: 8.0
2022-12-23 02:04:50,182 WARNING [train.py:1060] (2/4) Exclude cut with ID 2195-150901-0045-59933-0 from training. Duration: 20.65
2022-12-23 02:04:55,709 INFO [scaling.py:679] (2/4) Whitening: num_groups=8, num_channels=192, metric=1.93 vs. limit=2.0
2022-12-23 02:04:56,531 WARNING [train.py:1060] (2/4) Exclude cut with ID 5796-66357-0007-116447-0 from training. Duration: 21.46
2022-12-23 02:04:59,805 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.6862, 1.5579, 1.5141, 1.9528, 1.7699, 3.0077, 1.4823, 1.6152],
       device='cuda:2'), covar=tensor([0.0958, 0.1734, 0.1121, 0.0872, 0.1417, 0.0318, 0.1435, 0.1510],
       device='cuda:2'), in_proj_covar=tensor([0.0073, 0.0083, 0.0075, 0.0075, 0.0092, 0.0072, 0.0086, 0.0078],
       device='cuda:2'), out_proj_covar=tensor([0.0003, 0.0004, 0.0003, 0.0003, 0.0004, 0.0003, 0.0004, 0.0004],
       device='cuda:2')
2022-12-23 02:05:08,584 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 3.252e+02 5.274e+02 6.467e+02 7.854e+02 2.150e+03, threshold=1.293e+03, percent-clipped=5.0
2022-12-23 02:05:44,446 WARNING [train.py:1060] (2/4) Exclude cut with ID 3557-8342-0013-54691-0 from training. Duration: 0.92
2022-12-23 02:05:50,704 WARNING [train.py:1060] (2/4) Exclude cut with ID 8291-282929-0023-13010-0_sp0.9 from training. Duration: 23.7666875
2022-12-23 02:05:51,641 INFO [scaling.py:679] (2/4) Whitening: num_groups=8, num_channels=96, metric=1.36 vs. limit=2.0
2022-12-23 02:06:03,455 INFO [zipformer.py:660] (2/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=37161.0, num_to_drop=1, layers_to_drop={1}
2022-12-23 02:06:06,097 INFO [train.py:894] (2/4) Epoch 11, batch 2100, loss[loss=0.2186, simple_loss=0.2801, pruned_loss=0.07853, over 18597.00 frames. ], tot_loss[loss=0.2163, simple_loss=0.2908, pruned_loss=0.07088, over 3712754.83 frames. ], batch size: 41, lr: 1.05e-02, grad_scale: 8.0
2022-12-23 02:06:30,089 WARNING [train.py:1060] (2/4) Exclude cut with ID 8544-281189-0060-101339-0_sp0.9 from training. Duration: 20.861125
2022-12-23 02:06:40,397 WARNING [train.py:1060] (2/4) Exclude cut with ID 3488-65654-0031-41259-0_sp0.9 from training. Duration: 22.711125
2022-12-23 02:07:22,312 WARNING [train.py:1060] (2/4) Exclude cut with ID 6951-79737-0043-132310-0_sp1.1 from training. Duration: 22.986375
2022-12-23 02:07:23,582 INFO [train.py:894] (2/4) Epoch 11, batch 2150, loss[loss=0.211, simple_loss=0.2907, pruned_loss=0.06568, over 18647.00 frames. ], tot_loss[loss=0.2179, simple_loss=0.2918, pruned_loss=0.07197, over 3712549.60 frames. ], batch size: 60, lr: 1.05e-02, grad_scale: 8.0
2022-12-23 02:07:38,265 WARNING [train.py:1060] (2/4) Exclude cut with ID 8040-260924-0003-80960-0_sp0.9 from training. Duration: 22.07225
2022-12-23 02:07:43,048 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 3.116e+02 5.173e+02 6.018e+02 7.074e+02 1.637e+03, threshold=1.204e+03, percent-clipped=3.0
2022-12-23 02:07:43,095 WARNING [train.py:1060] (2/4) Exclude cut with ID 7699-105389-0045-26330-0_sp0.9 from training. Duration: 20.3055625
2022-12-23 02:07:46,741 WARNING [train.py:1060] (2/4) Exclude cut with ID 6356-271890-0060-94317-0_sp0.9 from training. Duration: 20.72225
2022-12-23 02:08:05,102 WARNING [train.py:1060] (2/4) Exclude cut with ID 3488-85273-0017-41203-0_sp1.1 from training. Duration: 22.4818125
2022-12-23 02:08:33,355 WARNING [train.py:1060] (2/4) Exclude cut with ID 4964-30587-0040-44509-0_sp0.9 from training. Duration: 25.0944375
2022-12-23 02:08:36,345 WARNING [train.py:1060] (2/4) Exclude cut with ID 6533-399-0047-104881-0 from training. Duration: 21.515
2022-12-23 02:08:36,576 INFO [zipformer.py:660] (2/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=37261.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 02:08:39,101 INFO [train.py:894] (2/4) Epoch 11, batch 2200, loss[loss=0.2236, simple_loss=0.302, pruned_loss=0.07261, over 18374.00 frames. ], tot_loss[loss=0.2172, simple_loss=0.291, pruned_loss=0.07169, over 3713491.31 frames. ], batch size: 51, lr: 1.05e-02, grad_scale: 8.0
2022-12-23 02:08:40,902 INFO [zipformer.py:660] (2/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=37264.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 02:08:42,422 WARNING [train.py:1060] (2/4) Exclude cut with ID 7357-94126-0009-15840-0_sp0.9 from training. Duration: 27.02225
2022-12-23 02:08:45,098 INFO [zipformer.py:660] (2/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=37267.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 02:08:46,419 WARNING [train.py:1060] (2/4) Exclude cut with ID 432-122774-0010-62480-0_sp0.9 from training. Duration: 22.22225
2022-12-23 02:08:46,706 INFO [zipformer.py:660] (2/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=37268.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 02:08:46,865 INFO [zipformer.py:660] (2/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=37268.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 02:08:52,576 WARNING [train.py:1060] (2/4) Exclude cut with ID 4964-30587-0085-44554-0_sp0.9 from training. Duration: 20.85
2022-12-23 02:09:17,527 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.8755, 1.8928, 1.3833, 1.9431, 2.0568, 1.7645, 2.6242, 1.9555],
       device='cuda:2'), covar=tensor([0.0820, 0.1384, 0.2423, 0.1701, 0.1569, 0.0869, 0.0913, 0.1022],
       device='cuda:2'), in_proj_covar=tensor([0.0173, 0.0191, 0.0230, 0.0281, 0.0222, 0.0181, 0.0201, 0.0192],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-23 02:09:26,716 WARNING [train.py:1060] (2/4) Exclude cut with ID 4295-39940-0007-92567-0 from training. Duration: 21.54
2022-12-23 02:09:32,170 WARNING [train.py:1060] (2/4) Exclude cut with ID 4964-30587-0040-44509-0_sp1.1 from training. Duration: 20.5318125
2022-12-23 02:09:41,230 WARNING [train.py:1060] (2/4) Exclude cut with ID 7255-291500-0012-134311-0_sp0.9 from training. Duration: 21.9333125
2022-12-23 02:09:44,281 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.6358, 1.0913, 1.9239, 2.9295, 2.0655, 2.3488, 0.6904, 2.0470],
       device='cuda:2'), covar=tensor([0.1764, 0.1887, 0.1389, 0.0621, 0.1207, 0.1320, 0.2418, 0.1222],
       device='cuda:2'), in_proj_covar=tensor([0.0102, 0.0115, 0.0128, 0.0127, 0.0104, 0.0132, 0.0130, 0.0108],
       device='cuda:2'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003],
       device='cuda:2')
2022-12-23 02:09:48,521 INFO [zipformer.py:660] (2/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=37309.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 02:09:53,209 INFO [zipformer.py:660] (2/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=37312.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 02:09:54,466 INFO [train.py:894] (2/4) Epoch 11, batch 2250, loss[loss=0.1823, simple_loss=0.2569, pruned_loss=0.05387, over 18531.00 frames. ], tot_loss[loss=0.2179, simple_loss=0.2915, pruned_loss=0.07218, over 3712965.73 frames. ], batch size: 44, lr: 1.05e-02, grad_scale: 8.0
2022-12-23 02:09:56,453 INFO [zipformer.py:660] (2/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=37314.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 02:09:59,321 INFO [zipformer.py:660] (2/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=37316.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 02:10:15,915 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 3.192e+02 5.513e+02 6.965e+02 8.801e+02 2.114e+03, threshold=1.393e+03, percent-clipped=10.0
2022-12-23 02:10:20,669 INFO [zipformer.py:660] (2/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=37329.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 02:10:27,354 WARNING [train.py:1060] (2/4) Exclude cut with ID 8631-249866-0025-130151-0_sp0.9 from training. Duration: 21.7944375
2022-12-23 02:10:38,288 WARNING [train.py:1060] (2/4) Exclude cut with ID 8291-282929-0002-12989-0_sp0.9 from training. Duration: 22.4666875
2022-12-23 02:10:43,075 INFO [scaling.py:679] (2/4) Whitening: num_groups=1, num_channels=384, metric=3.50 vs. limit=5.0
2022-12-23 02:10:45,131 WARNING [train.py:1060] (2/4) Exclude cut with ID 6121-9014-0076-24124-0 from training. Duration: 21.635
2022-12-23 02:10:51,353 WARNING [train.py:1060] (2/4) Exclude cut with ID 6121-9014-0076-24124-0_sp0.9 from training. Duration: 24.038875
2022-12-23 02:11:00,817 INFO [zipformer.py:660] (2/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=37356.0, num_to_drop=1, layers_to_drop={2}
2022-12-23 02:11:06,386 INFO [zipformer.py:660] (2/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=37360.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 02:11:10,405 INFO [train.py:894] (2/4) Epoch 11, batch 2300, loss[loss=0.1828, simple_loss=0.2606, pruned_loss=0.05246, over 18402.00 frames. ], tot_loss[loss=0.2184, simple_loss=0.2919, pruned_loss=0.07249, over 3713914.84 frames. ], batch size: 46, lr: 1.05e-02, grad_scale: 8.0
2022-12-23 02:11:12,061 INFO [zipformer.py:660] (2/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=37364.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 02:11:35,176 WARNING [train.py:1060] (2/4) Exclude cut with ID 7255-291500-0009-134308-0_sp1.1 from training. Duration: 21.786375
2022-12-23 02:11:36,060 INFO [scaling.py:679] (2/4) Whitening: num_groups=8, num_channels=96, metric=1.16 vs. limit=2.0
2022-12-23 02:11:48,609 WARNING [train.py:1060] (2/4) Exclude cut with ID 8291-282929-0002-12989-0 from training. Duration: 20.22
2022-12-23 02:12:12,852 INFO [zipformer.py:660] (2/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=37404.0, num_to_drop=1, layers_to_drop={0}
2022-12-23 02:12:25,850 INFO [train.py:894] (2/4) Epoch 11, batch 2350, loss[loss=0.2168, simple_loss=0.2979, pruned_loss=0.06786, over 18710.00 frames. ], tot_loss[loss=0.2191, simple_loss=0.2923, pruned_loss=0.07299, over 3713127.76 frames. ], batch size: 52, lr: 1.05e-02, grad_scale: 8.0
2022-12-23 02:12:45,422 INFO [zipformer.py:660] (2/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=37425.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 02:12:46,721 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 3.259e+02 5.576e+02 6.263e+02 7.534e+02 2.266e+03, threshold=1.253e+03, percent-clipped=1.0
2022-12-23 02:13:41,409 INFO [zipformer.py:660] (2/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=37461.0, num_to_drop=1, layers_to_drop={0}
2022-12-23 02:13:44,321 INFO [train.py:894] (2/4) Epoch 11, batch 2400, loss[loss=0.2126, simple_loss=0.2792, pruned_loss=0.07296, over 18521.00 frames. ], tot_loss[loss=0.218, simple_loss=0.2914, pruned_loss=0.07233, over 3714020.90 frames. ], batch size: 47, lr: 1.05e-02, grad_scale: 16.0
2022-12-23 02:13:48,965 WARNING [train.py:1060] (2/4) Exclude cut with ID 6951-79737-0043-132310-0 from training. Duration: 25.285
2022-12-23 02:13:58,168 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.5994, 1.0561, 0.8391, 1.1500, 1.9525, 0.9141, 1.3056, 1.4989],
       device='cuda:2'), covar=tensor([0.1576, 0.2282, 0.2311, 0.1700, 0.1842, 0.1753, 0.1550, 0.1697],
       device='cuda:2'), in_proj_covar=tensor([0.0091, 0.0100, 0.0119, 0.0096, 0.0114, 0.0091, 0.0098, 0.0096],
       device='cuda:2'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003],
       device='cuda:2')
2022-12-23 02:14:51,781 WARNING [train.py:1060] (2/4) Exclude cut with ID 811-130148-0001-63453-0_sp0.9 from training. Duration: 20.861125
2022-12-23 02:14:55,269 INFO [zipformer.py:660] (2/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=37509.0, num_to_drop=1, layers_to_drop={0}
2022-12-23 02:15:00,715 INFO [train.py:894] (2/4) Epoch 11, batch 2450, loss[loss=0.1879, simple_loss=0.2575, pruned_loss=0.05913, over 18475.00 frames. ], tot_loss[loss=0.219, simple_loss=0.2923, pruned_loss=0.07288, over 3715004.60 frames. ], batch size: 43, lr: 1.04e-02, grad_scale: 16.0
2022-12-23 02:15:13,199 WARNING [train.py:1060] (2/4) Exclude cut with ID 6010-56788-0055-90261-0 from training. Duration: 20.88
2022-12-23 02:15:20,403 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 3.193e+02 5.351e+02 6.733e+02 8.810e+02 1.600e+03, threshold=1.347e+03, percent-clipped=5.0
2022-12-23 02:15:44,367 WARNING [train.py:1060] (2/4) Exclude cut with ID 7357-94126-0045-15876-0_sp0.9 from training. Duration: 23.4166875
2022-12-23 02:16:16,706 INFO [train.py:894] (2/4) Epoch 11, batch 2500, loss[loss=0.2302, simple_loss=0.3018, pruned_loss=0.07929, over 18595.00 frames. ], tot_loss[loss=0.2185, simple_loss=0.2919, pruned_loss=0.07252, over 3714580.43 frames. ], batch size: 69, lr: 1.04e-02, grad_scale: 16.0
2022-12-23 02:16:23,482 INFO [zipformer.py:660] (2/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=37567.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 02:16:33,176 INFO [scaling.py:679] (2/4) Whitening: num_groups=1, num_channels=384, metric=3.89 vs. limit=5.0
2022-12-23 02:16:40,174 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([2.2146, 2.2545, 2.3413, 2.2308, 2.0291, 3.9956, 2.4368, 2.7955],
       device='cuda:2'), covar=tensor([0.2559, 0.1529, 0.1373, 0.1440, 0.1036, 0.0162, 0.1298, 0.0667],
       device='cuda:2'), in_proj_covar=tensor([0.0139, 0.0117, 0.0128, 0.0120, 0.0104, 0.0100, 0.0097, 0.0093],
       device='cuda:2'), out_proj_covar=tensor([0.0005, 0.0005, 0.0005, 0.0004, 0.0004, 0.0003, 0.0004, 0.0004],
       device='cuda:2')
2022-12-23 02:16:48,350 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([2.3747, 2.0937, 1.6118, 2.3550, 1.8400, 1.9044, 1.9026, 2.4256],
       device='cuda:2'), covar=tensor([0.1634, 0.2551, 0.1516, 0.2225, 0.2674, 0.0885, 0.2403, 0.0682],
       device='cuda:2'), in_proj_covar=tensor([0.0278, 0.0269, 0.0226, 0.0343, 0.0251, 0.0211, 0.0265, 0.0192],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0003, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-23 02:17:02,945 WARNING [train.py:1060] (2/4) Exclude cut with ID 7255-291500-0006-134305-0 from training. Duration: 21.24
2022-12-23 02:17:02,960 WARNING [train.py:1060] (2/4) Exclude cut with ID 6533-399-0047-104881-0_sp0.9 from training. Duration: 23.9055625
2022-12-23 02:17:34,318 INFO [train.py:894] (2/4) Epoch 11, batch 2550, loss[loss=0.1997, simple_loss=0.2854, pruned_loss=0.05703, over 18399.00 frames. ], tot_loss[loss=0.2184, simple_loss=0.2918, pruned_loss=0.07252, over 3715823.68 frames. ], batch size: 53, lr: 1.04e-02, grad_scale: 16.0
2022-12-23 02:17:36,096 INFO [zipformer.py:660] (2/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=37614.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 02:17:37,353 INFO [zipformer.py:660] (2/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=37615.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 02:17:38,946 WARNING [train.py:1060] (2/4) Exclude cut with ID 6758-72288-0033-108368-0_sp0.9 from training. Duration: 25.988875
2022-12-23 02:17:48,195 WARNING [train.py:1060] (2/4) Exclude cut with ID 7255-291500-0001-134300-0_sp0.9 from training. Duration: 20.67225
2022-12-23 02:17:51,506 INFO [zipformer.py:660] (2/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=37624.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 02:17:54,170 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 3.571e+02 5.424e+02 6.401e+02 7.705e+02 1.573e+03, threshold=1.280e+03, percent-clipped=3.0
2022-12-23 02:18:35,911 WARNING [train.py:1060] (2/4) Exclude cut with ID 3488-85273-0038-41224-0 from training. Duration: 20.34
2022-12-23 02:18:47,511 INFO [zipformer.py:660] (2/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=37660.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 02:18:50,115 INFO [zipformer.py:660] (2/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=37662.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 02:18:51,374 INFO [train.py:894] (2/4) Epoch 11, batch 2600, loss[loss=0.1974, simple_loss=0.2739, pruned_loss=0.0604, over 18559.00 frames. ], tot_loss[loss=0.2171, simple_loss=0.2904, pruned_loss=0.07195, over 3715268.69 frames. ], batch size: 49, lr: 1.04e-02, grad_scale: 16.0
2022-12-23 02:19:50,147 WARNING [train.py:1060] (2/4) Exclude cut with ID 7357-94126-0026-15857-0_sp0.9 from training. Duration: 25.061125
2022-12-23 02:20:01,296 WARNING [train.py:1060] (2/4) Exclude cut with ID 3033-130750-0096-55598-0 from training. Duration: 0.83
2022-12-23 02:20:01,495 INFO [zipformer.py:660] (2/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=37708.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 02:20:08,858 INFO [train.py:894] (2/4) Epoch 11, batch 2650, loss[loss=0.2526, simple_loss=0.3125, pruned_loss=0.09631, over 18663.00 frames. ], tot_loss[loss=0.2165, simple_loss=0.2899, pruned_loss=0.07158, over 3713811.45 frames. ], batch size: 186, lr: 1.04e-02, grad_scale: 8.0
2022-12-23 02:20:19,044 INFO [zipformer.py:660] (2/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=37720.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 02:20:25,199 WARNING [train.py:1060] (2/4) Exclude cut with ID 3488-85273-0017-41203-0 from training. Duration: 24.73
2022-12-23 02:20:29,877 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 3.225e+02 5.554e+02 6.745e+02 9.322e+02 2.077e+03, threshold=1.349e+03, percent-clipped=8.0
2022-12-23 02:20:37,003 WARNING [train.py:1060] (2/4) Exclude cut with ID 7255-291500-0009-134308-0 from training. Duration: 23.965
2022-12-23 02:20:46,960 WARNING [train.py:1060] (2/4) Exclude cut with ID 6978-92210-0030-146996-0_sp0.9 from training. Duration: 22.088875
2022-12-23 02:20:48,937 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([2.0407, 2.1895, 1.5214, 2.5775, 2.3406, 2.0162, 3.0235, 2.2158],
       device='cuda:2'), covar=tensor([0.0821, 0.1516, 0.2491, 0.1623, 0.1493, 0.0808, 0.0844, 0.1041],
       device='cuda:2'), in_proj_covar=tensor([0.0177, 0.0196, 0.0239, 0.0286, 0.0229, 0.0186, 0.0207, 0.0197],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-23 02:21:05,926 WARNING [train.py:1060] (2/4) Exclude cut with ID 7255-291500-0006-134305-0_sp0.9 from training. Duration: 23.6
2022-12-23 02:21:25,766 INFO [train.py:894] (2/4) Epoch 11, batch 2700, loss[loss=0.2282, simple_loss=0.302, pruned_loss=0.07719, over 18501.00 frames. ], tot_loss[loss=0.217, simple_loss=0.2904, pruned_loss=0.07177, over 3713922.17 frames. ], batch size: 58, lr: 1.04e-02, grad_scale: 8.0
2022-12-23 02:22:42,364 INFO [train.py:894] (2/4) Epoch 11, batch 2750, loss[loss=0.2351, simple_loss=0.3036, pruned_loss=0.08332, over 18490.00 frames. ], tot_loss[loss=0.2164, simple_loss=0.2898, pruned_loss=0.07156, over 3714320.70 frames. ], batch size: 77, lr: 1.04e-02, grad_scale: 8.0
2022-12-23 02:22:45,635 WARNING [train.py:1060] (2/4) Exclude cut with ID 8291-282929-0024-13011-0 from training. Duration: 23.795
2022-12-23 02:23:02,366 WARNING [train.py:1060] (2/4) Exclude cut with ID 8631-249866-0030-130156-0_sp1.1 from training. Duration: 21.5409375
2022-12-23 02:23:03,845 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 2.734e+02 4.899e+02 5.772e+02 7.300e+02 1.732e+03, threshold=1.154e+03, percent-clipped=2.0
2022-12-23 02:23:06,095 WARNING [train.py:1060] (2/4) Exclude cut with ID 6978-92210-0019-146985-0_sp0.9 from training. Duration: 24.97775
2022-12-23 02:23:18,541 WARNING [train.py:1060] (2/4) Exclude cut with ID 1085-156170-0017-128270-0_sp0.9 from training. Duration: 23.3444375
2022-12-23 02:23:44,588 WARNING [train.py:1060] (2/4) Exclude cut with ID 6010-56788-0055-90261-0_sp0.9 from training. Duration: 23.2
2022-12-23 02:23:48,218 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([2.1751, 1.4457, 1.7737, 1.8182, 2.1182, 1.9750, 1.9710, 1.5097],
       device='cuda:2'), covar=tensor([0.1739, 0.2696, 0.2101, 0.2227, 0.1464, 0.0856, 0.2343, 0.1080],
       device='cuda:2'), in_proj_covar=tensor([0.0256, 0.0288, 0.0264, 0.0296, 0.0280, 0.0238, 0.0307, 0.0226],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-23 02:23:50,729 WARNING [train.py:1060] (2/4) Exclude cut with ID 5653-46179-0060-117930-0_sp0.9 from training. Duration: 21.17225
2022-12-23 02:24:00,139 INFO [train.py:894] (2/4) Epoch 11, batch 2800, loss[loss=0.2892, simple_loss=0.3409, pruned_loss=0.1187, over 18617.00 frames. ], tot_loss[loss=0.2177, simple_loss=0.2912, pruned_loss=0.07211, over 3714596.05 frames. ], batch size: 185, lr: 1.04e-02, grad_scale: 8.0
2022-12-23 02:24:09,152 WARNING [train.py:1060] (2/4) Exclude cut with ID 7255-291500-0008-134307-0_sp0.9 from training. Duration: 24.6555625
2022-12-23 02:25:08,478 WARNING [train.py:1060] (2/4) Exclude cut with ID 3488-65654-0031-41259-0 from training. Duration: 20.44
2022-12-23 02:25:16,329 INFO [train.py:894] (2/4) Epoch 11, batch 2850, loss[loss=0.2047, simple_loss=0.2871, pruned_loss=0.06113, over 18465.00 frames. ], tot_loss[loss=0.2185, simple_loss=0.2919, pruned_loss=0.07253, over 3714290.87 frames. ], batch size: 54, lr: 1.04e-02, grad_scale: 8.0
2022-12-23 02:25:25,233 WARNING [train.py:1060] (2/4) Exclude cut with ID 6951-79737-0018-132285-0_sp0.9 from training. Duration: 23.45
2022-12-23 02:25:33,208 INFO [zipformer.py:660] (2/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=37924.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 02:25:37,330 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 3.605e+02 5.185e+02 6.329e+02 8.068e+02 1.822e+03, threshold=1.266e+03, percent-clipped=5.0
2022-12-23 02:25:56,717 WARNING [train.py:1060] (2/4) Exclude cut with ID 6945-60535-0076-12784-0_sp0.9 from training. Duration: 20.52225
2022-12-23 02:26:03,903 WARNING [train.py:1060] (2/4) Exclude cut with ID 7255-291500-0008-134307-0 from training. Duration: 22.19
2022-12-23 02:26:14,407 WARNING [train.py:1060] (2/4) Exclude cut with ID 7357-94126-0014-15845-0_sp1.1 from training. Duration: 25.3818125
2022-12-23 02:26:32,283 INFO [train.py:894] (2/4) Epoch 11, batch 2900, loss[loss=0.207, simple_loss=0.2793, pruned_loss=0.06735, over 18709.00 frames. ], tot_loss[loss=0.2176, simple_loss=0.2914, pruned_loss=0.07189, over 3713786.49 frames. ], batch size: 50, lr: 1.04e-02, grad_scale: 8.0
2022-12-23 02:26:32,357 WARNING [train.py:1060] (2/4) Exclude cut with ID 6951-79737-0043-132310-0_sp0.9 from training. Duration: 28.0944375
2022-12-23 02:26:38,753 WARNING [train.py:1060] (2/4) Exclude cut with ID 2195-150901-0045-59933-0_sp0.9 from training. Duration: 22.9444375
2022-12-23 02:26:46,218 WARNING [train.py:1060] (2/4) Exclude cut with ID 8291-282929-0024-13011-0_sp1.1 from training. Duration: 21.6318125
2022-12-23 02:26:46,367 INFO [zipformer.py:660] (2/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=37972.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 02:26:57,682 INFO [zipformer.py:660] (2/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=37979.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 02:27:01,944 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.3069, 0.9599, 1.2969, 2.1100, 1.4897, 2.2867, 0.6369, 1.5207],
       device='cuda:2'), covar=tensor([0.1831, 0.1998, 0.1470, 0.0828, 0.1381, 0.0921, 0.2153, 0.1293],
       device='cuda:2'), in_proj_covar=tensor([0.0104, 0.0116, 0.0129, 0.0130, 0.0105, 0.0134, 0.0129, 0.0110],
       device='cuda:2'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0004, 0.0003, 0.0003],
       device='cuda:2')
2022-12-23 02:27:06,592 WARNING [train.py:1060] (2/4) Exclude cut with ID 8631-249866-0030-130156-0 from training. Duration: 23.695
2022-12-23 02:27:34,743 WARNING [train.py:1060] (2/4) Exclude cut with ID 7699-105389-0094-26379-0 from training. Duration: 23.955
2022-12-23 02:27:50,585 INFO [train.py:894] (2/4) Epoch 11, batch 2950, loss[loss=0.2035, simple_loss=0.277, pruned_loss=0.06495, over 18477.00 frames. ], tot_loss[loss=0.2175, simple_loss=0.2914, pruned_loss=0.07174, over 3713970.87 frames. ], batch size: 50, lr: 1.04e-02, grad_scale: 8.0
2022-12-23 02:28:01,794 INFO [zipformer.py:660] (2/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=38020.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 02:28:03,321 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([2.7956, 2.5345, 2.0150, 0.9476, 2.1843, 2.0707, 1.8788, 2.0795],
       device='cuda:2'), covar=tensor([0.0658, 0.0486, 0.1170, 0.1687, 0.1170, 0.1465, 0.1390, 0.0801],
       device='cuda:2'), in_proj_covar=tensor([0.0168, 0.0182, 0.0202, 0.0195, 0.0206, 0.0192, 0.0207, 0.0195],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-23 02:28:08,879 WARNING [train.py:1060] (2/4) Exclude cut with ID 8291-282929-0024-13011-0_sp0.9 from training. Duration: 26.438875
2022-12-23 02:28:12,091 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 3.511e+02 5.339e+02 6.601e+02 8.126e+02 2.369e+03, threshold=1.320e+03, percent-clipped=4.0
2022-12-23 02:28:32,322 INFO [zipformer.py:660] (2/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=38040.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 02:28:41,005 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.8962, 1.2183, 0.8231, 1.2015, 2.1293, 1.1855, 1.6208, 1.8140],
       device='cuda:2'), covar=tensor([0.1620, 0.2414, 0.2752, 0.1829, 0.1839, 0.1947, 0.1543, 0.1656],
       device='cuda:2'), in_proj_covar=tensor([0.0091, 0.0102, 0.0120, 0.0097, 0.0115, 0.0091, 0.0098, 0.0096],
       device='cuda:2'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003],
       device='cuda:2')
2022-12-23 02:28:47,446 INFO [scaling.py:679] (2/4) Whitening: num_groups=1, num_channels=384, metric=4.05 vs. limit=5.0
2022-12-23 02:28:51,563 WARNING [train.py:1060] (2/4) Exclude cut with ID 7699-105389-0021-26306-0_sp0.9 from training. Duration: 21.2444375
2022-12-23 02:28:52,968 WARNING [train.py:1060] (2/4) Exclude cut with ID 7357-94126-0014-15845-0_sp0.9 from training. Duration: 31.02225
2022-12-23 02:29:02,026 WARNING [train.py:1060] (2/4) Exclude cut with ID 432-122774-0017-62487-0 from training. Duration: 22.395
2022-12-23 02:29:06,269 INFO [train.py:894] (2/4) Epoch 11, batch 3000, loss[loss=0.2179, simple_loss=0.2979, pruned_loss=0.06894, over 18482.00 frames. ], tot_loss[loss=0.2184, simple_loss=0.2918, pruned_loss=0.07253, over 3714734.88 frames. ], batch size: 64, lr: 1.04e-02, grad_scale: 8.0
2022-12-23 02:29:06,270 INFO [train.py:919] (2/4) Computing validation loss
2022-12-23 02:29:17,972 INFO [train.py:928] (2/4) Epoch 11, validation: loss=0.1727, simple_loss=0.2723, pruned_loss=0.03652, over 944034.00 frames. 
2022-12-23 02:29:17,972 INFO [train.py:929] (2/4) Maximum memory allocated so far is 24680MB
2022-12-23 02:29:25,796 INFO [zipformer.py:660] (2/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=38068.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 02:29:28,952 WARNING [train.py:1060] (2/4) Exclude cut with ID 7357-94126-0045-15876-0 from training. Duration: 21.075
2022-12-23 02:29:29,789 INFO [scaling.py:679] (2/4) Whitening: num_groups=8, num_channels=96, metric=1.32 vs. limit=2.0
2022-12-23 02:29:33,137 WARNING [train.py:1060] (2/4) Exclude cut with ID 6482-98857-0025-147532-0_sp0.9 from training. Duration: 20.0055625
2022-12-23 02:29:33,151 WARNING [train.py:1060] (2/4) Exclude cut with ID 6951-79737-0037-132304-0_sp0.9 from training. Duration: 22.05
2022-12-23 02:29:34,437 WARNING [train.py:1060] (2/4) Exclude cut with ID 7255-291500-0003-134302-0 from training. Duration: 26.8349375
2022-12-23 02:29:37,608 WARNING [train.py:1060] (2/4) Exclude cut with ID 7357-94126-0009-15840-0_sp1.1 from training. Duration: 22.1090625
2022-12-23 02:29:43,609 WARNING [train.py:1060] (2/4) Exclude cut with ID 7699-105389-0094-26379-0_sp0.9 from training. Duration: 26.6166875
2022-12-23 02:30:02,826 WARNING [train.py:1060] (2/4) Exclude cut with ID 2046-178027-0000-53705-0_sp0.9 from training. Duration: 20.3055625
2022-12-23 02:30:23,336 WARNING [train.py:1060] (2/4) Exclude cut with ID 7205-50138-0008-5373-0_sp0.9 from training. Duration: 20.7
2022-12-23 02:30:34,196 INFO [train.py:894] (2/4) Epoch 11, batch 3050, loss[loss=0.1992, simple_loss=0.2745, pruned_loss=0.06193, over 18442.00 frames. ], tot_loss[loss=0.2188, simple_loss=0.2918, pruned_loss=0.0729, over 3714986.38 frames. ], batch size: 48, lr: 1.04e-02, grad_scale: 8.0
2022-12-23 02:30:55,964 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 3.049e+02 5.489e+02 6.697e+02 9.055e+02 1.991e+03, threshold=1.339e+03, percent-clipped=3.0
2022-12-23 02:31:05,494 WARNING [train.py:1060] (2/4) Exclude cut with ID 6978-92210-0019-146985-0 from training. Duration: 22.48
2022-12-23 02:31:20,768 WARNING [train.py:1060] (2/4) Exclude cut with ID 7255-291500-0003-134302-0_sp0.9 from training. Duration: 29.816625
2022-12-23 02:31:41,871 WARNING [train.py:1060] (2/4) Exclude cut with ID 7255-291500-0005-134304-0_sp1.1 from training. Duration: 22.7590625
2022-12-23 02:31:46,183 WARNING [train.py:1060] (2/4) Exclude cut with ID 7357-94126-0026-15857-0 from training. Duration: 22.555
2022-12-23 02:31:49,461 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([3.1858, 2.4429, 1.8713, 3.1289, 2.2947, 2.1897, 2.5243, 3.6291],
       device='cuda:2'), covar=tensor([0.1802, 0.2788, 0.1666, 0.2656, 0.3229, 0.1022, 0.2558, 0.0544],
       device='cuda:2'), in_proj_covar=tensor([0.0286, 0.0275, 0.0233, 0.0353, 0.0256, 0.0219, 0.0271, 0.0195],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0003, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-23 02:31:50,332 INFO [train.py:894] (2/4) Epoch 11, batch 3100, loss[loss=0.2045, simple_loss=0.2855, pruned_loss=0.06179, over 18451.00 frames. ], tot_loss[loss=0.2171, simple_loss=0.2904, pruned_loss=0.07192, over 3714115.08 frames. ], batch size: 50, lr: 1.04e-02, grad_scale: 8.0
2022-12-23 02:31:58,899 INFO [scaling.py:679] (2/4) Whitening: num_groups=8, num_channels=96, metric=1.19 vs. limit=2.0
2022-12-23 02:32:07,427 WARNING [train.py:1060] (2/4) Exclude cut with ID 1250-135782-0005-25975-0_sp0.9 from training. Duration: 21.688875
2022-12-23 02:32:07,993 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([2.5178, 2.0362, 1.5673, 2.3201, 1.8784, 1.8959, 1.9858, 2.5421],
       device='cuda:2'), covar=tensor([0.1855, 0.2927, 0.1669, 0.2625, 0.3061, 0.0972, 0.2588, 0.0716],
       device='cuda:2'), in_proj_covar=tensor([0.0283, 0.0272, 0.0231, 0.0349, 0.0253, 0.0218, 0.0269, 0.0193],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0003, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-23 02:32:41,669 WARNING [train.py:1060] (2/4) Exclude cut with ID 3488-85273-0038-41224-0_sp0.9 from training. Duration: 22.6
2022-12-23 02:33:07,456 INFO [train.py:894] (2/4) Epoch 11, batch 3150, loss[loss=0.2182, simple_loss=0.2961, pruned_loss=0.07022, over 18629.00 frames. ], tot_loss[loss=0.2156, simple_loss=0.2891, pruned_loss=0.07101, over 3714766.30 frames. ], batch size: 53, lr: 1.04e-02, grad_scale: 8.0
2022-12-23 02:33:19,408 WARNING [train.py:1060] (2/4) Exclude cut with ID 7357-94126-0009-15840-0 from training. Duration: 24.32
2022-12-23 02:33:21,284 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([2.1731, 2.3636, 1.7396, 1.2002, 3.0064, 2.6607, 1.7495, 1.6406],
       device='cuda:2'), covar=tensor([0.0431, 0.0346, 0.0568, 0.0763, 0.0213, 0.0304, 0.0596, 0.0868],
       device='cuda:2'), in_proj_covar=tensor([0.0121, 0.0119, 0.0127, 0.0120, 0.0091, 0.0117, 0.0136, 0.0150],
       device='cuda:2'), out_proj_covar=tensor([0.0001, 0.0001, 0.0002, 0.0001, 0.0001, 0.0001, 0.0002, 0.0002],
       device='cuda:2')
2022-12-23 02:33:28,385 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 2.849e+02 5.249e+02 6.448e+02 8.290e+02 2.318e+03, threshold=1.290e+03, percent-clipped=4.0
2022-12-23 02:34:12,481 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([2.2733, 1.8373, 2.0571, 1.7012, 2.3445, 2.9766, 2.8533, 2.0761],
       device='cuda:2'), covar=tensor([0.0485, 0.0373, 0.0375, 0.0286, 0.0250, 0.0331, 0.0281, 0.0305],
       device='cuda:2'), in_proj_covar=tensor([0.0087, 0.0117, 0.0142, 0.0122, 0.0111, 0.0110, 0.0091, 0.0120],
       device='cuda:2'), out_proj_covar=tensor([7.3170e-05, 9.6602e-05, 1.2339e-04, 1.0196e-04, 9.4518e-05, 8.8095e-05,
        7.4949e-05, 9.9896e-05], device='cuda:2')
2022-12-23 02:34:13,855 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.6989, 1.0207, 1.6231, 2.7663, 2.0209, 2.4079, 0.6229, 1.9524],
       device='cuda:2'), covar=tensor([0.1678, 0.1966, 0.1570, 0.0678, 0.1181, 0.1057, 0.2479, 0.1172],
       device='cuda:2'), in_proj_covar=tensor([0.0104, 0.0115, 0.0129, 0.0131, 0.0105, 0.0135, 0.0129, 0.0108],
       device='cuda:2'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0004, 0.0003, 0.0003],
       device='cuda:2')
2022-12-23 02:34:16,607 WARNING [train.py:1060] (2/4) Exclude cut with ID 8291-276745-0093-13116-0_sp0.9 from training. Duration: 21.061125
2022-12-23 02:34:25,190 INFO [train.py:894] (2/4) Epoch 11, batch 3200, loss[loss=0.1896, simple_loss=0.2721, pruned_loss=0.05357, over 18564.00 frames. ], tot_loss[loss=0.2136, simple_loss=0.2876, pruned_loss=0.06982, over 3714850.64 frames. ], batch size: 51, lr: 1.03e-02, grad_scale: 8.0
2022-12-23 02:34:33,987 WARNING [train.py:1060] (2/4) Exclude cut with ID 7357-94126-0024-15855-0_sp0.9 from training. Duration: 20.32225
2022-12-23 02:34:44,726 WARNING [train.py:1060] (2/4) Exclude cut with ID 3033-130750-0096-55598-0_sp1.1 from training. Duration: 0.7545625
2022-12-23 02:35:01,804 WARNING [train.py:1060] (2/4) Exclude cut with ID 4295-39940-0007-92567-0_sp0.9 from training. Duration: 23.9333125
2022-12-23 02:35:22,253 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([2.5607, 2.7834, 1.7878, 2.9539, 2.7486, 2.3528, 4.0286, 2.5688],
       device='cuda:2'), covar=tensor([0.0747, 0.1495, 0.2448, 0.1910, 0.1520, 0.0776, 0.0791, 0.1012],
       device='cuda:2'), in_proj_covar=tensor([0.0175, 0.0194, 0.0237, 0.0283, 0.0228, 0.0183, 0.0205, 0.0196],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-23 02:35:34,596 WARNING [train.py:1060] (2/4) Exclude cut with ID 7255-291500-0008-134307-0_sp1.1 from training. Duration: 20.17275
2022-12-23 02:35:39,695 WARNING [train.py:1060] (2/4) Exclude cut with ID 6978-92210-0019-146985-0_sp1.1 from training. Duration: 20.436375
2022-12-23 02:35:41,180 INFO [train.py:894] (2/4) Epoch 11, batch 3250, loss[loss=0.2018, simple_loss=0.2712, pruned_loss=0.06615, over 18395.00 frames. ], tot_loss[loss=0.2124, simple_loss=0.2868, pruned_loss=0.06898, over 3714543.10 frames. ], batch size: 46, lr: 1.03e-02, grad_scale: 8.0
2022-12-23 02:36:02,550 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 3.401e+02 5.639e+02 6.485e+02 8.508e+02 1.298e+03, threshold=1.297e+03, percent-clipped=2.0
2022-12-23 02:36:16,071 INFO [zipformer.py:660] (2/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=38335.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 02:36:53,596 INFO [zipformer.py:660] (2/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=38360.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 02:36:57,181 INFO [train.py:894] (2/4) Epoch 11, batch 3300, loss[loss=0.1741, simple_loss=0.2461, pruned_loss=0.05106, over 18587.00 frames. ], tot_loss[loss=0.2132, simple_loss=0.2876, pruned_loss=0.06942, over 3713934.55 frames. ], batch size: 45, lr: 1.03e-02, grad_scale: 8.0
2022-12-23 02:36:59,824 WARNING [train.py:1060] (2/4) Exclude cut with ID 4234-40345-0022-142709-0_sp0.9 from training. Duration: 23.1055625
2022-12-23 02:37:02,592 WARNING [train.py:1060] (2/4) Exclude cut with ID 8291-282929-0007-12994-0_sp1.1 from training. Duration: 23.5
2022-12-23 02:37:12,973 WARNING [train.py:1060] (2/4) Exclude cut with ID 7255-291500-0009-134308-0_sp0.9 from training. Duration: 26.62775
2022-12-23 02:37:26,283 WARNING [train.py:1060] (2/4) Exclude cut with ID 6951-79737-0018-132285-0 from training. Duration: 21.105
2022-12-23 02:37:30,560 WARNING [train.py:1060] (2/4) Exclude cut with ID 4511-76322-0006-80011-0_sp0.9 from training. Duration: 24.411125
2022-12-23 02:37:35,586 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([0.1032, 1.1368, 1.2151, 0.3385, 0.5564, 1.2816, 1.2297, 1.1779],
       device='cuda:2'), covar=tensor([0.0533, 0.0241, 0.0276, 0.0332, 0.0397, 0.0372, 0.0214, 0.0465],
       device='cuda:2'), in_proj_covar=tensor([0.0136, 0.0157, 0.0115, 0.0132, 0.0138, 0.0129, 0.0146, 0.0148],
       device='cuda:2'), out_proj_covar=tensor([1.1140e-04, 1.2901e-04, 9.3064e-05, 1.0550e-04, 1.1141e-04, 1.0731e-04,
        1.2199e-04, 1.2145e-04], device='cuda:2')
2022-12-23 02:37:55,857 WARNING [train.py:1060] (2/4) Exclude cut with ID 6758-72288-0033-108368-0_sp1.1 from training. Duration: 21.263625
2022-12-23 02:38:13,626 INFO [train.py:894] (2/4) Epoch 11, batch 3350, loss[loss=0.2139, simple_loss=0.2829, pruned_loss=0.07249, over 18420.00 frames. ], tot_loss[loss=0.2133, simple_loss=0.2875, pruned_loss=0.06959, over 3713013.31 frames. ], batch size: 48, lr: 1.03e-02, grad_scale: 8.0
2022-12-23 02:38:26,261 INFO [zipformer.py:660] (2/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=38421.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 02:38:30,620 WARNING [train.py:1060] (2/4) Exclude cut with ID 4234-40345-0022-142709-0 from training. Duration: 20.795
2022-12-23 02:38:35,720 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 3.061e+02 5.055e+02 5.996e+02 7.502e+02 1.899e+03, threshold=1.199e+03, percent-clipped=4.0
2022-12-23 02:38:41,808 WARNING [train.py:1060] (2/4) Exclude cut with ID 7357-94126-0021-15852-0 from training. Duration: 24.76
2022-12-23 02:38:43,155 WARNING [train.py:1060] (2/4) Exclude cut with ID 3867-173237-0077-144769-0_sp0.9 from training. Duration: 22.25
2022-12-23 02:38:45,178 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([4.1587, 3.5965, 3.5851, 4.0632, 3.6962, 3.6885, 4.2739, 1.1888],
       device='cuda:2'), covar=tensor([0.0802, 0.0685, 0.0672, 0.0805, 0.1614, 0.1127, 0.0594, 0.5192],
       device='cuda:2'), in_proj_covar=tensor([0.0312, 0.0207, 0.0214, 0.0226, 0.0298, 0.0245, 0.0255, 0.0266],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0001, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-23 02:39:08,252 WARNING [train.py:1060] (2/4) Exclude cut with ID 7357-94126-0026-15857-0_sp1.1 from training. Duration: 20.5045625
2022-12-23 02:39:11,691 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([2.2610, 1.8394, 1.7412, 1.5411, 2.1460, 2.7445, 2.6514, 1.6987],
       device='cuda:2'), covar=tensor([0.0411, 0.0397, 0.0451, 0.0323, 0.0290, 0.0287, 0.0319, 0.0388],
       device='cuda:2'), in_proj_covar=tensor([0.0088, 0.0118, 0.0145, 0.0123, 0.0112, 0.0112, 0.0092, 0.0122],
       device='cuda:2'), out_proj_covar=tensor([7.3684e-05, 9.7781e-05, 1.2587e-04, 1.0321e-04, 9.5662e-05, 8.9729e-05,
        7.6063e-05, 1.0105e-04], device='cuda:2')
2022-12-23 02:39:18,765 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.5600, 1.6587, 1.6926, 1.6349, 1.1671, 3.8726, 1.7019, 2.2080],
       device='cuda:2'), covar=tensor([0.3321, 0.1939, 0.1907, 0.2007, 0.1505, 0.0165, 0.1540, 0.0852],
       device='cuda:2'), in_proj_covar=tensor([0.0139, 0.0117, 0.0128, 0.0120, 0.0104, 0.0102, 0.0097, 0.0092],
       device='cuda:2'), out_proj_covar=tensor([0.0005, 0.0005, 0.0005, 0.0004, 0.0004, 0.0003, 0.0004, 0.0004],
       device='cuda:2')
2022-12-23 02:39:28,868 INFO [train.py:894] (2/4) Epoch 11, batch 3400, loss[loss=0.2451, simple_loss=0.3169, pruned_loss=0.08667, over 18684.00 frames. ], tot_loss[loss=0.213, simple_loss=0.2876, pruned_loss=0.06925, over 3712920.01 frames. ], batch size: 97, lr: 1.03e-02, grad_scale: 8.0
2022-12-23 02:40:42,364 INFO [train.py:894] (2/4) Epoch 11, batch 3450, loss[loss=0.1864, simple_loss=0.275, pruned_loss=0.04887, over 18588.00 frames. ], tot_loss[loss=0.2141, simple_loss=0.2882, pruned_loss=0.07002, over 3713578.77 frames. ], batch size: 77, lr: 1.03e-02, grad_scale: 8.0
2022-12-23 02:40:43,185 INFO [scaling.py:679] (2/4) Whitening: num_groups=8, num_channels=96, metric=1.26 vs. limit=2.0
2022-12-23 02:41:02,164 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 3.214e+02 4.989e+02 6.610e+02 8.130e+02 1.422e+03, threshold=1.322e+03, percent-clipped=2.0
2022-12-23 02:41:27,416 INFO [scaling.py:679] (2/4) Whitening: num_groups=8, num_channels=96, metric=1.26 vs. limit=2.0
2022-12-23 02:41:55,276 INFO [train.py:894] (2/4) Epoch 11, batch 3500, loss[loss=0.27, simple_loss=0.327, pruned_loss=0.1065, over 18678.00 frames. ], tot_loss[loss=0.2161, simple_loss=0.2897, pruned_loss=0.07122, over 3714858.25 frames. ], batch size: 184, lr: 1.03e-02, grad_scale: 8.0
2022-12-23 02:42:16,643 WARNING [train.py:1060] (2/4) Exclude cut with ID 7859-102521-0017-7548-0_sp1.1 from training. Duration: 22.2954375
2022-12-23 02:42:26,115 INFO [train.py:894] (2/4) Epoch 12, batch 0, loss[loss=0.1876, simple_loss=0.2706, pruned_loss=0.05232, over 18459.00 frames. ], tot_loss[loss=0.1876, simple_loss=0.2706, pruned_loss=0.05232, over 18459.00 frames. ], batch size: 50, lr: 9.87e-03, grad_scale: 8.0
2022-12-23 02:42:26,116 INFO [train.py:919] (2/4) Computing validation loss
2022-12-23 02:42:32,538 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.6640, 1.5960, 1.7015, 1.7461, 1.2462, 3.5122, 1.7018, 2.1430],
       device='cuda:2'), covar=tensor([0.3744, 0.2396, 0.2047, 0.2154, 0.1571, 0.0234, 0.1679, 0.0863],
       device='cuda:2'), in_proj_covar=tensor([0.0138, 0.0118, 0.0127, 0.0120, 0.0104, 0.0101, 0.0097, 0.0092],
       device='cuda:2'), out_proj_covar=tensor([0.0005, 0.0005, 0.0005, 0.0004, 0.0004, 0.0003, 0.0004, 0.0004],
       device='cuda:2')
2022-12-23 02:42:36,115 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([3.7184, 3.3223, 3.1701, 3.6340, 3.2598, 3.2787, 3.8078, 1.3033],
       device='cuda:2'), covar=tensor([0.0863, 0.0696, 0.0877, 0.0844, 0.1670, 0.1176, 0.0561, 0.5335],
       device='cuda:2'), in_proj_covar=tensor([0.0305, 0.0202, 0.0209, 0.0221, 0.0291, 0.0242, 0.0251, 0.0260],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0001, 0.0001, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-23 02:42:36,864 INFO [train.py:928] (2/4) Epoch 12, validation: loss=0.1736, simple_loss=0.273, pruned_loss=0.03713, over 944034.00 frames. 
2022-12-23 02:42:36,864 INFO [train.py:929] (2/4) Maximum memory allocated so far is 24680MB
2022-12-23 02:43:26,943 WARNING [train.py:1060] (2/4) Exclude cut with ID 298-126791-0067-24026-0_sp0.9 from training. Duration: 21.438875
2022-12-23 02:43:31,231 WARNING [train.py:1060] (2/4) Exclude cut with ID 5652-39938-0025-23684-0_sp0.9 from training. Duration: 22.2055625
2022-12-23 02:43:54,088 INFO [train.py:894] (2/4) Epoch 12, batch 50, loss[loss=0.2392, simple_loss=0.3238, pruned_loss=0.07728, over 18504.00 frames. ], tot_loss[loss=0.2025, simple_loss=0.2861, pruned_loss=0.05945, over 838080.00 frames. ], batch size: 58, lr: 9.87e-03, grad_scale: 8.0
2022-12-23 02:44:05,893 INFO [scaling.py:679] (2/4) Whitening: num_groups=1, num_channels=384, metric=3.91 vs. limit=5.0
2022-12-23 02:44:06,424 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 2.798e+02 4.356e+02 5.737e+02 7.037e+02 2.274e+03, threshold=1.147e+03, percent-clipped=3.0
2022-12-23 02:44:11,618 INFO [scaling.py:679] (2/4) Whitening: num_groups=8, num_channels=96, metric=1.26 vs. limit=2.0
2022-12-23 02:44:18,574 INFO [zipformer.py:660] (2/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=38635.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 02:44:23,423 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([4.4539, 3.8248, 3.8634, 4.3218, 4.0663, 4.0096, 4.5519, 1.6602],
       device='cuda:2'), covar=tensor([0.0681, 0.0630, 0.0592, 0.0733, 0.1387, 0.1045, 0.0570, 0.4646],
       device='cuda:2'), in_proj_covar=tensor([0.0303, 0.0202, 0.0208, 0.0221, 0.0289, 0.0241, 0.0250, 0.0261],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0001, 0.0001, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-23 02:45:09,835 INFO [train.py:894] (2/4) Epoch 12, batch 100, loss[loss=0.2016, simple_loss=0.2867, pruned_loss=0.05825, over 18719.00 frames. ], tot_loss[loss=0.1992, simple_loss=0.2834, pruned_loss=0.05754, over 1475815.07 frames. ], batch size: 60, lr: 9.86e-03, grad_scale: 8.0
2022-12-23 02:45:31,044 INFO [zipformer.py:660] (2/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=38683.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 02:46:20,171 INFO [zipformer.py:660] (2/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=38716.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 02:46:24,467 INFO [train.py:894] (2/4) Epoch 12, batch 150, loss[loss=0.1794, simple_loss=0.257, pruned_loss=0.05092, over 18544.00 frames. ], tot_loss[loss=0.1955, simple_loss=0.2801, pruned_loss=0.05549, over 1971619.98 frames. ], batch size: 47, lr: 9.85e-03, grad_scale: 8.0
2022-12-23 02:46:31,302 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.8784, 1.8314, 1.9056, 2.2590, 2.1911, 4.6362, 1.9912, 1.8809],
       device='cuda:2'), covar=tensor([0.0860, 0.1667, 0.0983, 0.0918, 0.1260, 0.0148, 0.1245, 0.1549],
       device='cuda:2'), in_proj_covar=tensor([0.0073, 0.0081, 0.0074, 0.0074, 0.0090, 0.0071, 0.0085, 0.0077],
       device='cuda:2'), out_proj_covar=tensor([0.0003, 0.0004, 0.0003, 0.0003, 0.0004, 0.0003, 0.0004, 0.0004],
       device='cuda:2')
2022-12-23 02:46:32,271 WARNING [train.py:1060] (2/4) Exclude cut with ID 7859-102521-0017-7548-0 from training. Duration: 24.525
2022-12-23 02:46:36,582 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 2.508e+02 3.672e+02 4.513e+02 6.038e+02 1.379e+03, threshold=9.027e+02, percent-clipped=2.0
2022-12-23 02:47:03,517 WARNING [train.py:1060] (2/4) Exclude cut with ID 3699-47246-0007-3408-0_sp0.9 from training. Duration: 20.26675
2022-12-23 02:47:06,889 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.1415, 1.1897, 0.6047, 1.1829, 1.4941, 2.5001, 1.1912, 1.3769],
       device='cuda:2'), covar=tensor([0.0967, 0.1868, 0.1275, 0.1003, 0.1476, 0.0290, 0.1500, 0.1672],
       device='cuda:2'), in_proj_covar=tensor([0.0073, 0.0081, 0.0074, 0.0074, 0.0090, 0.0071, 0.0085, 0.0076],
       device='cuda:2'), out_proj_covar=tensor([0.0003, 0.0004, 0.0003, 0.0003, 0.0004, 0.0003, 0.0004, 0.0004],
       device='cuda:2')
2022-12-23 02:47:18,680 WARNING [train.py:1060] (2/4) Exclude cut with ID 7859-102521-0017-7548-0_sp0.9 from training. Duration: 27.25
2022-12-23 02:47:39,871 INFO [train.py:894] (2/4) Epoch 12, batch 200, loss[loss=0.1715, simple_loss=0.2533, pruned_loss=0.04481, over 18548.00 frames. ], tot_loss[loss=0.1944, simple_loss=0.2785, pruned_loss=0.0552, over 2358465.07 frames. ], batch size: 44, lr: 9.85e-03, grad_scale: 8.0
2022-12-23 02:48:34,277 WARNING [train.py:1060] (2/4) Exclude cut with ID 6426-64292-0017-15984-0 from training. Duration: 21.68
2022-12-23 02:48:36,406 INFO [scaling.py:679] (2/4) Whitening: num_groups=8, num_channels=96, metric=1.21 vs. limit=2.0
2022-12-23 02:48:44,318 WARNING [train.py:1060] (2/4) Exclude cut with ID 4278-13270-0007-59342-0 from training. Duration: 21.6300625
2022-12-23 02:48:45,337 INFO [scaling.py:679] (2/4) Whitening: num_groups=8, num_channels=192, metric=1.92 vs. limit=2.0
2022-12-23 02:48:55,432 INFO [train.py:894] (2/4) Epoch 12, batch 250, loss[loss=0.1992, simple_loss=0.287, pruned_loss=0.05572, over 18640.00 frames. ], tot_loss[loss=0.1939, simple_loss=0.2779, pruned_loss=0.05496, over 2659465.55 frames. ], batch size: 53, lr: 9.84e-03, grad_scale: 8.0
2022-12-23 02:49:02,113 INFO [zipformer.py:660] (2/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=38823.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 02:49:07,342 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 2.354e+02 3.798e+02 4.673e+02 6.319e+02 1.410e+03, threshold=9.347e+02, percent-clipped=6.0
2022-12-23 02:49:10,346 WARNING [train.py:1060] (2/4) Exclude cut with ID 4278-13270-0007-59342-0_sp0.9 from training. Duration: 24.033375
2022-12-23 02:49:15,405 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.6079, 1.3127, 1.4299, 1.4259, 1.8598, 1.7406, 1.8360, 1.1440],
       device='cuda:2'), covar=tensor([0.0277, 0.0234, 0.0397, 0.0194, 0.0155, 0.0302, 0.0222, 0.0267],
       device='cuda:2'), in_proj_covar=tensor([0.0086, 0.0116, 0.0143, 0.0121, 0.0110, 0.0110, 0.0092, 0.0119],
       device='cuda:2'), out_proj_covar=tensor([7.1814e-05, 9.6304e-05, 1.2422e-04, 1.0110e-04, 9.3643e-05, 8.7717e-05,
        7.5765e-05, 9.8289e-05], device='cuda:2')
2022-12-23 02:50:09,347 INFO [train.py:894] (2/4) Epoch 12, batch 300, loss[loss=0.1912, simple_loss=0.2808, pruned_loss=0.05083, over 18582.00 frames. ], tot_loss[loss=0.1931, simple_loss=0.277, pruned_loss=0.05461, over 2893389.47 frames. ], batch size: 56, lr: 9.83e-03, grad_scale: 8.0
2022-12-23 02:50:09,409 WARNING [train.py:1060] (2/4) Exclude cut with ID 4278-13270-0009-59344-0 from training. Duration: 22.905
2022-12-23 02:50:09,453 WARNING [train.py:1060] (2/4) Exclude cut with ID 5622-44585-0006-90525-0_sp1.1 from training. Duration: 23.4318125
2022-12-23 02:50:16,237 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.5809, 1.4834, 1.3746, 0.7607, 1.8829, 1.5318, 1.3800, 1.1385],
       device='cuda:2'), covar=tensor([0.0313, 0.0407, 0.0432, 0.0700, 0.0283, 0.0307, 0.0431, 0.0937],
       device='cuda:2'), in_proj_covar=tensor([0.0118, 0.0117, 0.0125, 0.0118, 0.0089, 0.0115, 0.0132, 0.0149],
       device='cuda:2'), out_proj_covar=tensor([0.0001, 0.0001, 0.0002, 0.0001, 0.0001, 0.0001, 0.0002, 0.0002],
       device='cuda:2')
2022-12-23 02:50:32,578 INFO [zipformer.py:660] (2/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=38884.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 02:50:58,276 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([2.6087, 2.1907, 1.8691, 0.8374, 1.8835, 1.8248, 1.5835, 1.9164],
       device='cuda:2'), covar=tensor([0.0599, 0.0558, 0.1221, 0.1628, 0.1242, 0.1640, 0.1617, 0.0858],
       device='cuda:2'), in_proj_covar=tensor([0.0164, 0.0177, 0.0197, 0.0192, 0.0202, 0.0192, 0.0202, 0.0192],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-23 02:51:24,894 INFO [train.py:894] (2/4) Epoch 12, batch 350, loss[loss=0.1764, simple_loss=0.2698, pruned_loss=0.04149, over 18720.00 frames. ], tot_loss[loss=0.1956, simple_loss=0.2791, pruned_loss=0.05603, over 3075549.29 frames. ], batch size: 54, lr: 9.83e-03, grad_scale: 8.0
2022-12-23 02:51:36,876 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 2.329e+02 4.042e+02 5.093e+02 6.152e+02 1.114e+03, threshold=1.019e+03, percent-clipped=2.0
2022-12-23 02:52:06,907 WARNING [train.py:1060] (2/4) Exclude cut with ID 4278-13270-0009-59344-0_sp1.1 from training. Duration: 20.82275
2022-12-23 02:52:08,514 WARNING [train.py:1060] (2/4) Exclude cut with ID 4278-13270-0009-59344-0_sp0.9 from training. Duration: 25.45
2022-12-23 02:52:32,950 INFO [scaling.py:679] (2/4) Whitening: num_groups=8, num_channels=192, metric=1.96 vs. limit=2.0
2022-12-23 02:52:39,572 INFO [train.py:894] (2/4) Epoch 12, batch 400, loss[loss=0.1953, simple_loss=0.2772, pruned_loss=0.05668, over 18565.00 frames. ], tot_loss[loss=0.1964, simple_loss=0.2798, pruned_loss=0.05653, over 3216659.57 frames. ], batch size: 49, lr: 9.82e-03, grad_scale: 8.0
2022-12-23 02:53:08,747 WARNING [train.py:1060] (2/4) Exclude cut with ID 5622-44585-0006-90525-0 from training. Duration: 25.775
2022-12-23 02:53:31,142 WARNING [train.py:1060] (2/4) Exclude cut with ID 2929-85685-0071-62375-0_sp0.9 from training. Duration: 22.25
2022-12-23 02:53:51,610 INFO [zipformer.py:660] (2/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=39016.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 02:53:55,552 INFO [train.py:894] (2/4) Epoch 12, batch 450, loss[loss=0.2248, simple_loss=0.3092, pruned_loss=0.07022, over 18459.00 frames. ], tot_loss[loss=0.1973, simple_loss=0.2807, pruned_loss=0.05692, over 3326367.24 frames. ], batch size: 64, lr: 9.82e-03, grad_scale: 8.0
2022-12-23 02:54:00,438 WARNING [train.py:1060] (2/4) Exclude cut with ID 3972-170212-0014-23379-0 from training. Duration: 26.205
2022-12-23 02:54:02,370 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.6376, 1.3218, 0.8913, 1.2019, 2.0654, 0.9675, 1.4887, 1.6460],
       device='cuda:2'), covar=tensor([0.1649, 0.2014, 0.2180, 0.1551, 0.1606, 0.1714, 0.1318, 0.1598],
       device='cuda:2'), in_proj_covar=tensor([0.0090, 0.0099, 0.0117, 0.0094, 0.0112, 0.0090, 0.0097, 0.0094],
       device='cuda:2'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003],
       device='cuda:2')
2022-12-23 02:54:08,262 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 2.414e+02 3.857e+02 4.585e+02 5.760e+02 1.375e+03, threshold=9.170e+02, percent-clipped=1.0
2022-12-23 02:54:15,661 WARNING [train.py:1060] (2/4) Exclude cut with ID 5239-32139-0047-9341-0_sp0.9 from training. Duration: 30.1555625
2022-12-23 02:54:21,287 WARNING [train.py:1060] (2/4) Exclude cut with ID 1265-135635-0050-6781-0_sp0.9 from training. Duration: 21.8333125
2022-12-23 02:54:30,331 WARNING [train.py:1060] (2/4) Exclude cut with ID 1914-133440-0024-94914-0_sp1.1 from training. Duration: 20.6545625
2022-12-23 02:55:03,933 INFO [zipformer.py:660] (2/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=39064.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 02:55:11,088 INFO [train.py:894] (2/4) Epoch 12, batch 500, loss[loss=0.2183, simple_loss=0.2975, pruned_loss=0.06957, over 18677.00 frames. ], tot_loss[loss=0.1981, simple_loss=0.2819, pruned_loss=0.05719, over 3412761.43 frames. ], batch size: 62, lr: 9.81e-03, grad_scale: 8.0
2022-12-23 02:55:14,556 WARNING [train.py:1060] (2/4) Exclude cut with ID 7395-89880-0045-39920-0_sp0.9 from training. Duration: 20.52225
2022-12-23 02:55:34,298 WARNING [train.py:1060] (2/4) Exclude cut with ID 3972-170212-0014-23379-0_sp0.9 from training. Duration: 29.1166875
2022-12-23 02:56:13,685 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([2.2198, 2.1817, 1.4637, 2.6114, 2.4953, 2.1550, 3.1607, 2.1200],
       device='cuda:2'), covar=tensor([0.0774, 0.1624, 0.2628, 0.1663, 0.1449, 0.0768, 0.0784, 0.1088],
       device='cuda:2'), in_proj_covar=tensor([0.0170, 0.0193, 0.0234, 0.0278, 0.0224, 0.0181, 0.0203, 0.0192],
       device='cuda:2'), out_proj_covar=tensor([0.0001, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-23 02:56:26,649 INFO [train.py:894] (2/4) Epoch 12, batch 550, loss[loss=0.2161, simple_loss=0.3019, pruned_loss=0.06515, over 18580.00 frames. ], tot_loss[loss=0.1991, simple_loss=0.2832, pruned_loss=0.05748, over 3480331.10 frames. ], batch size: 56, lr: 9.80e-03, grad_scale: 8.0
2022-12-23 02:56:32,941 WARNING [train.py:1060] (2/4) Exclude cut with ID 543-133211-0007-59831-0_sp0.9 from training. Duration: 21.388875
2022-12-23 02:56:33,320 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([2.6013, 2.7095, 2.1103, 2.0736, 3.1728, 3.1839, 2.7262, 2.2136],
       device='cuda:2'), covar=tensor([0.0384, 0.0277, 0.0479, 0.0543, 0.0158, 0.0222, 0.0366, 0.0588],
       device='cuda:2'), in_proj_covar=tensor([0.0123, 0.0121, 0.0130, 0.0121, 0.0090, 0.0119, 0.0137, 0.0153],
       device='cuda:2'), out_proj_covar=tensor([0.0001, 0.0001, 0.0002, 0.0001, 0.0001, 0.0001, 0.0002, 0.0002],
       device='cuda:2')
2022-12-23 02:56:39,179 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 2.465e+02 4.214e+02 4.900e+02 6.411e+02 1.265e+03, threshold=9.799e+02, percent-clipped=3.0
2022-12-23 02:57:09,752 WARNING [train.py:1060] (2/4) Exclude cut with ID 1914-133440-0024-94914-0 from training. Duration: 22.72
2022-12-23 02:57:11,182 WARNING [train.py:1060] (2/4) Exclude cut with ID 1914-133440-0031-94921-0_sp0.9 from training. Duration: 22.7444375
2022-12-23 02:57:41,897 INFO [train.py:894] (2/4) Epoch 12, batch 600, loss[loss=0.1541, simple_loss=0.2342, pruned_loss=0.03704, over 18425.00 frames. ], tot_loss[loss=0.1988, simple_loss=0.2826, pruned_loss=0.05748, over 3531658.11 frames. ], batch size: 42, lr: 9.80e-03, grad_scale: 8.0
2022-12-23 02:57:53,063 WARNING [train.py:1060] (2/4) Exclude cut with ID 4133-6541-0027-40495-0_sp1.1 from training. Duration: 0.9681875
2022-12-23 02:57:56,151 WARNING [train.py:1060] (2/4) Exclude cut with ID 6330-62851-0022-91297-0_sp0.9 from training. Duration: 22.3166875
2022-12-23 02:57:57,768 INFO [zipformer.py:660] (2/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=39179.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 02:58:02,082 WARNING [train.py:1060] (2/4) Exclude cut with ID 543-133212-0015-59917-0_sp0.9 from training. Duration: 21.8166875
2022-12-23 02:58:06,608 INFO [zipformer.py:660] (2/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=39185.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 02:58:09,372 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([2.1796, 1.5039, 1.8908, 1.9068, 2.1761, 2.1404, 1.9996, 1.6298],
       device='cuda:2'), covar=tensor([0.1841, 0.2903, 0.2125, 0.2320, 0.1528, 0.0781, 0.2423, 0.1059],
       device='cuda:2'), in_proj_covar=tensor([0.0252, 0.0286, 0.0262, 0.0296, 0.0280, 0.0237, 0.0305, 0.0223],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-23 02:58:57,143 INFO [train.py:894] (2/4) Epoch 12, batch 650, loss[loss=0.1661, simple_loss=0.2481, pruned_loss=0.042, over 18515.00 frames. ], tot_loss[loss=0.1978, simple_loss=0.2812, pruned_loss=0.05719, over 3571524.30 frames. ], batch size: 44, lr: 9.79e-03, grad_scale: 8.0
2022-12-23 02:59:09,407 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 2.588e+02 3.961e+02 5.057e+02 6.119e+02 1.297e+03, threshold=1.011e+03, percent-clipped=1.0
2022-12-23 02:59:39,011 INFO [zipformer.py:660] (2/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=39246.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 02:59:44,323 WARNING [train.py:1060] (2/4) Exclude cut with ID 4957-30119-0041-23990-0_sp0.9 from training. Duration: 20.22775
2022-12-23 02:59:53,431 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([2.3588, 3.8955, 3.7433, 1.5715, 3.8060, 2.9211, 0.5879, 2.4712],
       device='cuda:2'), covar=tensor([0.2318, 0.0872, 0.1404, 0.4133, 0.0819, 0.1058, 0.6143, 0.1815],
       device='cuda:2'), in_proj_covar=tensor([0.0136, 0.0125, 0.0150, 0.0121, 0.0130, 0.0105, 0.0142, 0.0110],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0002, 0.0003, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-23 03:00:13,405 INFO [train.py:894] (2/4) Epoch 12, batch 700, loss[loss=0.1986, simple_loss=0.2911, pruned_loss=0.05308, over 18634.00 frames. ], tot_loss[loss=0.1982, simple_loss=0.2817, pruned_loss=0.0574, over 3603019.68 frames. ], batch size: 53, lr: 9.78e-03, grad_scale: 8.0
2022-12-23 03:00:28,245 WARNING [train.py:1060] (2/4) Exclude cut with ID 5239-32139-0047-9341-0_sp1.1 from training. Duration: 24.67275
2022-12-23 03:00:35,604 INFO [zipformer.py:660] (2/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=39284.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 03:00:54,788 WARNING [train.py:1060] (2/4) Exclude cut with ID 3082-165428-0081-50734-0_sp0.9 from training. Duration: 21.8055625
2022-12-23 03:01:27,310 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.1366, 2.7375, 3.2507, 1.2207, 2.6138, 3.4794, 2.1728, 2.9130],
       device='cuda:2'), covar=tensor([0.0863, 0.0413, 0.0276, 0.0476, 0.0366, 0.0373, 0.0441, 0.0502],
       device='cuda:2'), in_proj_covar=tensor([0.0136, 0.0157, 0.0113, 0.0129, 0.0137, 0.0129, 0.0146, 0.0148],
       device='cuda:2'), out_proj_covar=tensor([1.1047e-04, 1.2885e-04, 9.1057e-05, 1.0267e-04, 1.1031e-04, 1.0704e-04,
        1.2133e-04, 1.2060e-04], device='cuda:2')
2022-12-23 03:01:28,248 INFO [train.py:894] (2/4) Epoch 12, batch 750, loss[loss=0.1848, simple_loss=0.2657, pruned_loss=0.05194, over 18691.00 frames. ], tot_loss[loss=0.1981, simple_loss=0.2815, pruned_loss=0.05739, over 3627622.60 frames. ], batch size: 48, lr: 9.78e-03, grad_scale: 8.0
2022-12-23 03:01:32,588 WARNING [train.py:1060] (2/4) Exclude cut with ID 3340-169293-0054-76830-0_sp0.9 from training. Duration: 22.6666875
2022-12-23 03:01:39,379 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 2.862e+02 4.095e+02 4.892e+02 6.005e+02 1.372e+03, threshold=9.785e+02, percent-clipped=2.0
2022-12-23 03:02:07,082 INFO [zipformer.py:660] (2/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=39345.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 03:02:33,331 WARNING [train.py:1060] (2/4) Exclude cut with ID 2411-132532-0017-82279-0_sp1.1 from training. Duration: 0.9681875
2022-12-23 03:02:43,262 INFO [train.py:894] (2/4) Epoch 12, batch 800, loss[loss=0.1783, simple_loss=0.2587, pruned_loss=0.04891, over 18532.00 frames. ], tot_loss[loss=0.1985, simple_loss=0.2821, pruned_loss=0.05745, over 3647232.85 frames. ], batch size: 47, lr: 9.77e-03, grad_scale: 8.0
2022-12-23 03:02:59,709 WARNING [train.py:1060] (2/4) Exclude cut with ID 6330-62850-0007-91323-0 from training. Duration: 22.485
2022-12-23 03:03:37,040 WARNING [train.py:1060] (2/4) Exclude cut with ID 3972-170212-0014-23379-0_sp1.1 from training. Duration: 23.82275
2022-12-23 03:03:51,563 WARNING [train.py:1060] (2/4) Exclude cut with ID 4860-13185-0032-76709-0 from training. Duration: 20.77
2022-12-23 03:03:58,236 INFO [train.py:894] (2/4) Epoch 12, batch 850, loss[loss=0.2558, simple_loss=0.3187, pruned_loss=0.09645, over 18709.00 frames. ], tot_loss[loss=0.1983, simple_loss=0.2818, pruned_loss=0.05739, over 3660913.51 frames. ], batch size: 52, lr: 9.77e-03, grad_scale: 8.0
2022-12-23 03:03:59,620 WARNING [train.py:1060] (2/4) Exclude cut with ID 6426-64292-0017-15984-0_sp0.9 from training. Duration: 24.088875
2022-12-23 03:04:07,790 INFO [scaling.py:679] (2/4) Whitening: num_groups=8, num_channels=96, metric=1.32 vs. limit=2.0
2022-12-23 03:04:09,524 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 2.336e+02 4.122e+02 4.845e+02 6.172e+02 1.351e+03, threshold=9.691e+02, percent-clipped=1.0
2022-12-23 03:04:18,467 INFO [scaling.py:679] (2/4) Whitening: num_groups=1, num_channels=384, metric=4.76 vs. limit=5.0
2022-12-23 03:04:30,718 WARNING [train.py:1060] (2/4) Exclude cut with ID 6330-62850-0007-91323-0_sp1.1 from training. Duration: 20.4409375
2022-12-23 03:05:04,146 INFO [zipformer.py:660] (2/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=39463.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 03:05:12,888 INFO [train.py:894] (2/4) Epoch 12, batch 900, loss[loss=0.1909, simple_loss=0.2775, pruned_loss=0.05209, over 18561.00 frames. ], tot_loss[loss=0.1984, simple_loss=0.282, pruned_loss=0.05742, over 3673124.81 frames. ], batch size: 49, lr: 9.76e-03, grad_scale: 8.0
2022-12-23 03:05:27,830 INFO [zipformer.py:660] (2/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=39479.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 03:05:35,079 INFO [scaling.py:679] (2/4) Whitening: num_groups=8, num_channels=192, metric=2.04 vs. limit=2.0
2022-12-23 03:05:47,510 WARNING [train.py:1060] (2/4) Exclude cut with ID 2929-85685-0062-62366-0_sp0.9 from training. Duration: 22.511125
2022-12-23 03:05:47,536 WARNING [train.py:1060] (2/4) Exclude cut with ID 7395-89880-0031-39906-0 from training. Duration: 20.675
2022-12-23 03:06:15,929 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.5113, 1.4115, 1.3552, 0.7998, 1.8115, 1.4883, 1.3804, 1.2259],
       device='cuda:2'), covar=tensor([0.0380, 0.0471, 0.0495, 0.0713, 0.0326, 0.0389, 0.0473, 0.0881],
       device='cuda:2'), in_proj_covar=tensor([0.0121, 0.0119, 0.0127, 0.0120, 0.0090, 0.0117, 0.0134, 0.0151],
       device='cuda:2'), out_proj_covar=tensor([0.0001, 0.0001, 0.0002, 0.0001, 0.0001, 0.0001, 0.0002, 0.0002],
       device='cuda:2')
2022-12-23 03:06:30,620 INFO [train.py:894] (2/4) Epoch 12, batch 950, loss[loss=0.2256, simple_loss=0.307, pruned_loss=0.07209, over 18561.00 frames. ], tot_loss[loss=0.1981, simple_loss=0.2818, pruned_loss=0.05723, over 3681964.02 frames. ], batch size: 77, lr: 9.75e-03, grad_scale: 8.0
2022-12-23 03:06:38,390 INFO [zipformer.py:660] (2/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=39524.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 03:06:42,894 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 2.364e+02 4.631e+02 5.225e+02 6.813e+02 1.904e+03, threshold=1.045e+03, percent-clipped=5.0
2022-12-23 03:06:43,081 INFO [zipformer.py:660] (2/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=39527.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 03:06:54,798 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.4523, 1.0570, 1.6735, 2.5535, 1.8860, 2.1928, 0.8415, 1.8117],
       device='cuda:2'), covar=tensor([0.1849, 0.1967, 0.1528, 0.0723, 0.1203, 0.1176, 0.2221, 0.1315],
       device='cuda:2'), in_proj_covar=tensor([0.0101, 0.0114, 0.0130, 0.0131, 0.0104, 0.0134, 0.0128, 0.0107],
       device='cuda:2'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0004, 0.0003, 0.0003],
       device='cuda:2')
2022-12-23 03:06:55,005 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.9519, 1.8913, 1.4061, 1.9432, 2.0977, 1.8663, 2.5521, 2.0348],
       device='cuda:2'), covar=tensor([0.0886, 0.1632, 0.2736, 0.1812, 0.1712, 0.0853, 0.1057, 0.1107],
       device='cuda:2'), in_proj_covar=tensor([0.0173, 0.0195, 0.0236, 0.0281, 0.0224, 0.0182, 0.0206, 0.0195],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-23 03:07:03,399 INFO [zipformer.py:660] (2/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=39541.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 03:07:28,477 WARNING [train.py:1060] (2/4) Exclude cut with ID 6330-62850-0007-91323-0_sp0.9 from training. Duration: 24.9833125
2022-12-23 03:07:42,099 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.7304, 1.3314, 0.5993, 1.1998, 2.1963, 1.3303, 1.6848, 1.7529],
       device='cuda:2'), covar=tensor([0.1667, 0.2157, 0.2847, 0.1745, 0.1711, 0.1707, 0.1539, 0.1725],
       device='cuda:2'), in_proj_covar=tensor([0.0091, 0.0099, 0.0119, 0.0095, 0.0114, 0.0091, 0.0098, 0.0095],
       device='cuda:2'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003],
       device='cuda:2')
2022-12-23 03:07:46,163 INFO [train.py:894] (2/4) Epoch 12, batch 1000, loss[loss=0.2027, simple_loss=0.2896, pruned_loss=0.05787, over 18646.00 frames. ], tot_loss[loss=0.198, simple_loss=0.2817, pruned_loss=0.05714, over 3689318.12 frames. ], batch size: 69, lr: 9.75e-03, grad_scale: 8.0
2022-12-23 03:07:58,907 WARNING [train.py:1060] (2/4) Exclude cut with ID 5239-32139-0047-9341-0 from training. Duration: 27.14
2022-12-23 03:08:14,357 WARNING [train.py:1060] (2/4) Exclude cut with ID 2929-85685-0044-62348-0 from training. Duration: 22.44
2022-12-23 03:08:49,068 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([2.7118, 2.1712, 1.8067, 2.4367, 1.9062, 2.1945, 2.0375, 2.5769],
       device='cuda:2'), covar=tensor([0.1550, 0.2891, 0.1557, 0.2310, 0.3166, 0.0871, 0.2528, 0.0667],
       device='cuda:2'), in_proj_covar=tensor([0.0276, 0.0271, 0.0227, 0.0340, 0.0251, 0.0214, 0.0266, 0.0191],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0003, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-23 03:09:01,597 INFO [train.py:894] (2/4) Epoch 12, batch 1050, loss[loss=0.1933, simple_loss=0.2915, pruned_loss=0.04751, over 18582.00 frames. ], tot_loss[loss=0.1975, simple_loss=0.2812, pruned_loss=0.05691, over 3694068.36 frames. ], batch size: 56, lr: 9.74e-03, grad_scale: 8.0
2022-12-23 03:09:12,770 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([0.2121, 1.3529, 1.6964, 1.0038, 0.9564, 1.7322, 1.6068, 1.5681],
       device='cuda:2'), covar=tensor([0.0677, 0.0352, 0.0271, 0.0284, 0.0417, 0.0362, 0.0231, 0.0502],
       device='cuda:2'), in_proj_covar=tensor([0.0137, 0.0158, 0.0114, 0.0129, 0.0138, 0.0130, 0.0146, 0.0148],
       device='cuda:2'), out_proj_covar=tensor([1.1091e-04, 1.2928e-04, 9.2039e-05, 1.0297e-04, 1.1087e-04, 1.0801e-04,
        1.2142e-04, 1.2059e-04], device='cuda:2')
2022-12-23 03:09:13,699 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 2.640e+02 3.885e+02 4.737e+02 5.602e+02 8.363e+02, threshold=9.474e+02, percent-clipped=0.0
2022-12-23 03:09:31,209 WARNING [train.py:1060] (2/4) Exclude cut with ID 2929-85685-0060-62364-0_sp0.9 from training. Duration: 21.361125
2022-12-23 03:09:31,536 INFO [zipformer.py:660] (2/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=39639.0, num_to_drop=1, layers_to_drop={1}
2022-12-23 03:09:32,776 INFO [zipformer.py:660] (2/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=39640.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 03:09:37,374 WARNING [train.py:1060] (2/4) Exclude cut with ID 2929-85685-0079-62383-0_sp1.1 from training. Duration: 27.0318125
2022-12-23 03:09:45,951 WARNING [train.py:1060] (2/4) Exclude cut with ID 5622-44585-0006-90525-0_sp0.9 from training. Duration: 28.638875
2022-12-23 03:10:03,120 WARNING [train.py:1060] (2/4) Exclude cut with ID 3340-169293-0054-76830-0 from training. Duration: 20.4
2022-12-23 03:10:16,629 INFO [train.py:894] (2/4) Epoch 12, batch 1100, loss[loss=0.2075, simple_loss=0.297, pruned_loss=0.05897, over 18726.00 frames. ], tot_loss[loss=0.198, simple_loss=0.2816, pruned_loss=0.0572, over 3697439.51 frames. ], batch size: 54, lr: 9.74e-03, grad_scale: 8.0
2022-12-23 03:10:20,481 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.8339, 1.2525, 2.2972, 3.0460, 2.3242, 2.4393, 1.2432, 2.1500],
       device='cuda:2'), covar=tensor([0.1599, 0.1831, 0.1280, 0.0613, 0.1019, 0.1440, 0.2062, 0.1100],
       device='cuda:2'), in_proj_covar=tensor([0.0103, 0.0114, 0.0130, 0.0131, 0.0105, 0.0134, 0.0129, 0.0108],
       device='cuda:2'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0004, 0.0003, 0.0003],
       device='cuda:2')
2022-12-23 03:10:36,336 WARNING [train.py:1060] (2/4) Exclude cut with ID 2929-85685-0071-62375-0 from training. Duration: 20.025
2022-12-23 03:10:36,344 WARNING [train.py:1060] (2/4) Exclude cut with ID 2364-131735-0112-64612-0_sp0.9 from training. Duration: 20.488875
2022-12-23 03:10:40,531 WARNING [train.py:1060] (2/4) Exclude cut with ID 2929-85685-0079-62383-0 from training. Duration: 29.735
2022-12-23 03:11:04,415 INFO [zipformer.py:660] (2/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=39700.0, num_to_drop=1, layers_to_drop={3}
2022-12-23 03:11:33,402 INFO [train.py:894] (2/4) Epoch 12, batch 1150, loss[loss=0.1605, simple_loss=0.2447, pruned_loss=0.03817, over 18408.00 frames. ], tot_loss[loss=0.1986, simple_loss=0.2822, pruned_loss=0.05747, over 3700764.50 frames. ], batch size: 42, lr: 9.73e-03, grad_scale: 8.0
2022-12-23 03:11:46,695 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 2.615e+02 4.757e+02 5.682e+02 7.291e+02 2.564e+03, threshold=1.136e+03, percent-clipped=13.0
2022-12-23 03:11:59,925 WARNING [train.py:1060] (2/4) Exclude cut with ID 7276-92427-0014-12983-0_sp0.9 from training. Duration: 21.3055625
2022-12-23 03:12:01,319 WARNING [train.py:1060] (2/4) Exclude cut with ID 1025-75365-0008-79168-0_sp0.9 from training. Duration: 22.0666875
2022-12-23 03:12:46,057 INFO [zipformer.py:660] (2/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=39767.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 03:12:48,821 INFO [train.py:894] (2/4) Epoch 12, batch 1200, loss[loss=0.1799, simple_loss=0.2681, pruned_loss=0.04587, over 18429.00 frames. ], tot_loss[loss=0.198, simple_loss=0.2818, pruned_loss=0.05709, over 3703902.04 frames. ], batch size: 48, lr: 9.72e-03, grad_scale: 8.0
2022-12-23 03:13:46,084 WARNING [train.py:1060] (2/4) Exclude cut with ID 2929-85685-0062-62366-0 from training. Duration: 20.26
2022-12-23 03:13:49,380 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.1136, 2.5006, 3.0695, 1.1192, 2.4965, 3.2844, 2.2828, 2.6507],
       device='cuda:2'), covar=tensor([0.0830, 0.0371, 0.0237, 0.0470, 0.0355, 0.0270, 0.0349, 0.0500],
       device='cuda:2'), in_proj_covar=tensor([0.0137, 0.0158, 0.0114, 0.0130, 0.0138, 0.0130, 0.0146, 0.0148],
       device='cuda:2'), out_proj_covar=tensor([1.1132e-04, 1.2911e-04, 9.2012e-05, 1.0344e-04, 1.1040e-04, 1.0765e-04,
        1.2105e-04, 1.2018e-04], device='cuda:2')
2022-12-23 03:13:57,087 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([2.8003, 2.3614, 1.9053, 0.7938, 1.8629, 2.2298, 1.8613, 1.9878],
       device='cuda:2'), covar=tensor([0.0591, 0.0486, 0.1072, 0.1616, 0.1283, 0.1199, 0.1318, 0.0772],
       device='cuda:2'), in_proj_covar=tensor([0.0164, 0.0176, 0.0196, 0.0193, 0.0202, 0.0187, 0.0202, 0.0191],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-23 03:13:58,553 INFO [zipformer.py:660] (2/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=39815.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 03:14:00,965 WARNING [train.py:1060] (2/4) Exclude cut with ID 5239-32139-0030-9324-0_sp0.9 from training. Duration: 21.3444375
2022-12-23 03:14:03,875 INFO [train.py:894] (2/4) Epoch 12, batch 1250, loss[loss=0.1697, simple_loss=0.2601, pruned_loss=0.03961, over 18688.00 frames. ], tot_loss[loss=0.198, simple_loss=0.2821, pruned_loss=0.05696, over 3706194.48 frames. ], batch size: 48, lr: 9.72e-03, grad_scale: 8.0
2022-12-23 03:14:04,032 INFO [zipformer.py:660] (2/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=39819.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 03:14:16,865 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 2.560e+02 3.885e+02 4.860e+02 6.199e+02 1.184e+03, threshold=9.720e+02, percent-clipped=2.0
2022-12-23 03:14:17,227 INFO [zipformer.py:660] (2/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=39828.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 03:14:36,681 INFO [zipformer.py:660] (2/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=39841.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 03:14:56,762 WARNING [train.py:1060] (2/4) Exclude cut with ID 497-129325-0061-62254-0_sp1.1 from training. Duration: 0.97725
2022-12-23 03:15:19,543 INFO [train.py:894] (2/4) Epoch 12, batch 1300, loss[loss=0.2324, simple_loss=0.3172, pruned_loss=0.07373, over 18596.00 frames. ], tot_loss[loss=0.198, simple_loss=0.2821, pruned_loss=0.05697, over 3708070.21 frames. ], batch size: 98, lr: 9.71e-03, grad_scale: 8.0
2022-12-23 03:15:29,797 INFO [zipformer.py:660] (2/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=39876.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 03:15:40,604 WARNING [train.py:1060] (2/4) Exclude cut with ID 7395-89880-0031-39906-0_sp0.9 from training. Duration: 22.97225
2022-12-23 03:15:49,166 INFO [zipformer.py:660] (2/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=39889.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 03:16:12,687 WARNING [train.py:1060] (2/4) Exclude cut with ID 7395-89880-0047-39922-0_sp0.9 from training. Duration: 21.97775
2022-12-23 03:16:27,543 WARNING [train.py:1060] (2/4) Exclude cut with ID 1112-1043-0006-89194-0_sp0.9 from training. Duration: 21.8333125
2022-12-23 03:16:33,476 INFO [train.py:894] (2/4) Epoch 12, batch 1350, loss[loss=0.1941, simple_loss=0.2803, pruned_loss=0.05389, over 18545.00 frames. ], tot_loss[loss=0.1974, simple_loss=0.2814, pruned_loss=0.05666, over 3708973.43 frames. ], batch size: 55, lr: 9.71e-03, grad_scale: 8.0
2022-12-23 03:16:39,446 WARNING [train.py:1060] (2/4) Exclude cut with ID 1914-133440-0031-94921-0 from training. Duration: 20.47
2022-12-23 03:16:46,694 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 2.753e+02 3.746e+02 4.276e+02 5.136e+02 9.352e+02, threshold=8.553e+02, percent-clipped=0.0
2022-12-23 03:17:04,988 INFO [zipformer.py:660] (2/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=39940.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 03:17:05,291 INFO [scaling.py:679] (2/4) Whitening: num_groups=1, num_channels=384, metric=2.62 vs. limit=5.0
2022-12-23 03:17:44,786 WARNING [train.py:1060] (2/4) Exclude cut with ID 7395-89880-0037-39912-0_sp0.9 from training. Duration: 20.67225
2022-12-23 03:17:47,788 INFO [train.py:894] (2/4) Epoch 12, batch 1400, loss[loss=0.1933, simple_loss=0.2813, pruned_loss=0.05263, over 18541.00 frames. ], tot_loss[loss=0.1977, simple_loss=0.2817, pruned_loss=0.05683, over 3709510.13 frames. ], batch size: 47, lr: 9.70e-03, grad_scale: 8.0
2022-12-23 03:18:03,479 WARNING [train.py:1060] (2/4) Exclude cut with ID 1914-133440-0024-94914-0_sp0.9 from training. Duration: 25.2444375
2022-12-23 03:18:16,683 INFO [zipformer.py:660] (2/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=39988.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 03:18:23,420 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.3575, 1.3366, 1.0000, 1.7652, 1.5888, 2.9471, 1.2152, 1.4210],
       device='cuda:2'), covar=tensor([0.0934, 0.1811, 0.1210, 0.0873, 0.1445, 0.0238, 0.1422, 0.1512],
       device='cuda:2'), in_proj_covar=tensor([0.0073, 0.0082, 0.0074, 0.0074, 0.0091, 0.0071, 0.0086, 0.0076],
       device='cuda:2'), out_proj_covar=tensor([0.0003, 0.0004, 0.0003, 0.0003, 0.0004, 0.0003, 0.0004, 0.0004],
       device='cuda:2')
2022-12-23 03:18:26,045 WARNING [train.py:1060] (2/4) Exclude cut with ID 3340-169293-0021-76797-0_sp0.9 from training. Duration: 21.1445
2022-12-23 03:18:28,271 INFO [zipformer.py:660] (2/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=39995.0, num_to_drop=1, layers_to_drop={3}
2022-12-23 03:18:34,695 INFO [scaling.py:679] (2/4) Whitening: num_groups=8, num_channels=96, metric=1.18 vs. limit=2.0
2022-12-23 03:19:07,131 INFO [train.py:894] (2/4) Epoch 12, batch 1450, loss[loss=0.221, simple_loss=0.3024, pruned_loss=0.06974, over 18657.00 frames. ], tot_loss[loss=0.1966, simple_loss=0.2811, pruned_loss=0.05604, over 3710403.92 frames. ], batch size: 164, lr: 9.69e-03, grad_scale: 8.0
2022-12-23 03:19:20,694 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 2.732e+02 3.983e+02 4.941e+02 6.040e+02 1.228e+03, threshold=9.881e+02, percent-clipped=7.0
2022-12-23 03:19:44,121 WARNING [train.py:1060] (2/4) Exclude cut with ID 2929-85685-0079-62383-0_sp0.9 from training. Duration: 33.038875
2022-12-23 03:20:17,954 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([3.5488, 2.7466, 2.4475, 1.2383, 2.4847, 2.2022, 1.6675, 2.2615],
       device='cuda:2'), covar=tensor([0.0611, 0.0821, 0.1657, 0.1995, 0.1843, 0.1556, 0.1936, 0.1254],
       device='cuda:2'), in_proj_covar=tensor([0.0165, 0.0180, 0.0199, 0.0195, 0.0207, 0.0192, 0.0205, 0.0193],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-23 03:20:22,050 INFO [train.py:894] (2/4) Epoch 12, batch 1500, loss[loss=0.2006, simple_loss=0.2904, pruned_loss=0.05543, over 18665.00 frames. ], tot_loss[loss=0.1968, simple_loss=0.2811, pruned_loss=0.05627, over 3711503.32 frames. ], batch size: 98, lr: 9.69e-03, grad_scale: 8.0
2022-12-23 03:20:22,088 WARNING [train.py:1060] (2/4) Exclude cut with ID 6426-64291-0000-16059-0_sp0.9 from training. Duration: 20.0944375
2022-12-23 03:20:37,051 WARNING [train.py:1060] (2/4) Exclude cut with ID 2929-85685-0044-62348-0_sp1.1 from training. Duration: 20.4
2022-12-23 03:20:44,922 WARNING [train.py:1060] (2/4) Exclude cut with ID 6330-62851-0022-91297-0 from training. Duration: 20.085
2022-12-23 03:20:55,614 WARNING [train.py:1060] (2/4) Exclude cut with ID 4860-13185-0032-76709-0_sp0.9 from training. Duration: 23.07775
2022-12-23 03:21:37,404 INFO [train.py:894] (2/4) Epoch 12, batch 1550, loss[loss=0.1838, simple_loss=0.2812, pruned_loss=0.04317, over 18537.00 frames. ], tot_loss[loss=0.1946, simple_loss=0.2791, pruned_loss=0.05501, over 3710632.73 frames. ], batch size: 55, lr: 9.68e-03, grad_scale: 8.0
2022-12-23 03:21:38,426 INFO [zipformer.py:660] (2/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=40119.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 03:21:42,621 WARNING [train.py:1060] (2/4) Exclude cut with ID 2929-85685-0044-62348-0_sp0.9 from training. Duration: 24.9333125
2022-12-23 03:21:44,061 INFO [zipformer.py:660] (2/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=40123.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 03:21:51,315 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 2.523e+02 3.928e+02 5.193e+02 6.914e+02 1.537e+03, threshold=1.039e+03, percent-clipped=5.0
2022-12-23 03:22:25,101 WARNING [train.py:1060] (2/4) Exclude cut with ID 5118-111612-0016-124680-0_sp0.9 from training. Duration: 20.388875
2022-12-23 03:22:31,715 WARNING [train.py:1060] (2/4) Exclude cut with ID 432-122774-0017-62487-0_sp1.1 from training. Duration: 20.3590625
2022-12-23 03:22:51,399 INFO [zipformer.py:660] (2/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=40167.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 03:22:54,111 INFO [train.py:894] (2/4) Epoch 12, batch 1600, loss[loss=0.1915, simple_loss=0.2766, pruned_loss=0.05323, over 18584.00 frames. ], tot_loss[loss=0.1954, simple_loss=0.2804, pruned_loss=0.05525, over 3711718.71 frames. ], batch size: 56, lr: 9.68e-03, grad_scale: 8.0
2022-12-23 03:22:57,568 INFO [zipformer.py:660] (2/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=40171.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 03:23:33,955 INFO [zipformer.py:660] (2/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=40195.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 03:23:40,949 WARNING [train.py:1060] (2/4) Exclude cut with ID 3557-8342-0013-54691-0_sp1.1 from training. Duration: 0.836375
2022-12-23 03:24:08,544 INFO [zipformer.py:660] (2/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=40218.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 03:24:09,531 INFO [train.py:894] (2/4) Epoch 12, batch 1650, loss[loss=0.2261, simple_loss=0.3074, pruned_loss=0.0724, over 18667.00 frames. ], tot_loss[loss=0.197, simple_loss=0.2812, pruned_loss=0.05642, over 3712552.21 frames. ], batch size: 99, lr: 9.67e-03, grad_scale: 8.0
2022-12-23 03:24:23,015 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 2.557e+02 4.207e+02 4.862e+02 6.254e+02 1.082e+03, threshold=9.723e+02, percent-clipped=1.0
2022-12-23 03:24:23,100 WARNING [train.py:1060] (2/4) Exclude cut with ID 8565-290391-0049-67394-0_sp0.9 from training. Duration: 21.3166875
2022-12-23 03:24:53,725 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([3.7583, 3.0236, 2.7981, 1.2509, 2.5679, 2.6206, 2.0143, 2.2568],
       device='cuda:2'), covar=tensor([0.0479, 0.0545, 0.1185, 0.1540, 0.1452, 0.1032, 0.1326, 0.0966],
       device='cuda:2'), in_proj_covar=tensor([0.0164, 0.0178, 0.0199, 0.0193, 0.0206, 0.0190, 0.0204, 0.0192],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-23 03:24:55,378 WARNING [train.py:1060] (2/4) Exclude cut with ID 6533-399-0029-104863-0_sp0.9 from training. Duration: 22.1055625
2022-12-23 03:25:05,099 WARNING [train.py:1060] (2/4) Exclude cut with ID 7699-105389-0094-26379-0_sp1.1 from training. Duration: 21.77725
2022-12-23 03:25:05,446 INFO [zipformer.py:660] (2/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=40256.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 03:25:24,593 INFO [train.py:894] (2/4) Epoch 12, batch 1700, loss[loss=0.1974, simple_loss=0.2821, pruned_loss=0.05633, over 18583.00 frames. ], tot_loss[loss=0.1998, simple_loss=0.283, pruned_loss=0.05831, over 3713009.84 frames. ], batch size: 51, lr: 9.66e-03, grad_scale: 8.0
2022-12-23 03:25:27,472 WARNING [train.py:1060] (2/4) Exclude cut with ID 7255-291500-0005-134304-0_sp0.9 from training. Duration: 27.8166875
2022-12-23 03:25:29,458 INFO [zipformer.py:660] (2/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=40272.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 03:25:39,584 INFO [zipformer.py:660] (2/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=40279.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 03:25:51,912 WARNING [train.py:1060] (2/4) Exclude cut with ID 7357-94126-0021-15852-0_sp1.1 from training. Duration: 22.5090625
2022-12-23 03:25:55,982 INFO [scaling.py:679] (2/4) Whitening: num_groups=1, num_channels=384, metric=3.46 vs. limit=5.0
2022-12-23 03:25:59,924 WARNING [train.py:1060] (2/4) Exclude cut with ID 7255-291500-0005-134304-0 from training. Duration: 25.035
2022-12-23 03:26:04,178 INFO [zipformer.py:660] (2/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=40295.0, num_to_drop=1, layers_to_drop={2}
2022-12-23 03:26:16,640 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([0.2789, 1.6676, 1.9484, 0.9284, 1.1128, 2.0460, 1.8100, 1.5710],
       device='cuda:2'), covar=tensor([0.0609, 0.0270, 0.0256, 0.0313, 0.0342, 0.0340, 0.0192, 0.0578],
       device='cuda:2'), in_proj_covar=tensor([0.0137, 0.0158, 0.0114, 0.0131, 0.0139, 0.0129, 0.0144, 0.0149],
       device='cuda:2'), out_proj_covar=tensor([1.1092e-04, 1.2891e-04, 9.1579e-05, 1.0394e-04, 1.1134e-04, 1.0659e-04,
        1.1943e-04, 1.2109e-04], device='cuda:2')
2022-12-23 03:26:17,706 WARNING [train.py:1060] (2/4) Exclude cut with ID 774-127930-0014-10412-0_sp1.1 from training. Duration: 0.95
2022-12-23 03:26:36,135 WARNING [train.py:1060] (2/4) Exclude cut with ID 3033-130750-0096-55598-0_sp0.9 from training. Duration: 0.92225
2022-12-23 03:26:40,280 INFO [train.py:894] (2/4) Epoch 12, batch 1750, loss[loss=0.2488, simple_loss=0.3204, pruned_loss=0.08853, over 18585.00 frames. ], tot_loss[loss=0.2032, simple_loss=0.2845, pruned_loss=0.06096, over 3713818.58 frames. ], batch size: 56, lr: 9.66e-03, grad_scale: 8.0
2022-12-23 03:26:53,120 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 2.694e+02 5.133e+02 6.045e+02 7.323e+02 1.639e+03, threshold=1.209e+03, percent-clipped=7.0
2022-12-23 03:26:56,479 INFO [zipformer.py:660] (2/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=40330.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 03:26:59,599 WARNING [train.py:1060] (2/4) Exclude cut with ID 4511-76322-0006-80011-0 from training. Duration: 21.97
2022-12-23 03:27:01,452 INFO [zipformer.py:660] (2/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=40333.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 03:27:16,693 INFO [zipformer.py:660] (2/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=40343.0, num_to_drop=1, layers_to_drop={1}
2022-12-23 03:27:20,135 WARNING [train.py:1060] (2/4) Exclude cut with ID 7492-105653-0055-62765-0_sp0.9 from training. Duration: 21.97225
2022-12-23 03:27:21,506 WARNING [train.py:1060] (2/4) Exclude cut with ID 453-131332-0000-47844-0_sp0.9 from training. Duration: 25.3333125
2022-12-23 03:27:23,629 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([0.5946, 1.7857, 2.3242, 0.8663, 1.3642, 2.4617, 1.8053, 1.7655],
       device='cuda:2'), covar=tensor([0.0690, 0.0372, 0.0262, 0.0421, 0.0363, 0.0296, 0.0279, 0.0599],
       device='cuda:2'), in_proj_covar=tensor([0.0138, 0.0160, 0.0115, 0.0132, 0.0141, 0.0131, 0.0146, 0.0150],
       device='cuda:2'), out_proj_covar=tensor([1.1197e-04, 1.3091e-04, 9.2382e-05, 1.0498e-04, 1.1288e-04, 1.0804e-04,
        1.2089e-04, 1.2208e-04], device='cuda:2')
2022-12-23 03:27:32,381 WARNING [train.py:1060] (2/4) Exclude cut with ID 5172-29468-0015-19128-0_sp0.9 from training. Duration: 21.5055625
2022-12-23 03:27:42,279 WARNING [train.py:1060] (2/4) Exclude cut with ID 453-131332-0000-47844-0_sp1.1 from training. Duration: 20.72725
2022-12-23 03:27:57,162 INFO [train.py:894] (2/4) Epoch 12, batch 1800, loss[loss=0.1975, simple_loss=0.2789, pruned_loss=0.0581, over 18642.00 frames. ], tot_loss[loss=0.2054, simple_loss=0.2848, pruned_loss=0.06305, over 3713270.74 frames. ], batch size: 53, lr: 9.65e-03, grad_scale: 8.0
2022-12-23 03:28:16,481 WARNING [train.py:1060] (2/4) Exclude cut with ID 8631-249866-0030-130156-0_sp0.9 from training. Duration: 26.32775
2022-12-23 03:28:30,911 INFO [zipformer.py:660] (2/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=40391.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 03:28:33,783 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.7713, 1.2235, 0.7692, 1.3352, 2.1164, 1.4084, 1.6214, 1.8287],
       device='cuda:2'), covar=tensor([0.1706, 0.2451, 0.2500, 0.1677, 0.1746, 0.1702, 0.1578, 0.1710],
       device='cuda:2'), in_proj_covar=tensor([0.0092, 0.0100, 0.0118, 0.0096, 0.0114, 0.0091, 0.0098, 0.0095],
       device='cuda:2'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003],
       device='cuda:2')
2022-12-23 03:28:50,248 WARNING [train.py:1060] (2/4) Exclude cut with ID 3867-173237-0077-144769-0 from training. Duration: 20.025
2022-12-23 03:28:56,651 WARNING [train.py:1060] (2/4) Exclude cut with ID 6709-74022-0004-86860-0_sp1.1 from training. Duration: 0.9409375
2022-12-23 03:28:56,661 WARNING [train.py:1060] (2/4) Exclude cut with ID 4757-1811-0023-62229-0_sp0.9 from training. Duration: 21.37775
2022-12-23 03:29:11,833 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([2.6913, 2.2430, 1.5672, 2.6199, 1.9396, 1.9976, 2.1710, 2.8730],
       device='cuda:2'), covar=tensor([0.1594, 0.2520, 0.1560, 0.2335, 0.3001, 0.0841, 0.2338, 0.0577],
       device='cuda:2'), in_proj_covar=tensor([0.0279, 0.0273, 0.0230, 0.0341, 0.0254, 0.0215, 0.0267, 0.0192],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0003, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-23 03:29:12,702 INFO [train.py:894] (2/4) Epoch 12, batch 1850, loss[loss=0.1999, simple_loss=0.2792, pruned_loss=0.06031, over 18455.00 frames. ], tot_loss[loss=0.2085, simple_loss=0.2867, pruned_loss=0.06517, over 3713881.26 frames. ], batch size: 54, lr: 9.65e-03, grad_scale: 8.0
2022-12-23 03:29:17,331 WARNING [train.py:1060] (2/4) Exclude cut with ID 1250-135782-0004-25974-0_sp0.9 from training. Duration: 21.17225
2022-12-23 03:29:17,344 WARNING [train.py:1060] (2/4) Exclude cut with ID 7357-94126-0021-15852-0_sp0.9 from training. Duration: 27.511125
2022-12-23 03:29:19,087 INFO [zipformer.py:660] (2/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=40423.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 03:29:26,480 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 3.348e+02 4.947e+02 6.136e+02 7.915e+02 1.614e+03, threshold=1.227e+03, percent-clipped=1.0
2022-12-23 03:29:49,755 INFO [scaling.py:679] (2/4) Whitening: num_groups=8, num_channels=192, metric=1.97 vs. limit=2.0
2022-12-23 03:29:50,452 WARNING [train.py:1060] (2/4) Exclude cut with ID 453-131332-0000-47844-0 from training. Duration: 22.8
2022-12-23 03:29:51,375 INFO [scaling.py:679] (2/4) Whitening: num_groups=8, num_channels=96, metric=1.23 vs. limit=2.0
2022-12-23 03:29:55,026 WARNING [train.py:1060] (2/4) Exclude cut with ID 4964-30587-0040-44509-0 from training. Duration: 22.585
2022-12-23 03:30:26,407 WARNING [train.py:1060] (2/4) Exclude cut with ID 6978-92210-0001-146967-0_sp0.9 from training. Duration: 22.0166875
2022-12-23 03:30:27,697 INFO [train.py:894] (2/4) Epoch 12, batch 1900, loss[loss=0.2138, simple_loss=0.2874, pruned_loss=0.07012, over 18559.00 frames. ], tot_loss[loss=0.2099, simple_loss=0.2871, pruned_loss=0.06631, over 3713688.82 frames. ], batch size: 49, lr: 9.64e-03, grad_scale: 8.0
2022-12-23 03:30:30,649 INFO [zipformer.py:660] (2/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=40471.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 03:30:30,757 INFO [zipformer.py:660] (2/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=40471.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 03:30:42,848 WARNING [train.py:1060] (2/4) Exclude cut with ID 7255-291500-0003-134302-0_sp1.1 from training. Duration: 24.395375
2022-12-23 03:30:48,115 INFO [scaling.py:679] (2/4) Whitening: num_groups=8, num_channels=192, metric=1.90 vs. limit=2.0
2022-12-23 03:30:50,559 WARNING [train.py:1060] (2/4) Exclude cut with ID 3488-85273-0017-41203-0_sp0.9 from training. Duration: 27.47775
2022-12-23 03:30:54,851 WARNING [train.py:1060] (2/4) Exclude cut with ID 432-122774-0017-62487-0_sp0.9 from training. Duration: 24.8833125
2022-12-23 03:30:57,698 WARNING [train.py:1060] (2/4) Exclude cut with ID 6758-72288-0033-108368-0 from training. Duration: 23.39
2022-12-23 03:31:02,670 INFO [zipformer.py:660] (2/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=40492.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 03:31:03,858 WARNING [train.py:1060] (2/4) Exclude cut with ID 8291-282929-0007-12994-0_sp0.9 from training. Duration: 28.72225
2022-12-23 03:31:07,785 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([2.5333, 2.1051, 2.0702, 1.8291, 2.3755, 2.9966, 3.0355, 1.9328],
       device='cuda:2'), covar=tensor([0.0310, 0.0281, 0.0347, 0.0250, 0.0215, 0.0292, 0.0263, 0.0299],
       device='cuda:2'), in_proj_covar=tensor([0.0087, 0.0119, 0.0144, 0.0121, 0.0110, 0.0111, 0.0094, 0.0123],
       device='cuda:2'), out_proj_covar=tensor([7.2356e-05, 9.7913e-05, 1.2471e-04, 1.0083e-04, 9.3418e-05, 8.9052e-05,
        7.7163e-05, 1.0102e-04], device='cuda:2')
2022-12-23 03:31:08,967 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([2.8922, 5.3542, 4.9352, 2.3649, 5.4090, 3.9629, 0.9958, 3.5801],
       device='cuda:2'), covar=tensor([0.1784, 0.0759, 0.1062, 0.3044, 0.0653, 0.0748, 0.4802, 0.1186],
       device='cuda:2'), in_proj_covar=tensor([0.0135, 0.0125, 0.0148, 0.0120, 0.0128, 0.0105, 0.0142, 0.0109],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-23 03:31:13,139 WARNING [train.py:1060] (2/4) Exclude cut with ID 585-294811-0110-133686-0_sp0.9 from training. Duration: 20.8944375
2022-12-23 03:31:26,807 WARNING [train.py:1060] (2/4) Exclude cut with ID 5796-66357-0007-116447-0_sp0.9 from training. Duration: 23.8444375
2022-12-23 03:31:42,939 INFO [train.py:894] (2/4) Epoch 12, batch 1950, loss[loss=0.1853, simple_loss=0.2635, pruned_loss=0.05353, over 18688.00 frames. ], tot_loss[loss=0.2118, simple_loss=0.2881, pruned_loss=0.06779, over 3714052.10 frames. ], batch size: 46, lr: 9.64e-03, grad_scale: 8.0
2022-12-23 03:31:43,119 INFO [zipformer.py:660] (2/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=40519.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 03:31:52,554 WARNING [train.py:1060] (2/4) Exclude cut with ID 8291-282929-0007-12994-0 from training. Duration: 25.85
2022-12-23 03:31:52,564 WARNING [train.py:1060] (2/4) Exclude cut with ID 8291-282929-0023-13010-0 from training. Duration: 21.39
2022-12-23 03:31:56,032 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.7851, 1.6710, 2.1334, 1.4625, 2.0197, 2.0759, 1.6264, 2.2363],
       device='cuda:2'), covar=tensor([0.0935, 0.1206, 0.1109, 0.1340, 0.0553, 0.0837, 0.1499, 0.0416],
       device='cuda:2'), in_proj_covar=tensor([0.0191, 0.0196, 0.0198, 0.0190, 0.0174, 0.0208, 0.0204, 0.0177],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-23 03:31:57,003 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 3.661e+02 5.030e+02 6.516e+02 8.267e+02 2.273e+03, threshold=1.303e+03, percent-clipped=7.0
2022-12-23 03:32:03,765 WARNING [train.py:1060] (2/4) Exclude cut with ID 7357-94126-0014-15845-0 from training. Duration: 27.92
2022-12-23 03:32:08,638 INFO [scaling.py:679] (2/4) Whitening: num_groups=1, num_channels=384, metric=2.84 vs. limit=5.0
2022-12-23 03:32:32,052 WARNING [train.py:1060] (2/4) Exclude cut with ID 8631-249866-0039-130165-0_sp0.9 from training. Duration: 20.661125
2022-12-23 03:32:32,207 INFO [zipformer.py:660] (2/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=40551.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 03:32:35,266 INFO [zipformer.py:660] (2/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=40553.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 03:32:55,380 WARNING [train.py:1060] (2/4) Exclude cut with ID 7357-94126-0043-15874-0_sp0.9 from training. Duration: 20.07225
2022-12-23 03:32:58,140 INFO [train.py:894] (2/4) Epoch 12, batch 2000, loss[loss=0.2249, simple_loss=0.3006, pruned_loss=0.07459, over 18670.00 frames. ], tot_loss[loss=0.2134, simple_loss=0.2889, pruned_loss=0.06899, over 3713837.31 frames. ], batch size: 98, lr: 9.63e-03, grad_scale: 8.0
2022-12-23 03:33:01,673 WARNING [train.py:1060] (2/4) Exclude cut with ID 1085-156170-0017-128270-0 from training. Duration: 21.01
2022-12-23 03:33:02,539 INFO [scaling.py:679] (2/4) Whitening: num_groups=8, num_channels=96, metric=1.19 vs. limit=2.0
2022-12-23 03:33:05,224 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.8254, 1.7286, 1.9949, 1.9254, 1.3301, 4.6710, 2.1729, 2.5921],
       device='cuda:2'), covar=tensor([0.3070, 0.2003, 0.1695, 0.1894, 0.1376, 0.0122, 0.1379, 0.0796],
       device='cuda:2'), in_proj_covar=tensor([0.0138, 0.0117, 0.0127, 0.0121, 0.0104, 0.0100, 0.0096, 0.0093],
       device='cuda:2'), out_proj_covar=tensor([0.0005, 0.0005, 0.0005, 0.0004, 0.0004, 0.0003, 0.0004, 0.0004],
       device='cuda:2')
2022-12-23 03:33:06,418 INFO [zipformer.py:660] (2/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=40574.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 03:33:08,306 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.4161, 1.7758, 1.3069, 2.0981, 2.5795, 1.4744, 1.3614, 1.1524],
       device='cuda:2'), covar=tensor([0.2011, 0.1749, 0.1617, 0.0954, 0.1204, 0.1153, 0.2073, 0.1545],
       device='cuda:2'), in_proj_covar=tensor([0.0236, 0.0211, 0.0200, 0.0184, 0.0250, 0.0188, 0.0208, 0.0186],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-23 03:33:49,178 INFO [scaling.py:679] (2/4) Whitening: num_groups=8, num_channels=96, metric=1.82 vs. limit=2.0
2022-12-23 03:33:51,690 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.0634, 0.8087, 0.9470, 1.0879, 1.2517, 1.1348, 1.0780, 0.9557],
       device='cuda:2'), covar=tensor([0.0232, 0.0211, 0.0406, 0.0184, 0.0177, 0.0281, 0.0214, 0.0223],
       device='cuda:2'), in_proj_covar=tensor([0.0086, 0.0118, 0.0143, 0.0120, 0.0110, 0.0110, 0.0093, 0.0121],
       device='cuda:2'), out_proj_covar=tensor([7.1760e-05, 9.7364e-05, 1.2389e-04, 9.9928e-05, 9.2931e-05, 8.7734e-05,
        7.5940e-05, 9.9591e-05], device='cuda:2')
2022-12-23 03:34:12,010 WARNING [train.py:1060] (2/4) Exclude cut with ID 2195-150901-0045-59933-0 from training. Duration: 20.65
2022-12-23 03:34:14,818 INFO [train.py:894] (2/4) Epoch 12, batch 2050, loss[loss=0.1855, simple_loss=0.2588, pruned_loss=0.05606, over 18686.00 frames. ], tot_loss[loss=0.214, simple_loss=0.2892, pruned_loss=0.06936, over 3714291.28 frames. ], batch size: 46, lr: 9.62e-03, grad_scale: 8.0
2022-12-23 03:34:19,090 WARNING [train.py:1060] (2/4) Exclude cut with ID 5796-66357-0007-116447-0 from training. Duration: 21.46
2022-12-23 03:34:28,295 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 2.998e+02 4.881e+02 5.858e+02 6.986e+02 1.392e+03, threshold=1.172e+03, percent-clipped=2.0
2022-12-23 03:34:28,641 INFO [zipformer.py:660] (2/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=40628.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 03:35:00,425 WARNING [train.py:1060] (2/4) Exclude cut with ID 3557-8342-0013-54691-0 from training. Duration: 0.92
2022-12-23 03:35:04,242 INFO [zipformer.py:660] (2/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=40651.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 03:35:06,935 WARNING [train.py:1060] (2/4) Exclude cut with ID 8291-282929-0023-13010-0_sp0.9 from training. Duration: 23.7666875
2022-12-23 03:35:32,192 INFO [train.py:894] (2/4) Epoch 12, batch 2100, loss[loss=0.266, simple_loss=0.3253, pruned_loss=0.1034, over 18607.00 frames. ], tot_loss[loss=0.2148, simple_loss=0.2896, pruned_loss=0.06998, over 3714736.69 frames. ], batch size: 180, lr: 9.62e-03, grad_scale: 8.0
2022-12-23 03:35:45,365 WARNING [train.py:1060] (2/4) Exclude cut with ID 8544-281189-0060-101339-0_sp0.9 from training. Duration: 20.861125
2022-12-23 03:35:57,729 WARNING [train.py:1060] (2/4) Exclude cut with ID 3488-65654-0031-41259-0_sp0.9 from training. Duration: 22.711125
2022-12-23 03:35:57,888 INFO [zipformer.py:660] (2/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=40686.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 03:36:36,102 INFO [zipformer.py:660] (2/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=40712.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 03:36:40,147 WARNING [train.py:1060] (2/4) Exclude cut with ID 6951-79737-0043-132310-0_sp1.1 from training. Duration: 22.986375
2022-12-23 03:36:45,602 INFO [train.py:894] (2/4) Epoch 12, batch 2150, loss[loss=0.2076, simple_loss=0.2662, pruned_loss=0.07453, over 18434.00 frames. ], tot_loss[loss=0.2137, simple_loss=0.2883, pruned_loss=0.06952, over 3714154.10 frames. ], batch size: 42, lr: 9.61e-03, grad_scale: 8.0
2022-12-23 03:36:55,102 WARNING [train.py:1060] (2/4) Exclude cut with ID 8040-260924-0003-80960-0_sp0.9 from training. Duration: 22.07225
2022-12-23 03:36:59,399 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 3.466e+02 5.161e+02 6.497e+02 7.799e+02 1.996e+03, threshold=1.299e+03, percent-clipped=5.0
2022-12-23 03:37:01,152 WARNING [train.py:1060] (2/4) Exclude cut with ID 7699-105389-0045-26330-0_sp0.9 from training. Duration: 20.3055625
2022-12-23 03:37:04,729 WARNING [train.py:1060] (2/4) Exclude cut with ID 6356-271890-0060-94317-0_sp0.9 from training. Duration: 20.72225
2022-12-23 03:37:19,957 WARNING [train.py:1060] (2/4) Exclude cut with ID 3488-85273-0017-41203-0_sp1.1 from training. Duration: 22.4818125
2022-12-23 03:37:44,132 WARNING [train.py:1060] (2/4) Exclude cut with ID 4964-30587-0040-44509-0_sp0.9 from training. Duration: 25.0944375
2022-12-23 03:37:48,584 WARNING [train.py:1060] (2/4) Exclude cut with ID 6533-399-0047-104881-0 from training. Duration: 21.515
2022-12-23 03:37:52,995 WARNING [train.py:1060] (2/4) Exclude cut with ID 7357-94126-0009-15840-0_sp0.9 from training. Duration: 27.02225
2022-12-23 03:37:59,487 WARNING [train.py:1060] (2/4) Exclude cut with ID 432-122774-0010-62480-0_sp0.9 from training. Duration: 22.22225
2022-12-23 03:38:02,639 INFO [train.py:894] (2/4) Epoch 12, batch 2200, loss[loss=0.198, simple_loss=0.2594, pruned_loss=0.06828, over 18577.00 frames. ], tot_loss[loss=0.2141, simple_loss=0.289, pruned_loss=0.0696, over 3714064.99 frames. ], batch size: 45, lr: 9.61e-03, grad_scale: 8.0
2022-12-23 03:38:04,182 WARNING [train.py:1060] (2/4) Exclude cut with ID 4964-30587-0085-44554-0_sp0.9 from training. Duration: 20.85
2022-12-23 03:38:41,185 WARNING [train.py:1060] (2/4) Exclude cut with ID 4295-39940-0007-92567-0 from training. Duration: 21.54
2022-12-23 03:38:44,217 WARNING [train.py:1060] (2/4) Exclude cut with ID 4964-30587-0040-44509-0_sp1.1 from training. Duration: 20.5318125
2022-12-23 03:38:53,646 WARNING [train.py:1060] (2/4) Exclude cut with ID 7255-291500-0012-134311-0_sp0.9 from training. Duration: 21.9333125
2022-12-23 03:39:19,321 INFO [train.py:894] (2/4) Epoch 12, batch 2250, loss[loss=0.1819, simple_loss=0.2664, pruned_loss=0.04872, over 18703.00 frames. ], tot_loss[loss=0.214, simple_loss=0.2887, pruned_loss=0.06967, over 3713936.60 frames. ], batch size: 50, lr: 9.60e-03, grad_scale: 8.0
2022-12-23 03:39:33,366 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 3.699e+02 5.281e+02 6.293e+02 7.609e+02 1.404e+03, threshold=1.259e+03, percent-clipped=2.0
2022-12-23 03:39:44,657 WARNING [train.py:1060] (2/4) Exclude cut with ID 8631-249866-0025-130151-0_sp0.9 from training. Duration: 21.7944375
2022-12-23 03:39:49,356 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([0.5279, 1.8233, 2.1687, 0.9307, 1.3563, 2.2721, 1.8876, 1.6283],
       device='cuda:2'), covar=tensor([0.0581, 0.0321, 0.0295, 0.0335, 0.0339, 0.0350, 0.0219, 0.0551],
       device='cuda:2'), in_proj_covar=tensor([0.0137, 0.0158, 0.0113, 0.0130, 0.0140, 0.0131, 0.0144, 0.0148],
       device='cuda:2'), out_proj_covar=tensor([1.1086e-04, 1.2883e-04, 9.0880e-05, 1.0292e-04, 1.1177e-04, 1.0803e-04,
        1.1905e-04, 1.1986e-04], device='cuda:2')
2022-12-23 03:39:56,169 WARNING [train.py:1060] (2/4) Exclude cut with ID 8291-282929-0002-12989-0_sp0.9 from training. Duration: 22.4666875
2022-12-23 03:40:02,689 INFO [zipformer.py:660] (2/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=40848.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 03:40:03,988 WARNING [train.py:1060] (2/4) Exclude cut with ID 6121-9014-0076-24124-0 from training. Duration: 21.635
2022-12-23 03:40:07,017 INFO [zipformer.py:660] (2/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=40851.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 03:40:09,793 WARNING [train.py:1060] (2/4) Exclude cut with ID 6121-9014-0076-24124-0_sp0.9 from training. Duration: 24.038875
2022-12-23 03:40:28,375 INFO [zipformer.py:660] (2/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=40865.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 03:40:34,338 INFO [train.py:894] (2/4) Epoch 12, batch 2300, loss[loss=0.2454, simple_loss=0.3158, pruned_loss=0.08746, over 18665.00 frames. ], tot_loss[loss=0.2157, simple_loss=0.2902, pruned_loss=0.07061, over 3714051.99 frames. ], batch size: 98, lr: 9.59e-03, grad_scale: 8.0
2022-12-23 03:40:42,597 INFO [zipformer.py:660] (2/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=40874.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 03:40:55,545 WARNING [train.py:1060] (2/4) Exclude cut with ID 7255-291500-0009-134308-0_sp1.1 from training. Duration: 21.786375
2022-12-23 03:40:57,953 INFO [scaling.py:679] (2/4) Whitening: num_groups=8, num_channels=96, metric=1.47 vs. limit=2.0
2022-12-23 03:41:06,387 WARNING [train.py:1060] (2/4) Exclude cut with ID 8291-282929-0002-12989-0 from training. Duration: 20.22
2022-12-23 03:41:20,694 INFO [zipformer.py:660] (2/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=40899.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 03:41:22,411 INFO [zipformer.py:660] (2/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=40900.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 03:41:39,361 INFO [scaling.py:679] (2/4) Whitening: num_groups=8, num_channels=192, metric=1.93 vs. limit=2.0
2022-12-23 03:41:41,820 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([2.4028, 1.9500, 1.9473, 2.1907, 2.3134, 2.2178, 2.1979, 1.6335],
       device='cuda:2'), covar=tensor([0.1725, 0.2649, 0.2051, 0.2319, 0.1458, 0.0779, 0.2522, 0.1049],
       device='cuda:2'), in_proj_covar=tensor([0.0255, 0.0288, 0.0264, 0.0298, 0.0282, 0.0238, 0.0309, 0.0224],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-23 03:41:51,437 INFO [train.py:894] (2/4) Epoch 12, batch 2350, loss[loss=0.2082, simple_loss=0.2954, pruned_loss=0.06047, over 18521.00 frames. ], tot_loss[loss=0.2142, simple_loss=0.2892, pruned_loss=0.06964, over 3714490.40 frames. ], batch size: 58, lr: 9.59e-03, grad_scale: 8.0
2022-12-23 03:41:56,378 INFO [zipformer.py:660] (2/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=40922.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 03:42:02,547 INFO [zipformer.py:660] (2/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=40926.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 03:42:04,878 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 3.147e+02 5.320e+02 6.167e+02 7.693e+02 1.883e+03, threshold=1.233e+03, percent-clipped=4.0
2022-12-23 03:42:05,261 INFO [zipformer.py:660] (2/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=40928.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 03:42:06,537 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([2.9517, 5.4593, 4.9402, 2.3497, 5.5518, 4.0763, 1.0556, 3.8212],
       device='cuda:2'), covar=tensor([0.1747, 0.0741, 0.1096, 0.3082, 0.0547, 0.0837, 0.4909, 0.1164],
       device='cuda:2'), in_proj_covar=tensor([0.0135, 0.0126, 0.0149, 0.0119, 0.0130, 0.0105, 0.0141, 0.0109],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-23 03:42:55,830 INFO [zipformer.py:660] (2/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=40961.0, num_to_drop=1, layers_to_drop={1}
2022-12-23 03:43:04,990 WARNING [train.py:1060] (2/4) Exclude cut with ID 6951-79737-0043-132310-0 from training. Duration: 25.285
2022-12-23 03:43:07,760 INFO [train.py:894] (2/4) Epoch 12, batch 2400, loss[loss=0.2933, simple_loss=0.3397, pruned_loss=0.1234, over 18608.00 frames. ], tot_loss[loss=0.2158, simple_loss=0.2905, pruned_loss=0.07059, over 3715526.96 frames. ], batch size: 176, lr: 9.58e-03, grad_scale: 8.0
2022-12-23 03:43:09,761 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([2.9120, 2.5402, 1.9782, 1.0123, 2.0375, 2.2816, 1.7033, 2.0407],
       device='cuda:2'), covar=tensor([0.0507, 0.0502, 0.1053, 0.1434, 0.1214, 0.1122, 0.1400, 0.0784],
       device='cuda:2'), in_proj_covar=tensor([0.0165, 0.0181, 0.0201, 0.0195, 0.0206, 0.0191, 0.0206, 0.0195],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-23 03:43:17,935 INFO [zipformer.py:660] (2/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=40976.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 03:43:32,739 INFO [zipformer.py:660] (2/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=40986.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 03:43:47,722 INFO [zipformer.py:660] (2/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=40997.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 03:44:04,252 INFO [zipformer.py:660] (2/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=41007.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 03:44:09,785 WARNING [train.py:1060] (2/4) Exclude cut with ID 811-130148-0001-63453-0_sp0.9 from training. Duration: 20.861125
2022-12-23 03:44:21,816 INFO [train.py:894] (2/4) Epoch 12, batch 2450, loss[loss=0.2344, simple_loss=0.3053, pruned_loss=0.08177, over 18722.00 frames. ], tot_loss[loss=0.2152, simple_loss=0.2897, pruned_loss=0.07033, over 3715004.49 frames. ], batch size: 52, lr: 9.58e-03, grad_scale: 8.0
2022-12-23 03:44:32,021 WARNING [train.py:1060] (2/4) Exclude cut with ID 6010-56788-0055-90261-0 from training. Duration: 20.88
2022-12-23 03:44:34,730 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 2.360e+02 5.211e+02 6.060e+02 7.407e+02 1.596e+03, threshold=1.212e+03, percent-clipped=4.0
2022-12-23 03:44:44,635 INFO [zipformer.py:660] (2/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=41034.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 03:45:01,325 INFO [zipformer.py:660] (2/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=41045.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 03:45:02,388 WARNING [train.py:1060] (2/4) Exclude cut with ID 7357-94126-0045-15876-0_sp0.9 from training. Duration: 23.4166875
2022-12-23 03:45:20,442 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([3.1606, 2.3928, 1.8124, 2.8010, 2.1258, 2.2608, 2.3668, 3.4623],
       device='cuda:2'), covar=tensor([0.1750, 0.3063, 0.1690, 0.3026, 0.3572, 0.0902, 0.2976, 0.0611],
       device='cuda:2'), in_proj_covar=tensor([0.0284, 0.0276, 0.0233, 0.0349, 0.0255, 0.0218, 0.0271, 0.0196],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0003, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-23 03:45:21,722 INFO [zipformer.py:660] (2/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=41058.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 03:45:38,343 INFO [train.py:894] (2/4) Epoch 12, batch 2500, loss[loss=0.193, simple_loss=0.2812, pruned_loss=0.05242, over 18547.00 frames. ], tot_loss[loss=0.2139, simple_loss=0.2888, pruned_loss=0.06948, over 3716137.51 frames. ], batch size: 55, lr: 9.57e-03, grad_scale: 8.0
2022-12-23 03:46:07,411 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([2.4895, 2.1248, 1.7544, 1.6548, 2.1439, 2.8797, 2.8397, 1.8897],
       device='cuda:2'), covar=tensor([0.0306, 0.0290, 0.0444, 0.0281, 0.0246, 0.0287, 0.0254, 0.0314],
       device='cuda:2'), in_proj_covar=tensor([0.0088, 0.0121, 0.0146, 0.0122, 0.0113, 0.0113, 0.0094, 0.0123],
       device='cuda:2'), out_proj_covar=tensor([7.3022e-05, 9.9313e-05, 1.2650e-04, 1.0116e-04, 9.5770e-05, 8.9615e-05,
        7.6579e-05, 1.0113e-04], device='cuda:2')
2022-12-23 03:46:19,341 WARNING [train.py:1060] (2/4) Exclude cut with ID 7255-291500-0006-134305-0 from training. Duration: 21.24
2022-12-23 03:46:19,352 WARNING [train.py:1060] (2/4) Exclude cut with ID 6533-399-0047-104881-0_sp0.9 from training. Duration: 23.9055625
2022-12-23 03:46:30,236 INFO [zipformer.py:660] (2/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=41102.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 03:46:36,730 INFO [zipformer.py:660] (2/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=41106.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 03:46:46,652 INFO [scaling.py:679] (2/4) Whitening: num_groups=1, num_channels=384, metric=4.42 vs. limit=5.0
2022-12-23 03:46:54,188 WARNING [train.py:1060] (2/4) Exclude cut with ID 6758-72288-0033-108368-0_sp0.9 from training. Duration: 25.988875
2022-12-23 03:46:55,597 INFO [train.py:894] (2/4) Epoch 12, batch 2550, loss[loss=0.2401, simple_loss=0.3168, pruned_loss=0.08172, over 18504.00 frames. ], tot_loss[loss=0.2129, simple_loss=0.2883, pruned_loss=0.06876, over 3715823.58 frames. ], batch size: 58, lr: 9.57e-03, grad_scale: 8.0
2022-12-23 03:47:03,537 WARNING [train.py:1060] (2/4) Exclude cut with ID 7255-291500-0001-134300-0_sp0.9 from training. Duration: 20.67225
2022-12-23 03:47:09,332 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 3.299e+02 5.132e+02 6.345e+02 7.897e+02 1.932e+03, threshold=1.269e+03, percent-clipped=6.0
2022-12-23 03:47:39,639 INFO [zipformer.py:660] (2/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=41148.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 03:47:49,991 WARNING [train.py:1060] (2/4) Exclude cut with ID 3488-85273-0038-41224-0 from training. Duration: 20.34
2022-12-23 03:48:01,673 INFO [zipformer.py:660] (2/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=41163.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 03:48:10,558 INFO [train.py:894] (2/4) Epoch 12, batch 2600, loss[loss=0.1882, simple_loss=0.2522, pruned_loss=0.06206, over 18546.00 frames. ], tot_loss[loss=0.213, simple_loss=0.2884, pruned_loss=0.06883, over 3714468.37 frames. ], batch size: 44, lr: 9.56e-03, grad_scale: 8.0
2022-12-23 03:48:52,564 INFO [zipformer.py:660] (2/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=41196.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 03:48:58,647 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([2.4472, 1.8480, 2.0451, 2.2305, 2.3701, 2.3325, 2.3262, 1.6363],
       device='cuda:2'), covar=tensor([0.1696, 0.2714, 0.2003, 0.2261, 0.1467, 0.0752, 0.2368, 0.1076],
       device='cuda:2'), in_proj_covar=tensor([0.0254, 0.0288, 0.0263, 0.0297, 0.0282, 0.0237, 0.0306, 0.0225],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-23 03:49:02,391 WARNING [train.py:1060] (2/4) Exclude cut with ID 7357-94126-0026-15857-0_sp0.9 from training. Duration: 25.061125
2022-12-23 03:49:12,578 WARNING [train.py:1060] (2/4) Exclude cut with ID 3033-130750-0096-55598-0 from training. Duration: 0.83
2022-12-23 03:49:26,844 INFO [train.py:894] (2/4) Epoch 12, batch 2650, loss[loss=0.1586, simple_loss=0.2338, pruned_loss=0.04174, over 18404.00 frames. ], tot_loss[loss=0.213, simple_loss=0.2882, pruned_loss=0.06886, over 3714574.14 frames. ], batch size: 42, lr: 9.55e-03, grad_scale: 8.0
2022-12-23 03:49:30,149 INFO [zipformer.py:660] (2/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=41221.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 03:49:38,797 WARNING [train.py:1060] (2/4) Exclude cut with ID 3488-85273-0017-41203-0 from training. Duration: 24.73
2022-12-23 03:49:40,119 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 2.597e+02 5.175e+02 6.585e+02 8.509e+02 1.944e+03, threshold=1.317e+03, percent-clipped=3.0
2022-12-23 03:49:52,637 WARNING [train.py:1060] (2/4) Exclude cut with ID 7255-291500-0009-134308-0 from training. Duration: 23.965
2022-12-23 03:50:00,557 WARNING [train.py:1060] (2/4) Exclude cut with ID 6978-92210-0030-146996-0_sp0.9 from training. Duration: 22.088875
2022-12-23 03:50:18,607 WARNING [train.py:1060] (2/4) Exclude cut with ID 7255-291500-0006-134305-0_sp0.9 from training. Duration: 23.6
2022-12-23 03:50:23,225 INFO [zipformer.py:660] (2/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=41256.0, num_to_drop=1, layers_to_drop={0}
2022-12-23 03:50:39,136 INFO [scaling.py:679] (2/4) Whitening: num_groups=8, num_channels=96, metric=1.24 vs. limit=2.0
2022-12-23 03:50:42,834 INFO [train.py:894] (2/4) Epoch 12, batch 2700, loss[loss=0.2555, simple_loss=0.3242, pruned_loss=0.09343, over 18448.00 frames. ], tot_loss[loss=0.2137, simple_loss=0.289, pruned_loss=0.06924, over 3714391.17 frames. ], batch size: 64, lr: 9.55e-03, grad_scale: 8.0
2022-12-23 03:51:41,059 INFO [zipformer.py:660] (2/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=41307.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 03:51:59,145 INFO [train.py:894] (2/4) Epoch 12, batch 2750, loss[loss=0.2315, simple_loss=0.3073, pruned_loss=0.07783, over 18462.00 frames. ], tot_loss[loss=0.2125, simple_loss=0.2877, pruned_loss=0.06865, over 3714353.87 frames. ], batch size: 68, lr: 9.54e-03, grad_scale: 8.0
2022-12-23 03:52:00,731 WARNING [train.py:1060] (2/4) Exclude cut with ID 8291-282929-0024-13011-0 from training. Duration: 23.795
2022-12-23 03:52:12,939 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 2.607e+02 4.738e+02 5.981e+02 8.070e+02 1.875e+03, threshold=1.196e+03, percent-clipped=2.0
2022-12-23 03:52:18,620 WARNING [train.py:1060] (2/4) Exclude cut with ID 8631-249866-0030-130156-0_sp1.1 from training. Duration: 21.5409375
2022-12-23 03:52:20,342 WARNING [train.py:1060] (2/4) Exclude cut with ID 6978-92210-0019-146985-0_sp0.9 from training. Duration: 24.97775
2022-12-23 03:52:32,139 WARNING [train.py:1060] (2/4) Exclude cut with ID 1085-156170-0017-128270-0_sp0.9 from training. Duration: 23.3444375
2022-12-23 03:52:50,870 INFO [zipformer.py:660] (2/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=41353.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 03:52:54,301 INFO [zipformer.py:660] (2/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=41355.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 03:53:00,170 WARNING [train.py:1060] (2/4) Exclude cut with ID 6010-56788-0055-90261-0_sp0.9 from training. Duration: 23.2
2022-12-23 03:53:05,821 WARNING [train.py:1060] (2/4) Exclude cut with ID 5653-46179-0060-117930-0_sp0.9 from training. Duration: 21.17225
2022-12-23 03:53:14,850 INFO [train.py:894] (2/4) Epoch 12, batch 2800, loss[loss=0.2484, simple_loss=0.3241, pruned_loss=0.08639, over 18671.00 frames. ], tot_loss[loss=0.212, simple_loss=0.2874, pruned_loss=0.0683, over 3713956.31 frames. ], batch size: 60, lr: 9.54e-03, grad_scale: 8.0
2022-12-23 03:53:25,441 WARNING [train.py:1060] (2/4) Exclude cut with ID 7255-291500-0008-134307-0_sp0.9 from training. Duration: 24.6555625
2022-12-23 03:54:04,875 INFO [zipformer.py:660] (2/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=41401.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 03:54:21,644 WARNING [train.py:1060] (2/4) Exclude cut with ID 3488-65654-0031-41259-0 from training. Duration: 20.44
2022-12-23 03:54:32,428 INFO [train.py:894] (2/4) Epoch 12, batch 2850, loss[loss=0.2402, simple_loss=0.3125, pruned_loss=0.08393, over 18654.00 frames. ], tot_loss[loss=0.2117, simple_loss=0.2872, pruned_loss=0.06812, over 3714090.55 frames. ], batch size: 60, lr: 9.53e-03, grad_scale: 8.0
2022-12-23 03:54:36,841 WARNING [train.py:1060] (2/4) Exclude cut with ID 6951-79737-0018-132285-0_sp0.9 from training. Duration: 23.45
2022-12-23 03:54:46,857 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 3.602e+02 5.248e+02 6.316e+02 8.103e+02 1.885e+03, threshold=1.263e+03, percent-clipped=5.0
2022-12-23 03:55:06,127 WARNING [train.py:1060] (2/4) Exclude cut with ID 6945-60535-0076-12784-0_sp0.9 from training. Duration: 20.52225
2022-12-23 03:55:14,481 WARNING [train.py:1060] (2/4) Exclude cut with ID 7255-291500-0008-134307-0 from training. Duration: 22.19
2022-12-23 03:55:24,744 WARNING [train.py:1060] (2/4) Exclude cut with ID 7357-94126-0014-15845-0_sp1.1 from training. Duration: 25.3818125
2022-12-23 03:55:31,773 INFO [zipformer.py:660] (2/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=41458.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 03:55:41,487 WARNING [train.py:1060] (2/4) Exclude cut with ID 6951-79737-0043-132310-0_sp0.9 from training. Duration: 28.0944375
2022-12-23 03:55:47,918 INFO [train.py:894] (2/4) Epoch 12, batch 2900, loss[loss=0.1546, simple_loss=0.2301, pruned_loss=0.03951, over 18409.00 frames. ], tot_loss[loss=0.2118, simple_loss=0.2875, pruned_loss=0.06809, over 3714560.05 frames. ], batch size: 42, lr: 9.53e-03, grad_scale: 8.0
2022-12-23 03:55:48,033 WARNING [train.py:1060] (2/4) Exclude cut with ID 2195-150901-0045-59933-0_sp0.9 from training. Duration: 22.9444375
2022-12-23 03:55:56,717 WARNING [train.py:1060] (2/4) Exclude cut with ID 8291-282929-0024-13011-0_sp1.1 from training. Duration: 21.6318125
2022-12-23 03:56:13,075 WARNING [train.py:1060] (2/4) Exclude cut with ID 8631-249866-0030-130156-0 from training. Duration: 23.695
2022-12-23 03:56:38,867 WARNING [train.py:1060] (2/4) Exclude cut with ID 7699-105389-0094-26379-0 from training. Duration: 23.955
2022-12-23 03:56:47,934 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([6.0516, 5.2583, 5.3769, 6.0388, 5.6608, 5.4500, 6.0822, 1.6328],
       device='cuda:2'), covar=tensor([0.0498, 0.0451, 0.0455, 0.0542, 0.1052, 0.0863, 0.0357, 0.4901],
       device='cuda:2'), in_proj_covar=tensor([0.0311, 0.0204, 0.0211, 0.0229, 0.0290, 0.0247, 0.0254, 0.0261],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0001, 0.0001, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-23 03:57:03,506 INFO [train.py:894] (2/4) Epoch 12, batch 2950, loss[loss=0.214, simple_loss=0.2977, pruned_loss=0.06515, over 18631.00 frames. ], tot_loss[loss=0.2112, simple_loss=0.2866, pruned_loss=0.06792, over 3715215.65 frames. ], batch size: 78, lr: 9.52e-03, grad_scale: 8.0
2022-12-23 03:57:07,387 INFO [zipformer.py:660] (2/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=41521.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 03:57:12,762 WARNING [train.py:1060] (2/4) Exclude cut with ID 8291-282929-0024-13011-0_sp0.9 from training. Duration: 26.438875
2022-12-23 03:57:17,824 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 2.973e+02 4.599e+02 6.243e+02 8.139e+02 2.376e+03, threshold=1.249e+03, percent-clipped=3.0
2022-12-23 03:57:57,884 WARNING [train.py:1060] (2/4) Exclude cut with ID 7699-105389-0021-26306-0_sp0.9 from training. Duration: 21.2444375
2022-12-23 03:57:59,298 WARNING [train.py:1060] (2/4) Exclude cut with ID 7357-94126-0014-15845-0_sp0.9 from training. Duration: 31.02225
2022-12-23 03:57:59,694 INFO [zipformer.py:660] (2/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=41556.0, num_to_drop=1, layers_to_drop={2}
2022-12-23 03:58:07,199 WARNING [train.py:1060] (2/4) Exclude cut with ID 432-122774-0017-62487-0 from training. Duration: 22.395
2022-12-23 03:58:20,152 INFO [train.py:894] (2/4) Epoch 12, batch 3000, loss[loss=0.1959, simple_loss=0.2655, pruned_loss=0.06318, over 18379.00 frames. ], tot_loss[loss=0.2107, simple_loss=0.2856, pruned_loss=0.06795, over 3714124.12 frames. ], batch size: 46, lr: 9.51e-03, grad_scale: 8.0
2022-12-23 03:58:20,153 INFO [train.py:919] (2/4) Computing validation loss
2022-12-23 03:58:30,341 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.9059, 1.5005, 2.1858, 2.7831, 2.0449, 2.6473, 1.4717, 2.0701],
       device='cuda:2'), covar=tensor([0.1548, 0.1644, 0.1260, 0.0495, 0.0905, 0.0947, 0.1770, 0.0999],
       device='cuda:2'), in_proj_covar=tensor([0.0102, 0.0115, 0.0132, 0.0132, 0.0105, 0.0135, 0.0129, 0.0109],
       device='cuda:2'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0004, 0.0003, 0.0003],
       device='cuda:2')
2022-12-23 03:58:31,130 INFO [train.py:928] (2/4) Epoch 12, validation: loss=0.1703, simple_loss=0.2698, pruned_loss=0.03541, over 944034.00 frames. 
2022-12-23 03:58:31,130 INFO [train.py:929] (2/4) Maximum memory allocated so far is 24680MB
2022-12-23 03:58:31,292 INFO [zipformer.py:660] (2/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=41569.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 03:58:35,607 WARNING [train.py:1060] (2/4) Exclude cut with ID 7357-94126-0045-15876-0 from training. Duration: 21.075
2022-12-23 03:58:40,123 WARNING [train.py:1060] (2/4) Exclude cut with ID 6482-98857-0025-147532-0_sp0.9 from training. Duration: 20.0055625
2022-12-23 03:58:41,443 WARNING [train.py:1060] (2/4) Exclude cut with ID 6951-79737-0037-132304-0_sp0.9 from training. Duration: 22.05
2022-12-23 03:58:41,452 WARNING [train.py:1060] (2/4) Exclude cut with ID 7255-291500-0003-134302-0 from training. Duration: 26.8349375
2022-12-23 03:58:44,296 WARNING [train.py:1060] (2/4) Exclude cut with ID 7357-94126-0009-15840-0_sp1.1 from training. Duration: 22.1090625
2022-12-23 03:58:51,896 WARNING [train.py:1060] (2/4) Exclude cut with ID 7699-105389-0094-26379-0_sp0.9 from training. Duration: 26.6166875
2022-12-23 03:59:08,945 WARNING [train.py:1060] (2/4) Exclude cut with ID 2046-178027-0000-53705-0_sp0.9 from training. Duration: 20.3055625
2022-12-23 03:59:25,867 INFO [zipformer.py:660] (2/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=41604.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 03:59:33,386 WARNING [train.py:1060] (2/4) Exclude cut with ID 7205-50138-0008-5373-0_sp0.9 from training. Duration: 20.7
2022-12-23 03:59:48,386 INFO [train.py:894] (2/4) Epoch 12, batch 3050, loss[loss=0.2027, simple_loss=0.2787, pruned_loss=0.06339, over 18463.00 frames. ], tot_loss[loss=0.2103, simple_loss=0.2854, pruned_loss=0.06757, over 3713684.72 frames. ], batch size: 50, lr: 9.51e-03, grad_scale: 8.0
2022-12-23 04:00:01,463 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 3.194e+02 4.690e+02 5.641e+02 7.048e+02 1.458e+03, threshold=1.128e+03, percent-clipped=1.0
2022-12-23 04:00:16,365 WARNING [train.py:1060] (2/4) Exclude cut with ID 6978-92210-0019-146985-0 from training. Duration: 22.48
2022-12-23 04:00:33,399 WARNING [train.py:1060] (2/4) Exclude cut with ID 7255-291500-0003-134302-0_sp0.9 from training. Duration: 29.816625
2022-12-23 04:00:40,241 INFO [zipformer.py:660] (2/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=41653.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 04:00:41,833 INFO [zipformer.py:660] (2/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=41654.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 04:00:53,260 WARNING [train.py:1060] (2/4) Exclude cut with ID 7255-291500-0005-134304-0_sp1.1 from training. Duration: 22.7590625
2022-12-23 04:00:59,667 WARNING [train.py:1060] (2/4) Exclude cut with ID 7357-94126-0026-15857-0 from training. Duration: 22.555
2022-12-23 04:01:02,816 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.5051, 1.3824, 1.5268, 1.5256, 1.0403, 3.0123, 1.3833, 1.6205],
       device='cuda:2'), covar=tensor([0.3249, 0.2181, 0.1930, 0.1973, 0.1389, 0.0238, 0.1547, 0.0887],
       device='cuda:2'), in_proj_covar=tensor([0.0135, 0.0115, 0.0126, 0.0119, 0.0101, 0.0098, 0.0094, 0.0090],
       device='cuda:2'), out_proj_covar=tensor([0.0005, 0.0005, 0.0005, 0.0004, 0.0004, 0.0003, 0.0004, 0.0004],
       device='cuda:2')
2022-12-23 04:01:03,873 INFO [train.py:894] (2/4) Epoch 12, batch 3100, loss[loss=0.2264, simple_loss=0.3017, pruned_loss=0.07557, over 18600.00 frames. ], tot_loss[loss=0.2109, simple_loss=0.2862, pruned_loss=0.06784, over 3714522.28 frames. ], batch size: 69, lr: 9.50e-03, grad_scale: 8.0
2022-12-23 04:01:20,292 WARNING [train.py:1060] (2/4) Exclude cut with ID 1250-135782-0005-25975-0_sp0.9 from training. Duration: 21.688875
2022-12-23 04:01:36,197 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.9619, 1.8896, 2.0781, 2.0305, 1.5009, 4.6060, 2.2112, 2.5775],
       device='cuda:2'), covar=tensor([0.2882, 0.1770, 0.1600, 0.1699, 0.1306, 0.0097, 0.1345, 0.0717],
       device='cuda:2'), in_proj_covar=tensor([0.0136, 0.0115, 0.0126, 0.0119, 0.0101, 0.0098, 0.0094, 0.0090],
       device='cuda:2'), out_proj_covar=tensor([0.0005, 0.0005, 0.0005, 0.0004, 0.0004, 0.0003, 0.0004, 0.0004],
       device='cuda:2')
2022-12-23 04:01:52,849 INFO [zipformer.py:660] (2/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=41701.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 04:01:52,945 INFO [zipformer.py:660] (2/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=41701.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 04:01:55,653 WARNING [train.py:1060] (2/4) Exclude cut with ID 3488-85273-0038-41224-0_sp0.9 from training. Duration: 22.6
2022-12-23 04:02:05,330 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([2.5617, 2.1251, 1.4072, 2.3560, 1.9624, 1.9040, 2.0436, 2.5631],
       device='cuda:2'), covar=tensor([0.1782, 0.2807, 0.1777, 0.2662, 0.3078, 0.0939, 0.2554, 0.0729],
       device='cuda:2'), in_proj_covar=tensor([0.0282, 0.0273, 0.0230, 0.0344, 0.0252, 0.0216, 0.0268, 0.0194],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0003, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-23 04:02:14,020 INFO [zipformer.py:660] (2/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=41715.0, num_to_drop=1, layers_to_drop={1}
2022-12-23 04:02:19,246 INFO [train.py:894] (2/4) Epoch 12, batch 3150, loss[loss=0.2743, simple_loss=0.3312, pruned_loss=0.1087, over 18574.00 frames. ], tot_loss[loss=0.2112, simple_loss=0.2867, pruned_loss=0.06782, over 3715486.81 frames. ], batch size: 97, lr: 9.50e-03, grad_scale: 16.0
2022-12-23 04:02:32,303 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 3.643e+02 4.988e+02 6.154e+02 7.740e+02 1.353e+03, threshold=1.231e+03, percent-clipped=2.0
2022-12-23 04:02:33,729 WARNING [train.py:1060] (2/4) Exclude cut with ID 7357-94126-0009-15840-0 from training. Duration: 24.32
2022-12-23 04:02:39,205 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([2.5156, 2.0632, 1.4916, 2.4786, 1.8851, 1.9231, 2.0430, 2.5269],
       device='cuda:2'), covar=tensor([0.1764, 0.2908, 0.1725, 0.2491, 0.3031, 0.0928, 0.2441, 0.0736],
       device='cuda:2'), in_proj_covar=tensor([0.0280, 0.0272, 0.0228, 0.0342, 0.0251, 0.0215, 0.0266, 0.0194],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0003, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-23 04:02:56,252 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.4879, 1.4183, 1.6195, 1.4956, 1.0185, 2.9758, 1.2167, 1.5912],
       device='cuda:2'), covar=tensor([0.3292, 0.2081, 0.1889, 0.2089, 0.1501, 0.0240, 0.1679, 0.0969],
       device='cuda:2'), in_proj_covar=tensor([0.0135, 0.0115, 0.0126, 0.0119, 0.0102, 0.0098, 0.0094, 0.0090],
       device='cuda:2'), out_proj_covar=tensor([0.0005, 0.0005, 0.0005, 0.0004, 0.0004, 0.0003, 0.0004, 0.0004],
       device='cuda:2')
2022-12-23 04:03:05,277 INFO [zipformer.py:660] (2/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=41749.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 04:03:09,218 INFO [scaling.py:679] (2/4) Whitening: num_groups=8, num_channels=96, metric=1.22 vs. limit=2.0
2022-12-23 04:03:19,903 INFO [zipformer.py:660] (2/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=41758.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 04:03:29,765 WARNING [train.py:1060] (2/4) Exclude cut with ID 8291-276745-0093-13116-0_sp0.9 from training. Duration: 21.061125
2022-12-23 04:03:35,744 INFO [train.py:894] (2/4) Epoch 12, batch 3200, loss[loss=0.235, simple_loss=0.3102, pruned_loss=0.0799, over 18712.00 frames. ], tot_loss[loss=0.2119, simple_loss=0.2874, pruned_loss=0.06817, over 3717120.34 frames. ], batch size: 52, lr: 9.49e-03, grad_scale: 16.0
2022-12-23 04:03:42,964 WARNING [train.py:1060] (2/4) Exclude cut with ID 7357-94126-0024-15855-0_sp0.9 from training. Duration: 20.32225
2022-12-23 04:03:43,361 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([2.8859, 2.4937, 1.8574, 1.0180, 1.9778, 2.3058, 1.6508, 1.9809],
       device='cuda:2'), covar=tensor([0.0464, 0.0521, 0.1189, 0.1602, 0.1186, 0.1137, 0.1590, 0.0779],
       device='cuda:2'), in_proj_covar=tensor([0.0165, 0.0183, 0.0204, 0.0197, 0.0208, 0.0194, 0.0209, 0.0195],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-23 04:03:54,512 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([0.0533, 0.9573, 1.1770, 0.3128, 0.5477, 1.2642, 1.2292, 1.1372],
       device='cuda:2'), covar=tensor([0.0717, 0.0392, 0.0305, 0.0414, 0.0474, 0.0469, 0.0279, 0.0592],
       device='cuda:2'), in_proj_covar=tensor([0.0137, 0.0158, 0.0113, 0.0129, 0.0138, 0.0131, 0.0146, 0.0150],
       device='cuda:2'), out_proj_covar=tensor([1.1070e-04, 1.2832e-04, 9.0202e-05, 1.0155e-04, 1.1045e-04, 1.0715e-04,
        1.2038e-04, 1.2075e-04], device='cuda:2')
2022-12-23 04:03:55,536 WARNING [train.py:1060] (2/4) Exclude cut with ID 3033-130750-0096-55598-0_sp1.1 from training. Duration: 0.7545625
2022-12-23 04:04:11,679 WARNING [train.py:1060] (2/4) Exclude cut with ID 4295-39940-0007-92567-0_sp0.9 from training. Duration: 23.9333125
2022-12-23 04:04:34,195 INFO [zipformer.py:660] (2/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=41806.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 04:04:36,558 INFO [scaling.py:679] (2/4) Whitening: num_groups=1, num_channels=384, metric=5.00 vs. limit=5.0
2022-12-23 04:04:44,440 WARNING [train.py:1060] (2/4) Exclude cut with ID 7255-291500-0008-134307-0_sp1.1 from training. Duration: 20.17275
2022-12-23 04:04:48,902 WARNING [train.py:1060] (2/4) Exclude cut with ID 6978-92210-0019-146985-0_sp1.1 from training. Duration: 20.436375
2022-12-23 04:04:53,939 INFO [train.py:894] (2/4) Epoch 12, batch 3250, loss[loss=0.2411, simple_loss=0.3099, pruned_loss=0.08612, over 18680.00 frames. ], tot_loss[loss=0.2107, simple_loss=0.286, pruned_loss=0.06764, over 3715569.68 frames. ], batch size: 78, lr: 9.49e-03, grad_scale: 16.0
2022-12-23 04:05:07,247 INFO [scaling.py:679] (2/4) Whitening: num_groups=8, num_channels=192, metric=2.01 vs. limit=2.0
2022-12-23 04:05:07,768 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 3.427e+02 5.153e+02 6.270e+02 8.432e+02 2.144e+03, threshold=1.254e+03, percent-clipped=7.0
2022-12-23 04:05:45,759 INFO [scaling.py:679] (2/4) Whitening: num_groups=1, num_channels=384, metric=3.73 vs. limit=5.0
2022-12-23 04:06:08,335 INFO [train.py:894] (2/4) Epoch 12, batch 3300, loss[loss=0.2336, simple_loss=0.3067, pruned_loss=0.08027, over 18580.00 frames. ], tot_loss[loss=0.2102, simple_loss=0.2861, pruned_loss=0.06718, over 3716261.08 frames. ], batch size: 78, lr: 9.48e-03, grad_scale: 16.0
2022-12-23 04:06:08,369 WARNING [train.py:1060] (2/4) Exclude cut with ID 4234-40345-0022-142709-0_sp0.9 from training. Duration: 23.1055625
2022-12-23 04:06:09,872 WARNING [train.py:1060] (2/4) Exclude cut with ID 8291-282929-0007-12994-0_sp1.1 from training. Duration: 23.5
2022-12-23 04:06:13,918 INFO [scaling.py:679] (2/4) Whitening: num_groups=8, num_channels=192, metric=1.76 vs. limit=2.0
2022-12-23 04:06:20,374 WARNING [train.py:1060] (2/4) Exclude cut with ID 7255-291500-0009-134308-0_sp0.9 from training. Duration: 26.62775
2022-12-23 04:06:34,507 WARNING [train.py:1060] (2/4) Exclude cut with ID 6951-79737-0018-132285-0 from training. Duration: 21.105
2022-12-23 04:06:37,494 WARNING [train.py:1060] (2/4) Exclude cut with ID 4511-76322-0006-80011-0_sp0.9 from training. Duration: 24.411125
2022-12-23 04:07:04,349 WARNING [train.py:1060] (2/4) Exclude cut with ID 6758-72288-0033-108368-0_sp1.1 from training. Duration: 21.263625
2022-12-23 04:07:04,752 INFO [zipformer.py:660] (2/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=41906.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 04:07:24,849 INFO [train.py:894] (2/4) Epoch 12, batch 3350, loss[loss=0.2413, simple_loss=0.313, pruned_loss=0.08484, over 18626.00 frames. ], tot_loss[loss=0.2112, simple_loss=0.2864, pruned_loss=0.06802, over 3715375.52 frames. ], batch size: 187, lr: 9.48e-03, grad_scale: 16.0
2022-12-23 04:07:37,875 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 2.521e+02 5.179e+02 6.368e+02 8.514e+02 1.589e+03, threshold=1.274e+03, percent-clipped=3.0
2022-12-23 04:07:37,937 WARNING [train.py:1060] (2/4) Exclude cut with ID 4234-40345-0022-142709-0 from training. Duration: 20.795
2022-12-23 04:07:49,278 WARNING [train.py:1060] (2/4) Exclude cut with ID 7357-94126-0021-15852-0 from training. Duration: 24.76
2022-12-23 04:07:49,292 WARNING [train.py:1060] (2/4) Exclude cut with ID 3867-173237-0077-144769-0_sp0.9 from training. Duration: 22.25
2022-12-23 04:08:08,794 INFO [zipformer.py:660] (2/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=41948.0, num_to_drop=1, layers_to_drop={1}
2022-12-23 04:08:13,370 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.8705, 1.3410, 0.6108, 1.3573, 2.1677, 1.2102, 1.5720, 2.0153],
       device='cuda:2'), covar=tensor([0.1506, 0.2136, 0.2573, 0.1489, 0.1637, 0.1579, 0.1443, 0.1441],
       device='cuda:2'), in_proj_covar=tensor([0.0092, 0.0100, 0.0119, 0.0096, 0.0116, 0.0091, 0.0098, 0.0095],
       device='cuda:2'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0003, 0.0004, 0.0003, 0.0003, 0.0003],
       device='cuda:2')
2022-12-23 04:08:14,574 WARNING [train.py:1060] (2/4) Exclude cut with ID 7357-94126-0026-15857-0_sp1.1 from training. Duration: 20.5045625
2022-12-23 04:08:37,764 INFO [zipformer.py:660] (2/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=41967.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 04:08:40,235 INFO [train.py:894] (2/4) Epoch 12, batch 3400, loss[loss=0.1949, simple_loss=0.2693, pruned_loss=0.06029, over 18568.00 frames. ], tot_loss[loss=0.2105, simple_loss=0.286, pruned_loss=0.06755, over 3715965.91 frames. ], batch size: 49, lr: 9.47e-03, grad_scale: 16.0
2022-12-23 04:08:55,154 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.4747, 1.9182, 1.4756, 2.2870, 2.5622, 1.5459, 1.5778, 1.2464],
       device='cuda:2'), covar=tensor([0.2073, 0.1754, 0.1674, 0.0954, 0.1175, 0.1285, 0.1932, 0.1656],
       device='cuda:2'), in_proj_covar=tensor([0.0237, 0.0215, 0.0203, 0.0186, 0.0252, 0.0189, 0.0209, 0.0188],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-23 04:09:19,582 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.6576, 1.1531, 1.5452, 2.8450, 2.1148, 2.3179, 1.0275, 2.0076],
       device='cuda:2'), covar=tensor([0.1770, 0.1835, 0.1666, 0.0705, 0.1146, 0.1215, 0.1989, 0.1201],
       device='cuda:2'), in_proj_covar=tensor([0.0101, 0.0115, 0.0131, 0.0133, 0.0105, 0.0133, 0.0129, 0.0110],
       device='cuda:2'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0004, 0.0003, 0.0003],
       device='cuda:2')
2022-12-23 04:09:42,058 INFO [zipformer.py:660] (2/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=42009.0, num_to_drop=1, layers_to_drop={0}
2022-12-23 04:09:43,228 INFO [zipformer.py:660] (2/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=42010.0, num_to_drop=1, layers_to_drop={2}
2022-12-23 04:09:56,441 INFO [train.py:894] (2/4) Epoch 12, batch 3450, loss[loss=0.216, simple_loss=0.2968, pruned_loss=0.06757, over 18676.00 frames. ], tot_loss[loss=0.2101, simple_loss=0.2856, pruned_loss=0.06723, over 3714629.72 frames. ], batch size: 62, lr: 9.46e-03, grad_scale: 16.0
2022-12-23 04:10:09,836 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 2.367e+02 5.696e+02 6.525e+02 7.684e+02 1.344e+03, threshold=1.305e+03, percent-clipped=1.0
2022-12-23 04:10:58,990 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.0219, 0.9836, 1.1077, 0.9883, 0.6640, 1.6258, 0.7662, 1.0722],
       device='cuda:2'), covar=tensor([0.2580, 0.1687, 0.1535, 0.1662, 0.1180, 0.0323, 0.1599, 0.0770],
       device='cuda:2'), in_proj_covar=tensor([0.0137, 0.0116, 0.0127, 0.0120, 0.0103, 0.0099, 0.0094, 0.0091],
       device='cuda:2'), out_proj_covar=tensor([0.0005, 0.0005, 0.0005, 0.0004, 0.0004, 0.0003, 0.0004, 0.0004],
       device='cuda:2')
2022-12-23 04:11:10,033 INFO [train.py:894] (2/4) Epoch 12, batch 3500, loss[loss=0.2864, simple_loss=0.3396, pruned_loss=0.1166, over 18570.00 frames. ], tot_loss[loss=0.2107, simple_loss=0.2864, pruned_loss=0.06752, over 3715150.49 frames. ], batch size: 178, lr: 9.46e-03, grad_scale: 16.0
2022-12-23 04:11:30,158 WARNING [train.py:1060] (2/4) Exclude cut with ID 7859-102521-0017-7548-0_sp1.1 from training. Duration: 22.2954375
2022-12-23 04:11:41,996 INFO [train.py:894] (2/4) Epoch 13, batch 0, loss[loss=0.2088, simple_loss=0.2868, pruned_loss=0.06538, over 18674.00 frames. ], tot_loss[loss=0.2088, simple_loss=0.2868, pruned_loss=0.06538, over 18674.00 frames. ], batch size: 48, lr: 9.09e-03, grad_scale: 16.0
2022-12-23 04:11:41,996 INFO [train.py:919] (2/4) Computing validation loss
2022-12-23 04:11:52,900 INFO [train.py:928] (2/4) Epoch 13, validation: loss=0.1708, simple_loss=0.27, pruned_loss=0.03586, over 944034.00 frames. 
2022-12-23 04:11:52,901 INFO [train.py:929] (2/4) Maximum memory allocated so far is 24680MB
2022-12-23 04:12:44,107 WARNING [train.py:1060] (2/4) Exclude cut with ID 298-126791-0067-24026-0_sp0.9 from training. Duration: 21.438875
2022-12-23 04:12:50,438 WARNING [train.py:1060] (2/4) Exclude cut with ID 5652-39938-0025-23684-0_sp0.9 from training. Duration: 22.2055625
2022-12-23 04:13:10,458 INFO [train.py:894] (2/4) Epoch 13, batch 50, loss[loss=0.1736, simple_loss=0.2548, pruned_loss=0.04623, over 18667.00 frames. ], tot_loss[loss=0.1976, simple_loss=0.2832, pruned_loss=0.05606, over 838299.35 frames. ], batch size: 48, lr: 9.08e-03, grad_scale: 16.0
2022-12-23 04:13:14,625 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 2.917e+02 4.264e+02 5.110e+02 6.496e+02 1.876e+03, threshold=1.022e+03, percent-clipped=1.0
2022-12-23 04:14:15,686 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.4486, 1.1580, 1.7926, 2.8800, 2.0333, 2.3739, 0.7348, 1.9697],
       device='cuda:2'), covar=tensor([0.1834, 0.1814, 0.1450, 0.0548, 0.1131, 0.1090, 0.2281, 0.1197],
       device='cuda:2'), in_proj_covar=tensor([0.0100, 0.0114, 0.0129, 0.0131, 0.0103, 0.0130, 0.0127, 0.0109],
       device='cuda:2'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003],
       device='cuda:2')
2022-12-23 04:14:18,779 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.9634, 1.9131, 1.9936, 1.0741, 2.1820, 2.1724, 1.5203, 2.4996],
       device='cuda:2'), covar=tensor([0.1167, 0.1660, 0.1344, 0.2074, 0.0723, 0.1141, 0.2206, 0.0529],
       device='cuda:2'), in_proj_covar=tensor([0.0195, 0.0199, 0.0199, 0.0191, 0.0175, 0.0210, 0.0208, 0.0179],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-23 04:14:26,124 INFO [train.py:894] (2/4) Epoch 13, batch 100, loss[loss=0.2019, simple_loss=0.2953, pruned_loss=0.05426, over 18533.00 frames. ], tot_loss[loss=0.1943, simple_loss=0.2796, pruned_loss=0.05446, over 1476574.55 frames. ], batch size: 55, lr: 9.08e-03, grad_scale: 16.0
2022-12-23 04:15:42,618 INFO [train.py:894] (2/4) Epoch 13, batch 150, loss[loss=0.1922, simple_loss=0.28, pruned_loss=0.05224, over 18556.00 frames. ], tot_loss[loss=0.1921, simple_loss=0.2772, pruned_loss=0.05347, over 1972167.58 frames. ], batch size: 99, lr: 9.07e-03, grad_scale: 16.0
2022-12-23 04:15:47,720 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 2.470e+02 3.667e+02 4.662e+02 5.480e+02 1.007e+03, threshold=9.324e+02, percent-clipped=0.0
2022-12-23 04:15:52,243 WARNING [train.py:1060] (2/4) Exclude cut with ID 7859-102521-0017-7548-0 from training. Duration: 24.525
2022-12-23 04:16:01,915 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.5828, 1.3227, 1.3785, 1.8794, 1.5760, 3.4501, 1.3718, 1.4267],
       device='cuda:2'), covar=tensor([0.0844, 0.1731, 0.1063, 0.0847, 0.1466, 0.0185, 0.1296, 0.1495],
       device='cuda:2'), in_proj_covar=tensor([0.0072, 0.0082, 0.0073, 0.0073, 0.0090, 0.0071, 0.0084, 0.0075],
       device='cuda:2'), out_proj_covar=tensor([0.0003, 0.0004, 0.0003, 0.0003, 0.0004, 0.0003, 0.0004, 0.0004],
       device='cuda:2')
2022-12-23 04:16:27,853 WARNING [train.py:1060] (2/4) Exclude cut with ID 3699-47246-0007-3408-0_sp0.9 from training. Duration: 20.26675
2022-12-23 04:16:37,766 INFO [zipformer.py:660] (2/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=42262.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 04:16:41,787 WARNING [train.py:1060] (2/4) Exclude cut with ID 7859-102521-0017-7548-0_sp0.9 from training. Duration: 27.25
2022-12-23 04:16:56,932 INFO [train.py:894] (2/4) Epoch 13, batch 200, loss[loss=0.2055, simple_loss=0.2899, pruned_loss=0.06053, over 18721.00 frames. ], tot_loss[loss=0.1918, simple_loss=0.2771, pruned_loss=0.05323, over 2358483.23 frames. ], batch size: 54, lr: 9.07e-03, grad_scale: 16.0
2022-12-23 04:17:41,277 INFO [zipformer.py:660] (2/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=42304.0, num_to_drop=1, layers_to_drop={2}
2022-12-23 04:17:50,534 INFO [zipformer.py:660] (2/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=42310.0, num_to_drop=1, layers_to_drop={2}
2022-12-23 04:17:54,998 WARNING [train.py:1060] (2/4) Exclude cut with ID 6426-64292-0017-15984-0 from training. Duration: 21.68
2022-12-23 04:18:07,805 WARNING [train.py:1060] (2/4) Exclude cut with ID 4278-13270-0007-59342-0 from training. Duration: 21.6300625
2022-12-23 04:18:09,672 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.1763, 1.2572, 1.0669, 1.5071, 1.5628, 1.2367, 0.9108, 1.0543],
       device='cuda:2'), covar=tensor([0.1781, 0.1798, 0.1555, 0.1072, 0.1279, 0.1102, 0.2172, 0.1394],
       device='cuda:2'), in_proj_covar=tensor([0.0238, 0.0217, 0.0204, 0.0187, 0.0252, 0.0190, 0.0210, 0.0190],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-23 04:18:13,377 INFO [train.py:894] (2/4) Epoch 13, batch 250, loss[loss=0.2255, simple_loss=0.302, pruned_loss=0.07448, over 18732.00 frames. ], tot_loss[loss=0.1902, simple_loss=0.2757, pruned_loss=0.05237, over 2658878.12 frames. ], batch size: 54, lr: 9.06e-03, grad_scale: 16.0
2022-12-23 04:18:18,286 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 2.488e+02 3.659e+02 4.624e+02 5.942e+02 1.816e+03, threshold=9.248e+02, percent-clipped=5.0
2022-12-23 04:18:32,849 WARNING [train.py:1060] (2/4) Exclude cut with ID 4278-13270-0007-59342-0_sp0.9 from training. Duration: 24.033375
2022-12-23 04:19:02,831 INFO [zipformer.py:660] (2/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=42358.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 04:19:19,759 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([2.1535, 1.5338, 1.7429, 1.7987, 2.1191, 2.1126, 1.9205, 1.5751],
       device='cuda:2'), covar=tensor([0.1810, 0.2648, 0.2065, 0.2533, 0.1486, 0.0746, 0.2582, 0.1023],
       device='cuda:2'), in_proj_covar=tensor([0.0254, 0.0288, 0.0267, 0.0300, 0.0284, 0.0240, 0.0311, 0.0224],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-23 04:19:29,151 INFO [train.py:894] (2/4) Epoch 13, batch 300, loss[loss=0.1764, simple_loss=0.2624, pruned_loss=0.04525, over 18672.00 frames. ], tot_loss[loss=0.1891, simple_loss=0.2745, pruned_loss=0.05187, over 2892936.94 frames. ], batch size: 46, lr: 9.06e-03, grad_scale: 8.0
2022-12-23 04:19:29,223 WARNING [train.py:1060] (2/4) Exclude cut with ID 4278-13270-0009-59344-0 from training. Duration: 22.905
2022-12-23 04:19:30,600 WARNING [train.py:1060] (2/4) Exclude cut with ID 5622-44585-0006-90525-0_sp1.1 from training. Duration: 23.4318125
2022-12-23 04:19:35,743 INFO [zipformer.py:660] (2/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=42379.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 04:20:27,266 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([2.5036, 1.7729, 0.7903, 2.0510, 2.7301, 1.7021, 2.2594, 2.2909],
       device='cuda:2'), covar=tensor([0.1408, 0.1969, 0.2547, 0.1362, 0.1394, 0.1455, 0.1294, 0.1535],
       device='cuda:2'), in_proj_covar=tensor([0.0092, 0.0100, 0.0118, 0.0096, 0.0114, 0.0090, 0.0098, 0.0095],
       device='cuda:2'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003],
       device='cuda:2')
2022-12-23 04:20:46,370 INFO [train.py:894] (2/4) Epoch 13, batch 350, loss[loss=0.1632, simple_loss=0.2484, pruned_loss=0.03897, over 18703.00 frames. ], tot_loss[loss=0.1891, simple_loss=0.2743, pruned_loss=0.05198, over 3073848.33 frames. ], batch size: 46, lr: 9.05e-03, grad_scale: 8.0
2022-12-23 04:20:52,407 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 2.737e+02 3.882e+02 4.906e+02 5.915e+02 1.901e+03, threshold=9.812e+02, percent-clipped=4.0
2022-12-23 04:21:09,394 INFO [zipformer.py:660] (2/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=42440.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 04:21:27,868 WARNING [train.py:1060] (2/4) Exclude cut with ID 4278-13270-0009-59344-0_sp1.1 from training. Duration: 20.82275
2022-12-23 04:21:29,288 WARNING [train.py:1060] (2/4) Exclude cut with ID 4278-13270-0009-59344-0_sp0.9 from training. Duration: 25.45
2022-12-23 04:22:02,672 INFO [train.py:894] (2/4) Epoch 13, batch 400, loss[loss=0.1994, simple_loss=0.2878, pruned_loss=0.05554, over 18496.00 frames. ], tot_loss[loss=0.1905, simple_loss=0.2751, pruned_loss=0.0529, over 3216281.19 frames. ], batch size: 52, lr: 9.05e-03, grad_scale: 8.0
2022-12-23 04:22:11,804 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.4871, 1.9222, 1.3550, 2.3685, 2.5256, 1.5160, 1.4824, 1.1872],
       device='cuda:2'), covar=tensor([0.2027, 0.1755, 0.1692, 0.0878, 0.1166, 0.1233, 0.2109, 0.1616],
       device='cuda:2'), in_proj_covar=tensor([0.0241, 0.0218, 0.0206, 0.0188, 0.0253, 0.0192, 0.0213, 0.0192],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-23 04:22:21,895 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.6439, 1.5338, 1.6072, 1.5917, 1.1854, 3.3943, 1.5780, 1.9864],
       device='cuda:2'), covar=tensor([0.3209, 0.2079, 0.1836, 0.2023, 0.1372, 0.0178, 0.1438, 0.0793],
       device='cuda:2'), in_proj_covar=tensor([0.0137, 0.0117, 0.0127, 0.0120, 0.0103, 0.0099, 0.0095, 0.0091],
       device='cuda:2'), out_proj_covar=tensor([0.0005, 0.0005, 0.0005, 0.0004, 0.0004, 0.0003, 0.0004, 0.0004],
       device='cuda:2')
2022-12-23 04:22:30,185 WARNING [train.py:1060] (2/4) Exclude cut with ID 5622-44585-0006-90525-0 from training. Duration: 25.775
2022-12-23 04:22:48,558 INFO [zipformer.py:660] (2/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=42506.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 04:22:52,776 WARNING [train.py:1060] (2/4) Exclude cut with ID 2929-85685-0071-62375-0_sp0.9 from training. Duration: 22.25
2022-12-23 04:23:17,535 INFO [train.py:894] (2/4) Epoch 13, batch 450, loss[loss=0.1636, simple_loss=0.247, pruned_loss=0.04007, over 18520.00 frames. ], tot_loss[loss=0.1923, simple_loss=0.2768, pruned_loss=0.05388, over 3326366.14 frames. ], batch size: 44, lr: 9.04e-03, grad_scale: 8.0
2022-12-23 04:23:21,648 WARNING [train.py:1060] (2/4) Exclude cut with ID 3972-170212-0014-23379-0 from training. Duration: 26.205
2022-12-23 04:23:22,901 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 2.137e+02 3.894e+02 4.578e+02 5.727e+02 1.078e+03, threshold=9.156e+02, percent-clipped=2.0
2022-12-23 04:23:39,343 WARNING [train.py:1060] (2/4) Exclude cut with ID 5239-32139-0047-9341-0_sp0.9 from training. Duration: 30.1555625
2022-12-23 04:23:44,968 WARNING [train.py:1060] (2/4) Exclude cut with ID 1265-135635-0050-6781-0_sp0.9 from training. Duration: 21.8333125
2022-12-23 04:23:54,439 WARNING [train.py:1060] (2/4) Exclude cut with ID 1914-133440-0024-94914-0_sp1.1 from training. Duration: 20.6545625
2022-12-23 04:24:12,983 INFO [zipformer.py:660] (2/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=42562.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 04:24:13,081 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([2.7074, 2.2733, 1.8394, 0.7881, 1.8244, 2.0940, 1.6292, 1.9644],
       device='cuda:2'), covar=tensor([0.0548, 0.0547, 0.1221, 0.1699, 0.1332, 0.1362, 0.1594, 0.0732],
       device='cuda:2'), in_proj_covar=tensor([0.0167, 0.0183, 0.0204, 0.0197, 0.0209, 0.0195, 0.0208, 0.0194],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-23 04:24:21,386 INFO [zipformer.py:660] (2/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=42567.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 04:24:26,420 INFO [scaling.py:679] (2/4) Whitening: num_groups=1, num_channels=384, metric=2.73 vs. limit=5.0
2022-12-23 04:24:34,375 INFO [train.py:894] (2/4) Epoch 13, batch 500, loss[loss=0.2136, simple_loss=0.3001, pruned_loss=0.06353, over 18672.00 frames. ], tot_loss[loss=0.1935, simple_loss=0.2779, pruned_loss=0.05453, over 3412319.74 frames. ], batch size: 69, lr: 9.04e-03, grad_scale: 8.0
2022-12-23 04:24:34,449 WARNING [train.py:1060] (2/4) Exclude cut with ID 7395-89880-0045-39920-0_sp0.9 from training. Duration: 20.52225
2022-12-23 04:24:43,849 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([3.2125, 2.4110, 1.7489, 3.0189, 2.0950, 2.3356, 2.3825, 3.4779],
       device='cuda:2'), covar=tensor([0.1544, 0.2891, 0.1622, 0.2557, 0.3409, 0.0895, 0.2841, 0.0553],
       device='cuda:2'), in_proj_covar=tensor([0.0278, 0.0273, 0.0230, 0.0343, 0.0256, 0.0217, 0.0268, 0.0195],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0003, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-23 04:24:56,220 WARNING [train.py:1060] (2/4) Exclude cut with ID 3972-170212-0014-23379-0_sp0.9 from training. Duration: 29.1166875
2022-12-23 04:25:17,546 INFO [zipformer.py:660] (2/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=42604.0, num_to_drop=1, layers_to_drop={1}
2022-12-23 04:25:26,252 INFO [zipformer.py:660] (2/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=42610.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 04:25:33,149 INFO [scaling.py:679] (2/4) Whitening: num_groups=8, num_channels=192, metric=2.00 vs. limit=2.0
2022-12-23 04:25:49,360 INFO [train.py:894] (2/4) Epoch 13, batch 550, loss[loss=0.2032, simple_loss=0.2877, pruned_loss=0.05936, over 18526.00 frames. ], tot_loss[loss=0.194, simple_loss=0.2787, pruned_loss=0.0547, over 3478690.53 frames. ], batch size: 64, lr: 9.03e-03, grad_scale: 8.0
2022-12-23 04:25:52,841 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([2.0183, 2.3305, 1.4320, 2.5036, 2.3629, 2.0329, 3.1045, 2.1275],
       device='cuda:2'), covar=tensor([0.0879, 0.1501, 0.2606, 0.1670, 0.1507, 0.0861, 0.0869, 0.1135],
       device='cuda:2'), in_proj_covar=tensor([0.0172, 0.0194, 0.0235, 0.0277, 0.0224, 0.0183, 0.0205, 0.0191],
       device='cuda:2'), out_proj_covar=tensor([0.0001, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-23 04:25:55,203 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 2.626e+02 4.162e+02 5.238e+02 6.247e+02 1.388e+03, threshold=1.048e+03, percent-clipped=3.0
2022-12-23 04:25:55,291 WARNING [train.py:1060] (2/4) Exclude cut with ID 543-133211-0007-59831-0_sp0.9 from training. Duration: 21.388875
2022-12-23 04:26:14,057 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([2.4282, 1.8253, 1.9192, 2.0801, 2.2034, 2.1564, 2.2062, 1.5660],
       device='cuda:2'), covar=tensor([0.1667, 0.2716, 0.2003, 0.2488, 0.1646, 0.0887, 0.2599, 0.1143],
       device='cuda:2'), in_proj_covar=tensor([0.0251, 0.0287, 0.0263, 0.0296, 0.0283, 0.0238, 0.0310, 0.0224],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-23 04:26:24,831 INFO [zipformer.py:660] (2/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=42649.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 04:26:29,012 INFO [zipformer.py:660] (2/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=42652.0, num_to_drop=1, layers_to_drop={1}
2022-12-23 04:26:30,321 WARNING [train.py:1060] (2/4) Exclude cut with ID 1914-133440-0024-94914-0 from training. Duration: 22.72
2022-12-23 04:26:31,993 WARNING [train.py:1060] (2/4) Exclude cut with ID 1914-133440-0031-94921-0_sp0.9 from training. Duration: 22.7444375
2022-12-23 04:26:52,049 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.1790, 1.5477, 1.1702, 1.8052, 1.7096, 1.4262, 1.0154, 1.1710],
       device='cuda:2'), covar=tensor([0.1987, 0.1833, 0.1638, 0.1074, 0.1181, 0.1213, 0.2050, 0.1548],
       device='cuda:2'), in_proj_covar=tensor([0.0237, 0.0216, 0.0204, 0.0187, 0.0252, 0.0190, 0.0211, 0.0190],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-23 04:27:04,016 INFO [train.py:894] (2/4) Epoch 13, batch 600, loss[loss=0.2299, simple_loss=0.3108, pruned_loss=0.07448, over 18689.00 frames. ], tot_loss[loss=0.1953, simple_loss=0.2798, pruned_loss=0.05538, over 3531369.97 frames. ], batch size: 78, lr: 9.03e-03, grad_scale: 8.0
2022-12-23 04:27:13,856 WARNING [train.py:1060] (2/4) Exclude cut with ID 4133-6541-0027-40495-0_sp1.1 from training. Duration: 0.9681875
2022-12-23 04:27:18,139 WARNING [train.py:1060] (2/4) Exclude cut with ID 6330-62851-0022-91297-0_sp0.9 from training. Duration: 22.3166875
2022-12-23 04:27:24,054 WARNING [train.py:1060] (2/4) Exclude cut with ID 543-133212-0015-59917-0_sp0.9 from training. Duration: 21.8166875
2022-12-23 04:27:34,521 INFO [scaling.py:679] (2/4) Whitening: num_groups=8, num_channels=96, metric=1.40 vs. limit=2.0
2022-12-23 04:27:56,162 INFO [zipformer.py:660] (2/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=42710.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 04:27:59,660 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.8865, 1.8002, 2.0701, 1.1553, 2.0313, 2.2894, 1.4936, 2.5161],
       device='cuda:2'), covar=tensor([0.1035, 0.1517, 0.1117, 0.1812, 0.0686, 0.0980, 0.1924, 0.0489],
       device='cuda:2'), in_proj_covar=tensor([0.0195, 0.0202, 0.0202, 0.0192, 0.0176, 0.0211, 0.0208, 0.0180],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-23 04:28:18,785 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([2.4693, 2.5131, 2.7609, 1.4497, 2.7073, 3.2271, 1.9031, 3.3762],
       device='cuda:2'), covar=tensor([0.1128, 0.1476, 0.1428, 0.2175, 0.0772, 0.1057, 0.1999, 0.0483],
       device='cuda:2'), in_proj_covar=tensor([0.0197, 0.0203, 0.0203, 0.0194, 0.0177, 0.0212, 0.0209, 0.0181],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-23 04:28:19,777 INFO [train.py:894] (2/4) Epoch 13, batch 650, loss[loss=0.192, simple_loss=0.282, pruned_loss=0.05103, over 18538.00 frames. ], tot_loss[loss=0.1943, simple_loss=0.2788, pruned_loss=0.05493, over 3570618.67 frames. ], batch size: 58, lr: 9.02e-03, grad_scale: 8.0
2022-12-23 04:28:25,609 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 2.822e+02 4.139e+02 5.026e+02 6.152e+02 1.523e+03, threshold=1.005e+03, percent-clipped=4.0
2022-12-23 04:28:34,829 INFO [zipformer.py:660] (2/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=42735.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 04:29:07,469 WARNING [train.py:1060] (2/4) Exclude cut with ID 4957-30119-0041-23990-0_sp0.9 from training. Duration: 20.22775
2022-12-23 04:29:35,747 INFO [train.py:894] (2/4) Epoch 13, batch 700, loss[loss=0.1636, simple_loss=0.2458, pruned_loss=0.04071, over 18591.00 frames. ], tot_loss[loss=0.1942, simple_loss=0.2785, pruned_loss=0.05497, over 3602305.06 frames. ], batch size: 45, lr: 9.01e-03, grad_scale: 8.0
2022-12-23 04:29:52,973 WARNING [train.py:1060] (2/4) Exclude cut with ID 5239-32139-0047-9341-0_sp1.1 from training. Duration: 24.67275
2022-12-23 04:30:08,761 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.1972, 1.0981, 0.7097, 1.3162, 1.3504, 2.3808, 1.0715, 1.3205],
       device='cuda:2'), covar=tensor([0.1153, 0.2543, 0.1478, 0.1065, 0.1968, 0.0392, 0.2029, 0.2149],
       device='cuda:2'), in_proj_covar=tensor([0.0072, 0.0082, 0.0074, 0.0074, 0.0091, 0.0072, 0.0085, 0.0076],
       device='cuda:2'), out_proj_covar=tensor([0.0003, 0.0004, 0.0003, 0.0003, 0.0004, 0.0003, 0.0004, 0.0004],
       device='cuda:2')
2022-12-23 04:30:22,274 WARNING [train.py:1060] (2/4) Exclude cut with ID 3082-165428-0081-50734-0_sp0.9 from training. Duration: 21.8055625
2022-12-23 04:30:51,742 INFO [train.py:894] (2/4) Epoch 13, batch 750, loss[loss=0.1596, simple_loss=0.2369, pruned_loss=0.0412, over 18403.00 frames. ], tot_loss[loss=0.1938, simple_loss=0.2782, pruned_loss=0.0547, over 3626797.94 frames. ], batch size: 42, lr: 9.01e-03, grad_scale: 8.0
2022-12-23 04:30:57,837 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 2.780e+02 4.039e+02 4.835e+02 6.109e+02 2.370e+03, threshold=9.671e+02, percent-clipped=2.0
2022-12-23 04:31:00,761 WARNING [train.py:1060] (2/4) Exclude cut with ID 3340-169293-0054-76830-0_sp0.9 from training. Duration: 22.6666875
2022-12-23 04:31:43,102 INFO [zipformer.py:660] (2/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=42859.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 04:31:47,250 INFO [zipformer.py:660] (2/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=42862.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 04:31:59,931 WARNING [train.py:1060] (2/4) Exclude cut with ID 2411-132532-0017-82279-0_sp1.1 from training. Duration: 0.9681875
2022-12-23 04:32:03,684 INFO [scaling.py:679] (2/4) Whitening: num_groups=8, num_channels=96, metric=1.79 vs. limit=2.0
2022-12-23 04:32:07,701 INFO [train.py:894] (2/4) Epoch 13, batch 800, loss[loss=0.2406, simple_loss=0.315, pruned_loss=0.08314, over 18662.00 frames. ], tot_loss[loss=0.1952, simple_loss=0.2797, pruned_loss=0.05538, over 3647622.19 frames. ], batch size: 174, lr: 9.00e-03, grad_scale: 8.0
2022-12-23 04:32:13,990 INFO [zipformer.py:660] (2/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=42879.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 04:32:24,535 WARNING [train.py:1060] (2/4) Exclude cut with ID 6330-62850-0007-91323-0 from training. Duration: 22.485
2022-12-23 04:33:02,867 WARNING [train.py:1060] (2/4) Exclude cut with ID 3972-170212-0014-23379-0_sp1.1 from training. Duration: 23.82275
2022-12-23 04:33:14,997 INFO [zipformer.py:660] (2/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=42920.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 04:33:16,036 WARNING [train.py:1060] (2/4) Exclude cut with ID 4860-13185-0032-76709-0 from training. Duration: 20.77
2022-12-23 04:33:21,782 INFO [train.py:894] (2/4) Epoch 13, batch 850, loss[loss=0.1901, simple_loss=0.2812, pruned_loss=0.04955, over 18518.00 frames. ], tot_loss[loss=0.1947, simple_loss=0.2792, pruned_loss=0.05508, over 3661117.17 frames. ], batch size: 52, lr: 9.00e-03, grad_scale: 8.0
2022-12-23 04:33:23,477 WARNING [train.py:1060] (2/4) Exclude cut with ID 6426-64292-0017-15984-0_sp0.9 from training. Duration: 24.088875
2022-12-23 04:33:27,849 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 2.652e+02 4.085e+02 4.833e+02 5.843e+02 1.315e+03, threshold=9.667e+02, percent-clipped=6.0
2022-12-23 04:33:44,871 INFO [zipformer.py:660] (2/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=42940.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 04:33:54,520 WARNING [train.py:1060] (2/4) Exclude cut with ID 6330-62850-0007-91323-0_sp1.1 from training. Duration: 20.4409375
2022-12-23 04:34:32,782 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([0.0619, 0.9437, 1.1428, 0.5317, 0.5475, 1.2167, 1.2038, 1.1764],
       device='cuda:2'), covar=tensor([0.0689, 0.0324, 0.0324, 0.0342, 0.0442, 0.0435, 0.0262, 0.0472],
       device='cuda:2'), in_proj_covar=tensor([0.0140, 0.0159, 0.0115, 0.0130, 0.0141, 0.0132, 0.0151, 0.0150],
       device='cuda:2'), out_proj_covar=tensor([1.1275e-04, 1.2862e-04, 9.1442e-05, 1.0202e-04, 1.1281e-04, 1.0793e-04,
        1.2394e-04, 1.2024e-04], device='cuda:2')
2022-12-23 04:34:36,971 INFO [train.py:894] (2/4) Epoch 13, batch 900, loss[loss=0.1832, simple_loss=0.2571, pruned_loss=0.05459, over 18599.00 frames. ], tot_loss[loss=0.1945, simple_loss=0.279, pruned_loss=0.05499, over 3673278.98 frames. ], batch size: 45, lr: 8.99e-03, grad_scale: 8.0
2022-12-23 04:34:45,808 INFO [scaling.py:679] (2/4) Whitening: num_groups=8, num_channels=192, metric=1.93 vs. limit=2.0
2022-12-23 04:35:09,710 WARNING [train.py:1060] (2/4) Exclude cut with ID 2929-85685-0062-62366-0_sp0.9 from training. Duration: 22.511125
2022-12-23 04:35:11,179 WARNING [train.py:1060] (2/4) Exclude cut with ID 7395-89880-0031-39906-0 from training. Duration: 20.675
2022-12-23 04:35:22,827 INFO [zipformer.py:660] (2/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=43005.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 04:35:33,533 INFO [scaling.py:679] (2/4) Whitening: num_groups=1, num_channels=384, metric=3.28 vs. limit=5.0
2022-12-23 04:35:53,122 INFO [train.py:894] (2/4) Epoch 13, batch 950, loss[loss=0.1845, simple_loss=0.274, pruned_loss=0.04751, over 18720.00 frames. ], tot_loss[loss=0.195, simple_loss=0.2797, pruned_loss=0.05513, over 3681915.68 frames. ], batch size: 52, lr: 8.99e-03, grad_scale: 8.0
2022-12-23 04:35:59,524 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 2.591e+02 3.779e+02 4.609e+02 5.645e+02 1.005e+03, threshold=9.218e+02, percent-clipped=1.0
2022-12-23 04:36:08,604 INFO [zipformer.py:660] (2/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=43035.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 04:36:50,772 WARNING [train.py:1060] (2/4) Exclude cut with ID 6330-62850-0007-91323-0_sp0.9 from training. Duration: 24.9833125
2022-12-23 04:37:08,455 INFO [train.py:894] (2/4) Epoch 13, batch 1000, loss[loss=0.1973, simple_loss=0.2882, pruned_loss=0.05323, over 18650.00 frames. ], tot_loss[loss=0.1945, simple_loss=0.2795, pruned_loss=0.05476, over 3689210.98 frames. ], batch size: 60, lr: 8.98e-03, grad_scale: 8.0
2022-12-23 04:37:20,507 INFO [zipformer.py:660] (2/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=43083.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 04:37:23,219 WARNING [train.py:1060] (2/4) Exclude cut with ID 5239-32139-0047-9341-0 from training. Duration: 27.14
2022-12-23 04:37:37,434 WARNING [train.py:1060] (2/4) Exclude cut with ID 2929-85685-0044-62348-0 from training. Duration: 22.44
2022-12-23 04:37:50,200 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.6100, 1.0756, 0.6320, 1.2062, 1.9445, 0.8713, 1.2526, 1.6324],
       device='cuda:2'), covar=tensor([0.1663, 0.2214, 0.2345, 0.1629, 0.1773, 0.1757, 0.1497, 0.1530],
       device='cuda:2'), in_proj_covar=tensor([0.0092, 0.0100, 0.0118, 0.0097, 0.0115, 0.0091, 0.0097, 0.0095],
       device='cuda:2'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0003, 0.0004, 0.0003, 0.0003, 0.0003],
       device='cuda:2')
2022-12-23 04:38:23,698 INFO [train.py:894] (2/4) Epoch 13, batch 1050, loss[loss=0.1973, simple_loss=0.2808, pruned_loss=0.05693, over 18588.00 frames. ], tot_loss[loss=0.1948, simple_loss=0.2797, pruned_loss=0.05497, over 3694059.66 frames. ], batch size: 51, lr: 8.98e-03, grad_scale: 8.0
2022-12-23 04:38:25,688 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.8204, 1.1137, 1.4261, 1.6424, 1.8459, 1.8700, 1.7610, 1.3912],
       device='cuda:2'), covar=tensor([0.2527, 0.3449, 0.2880, 0.2771, 0.2107, 0.1171, 0.2969, 0.1520],
       device='cuda:2'), in_proj_covar=tensor([0.0254, 0.0289, 0.0265, 0.0301, 0.0285, 0.0239, 0.0312, 0.0224],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-23 04:38:29,586 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 2.711e+02 3.874e+02 4.915e+02 6.506e+02 1.252e+03, threshold=9.831e+02, percent-clipped=4.0
2022-12-23 04:38:44,963 INFO [scaling.py:679] (2/4) Whitening: num_groups=1, num_channels=384, metric=2.91 vs. limit=5.0
2022-12-23 04:38:57,184 WARNING [train.py:1060] (2/4) Exclude cut with ID 2929-85685-0060-62364-0_sp0.9 from training. Duration: 21.361125
2022-12-23 04:39:00,324 INFO [scaling.py:679] (2/4) Whitening: num_groups=8, num_channels=192, metric=1.90 vs. limit=2.0
2022-12-23 04:39:03,980 WARNING [train.py:1060] (2/4) Exclude cut with ID 2929-85685-0079-62383-0_sp1.1 from training. Duration: 27.0318125
2022-12-23 04:39:12,946 WARNING [train.py:1060] (2/4) Exclude cut with ID 5622-44585-0006-90525-0_sp0.9 from training. Duration: 28.638875
2022-12-23 04:39:19,105 INFO [zipformer.py:660] (2/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=43162.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 04:39:26,288 WARNING [train.py:1060] (2/4) Exclude cut with ID 3340-169293-0054-76830-0 from training. Duration: 20.4
2022-12-23 04:39:27,966 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([2.6250, 2.0372, 0.5550, 2.1639, 2.8835, 1.8883, 2.7481, 2.5685],
       device='cuda:2'), covar=tensor([0.1399, 0.1947, 0.2747, 0.1390, 0.1352, 0.1516, 0.1201, 0.1529],
       device='cuda:2'), in_proj_covar=tensor([0.0091, 0.0099, 0.0117, 0.0096, 0.0113, 0.0091, 0.0096, 0.0094],
       device='cuda:2'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003],
       device='cuda:2')
2022-12-23 04:39:38,923 INFO [train.py:894] (2/4) Epoch 13, batch 1100, loss[loss=0.2523, simple_loss=0.3248, pruned_loss=0.08996, over 18604.00 frames. ], tot_loss[loss=0.1948, simple_loss=0.2794, pruned_loss=0.05509, over 3697677.77 frames. ], batch size: 171, lr: 8.97e-03, grad_scale: 8.0
2022-12-23 04:39:57,003 WARNING [train.py:1060] (2/4) Exclude cut with ID 2929-85685-0071-62375-0 from training. Duration: 20.025
2022-12-23 04:39:57,011 WARNING [train.py:1060] (2/4) Exclude cut with ID 2364-131735-0112-64612-0_sp0.9 from training. Duration: 20.488875
2022-12-23 04:40:01,608 WARNING [train.py:1060] (2/4) Exclude cut with ID 2929-85685-0079-62383-0 from training. Duration: 29.735
2022-12-23 04:40:33,943 INFO [zipformer.py:660] (2/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=43210.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 04:40:41,300 INFO [zipformer.py:660] (2/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=43215.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 04:40:56,300 INFO [train.py:894] (2/4) Epoch 13, batch 1150, loss[loss=0.1921, simple_loss=0.2873, pruned_loss=0.04843, over 18601.00 frames. ], tot_loss[loss=0.1944, simple_loss=0.2795, pruned_loss=0.05462, over 3701382.07 frames. ], batch size: 56, lr: 8.97e-03, grad_scale: 8.0
2022-12-23 04:41:02,036 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 2.407e+02 3.889e+02 4.613e+02 6.003e+02 1.099e+03, threshold=9.226e+02, percent-clipped=3.0
2022-12-23 04:41:11,033 INFO [zipformer.py:660] (2/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=43235.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 04:41:26,138 WARNING [train.py:1060] (2/4) Exclude cut with ID 7276-92427-0014-12983-0_sp0.9 from training. Duration: 21.3055625
2022-12-23 04:41:28,111 WARNING [train.py:1060] (2/4) Exclude cut with ID 1025-75365-0008-79168-0_sp0.9 from training. Duration: 22.0666875
2022-12-23 04:41:56,405 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([2.4885, 3.6284, 3.5286, 1.2109, 3.6641, 2.6651, 0.6483, 2.3022],
       device='cuda:2'), covar=tensor([0.2029, 0.0846, 0.1280, 0.3829, 0.0809, 0.1035, 0.4829, 0.1587],
       device='cuda:2'), in_proj_covar=tensor([0.0136, 0.0127, 0.0148, 0.0120, 0.0128, 0.0104, 0.0139, 0.0109],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-23 04:42:11,536 INFO [train.py:894] (2/4) Epoch 13, batch 1200, loss[loss=0.2005, simple_loss=0.295, pruned_loss=0.05294, over 18624.00 frames. ], tot_loss[loss=0.1936, simple_loss=0.279, pruned_loss=0.05408, over 3704208.34 frames. ], batch size: 53, lr: 8.96e-03, grad_scale: 8.0
2022-12-23 04:42:57,139 INFO [zipformer.py:660] (2/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=43305.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 04:43:16,810 WARNING [train.py:1060] (2/4) Exclude cut with ID 2929-85685-0062-62366-0 from training. Duration: 20.26
2022-12-23 04:43:26,883 INFO [train.py:894] (2/4) Epoch 13, batch 1250, loss[loss=0.2215, simple_loss=0.3058, pruned_loss=0.06854, over 18672.00 frames. ], tot_loss[loss=0.1926, simple_loss=0.2781, pruned_loss=0.05362, over 3705813.10 frames. ], batch size: 62, lr: 8.96e-03, grad_scale: 8.0
2022-12-23 04:43:29,619 WARNING [train.py:1060] (2/4) Exclude cut with ID 5239-32139-0030-9324-0_sp0.9 from training. Duration: 21.3444375
2022-12-23 04:43:32,666 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 2.493e+02 3.857e+02 4.657e+02 6.544e+02 3.664e+03, threshold=9.315e+02, percent-clipped=7.0
2022-12-23 04:44:09,722 INFO [zipformer.py:660] (2/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=43353.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 04:44:27,111 WARNING [train.py:1060] (2/4) Exclude cut with ID 497-129325-0061-62254-0_sp1.1 from training. Duration: 0.97725
2022-12-23 04:44:42,517 INFO [train.py:894] (2/4) Epoch 13, batch 1300, loss[loss=0.1613, simple_loss=0.2467, pruned_loss=0.03797, over 18533.00 frames. ], tot_loss[loss=0.1923, simple_loss=0.2778, pruned_loss=0.05345, over 3706796.98 frames. ], batch size: 47, lr: 8.95e-03, grad_scale: 8.0
2022-12-23 04:45:09,787 WARNING [train.py:1060] (2/4) Exclude cut with ID 7395-89880-0031-39906-0_sp0.9 from training. Duration: 22.97225
2022-12-23 04:45:41,058 WARNING [train.py:1060] (2/4) Exclude cut with ID 7395-89880-0047-39922-0_sp0.9 from training. Duration: 21.97775
2022-12-23 04:45:41,897 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.8473, 1.5514, 1.7890, 2.2404, 1.8121, 4.6309, 1.6455, 1.9127],
       device='cuda:2'), covar=tensor([0.0875, 0.1734, 0.1071, 0.0924, 0.1523, 0.0149, 0.1338, 0.1472],
       device='cuda:2'), in_proj_covar=tensor([0.0073, 0.0082, 0.0074, 0.0075, 0.0091, 0.0072, 0.0084, 0.0076],
       device='cuda:2'), out_proj_covar=tensor([0.0003, 0.0004, 0.0004, 0.0003, 0.0004, 0.0003, 0.0004, 0.0004],
       device='cuda:2')
2022-12-23 04:45:52,566 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([2.2543, 1.5540, 1.8248, 1.9082, 2.1798, 2.0531, 2.0623, 1.6904],
       device='cuda:2'), covar=tensor([0.1751, 0.2675, 0.2023, 0.2486, 0.1464, 0.0808, 0.2349, 0.1023],
       device='cuda:2'), in_proj_covar=tensor([0.0254, 0.0286, 0.0263, 0.0301, 0.0285, 0.0239, 0.0312, 0.0225],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-23 04:45:52,707 INFO [scaling.py:679] (2/4) Whitening: num_groups=8, num_channels=96, metric=1.21 vs. limit=2.0
2022-12-23 04:45:55,075 WARNING [train.py:1060] (2/4) Exclude cut with ID 1112-1043-0006-89194-0_sp0.9 from training. Duration: 21.8333125
2022-12-23 04:45:57,948 INFO [train.py:894] (2/4) Epoch 13, batch 1350, loss[loss=0.1887, simple_loss=0.2617, pruned_loss=0.0578, over 18411.00 frames. ], tot_loss[loss=0.1919, simple_loss=0.2771, pruned_loss=0.05334, over 3707595.15 frames. ], batch size: 42, lr: 8.95e-03, grad_scale: 8.0
2022-12-23 04:46:03,897 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 2.083e+02 3.554e+02 4.386e+02 5.691e+02 1.592e+03, threshold=8.773e+02, percent-clipped=6.0
2022-12-23 04:46:06,828 WARNING [train.py:1060] (2/4) Exclude cut with ID 1914-133440-0031-94921-0 from training. Duration: 20.47
2022-12-23 04:46:10,377 INFO [scaling.py:679] (2/4) Whitening: num_groups=8, num_channels=96, metric=1.18 vs. limit=2.0
2022-12-23 04:47:01,741 INFO [scaling.py:679] (2/4) Whitening: num_groups=8, num_channels=192, metric=1.94 vs. limit=2.0
2022-12-23 04:47:12,538 WARNING [train.py:1060] (2/4) Exclude cut with ID 7395-89880-0037-39912-0_sp0.9 from training. Duration: 20.67225
2022-12-23 04:47:14,270 INFO [train.py:894] (2/4) Epoch 13, batch 1400, loss[loss=0.1884, simple_loss=0.2817, pruned_loss=0.04756, over 18581.00 frames. ], tot_loss[loss=0.1914, simple_loss=0.2767, pruned_loss=0.05302, over 3708517.84 frames. ], batch size: 96, lr: 8.94e-03, grad_scale: 8.0
2022-12-23 04:47:32,075 WARNING [train.py:1060] (2/4) Exclude cut with ID 1914-133440-0024-94914-0_sp0.9 from training. Duration: 25.2444375
2022-12-23 04:47:37,402 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([2.2595, 2.1628, 1.8274, 1.8293, 2.3249, 2.8946, 2.8825, 2.0738],
       device='cuda:2'), covar=tensor([0.0477, 0.0258, 0.0375, 0.0252, 0.0188, 0.0264, 0.0312, 0.0293],
       device='cuda:2'), in_proj_covar=tensor([0.0089, 0.0122, 0.0147, 0.0125, 0.0112, 0.0112, 0.0096, 0.0125],
       device='cuda:2'), out_proj_covar=tensor([7.3576e-05, 1.0026e-04, 1.2559e-04, 1.0318e-04, 9.4218e-05, 8.8453e-05,
        7.7931e-05, 1.0180e-04], device='cuda:2')
2022-12-23 04:47:55,146 WARNING [train.py:1060] (2/4) Exclude cut with ID 3340-169293-0021-76797-0_sp0.9 from training. Duration: 21.1445
2022-12-23 04:48:13,678 INFO [zipformer.py:660] (2/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=43514.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 04:48:15,093 INFO [zipformer.py:660] (2/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=43515.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 04:48:28,976 INFO [train.py:894] (2/4) Epoch 13, batch 1450, loss[loss=0.1991, simple_loss=0.2828, pruned_loss=0.05774, over 18650.00 frames. ], tot_loss[loss=0.1911, simple_loss=0.2766, pruned_loss=0.05285, over 3710744.01 frames. ], batch size: 96, lr: 8.94e-03, grad_scale: 8.0
2022-12-23 04:48:34,932 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 2.266e+02 3.498e+02 4.440e+02 5.498e+02 1.255e+03, threshold=8.879e+02, percent-clipped=4.0
2022-12-23 04:48:38,937 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([2.4733, 2.1623, 2.2152, 2.1224, 2.3205, 2.2766, 2.3938, 2.0606],
       device='cuda:2'), covar=tensor([0.1479, 0.2220, 0.1618, 0.2217, 0.1426, 0.0694, 0.2241, 0.0897],
       device='cuda:2'), in_proj_covar=tensor([0.0254, 0.0288, 0.0264, 0.0303, 0.0285, 0.0240, 0.0313, 0.0226],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-23 04:48:45,699 INFO [zipformer.py:660] (2/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=43535.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 04:49:09,029 WARNING [train.py:1060] (2/4) Exclude cut with ID 2929-85685-0079-62383-0_sp0.9 from training. Duration: 33.038875
2022-12-23 04:49:27,627 INFO [zipformer.py:660] (2/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=43563.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 04:49:36,607 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([2.1225, 1.3262, 1.7589, 1.7770, 2.1374, 2.0578, 1.9241, 1.6265],
       device='cuda:2'), covar=tensor([0.1818, 0.2847, 0.2138, 0.2442, 0.1535, 0.0770, 0.2512, 0.1027],
       device='cuda:2'), in_proj_covar=tensor([0.0252, 0.0285, 0.0261, 0.0300, 0.0283, 0.0238, 0.0310, 0.0224],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-23 04:49:44,995 INFO [train.py:894] (2/4) Epoch 13, batch 1500, loss[loss=0.1963, simple_loss=0.2883, pruned_loss=0.05217, over 18511.00 frames. ], tot_loss[loss=0.191, simple_loss=0.2762, pruned_loss=0.05296, over 3711380.47 frames. ], batch size: 52, lr: 8.93e-03, grad_scale: 8.0
2022-12-23 04:49:45,448 INFO [zipformer.py:660] (2/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=43575.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 04:49:47,963 WARNING [train.py:1060] (2/4) Exclude cut with ID 6426-64291-0000-16059-0_sp0.9 from training. Duration: 20.0944375
2022-12-23 04:49:57,679 INFO [zipformer.py:660] (2/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=43583.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 04:50:04,015 WARNING [train.py:1060] (2/4) Exclude cut with ID 2929-85685-0044-62348-0_sp1.1 from training. Duration: 20.4
2022-12-23 04:50:12,809 WARNING [train.py:1060] (2/4) Exclude cut with ID 6330-62851-0022-91297-0 from training. Duration: 20.085
2022-12-23 04:50:22,995 WARNING [train.py:1060] (2/4) Exclude cut with ID 4860-13185-0032-76709-0_sp0.9 from training. Duration: 23.07775
2022-12-23 04:50:45,249 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.5282, 1.4943, 1.6035, 1.5384, 1.0974, 3.3043, 1.4698, 1.9645],
       device='cuda:2'), covar=tensor([0.3285, 0.2080, 0.1934, 0.2050, 0.1463, 0.0193, 0.1508, 0.0853],
       device='cuda:2'), in_proj_covar=tensor([0.0135, 0.0116, 0.0126, 0.0119, 0.0103, 0.0097, 0.0093, 0.0090],
       device='cuda:2'), out_proj_covar=tensor([0.0005, 0.0005, 0.0005, 0.0004, 0.0004, 0.0003, 0.0004, 0.0004],
       device='cuda:2')
2022-12-23 04:51:00,793 INFO [train.py:894] (2/4) Epoch 13, batch 1550, loss[loss=0.1892, simple_loss=0.278, pruned_loss=0.05016, over 18589.00 frames. ], tot_loss[loss=0.1915, simple_loss=0.2766, pruned_loss=0.05317, over 3711444.89 frames. ], batch size: 51, lr: 8.93e-03, grad_scale: 8.0
2022-12-23 04:51:07,027 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 2.501e+02 4.191e+02 4.765e+02 6.382e+02 1.278e+03, threshold=9.530e+02, percent-clipped=5.0
2022-12-23 04:51:09,072 WARNING [train.py:1060] (2/4) Exclude cut with ID 2929-85685-0044-62348-0_sp0.9 from training. Duration: 24.9333125
2022-12-23 04:51:52,549 WARNING [train.py:1060] (2/4) Exclude cut with ID 5118-111612-0016-124680-0_sp0.9 from training. Duration: 20.388875
2022-12-23 04:51:58,392 WARNING [train.py:1060] (2/4) Exclude cut with ID 432-122774-0017-62487-0_sp1.1 from training. Duration: 20.3590625
2022-12-23 04:52:15,953 INFO [train.py:894] (2/4) Epoch 13, batch 1600, loss[loss=0.1851, simple_loss=0.2709, pruned_loss=0.04966, over 18587.00 frames. ], tot_loss[loss=0.1921, simple_loss=0.2774, pruned_loss=0.05338, over 3712124.08 frames. ], batch size: 51, lr: 8.92e-03, grad_scale: 8.0
2022-12-23 04:52:42,595 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.7301, 1.2051, 0.8691, 1.3443, 1.9449, 1.4158, 1.4169, 1.7732],
       device='cuda:2'), covar=tensor([0.1648, 0.2287, 0.2419, 0.1676, 0.1867, 0.1648, 0.1614, 0.1657],
       device='cuda:2'), in_proj_covar=tensor([0.0091, 0.0098, 0.0118, 0.0096, 0.0112, 0.0090, 0.0096, 0.0093],
       device='cuda:2'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003],
       device='cuda:2')
2022-12-23 04:53:07,464 WARNING [train.py:1060] (2/4) Exclude cut with ID 3557-8342-0013-54691-0_sp1.1 from training. Duration: 0.836375
2022-12-23 04:53:31,656 INFO [train.py:894] (2/4) Epoch 13, batch 1650, loss[loss=0.1885, simple_loss=0.2598, pruned_loss=0.05856, over 18605.00 frames. ], tot_loss[loss=0.1937, simple_loss=0.2782, pruned_loss=0.05463, over 3713671.44 frames. ], batch size: 41, lr: 8.92e-03, grad_scale: 8.0
2022-12-23 04:53:38,417 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 2.973e+02 4.370e+02 5.353e+02 6.332e+02 1.180e+03, threshold=1.071e+03, percent-clipped=2.0
2022-12-23 04:53:49,731 WARNING [train.py:1060] (2/4) Exclude cut with ID 8565-290391-0049-67394-0_sp0.9 from training. Duration: 21.3166875
2022-12-23 04:54:19,371 WARNING [train.py:1060] (2/4) Exclude cut with ID 6533-399-0029-104863-0_sp0.9 from training. Duration: 22.1055625
2022-12-23 04:54:29,655 WARNING [train.py:1060] (2/4) Exclude cut with ID 7699-105389-0094-26379-0_sp1.1 from training. Duration: 21.77725
2022-12-23 04:54:29,953 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.4761, 1.0613, 0.6653, 1.0805, 1.9029, 0.6121, 1.1219, 1.3980],
       device='cuda:2'), covar=tensor([0.1746, 0.2244, 0.2138, 0.1702, 0.1795, 0.1756, 0.1640, 0.1702],
       device='cuda:2'), in_proj_covar=tensor([0.0091, 0.0098, 0.0118, 0.0096, 0.0112, 0.0090, 0.0096, 0.0093],
       device='cuda:2'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003],
       device='cuda:2')
2022-12-23 04:54:47,433 INFO [train.py:894] (2/4) Epoch 13, batch 1700, loss[loss=0.238, simple_loss=0.3085, pruned_loss=0.08378, over 18621.00 frames. ], tot_loss[loss=0.1969, simple_loss=0.2801, pruned_loss=0.05688, over 3713353.84 frames. ], batch size: 78, lr: 8.91e-03, grad_scale: 8.0
2022-12-23 04:54:50,307 WARNING [train.py:1060] (2/4) Exclude cut with ID 7255-291500-0005-134304-0_sp0.9 from training. Duration: 27.8166875
2022-12-23 04:55:06,594 INFO [zipformer.py:660] (2/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=43787.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 04:55:15,042 WARNING [train.py:1060] (2/4) Exclude cut with ID 7357-94126-0021-15852-0_sp1.1 from training. Duration: 22.5090625
2022-12-23 04:55:23,067 WARNING [train.py:1060] (2/4) Exclude cut with ID 7255-291500-0005-134304-0 from training. Duration: 25.035
2022-12-23 04:55:39,635 WARNING [train.py:1060] (2/4) Exclude cut with ID 774-127930-0014-10412-0_sp1.1 from training. Duration: 0.95
2022-12-23 04:55:41,434 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.2766, 1.6218, 1.1897, 1.7849, 1.7583, 1.4797, 0.9052, 1.1975],
       device='cuda:2'), covar=tensor([0.1831, 0.1614, 0.1590, 0.1015, 0.1020, 0.1078, 0.1939, 0.1406],
       device='cuda:2'), in_proj_covar=tensor([0.0237, 0.0214, 0.0205, 0.0190, 0.0251, 0.0190, 0.0210, 0.0189],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-23 04:55:57,265 WARNING [train.py:1060] (2/4) Exclude cut with ID 3033-130750-0096-55598-0_sp0.9 from training. Duration: 0.92225
2022-12-23 04:56:04,337 INFO [train.py:894] (2/4) Epoch 13, batch 1750, loss[loss=0.2018, simple_loss=0.2802, pruned_loss=0.06164, over 18588.00 frames. ], tot_loss[loss=0.1996, simple_loss=0.2818, pruned_loss=0.05873, over 3713558.46 frames. ], batch size: 51, lr: 8.91e-03, grad_scale: 8.0
2022-12-23 04:56:07,702 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([2.5539, 2.3296, 1.7919, 0.6966, 1.7199, 2.0452, 1.7594, 2.0296],
       device='cuda:2'), covar=tensor([0.0576, 0.0458, 0.1221, 0.1616, 0.1274, 0.1317, 0.1409, 0.0815],
       device='cuda:2'), in_proj_covar=tensor([0.0165, 0.0179, 0.0201, 0.0191, 0.0206, 0.0193, 0.0206, 0.0192],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-23 04:56:10,720 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 2.858e+02 4.539e+02 5.407e+02 6.945e+02 1.348e+03, threshold=1.081e+03, percent-clipped=3.0
2022-12-23 04:56:25,770 WARNING [train.py:1060] (2/4) Exclude cut with ID 4511-76322-0006-80011-0 from training. Duration: 21.97
2022-12-23 04:56:40,242 INFO [zipformer.py:660] (2/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=43848.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 04:56:45,441 WARNING [train.py:1060] (2/4) Exclude cut with ID 7492-105653-0055-62765-0_sp0.9 from training. Duration: 21.97225
2022-12-23 04:56:47,249 WARNING [train.py:1060] (2/4) Exclude cut with ID 453-131332-0000-47844-0_sp0.9 from training. Duration: 25.3333125
2022-12-23 04:56:57,442 WARNING [train.py:1060] (2/4) Exclude cut with ID 5172-29468-0015-19128-0_sp0.9 from training. Duration: 21.5055625
2022-12-23 04:57:06,977 WARNING [train.py:1060] (2/4) Exclude cut with ID 453-131332-0000-47844-0_sp1.1 from training. Duration: 20.72725
2022-12-23 04:57:12,912 INFO [zipformer.py:660] (2/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=43870.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 04:57:20,807 INFO [train.py:894] (2/4) Epoch 13, batch 1800, loss[loss=0.2378, simple_loss=0.312, pruned_loss=0.08177, over 18674.00 frames. ], tot_loss[loss=0.2014, simple_loss=0.2824, pruned_loss=0.06024, over 3713605.17 frames. ], batch size: 180, lr: 8.90e-03, grad_scale: 8.0
2022-12-23 04:57:41,329 WARNING [train.py:1060] (2/4) Exclude cut with ID 8631-249866-0030-130156-0_sp0.9 from training. Duration: 26.32775
2022-12-23 04:58:12,187 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.3012, 1.1352, 1.6754, 2.9248, 2.0068, 2.3201, 0.5117, 2.0001],
       device='cuda:2'), covar=tensor([0.1931, 0.1761, 0.1470, 0.0582, 0.1112, 0.1240, 0.2435, 0.1130],
       device='cuda:2'), in_proj_covar=tensor([0.0101, 0.0114, 0.0130, 0.0131, 0.0103, 0.0132, 0.0129, 0.0108],
       device='cuda:2'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0004, 0.0003, 0.0003],
       device='cuda:2')
2022-12-23 04:58:12,242 INFO [zipformer.py:660] (2/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=43909.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 04:58:13,357 WARNING [train.py:1060] (2/4) Exclude cut with ID 3867-173237-0077-144769-0 from training. Duration: 20.025
2022-12-23 04:58:17,943 WARNING [train.py:1060] (2/4) Exclude cut with ID 6709-74022-0004-86860-0_sp1.1 from training. Duration: 0.9409375
2022-12-23 04:58:17,950 WARNING [train.py:1060] (2/4) Exclude cut with ID 4757-1811-0023-62229-0_sp0.9 from training. Duration: 21.37775
2022-12-23 04:58:36,322 INFO [train.py:894] (2/4) Epoch 13, batch 1850, loss[loss=0.2049, simple_loss=0.2781, pruned_loss=0.06583, over 18430.00 frames. ], tot_loss[loss=0.2045, simple_loss=0.2839, pruned_loss=0.06261, over 3713434.77 frames. ], batch size: 48, lr: 8.90e-03, grad_scale: 8.0
2022-12-23 04:58:41,368 WARNING [train.py:1060] (2/4) Exclude cut with ID 1250-135782-0004-25974-0_sp0.9 from training. Duration: 21.17225
2022-12-23 04:58:41,378 WARNING [train.py:1060] (2/4) Exclude cut with ID 7357-94126-0021-15852-0_sp0.9 from training. Duration: 27.511125
2022-12-23 04:58:42,664 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 3.374e+02 4.790e+02 5.650e+02 6.943e+02 1.599e+03, threshold=1.130e+03, percent-clipped=7.0
2022-12-23 04:59:18,526 WARNING [train.py:1060] (2/4) Exclude cut with ID 453-131332-0000-47844-0 from training. Duration: 22.8
2022-12-23 04:59:22,967 WARNING [train.py:1060] (2/4) Exclude cut with ID 4964-30587-0040-44509-0 from training. Duration: 22.585
2022-12-23 04:59:24,786 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([2.2501, 1.6948, 0.7492, 2.1602, 2.4794, 1.6284, 2.2133, 2.3428],
       device='cuda:2'), covar=tensor([0.1552, 0.1970, 0.2469, 0.1357, 0.1712, 0.1564, 0.1366, 0.1499],
       device='cuda:2'), in_proj_covar=tensor([0.0091, 0.0099, 0.0117, 0.0096, 0.0114, 0.0091, 0.0096, 0.0093],
       device='cuda:2'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003],
       device='cuda:2')
2022-12-23 04:59:27,737 INFO [zipformer.py:660] (2/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=43959.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 04:59:36,610 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([2.0942, 1.6574, 1.9993, 2.5517, 2.0498, 4.1994, 1.6436, 1.8202],
       device='cuda:2'), covar=tensor([0.0792, 0.1772, 0.1054, 0.0799, 0.1295, 0.0226, 0.1336, 0.1386],
       device='cuda:2'), in_proj_covar=tensor([0.0072, 0.0081, 0.0073, 0.0074, 0.0089, 0.0071, 0.0083, 0.0075],
       device='cuda:2'), out_proj_covar=tensor([0.0003, 0.0004, 0.0003, 0.0003, 0.0004, 0.0003, 0.0004, 0.0004],
       device='cuda:2')
2022-12-23 04:59:44,787 INFO [zipformer.py:660] (2/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=43970.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 04:59:47,584 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.4077, 1.9851, 1.7005, 2.5630, 2.2593, 1.5813, 1.5729, 1.2901],
       device='cuda:2'), covar=tensor([0.2191, 0.1713, 0.1456, 0.0820, 0.1465, 0.1288, 0.2002, 0.1613],
       device='cuda:2'), in_proj_covar=tensor([0.0239, 0.0216, 0.0204, 0.0189, 0.0250, 0.0190, 0.0211, 0.0189],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-23 04:59:51,439 INFO [train.py:894] (2/4) Epoch 13, batch 1900, loss[loss=0.1935, simple_loss=0.2795, pruned_loss=0.05378, over 18579.00 frames. ], tot_loss[loss=0.2056, simple_loss=0.2841, pruned_loss=0.06359, over 3712284.14 frames. ], batch size: 57, lr: 8.89e-03, grad_scale: 8.0
2022-12-23 04:59:56,577 WARNING [train.py:1060] (2/4) Exclude cut with ID 6978-92210-0001-146967-0_sp0.9 from training. Duration: 22.0166875
2022-12-23 05:00:14,001 WARNING [train.py:1060] (2/4) Exclude cut with ID 7255-291500-0003-134302-0_sp1.1 from training. Duration: 24.395375
2022-12-23 05:00:20,213 WARNING [train.py:1060] (2/4) Exclude cut with ID 3488-85273-0017-41203-0_sp0.9 from training. Duration: 27.47775
2022-12-23 05:00:24,656 WARNING [train.py:1060] (2/4) Exclude cut with ID 432-122774-0017-62487-0_sp0.9 from training. Duration: 24.8833125
2022-12-23 05:00:27,593 WARNING [train.py:1060] (2/4) Exclude cut with ID 6758-72288-0033-108368-0 from training. Duration: 23.39
2022-12-23 05:00:37,074 WARNING [train.py:1060] (2/4) Exclude cut with ID 8291-282929-0007-12994-0_sp0.9 from training. Duration: 28.72225
2022-12-23 05:00:45,761 WARNING [train.py:1060] (2/4) Exclude cut with ID 585-294811-0110-133686-0_sp0.9 from training. Duration: 20.8944375
2022-12-23 05:00:57,435 INFO [scaling.py:679] (2/4) Whitening: num_groups=8, num_channels=192, metric=1.81 vs. limit=2.0
2022-12-23 05:01:00,986 WARNING [train.py:1060] (2/4) Exclude cut with ID 5796-66357-0007-116447-0_sp0.9 from training. Duration: 23.8444375
2022-12-23 05:01:04,333 INFO [zipformer.py:660] (2/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=44020.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 05:01:11,924 INFO [train.py:894] (2/4) Epoch 13, batch 1950, loss[loss=0.198, simple_loss=0.2702, pruned_loss=0.06286, over 18696.00 frames. ], tot_loss[loss=0.2068, simple_loss=0.2846, pruned_loss=0.06452, over 3713278.48 frames. ], batch size: 50, lr: 8.89e-03, grad_scale: 8.0
2022-12-23 05:01:18,254 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 2.985e+02 4.904e+02 5.983e+02 6.955e+02 1.317e+03, threshold=1.197e+03, percent-clipped=1.0
2022-12-23 05:01:26,871 WARNING [train.py:1060] (2/4) Exclude cut with ID 8291-282929-0007-12994-0 from training. Duration: 25.85
2022-12-23 05:01:26,877 WARNING [train.py:1060] (2/4) Exclude cut with ID 8291-282929-0023-13010-0 from training. Duration: 21.39
2022-12-23 05:01:31,054 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.3361, 2.5232, 2.9504, 1.1477, 2.2633, 3.0982, 2.2046, 2.5201],
       device='cuda:2'), covar=tensor([0.0764, 0.0344, 0.0272, 0.0425, 0.0434, 0.0359, 0.0421, 0.0733],
       device='cuda:2'), in_proj_covar=tensor([0.0143, 0.0159, 0.0116, 0.0133, 0.0143, 0.0136, 0.0150, 0.0156],
       device='cuda:2'), out_proj_covar=tensor([1.1457e-04, 1.2834e-04, 9.2400e-05, 1.0343e-04, 1.1334e-04, 1.1058e-04,
        1.2271e-04, 1.2488e-04], device='cuda:2')
2022-12-23 05:01:38,022 WARNING [train.py:1060] (2/4) Exclude cut with ID 7357-94126-0014-15845-0 from training. Duration: 27.92
2022-12-23 05:02:03,801 WARNING [train.py:1060] (2/4) Exclude cut with ID 8631-249866-0039-130165-0_sp0.9 from training. Duration: 20.661125
2022-12-23 05:02:27,773 INFO [train.py:894] (2/4) Epoch 13, batch 2000, loss[loss=0.2271, simple_loss=0.3093, pruned_loss=0.07243, over 18546.00 frames. ], tot_loss[loss=0.2079, simple_loss=0.2849, pruned_loss=0.06546, over 3713443.95 frames. ], batch size: 57, lr: 8.88e-03, grad_scale: 8.0
2022-12-23 05:02:27,814 WARNING [train.py:1060] (2/4) Exclude cut with ID 7357-94126-0043-15874-0_sp0.9 from training. Duration: 20.07225
2022-12-23 05:02:36,660 WARNING [train.py:1060] (2/4) Exclude cut with ID 1085-156170-0017-128270-0 from training. Duration: 21.01
2022-12-23 05:03:30,387 INFO [scaling.py:679] (2/4) Whitening: num_groups=1, num_channels=384, metric=5.00 vs. limit=5.0
2022-12-23 05:03:34,100 INFO [scaling.py:679] (2/4) Whitening: num_groups=8, num_channels=96, metric=1.21 vs. limit=2.0
2022-12-23 05:03:41,972 WARNING [train.py:1060] (2/4) Exclude cut with ID 2195-150901-0045-59933-0 from training. Duration: 20.65
2022-12-23 05:03:44,900 INFO [train.py:894] (2/4) Epoch 13, batch 2050, loss[loss=0.1673, simple_loss=0.2392, pruned_loss=0.04773, over 18690.00 frames. ], tot_loss[loss=0.2085, simple_loss=0.2847, pruned_loss=0.06615, over 3713940.24 frames. ], batch size: 46, lr: 8.88e-03, grad_scale: 8.0
2022-12-23 05:03:51,452 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 2.848e+02 4.873e+02 5.730e+02 6.896e+02 1.341e+03, threshold=1.146e+03, percent-clipped=1.0
2022-12-23 05:03:51,510 WARNING [train.py:1060] (2/4) Exclude cut with ID 5796-66357-0007-116447-0 from training. Duration: 21.46
2022-12-23 05:04:04,036 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.4483, 2.0165, 1.4845, 2.3799, 2.5221, 1.5106, 1.5272, 1.1590],
       device='cuda:2'), covar=tensor([0.2141, 0.1704, 0.1660, 0.0960, 0.1307, 0.1299, 0.2055, 0.1674],
       device='cuda:2'), in_proj_covar=tensor([0.0238, 0.0215, 0.0203, 0.0190, 0.0252, 0.0189, 0.0211, 0.0189],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-23 05:04:12,927 INFO [zipformer.py:660] (2/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=44143.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 05:04:35,139 WARNING [train.py:1060] (2/4) Exclude cut with ID 3557-8342-0013-54691-0 from training. Duration: 0.92
2022-12-23 05:04:42,109 WARNING [train.py:1060] (2/4) Exclude cut with ID 8291-282929-0023-13010-0_sp0.9 from training. Duration: 23.7666875
2022-12-23 05:04:54,419 INFO [zipformer.py:660] (2/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=44170.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 05:05:01,741 INFO [train.py:894] (2/4) Epoch 13, batch 2100, loss[loss=0.1973, simple_loss=0.2757, pruned_loss=0.05941, over 18478.00 frames. ], tot_loss[loss=0.2085, simple_loss=0.2846, pruned_loss=0.06618, over 3714142.72 frames. ], batch size: 54, lr: 8.87e-03, grad_scale: 8.0
2022-12-23 05:05:02,127 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([3.3393, 3.1220, 1.8352, 1.3350, 3.6760, 3.6905, 2.8905, 2.3191],
       device='cuda:2'), covar=tensor([0.0339, 0.0321, 0.0634, 0.0779, 0.0183, 0.0258, 0.0423, 0.0766],
       device='cuda:2'), in_proj_covar=tensor([0.0120, 0.0118, 0.0127, 0.0117, 0.0088, 0.0114, 0.0130, 0.0151],
       device='cuda:2'), out_proj_covar=tensor([0.0001, 0.0001, 0.0002, 0.0001, 0.0001, 0.0001, 0.0002, 0.0002],
       device='cuda:2')
2022-12-23 05:05:19,010 WARNING [train.py:1060] (2/4) Exclude cut with ID 8544-281189-0060-101339-0_sp0.9 from training. Duration: 20.861125
2022-12-23 05:05:22,773 INFO [scaling.py:679] (2/4) Whitening: num_groups=1, num_channels=384, metric=3.09 vs. limit=5.0
2022-12-23 05:05:28,304 WARNING [train.py:1060] (2/4) Exclude cut with ID 3488-65654-0031-41259-0_sp0.9 from training. Duration: 22.711125
2022-12-23 05:05:50,162 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([2.9094, 2.6990, 2.0747, 1.1208, 2.0213, 2.2629, 1.9155, 2.2815],
       device='cuda:2'), covar=tensor([0.0552, 0.0405, 0.1152, 0.1525, 0.1229, 0.1223, 0.1383, 0.0747],
       device='cuda:2'), in_proj_covar=tensor([0.0167, 0.0182, 0.0201, 0.0193, 0.0206, 0.0194, 0.0208, 0.0194],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-23 05:06:06,773 INFO [zipformer.py:660] (2/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=44218.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 05:06:08,244 WARNING [train.py:1060] (2/4) Exclude cut with ID 6951-79737-0043-132310-0_sp1.1 from training. Duration: 22.986375
2022-12-23 05:06:17,674 INFO [train.py:894] (2/4) Epoch 13, batch 2150, loss[loss=0.1892, simple_loss=0.2728, pruned_loss=0.05279, over 18587.00 frames. ], tot_loss[loss=0.2072, simple_loss=0.2832, pruned_loss=0.06559, over 3713806.15 frames. ], batch size: 51, lr: 8.87e-03, grad_scale: 8.0
2022-12-23 05:06:23,827 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 3.378e+02 5.026e+02 6.284e+02 7.553e+02 1.602e+03, threshold=1.257e+03, percent-clipped=4.0
2022-12-23 05:06:26,717 WARNING [train.py:1060] (2/4) Exclude cut with ID 8040-260924-0003-80960-0_sp0.9 from training. Duration: 22.07225
2022-12-23 05:06:31,054 WARNING [train.py:1060] (2/4) Exclude cut with ID 7699-105389-0045-26330-0_sp0.9 from training. Duration: 20.3055625
2022-12-23 05:06:32,454 WARNING [train.py:1060] (2/4) Exclude cut with ID 6356-271890-0060-94317-0_sp0.9 from training. Duration: 20.72225
2022-12-23 05:06:39,560 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.5958, 1.0638, 1.9910, 3.0529, 2.1758, 2.4576, 0.8373, 2.1192],
       device='cuda:2'), covar=tensor([0.1931, 0.2046, 0.1509, 0.0609, 0.1206, 0.1108, 0.2379, 0.1190],
       device='cuda:2'), in_proj_covar=tensor([0.0102, 0.0115, 0.0131, 0.0134, 0.0105, 0.0134, 0.0130, 0.0109],
       device='cuda:2'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0004, 0.0003, 0.0003],
       device='cuda:2')
2022-12-23 05:06:51,025 WARNING [train.py:1060] (2/4) Exclude cut with ID 3488-85273-0017-41203-0_sp1.1 from training. Duration: 22.4818125
2022-12-23 05:06:51,409 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([3.0648, 3.1242, 1.7117, 1.3841, 3.8147, 3.6778, 2.8243, 2.2753],
       device='cuda:2'), covar=tensor([0.0403, 0.0312, 0.0641, 0.0746, 0.0152, 0.0271, 0.0437, 0.0738],
       device='cuda:2'), in_proj_covar=tensor([0.0119, 0.0117, 0.0126, 0.0116, 0.0087, 0.0114, 0.0129, 0.0150],
       device='cuda:2'), out_proj_covar=tensor([0.0001, 0.0001, 0.0002, 0.0001, 0.0001, 0.0001, 0.0002, 0.0002],
       device='cuda:2')
2022-12-23 05:07:18,064 WARNING [train.py:1060] (2/4) Exclude cut with ID 4964-30587-0040-44509-0_sp0.9 from training. Duration: 25.0944375
2022-12-23 05:07:18,216 INFO [zipformer.py:660] (2/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=44265.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 05:07:20,943 WARNING [train.py:1060] (2/4) Exclude cut with ID 6533-399-0047-104881-0 from training. Duration: 21.515
2022-12-23 05:07:26,018 WARNING [train.py:1060] (2/4) Exclude cut with ID 7357-94126-0009-15840-0_sp0.9 from training. Duration: 27.02225
2022-12-23 05:07:31,766 WARNING [train.py:1060] (2/4) Exclude cut with ID 432-122774-0010-62480-0_sp0.9 from training. Duration: 22.22225
2022-12-23 05:07:33,323 INFO [train.py:894] (2/4) Epoch 13, batch 2200, loss[loss=0.203, simple_loss=0.2878, pruned_loss=0.0591, over 18508.00 frames. ], tot_loss[loss=0.2078, simple_loss=0.2835, pruned_loss=0.06603, over 3713929.89 frames. ], batch size: 64, lr: 8.86e-03, grad_scale: 8.0
2022-12-23 05:07:39,402 WARNING [train.py:1060] (2/4) Exclude cut with ID 4964-30587-0085-44554-0_sp0.9 from training. Duration: 20.85
2022-12-23 05:07:45,658 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.7350, 1.5782, 1.5454, 1.5144, 1.8820, 2.0160, 2.1234, 1.3507],
       device='cuda:2'), covar=tensor([0.0365, 0.0296, 0.0405, 0.0227, 0.0189, 0.0306, 0.0247, 0.0313],
       device='cuda:2'), in_proj_covar=tensor([0.0089, 0.0122, 0.0144, 0.0123, 0.0111, 0.0112, 0.0095, 0.0123],
       device='cuda:2'), out_proj_covar=tensor([7.2684e-05, 9.9644e-05, 1.2352e-04, 1.0181e-04, 9.3536e-05, 8.8629e-05,
        7.6824e-05, 9.9712e-05], device='cuda:2')
2022-12-23 05:07:48,362 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.8858, 1.8940, 1.2660, 2.0748, 2.0760, 1.7296, 2.6135, 1.9417],
       device='cuda:2'), covar=tensor([0.0816, 0.1510, 0.2621, 0.1569, 0.1526, 0.0840, 0.0913, 0.1083],
       device='cuda:2'), in_proj_covar=tensor([0.0172, 0.0195, 0.0238, 0.0283, 0.0226, 0.0184, 0.0208, 0.0195],
       device='cuda:2'), out_proj_covar=tensor([0.0001, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-23 05:08:14,738 WARNING [train.py:1060] (2/4) Exclude cut with ID 4295-39940-0007-92567-0 from training. Duration: 21.54
2022-12-23 05:08:19,485 WARNING [train.py:1060] (2/4) Exclude cut with ID 4964-30587-0040-44509-0_sp1.1 from training. Duration: 20.5318125
2022-12-23 05:08:28,229 WARNING [train.py:1060] (2/4) Exclude cut with ID 7255-291500-0012-134311-0_sp0.9 from training. Duration: 21.9333125
2022-12-23 05:08:32,782 INFO [zipformer.py:660] (2/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=44315.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 05:08:48,070 INFO [train.py:894] (2/4) Epoch 13, batch 2250, loss[loss=0.1917, simple_loss=0.2795, pruned_loss=0.05201, over 18512.00 frames. ], tot_loss[loss=0.2074, simple_loss=0.283, pruned_loss=0.06585, over 3714379.04 frames. ], batch size: 55, lr: 8.86e-03, grad_scale: 8.0
2022-12-23 05:08:54,075 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 3.356e+02 5.266e+02 6.249e+02 7.363e+02 1.689e+03, threshold=1.250e+03, percent-clipped=3.0
2022-12-23 05:09:06,026 INFO [scaling.py:679] (2/4) Whitening: num_groups=8, num_channels=96, metric=1.22 vs. limit=2.0
2022-12-23 05:09:13,044 INFO [scaling.py:679] (2/4) Whitening: num_groups=1, num_channels=384, metric=4.10 vs. limit=5.0
2022-12-23 05:09:15,234 WARNING [train.py:1060] (2/4) Exclude cut with ID 8631-249866-0025-130151-0_sp0.9 from training. Duration: 21.7944375
2022-12-23 05:09:28,098 WARNING [train.py:1060] (2/4) Exclude cut with ID 8291-282929-0002-12989-0_sp0.9 from training. Duration: 22.4666875
2022-12-23 05:09:34,041 WARNING [train.py:1060] (2/4) Exclude cut with ID 6121-9014-0076-24124-0 from training. Duration: 21.635
2022-12-23 05:09:40,118 WARNING [train.py:1060] (2/4) Exclude cut with ID 6121-9014-0076-24124-0_sp0.9 from training. Duration: 24.038875
2022-12-23 05:10:03,929 INFO [train.py:894] (2/4) Epoch 13, batch 2300, loss[loss=0.246, simple_loss=0.309, pruned_loss=0.09151, over 18587.00 frames. ], tot_loss[loss=0.208, simple_loss=0.2834, pruned_loss=0.06632, over 3713195.59 frames. ], batch size: 57, lr: 8.85e-03, grad_scale: 16.0
2022-12-23 05:10:23,892 WARNING [train.py:1060] (2/4) Exclude cut with ID 7255-291500-0009-134308-0_sp1.1 from training. Duration: 21.786375
2022-12-23 05:10:36,272 WARNING [train.py:1060] (2/4) Exclude cut with ID 8291-282929-0002-12989-0 from training. Duration: 20.22
2022-12-23 05:10:55,900 INFO [scaling.py:679] (2/4) Whitening: num_groups=1, num_channels=384, metric=3.66 vs. limit=5.0
2022-12-23 05:10:57,887 INFO [scaling.py:679] (2/4) Whitening: num_groups=8, num_channels=96, metric=1.60 vs. limit=2.0
2022-12-23 05:11:19,832 INFO [train.py:894] (2/4) Epoch 13, batch 2350, loss[loss=0.2035, simple_loss=0.2851, pruned_loss=0.06092, over 18670.00 frames. ], tot_loss[loss=0.2079, simple_loss=0.2834, pruned_loss=0.06624, over 3712783.93 frames. ], batch size: 97, lr: 8.85e-03, grad_scale: 16.0
2022-12-23 05:11:26,063 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 3.295e+02 5.245e+02 6.691e+02 8.586e+02 1.798e+03, threshold=1.338e+03, percent-clipped=9.0
2022-12-23 05:11:47,302 INFO [zipformer.py:660] (2/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=44443.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 05:12:21,390 INFO [scaling.py:679] (2/4) Whitening: num_groups=8, num_channels=192, metric=1.68 vs. limit=2.0
2022-12-23 05:12:33,959 WARNING [train.py:1060] (2/4) Exclude cut with ID 6951-79737-0043-132310-0 from training. Duration: 25.285
2022-12-23 05:12:35,370 INFO [train.py:894] (2/4) Epoch 13, batch 2400, loss[loss=0.2391, simple_loss=0.3079, pruned_loss=0.08514, over 18591.00 frames. ], tot_loss[loss=0.2076, simple_loss=0.2831, pruned_loss=0.06604, over 3713355.87 frames. ], batch size: 57, lr: 8.84e-03, grad_scale: 16.0
2022-12-23 05:12:40,810 INFO [scaling.py:679] (2/4) Whitening: num_groups=8, num_channels=192, metric=1.88 vs. limit=2.0
2022-12-23 05:13:00,022 INFO [zipformer.py:660] (2/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=44491.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 05:13:38,786 WARNING [train.py:1060] (2/4) Exclude cut with ID 811-130148-0001-63453-0_sp0.9 from training. Duration: 20.861125
2022-12-23 05:13:49,407 INFO [train.py:894] (2/4) Epoch 13, batch 2450, loss[loss=0.2017, simple_loss=0.2869, pruned_loss=0.05819, over 18454.00 frames. ], tot_loss[loss=0.2084, simple_loss=0.2841, pruned_loss=0.06635, over 3713543.15 frames. ], batch size: 50, lr: 8.84e-03, grad_scale: 8.0
2022-12-23 05:13:49,946 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.8245, 1.8068, 1.4957, 1.7549, 1.6926, 1.7557, 1.6118, 1.9458],
       device='cuda:2'), covar=tensor([0.1967, 0.2718, 0.1691, 0.2236, 0.2704, 0.1029, 0.2325, 0.0755],
       device='cuda:2'), in_proj_covar=tensor([0.0279, 0.0273, 0.0231, 0.0346, 0.0254, 0.0215, 0.0268, 0.0192],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0003, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-23 05:13:53,004 INFO [zipformer.py:660] (2/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=44527.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 05:13:56,729 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 3.765e+02 5.091e+02 6.317e+02 7.743e+02 2.009e+03, threshold=1.263e+03, percent-clipped=3.0
2022-12-23 05:14:01,006 WARNING [train.py:1060] (2/4) Exclude cut with ID 6010-56788-0055-90261-0 from training. Duration: 20.88
2022-12-23 05:14:27,728 INFO [zipformer.py:660] (2/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=44550.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 05:14:33,497 WARNING [train.py:1060] (2/4) Exclude cut with ID 7357-94126-0045-15876-0_sp0.9 from training. Duration: 23.4166875
2022-12-23 05:14:50,325 INFO [zipformer.py:660] (2/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=44565.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 05:15:05,141 INFO [train.py:894] (2/4) Epoch 13, batch 2500, loss[loss=0.2422, simple_loss=0.3152, pruned_loss=0.08461, over 18605.00 frames. ], tot_loss[loss=0.2085, simple_loss=0.2842, pruned_loss=0.06639, over 3713702.15 frames. ], batch size: 56, lr: 8.83e-03, grad_scale: 8.0
2022-12-23 05:15:26,527 INFO [zipformer.py:660] (2/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=44588.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 05:15:40,645 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.2815, 1.2182, 1.1325, 1.2206, 1.3302, 1.3002, 1.4122, 1.0259],
       device='cuda:2'), covar=tensor([0.0262, 0.0223, 0.0350, 0.0197, 0.0231, 0.0343, 0.0177, 0.0261],
       device='cuda:2'), in_proj_covar=tensor([0.0088, 0.0123, 0.0145, 0.0123, 0.0112, 0.0113, 0.0094, 0.0123],
       device='cuda:2'), out_proj_covar=tensor([7.2530e-05, 1.0006e-04, 1.2429e-04, 1.0144e-04, 9.3977e-05, 8.9456e-05,
        7.5996e-05, 9.9897e-05], device='cuda:2')
2022-12-23 05:15:51,039 WARNING [train.py:1060] (2/4) Exclude cut with ID 7255-291500-0006-134305-0 from training. Duration: 21.24
2022-12-23 05:15:51,052 WARNING [train.py:1060] (2/4) Exclude cut with ID 6533-399-0047-104881-0_sp0.9 from training. Duration: 23.9055625
2022-12-23 05:15:52,803 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.5028, 1.3976, 1.2835, 0.8608, 1.7623, 1.5018, 1.4083, 1.1933],
       device='cuda:2'), covar=tensor([0.0450, 0.0479, 0.0494, 0.0693, 0.0305, 0.0367, 0.0454, 0.0915],
       device='cuda:2'), in_proj_covar=tensor([0.0121, 0.0120, 0.0128, 0.0118, 0.0088, 0.0117, 0.0131, 0.0151],
       device='cuda:2'), out_proj_covar=tensor([0.0001, 0.0001, 0.0002, 0.0001, 0.0001, 0.0001, 0.0002, 0.0002],
       device='cuda:2')
2022-12-23 05:16:01,736 INFO [zipformer.py:660] (2/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=44611.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 05:16:03,945 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.2173, 1.4522, 1.1566, 1.6806, 1.6669, 1.2958, 1.1205, 1.1116],
       device='cuda:2'), covar=tensor([0.1514, 0.1185, 0.1194, 0.0686, 0.1131, 0.0863, 0.1711, 0.1156],
       device='cuda:2'), in_proj_covar=tensor([0.0237, 0.0212, 0.0201, 0.0189, 0.0250, 0.0188, 0.0211, 0.0189],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-23 05:16:05,153 INFO [zipformer.py:660] (2/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=44613.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 05:16:08,142 INFO [zipformer.py:660] (2/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=44615.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 05:16:22,383 INFO [train.py:894] (2/4) Epoch 13, batch 2550, loss[loss=0.2257, simple_loss=0.299, pruned_loss=0.07618, over 18445.00 frames. ], tot_loss[loss=0.2089, simple_loss=0.2849, pruned_loss=0.0665, over 3714104.02 frames. ], batch size: 50, lr: 8.83e-03, grad_scale: 8.0
2022-12-23 05:16:25,872 WARNING [train.py:1060] (2/4) Exclude cut with ID 6758-72288-0033-108368-0_sp0.9 from training. Duration: 25.988875
2022-12-23 05:16:30,044 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 3.522e+02 4.934e+02 5.713e+02 7.366e+02 1.644e+03, threshold=1.143e+03, percent-clipped=2.0
2022-12-23 05:16:34,532 WARNING [train.py:1060] (2/4) Exclude cut with ID 7255-291500-0001-134300-0_sp0.9 from training. Duration: 20.67225
2022-12-23 05:17:16,416 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([2.4368, 1.7951, 1.9618, 2.1691, 2.3160, 2.1491, 2.2630, 1.6405],
       device='cuda:2'), covar=tensor([0.1812, 0.2842, 0.2232, 0.2553, 0.1539, 0.0836, 0.2732, 0.1101],
       device='cuda:2'), in_proj_covar=tensor([0.0254, 0.0292, 0.0265, 0.0302, 0.0287, 0.0239, 0.0315, 0.0228],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-23 05:17:20,714 WARNING [train.py:1060] (2/4) Exclude cut with ID 3488-85273-0038-41224-0 from training. Duration: 20.34
2022-12-23 05:17:20,807 INFO [zipformer.py:660] (2/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=44663.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 05:17:38,194 INFO [scaling.py:679] (2/4) Whitening: num_groups=1, num_channels=384, metric=4.14 vs. limit=5.0
2022-12-23 05:17:38,663 INFO [train.py:894] (2/4) Epoch 13, batch 2600, loss[loss=0.2034, simple_loss=0.2854, pruned_loss=0.06075, over 18587.00 frames. ], tot_loss[loss=0.2097, simple_loss=0.2854, pruned_loss=0.06695, over 3714200.57 frames. ], batch size: 49, lr: 8.82e-03, grad_scale: 8.0
2022-12-23 05:18:20,317 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([2.3483, 2.0123, 1.3872, 2.2557, 1.7403, 1.8677, 1.8393, 2.4204],
       device='cuda:2'), covar=tensor([0.1919, 0.2872, 0.1918, 0.2531, 0.3268, 0.1036, 0.2614, 0.0814],
       device='cuda:2'), in_proj_covar=tensor([0.0277, 0.0269, 0.0228, 0.0341, 0.0250, 0.0213, 0.0265, 0.0189],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0003, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-23 05:18:33,345 WARNING [train.py:1060] (2/4) Exclude cut with ID 7357-94126-0026-15857-0_sp0.9 from training. Duration: 25.061125
2022-12-23 05:18:46,959 WARNING [train.py:1060] (2/4) Exclude cut with ID 3033-130750-0096-55598-0 from training. Duration: 0.83
2022-12-23 05:18:54,358 INFO [train.py:894] (2/4) Epoch 13, batch 2650, loss[loss=0.173, simple_loss=0.2502, pruned_loss=0.04793, over 18571.00 frames. ], tot_loss[loss=0.2078, simple_loss=0.2835, pruned_loss=0.06608, over 3714229.29 frames. ], batch size: 49, lr: 8.82e-03, grad_scale: 8.0
2022-12-23 05:19:02,485 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 2.920e+02 4.661e+02 5.769e+02 6.751e+02 1.660e+03, threshold=1.154e+03, percent-clipped=4.0
2022-12-23 05:19:11,977 WARNING [train.py:1060] (2/4) Exclude cut with ID 3488-85273-0017-41203-0 from training. Duration: 24.73
2022-12-23 05:19:22,472 WARNING [train.py:1060] (2/4) Exclude cut with ID 7255-291500-0009-134308-0 from training. Duration: 23.965
2022-12-23 05:19:31,320 WARNING [train.py:1060] (2/4) Exclude cut with ID 6978-92210-0030-146996-0_sp0.9 from training. Duration: 22.088875
2022-12-23 05:19:48,776 WARNING [train.py:1060] (2/4) Exclude cut with ID 7255-291500-0006-134305-0_sp0.9 from training. Duration: 23.6
2022-12-23 05:20:10,130 INFO [train.py:894] (2/4) Epoch 13, batch 2700, loss[loss=0.2029, simple_loss=0.2877, pruned_loss=0.05908, over 18487.00 frames. ], tot_loss[loss=0.2074, simple_loss=0.2837, pruned_loss=0.0656, over 3713910.45 frames. ], batch size: 64, lr: 8.81e-03, grad_scale: 8.0
2022-12-23 05:20:33,471 INFO [scaling.py:679] (2/4) Whitening: num_groups=8, num_channels=96, metric=1.17 vs. limit=2.0
2022-12-23 05:21:25,755 INFO [train.py:894] (2/4) Epoch 13, batch 2750, loss[loss=0.1937, simple_loss=0.2606, pruned_loss=0.0634, over 18474.00 frames. ], tot_loss[loss=0.2076, simple_loss=0.2836, pruned_loss=0.06581, over 3714041.75 frames. ], batch size: 43, lr: 8.81e-03, grad_scale: 8.0
2022-12-23 05:21:32,106 WARNING [train.py:1060] (2/4) Exclude cut with ID 8291-282929-0024-13011-0 from training. Duration: 23.795
2022-12-23 05:21:33,368 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 2.534e+02 5.060e+02 5.932e+02 7.826e+02 1.329e+03, threshold=1.186e+03, percent-clipped=3.0
2022-12-23 05:21:47,348 WARNING [train.py:1060] (2/4) Exclude cut with ID 8631-249866-0030-130156-0_sp1.1 from training. Duration: 21.5409375
2022-12-23 05:21:50,084 WARNING [train.py:1060] (2/4) Exclude cut with ID 6978-92210-0019-146985-0_sp0.9 from training. Duration: 24.97775
2022-12-23 05:22:02,420 WARNING [train.py:1060] (2/4) Exclude cut with ID 1085-156170-0017-128270-0_sp0.9 from training. Duration: 23.3444375
2022-12-23 05:22:12,474 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.4982, 1.0917, 1.3637, 1.3789, 1.6886, 1.5288, 1.5322, 1.1771],
       device='cuda:2'), covar=tensor([0.0244, 0.0250, 0.0451, 0.0210, 0.0194, 0.0332, 0.0226, 0.0267],
       device='cuda:2'), in_proj_covar=tensor([0.0088, 0.0122, 0.0144, 0.0123, 0.0111, 0.0111, 0.0093, 0.0122],
       device='cuda:2'), out_proj_covar=tensor([7.2047e-05, 9.9822e-05, 1.2340e-04, 1.0084e-04, 9.3212e-05, 8.7864e-05,
        7.5179e-05, 9.9480e-05], device='cuda:2')
2022-12-23 05:22:29,078 WARNING [train.py:1060] (2/4) Exclude cut with ID 6010-56788-0055-90261-0_sp0.9 from training. Duration: 23.2
2022-12-23 05:22:35,355 WARNING [train.py:1060] (2/4) Exclude cut with ID 5653-46179-0060-117930-0_sp0.9 from training. Duration: 21.17225
2022-12-23 05:22:41,454 INFO [train.py:894] (2/4) Epoch 13, batch 2800, loss[loss=0.191, simple_loss=0.2713, pruned_loss=0.0553, over 18708.00 frames. ], tot_loss[loss=0.2071, simple_loss=0.2833, pruned_loss=0.06549, over 3714072.82 frames. ], batch size: 50, lr: 8.80e-03, grad_scale: 8.0
2022-12-23 05:22:53,789 INFO [zipformer.py:660] (2/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=44883.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 05:22:54,013 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.3973, 1.3616, 1.4072, 1.3065, 0.7884, 2.2231, 0.9194, 1.4363],
       device='cuda:2'), covar=tensor([0.3445, 0.2004, 0.2069, 0.2146, 0.1445, 0.0363, 0.1631, 0.0911],
       device='cuda:2'), in_proj_covar=tensor([0.0135, 0.0116, 0.0126, 0.0119, 0.0102, 0.0098, 0.0093, 0.0091],
       device='cuda:2'), out_proj_covar=tensor([0.0005, 0.0005, 0.0005, 0.0004, 0.0004, 0.0003, 0.0004, 0.0004],
       device='cuda:2')
2022-12-23 05:22:55,125 WARNING [train.py:1060] (2/4) Exclude cut with ID 7255-291500-0008-134307-0_sp0.9 from training. Duration: 24.6555625
2022-12-23 05:23:28,761 INFO [zipformer.py:660] (2/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=44906.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 05:23:47,887 WARNING [train.py:1060] (2/4) Exclude cut with ID 3488-65654-0031-41259-0 from training. Duration: 20.44
2022-12-23 05:23:57,007 INFO [train.py:894] (2/4) Epoch 13, batch 2850, loss[loss=0.1793, simple_loss=0.2644, pruned_loss=0.04716, over 18424.00 frames. ], tot_loss[loss=0.2073, simple_loss=0.2833, pruned_loss=0.06566, over 3714165.60 frames. ], batch size: 48, lr: 8.80e-03, grad_scale: 8.0
2022-12-23 05:24:03,963 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 3.255e+02 5.409e+02 6.227e+02 8.445e+02 1.567e+03, threshold=1.245e+03, percent-clipped=8.0
2022-12-23 05:24:04,009 WARNING [train.py:1060] (2/4) Exclude cut with ID 6951-79737-0018-132285-0_sp0.9 from training. Duration: 23.45
2022-12-23 05:24:33,664 WARNING [train.py:1060] (2/4) Exclude cut with ID 6945-60535-0076-12784-0_sp0.9 from training. Duration: 20.52225
2022-12-23 05:24:42,410 WARNING [train.py:1060] (2/4) Exclude cut with ID 7255-291500-0008-134307-0 from training. Duration: 22.19
2022-12-23 05:24:51,319 WARNING [train.py:1060] (2/4) Exclude cut with ID 7357-94126-0014-15845-0_sp1.1 from training. Duration: 25.3818125
2022-12-23 05:24:53,586 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([2.3810, 1.8979, 1.4237, 1.4024, 2.2523, 2.6442, 2.6938, 1.8194],
       device='cuda:2'), covar=tensor([0.0296, 0.0317, 0.0542, 0.0313, 0.0230, 0.0304, 0.0220, 0.0343],
       device='cuda:2'), in_proj_covar=tensor([0.0089, 0.0123, 0.0146, 0.0124, 0.0113, 0.0112, 0.0094, 0.0124],
       device='cuda:2'), out_proj_covar=tensor([7.2867e-05, 1.0030e-04, 1.2466e-04, 1.0186e-04, 9.4602e-05, 8.8922e-05,
        7.5816e-05, 1.0101e-04], device='cuda:2')
2022-12-23 05:25:10,181 WARNING [train.py:1060] (2/4) Exclude cut with ID 6951-79737-0043-132310-0_sp0.9 from training. Duration: 28.0944375
2022-12-23 05:25:10,516 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.8289, 1.4699, 1.5496, 2.0157, 1.5398, 3.5609, 1.3302, 1.5992],
       device='cuda:2'), covar=tensor([0.0811, 0.1749, 0.1066, 0.0835, 0.1496, 0.0226, 0.1390, 0.1492],
       device='cuda:2'), in_proj_covar=tensor([0.0072, 0.0081, 0.0073, 0.0073, 0.0090, 0.0073, 0.0084, 0.0076],
       device='cuda:2'), out_proj_covar=tensor([0.0003, 0.0004, 0.0003, 0.0003, 0.0004, 0.0003, 0.0004, 0.0004],
       device='cuda:2')
2022-12-23 05:25:12,978 INFO [train.py:894] (2/4) Epoch 13, batch 2900, loss[loss=0.1896, simple_loss=0.2646, pruned_loss=0.0573, over 18535.00 frames. ], tot_loss[loss=0.2066, simple_loss=0.2829, pruned_loss=0.0652, over 3714430.90 frames. ], batch size: 47, lr: 8.79e-03, grad_scale: 8.0
2022-12-23 05:25:14,305 WARNING [train.py:1060] (2/4) Exclude cut with ID 2195-150901-0045-59933-0_sp0.9 from training. Duration: 22.9444375
2022-12-23 05:25:23,946 WARNING [train.py:1060] (2/4) Exclude cut with ID 8291-282929-0024-13011-0_sp1.1 from training. Duration: 21.6318125
2022-12-23 05:25:38,811 WARNING [train.py:1060] (2/4) Exclude cut with ID 8631-249866-0030-130156-0 from training. Duration: 23.695
2022-12-23 05:26:03,970 WARNING [train.py:1060] (2/4) Exclude cut with ID 7699-105389-0094-26379-0 from training. Duration: 23.955
2022-12-23 05:26:29,265 INFO [train.py:894] (2/4) Epoch 13, batch 2950, loss[loss=0.2346, simple_loss=0.2954, pruned_loss=0.08691, over 18676.00 frames. ], tot_loss[loss=0.2071, simple_loss=0.2829, pruned_loss=0.06561, over 3714661.73 frames. ], batch size: 46, lr: 8.79e-03, grad_scale: 8.0
2022-12-23 05:26:37,234 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 2.731e+02 4.907e+02 6.071e+02 7.847e+02 1.453e+03, threshold=1.214e+03, percent-clipped=2.0
2022-12-23 05:26:38,609 WARNING [train.py:1060] (2/4) Exclude cut with ID 8291-282929-0024-13011-0_sp0.9 from training. Duration: 26.438875
2022-12-23 05:27:02,283 INFO [scaling.py:679] (2/4) Whitening: num_groups=1, num_channels=384, metric=4.90 vs. limit=5.0
2022-12-23 05:27:16,074 INFO [zipformer.py:660] (2/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=45056.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 05:27:21,598 WARNING [train.py:1060] (2/4) Exclude cut with ID 7699-105389-0021-26306-0_sp0.9 from training. Duration: 21.2444375
2022-12-23 05:27:21,625 WARNING [train.py:1060] (2/4) Exclude cut with ID 7357-94126-0014-15845-0_sp0.9 from training. Duration: 31.02225
2022-12-23 05:27:32,324 WARNING [train.py:1060] (2/4) Exclude cut with ID 432-122774-0017-62487-0 from training. Duration: 22.395
2022-12-23 05:27:44,693 INFO [train.py:894] (2/4) Epoch 13, batch 3000, loss[loss=0.1983, simple_loss=0.2865, pruned_loss=0.05507, over 18507.00 frames. ], tot_loss[loss=0.2079, simple_loss=0.2836, pruned_loss=0.06614, over 3714381.36 frames. ], batch size: 52, lr: 8.78e-03, grad_scale: 8.0
2022-12-23 05:27:44,693 INFO [train.py:919] (2/4) Computing validation loss
2022-12-23 05:27:55,635 INFO [train.py:928] (2/4) Epoch 13, validation: loss=0.1686, simple_loss=0.268, pruned_loss=0.03461, over 944034.00 frames. 
2022-12-23 05:27:55,636 INFO [train.py:929] (2/4) Maximum memory allocated so far is 24680MB
2022-12-23 05:28:01,389 WARNING [train.py:1060] (2/4) Exclude cut with ID 7357-94126-0045-15876-0 from training. Duration: 21.075
2022-12-23 05:28:04,789 INFO [zipformer.py:660] (2/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=45081.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 05:28:05,877 WARNING [train.py:1060] (2/4) Exclude cut with ID 6482-98857-0025-147532-0_sp0.9 from training. Duration: 20.0055625
2022-12-23 05:28:05,884 WARNING [train.py:1060] (2/4) Exclude cut with ID 6951-79737-0037-132304-0_sp0.9 from training. Duration: 22.05
2022-12-23 05:28:05,895 WARNING [train.py:1060] (2/4) Exclude cut with ID 7255-291500-0003-134302-0 from training. Duration: 26.8349375
2022-12-23 05:28:08,003 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([2.5712, 1.5198, 1.1073, 2.3339, 2.7282, 1.7099, 2.3374, 2.4932],
       device='cuda:2'), covar=tensor([0.1360, 0.2193, 0.2244, 0.1309, 0.1427, 0.1586, 0.1299, 0.1452],
       device='cuda:2'), in_proj_covar=tensor([0.0090, 0.0098, 0.0118, 0.0096, 0.0113, 0.0090, 0.0097, 0.0093],
       device='cuda:2'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003],
       device='cuda:2')
2022-12-23 05:28:10,233 WARNING [train.py:1060] (2/4) Exclude cut with ID 7357-94126-0009-15840-0_sp1.1 from training. Duration: 22.1090625
2022-12-23 05:28:11,306 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([2.2117, 2.0861, 1.6793, 0.9566, 2.6674, 2.3013, 2.0050, 1.5417],
       device='cuda:2'), covar=tensor([0.0345, 0.0358, 0.0500, 0.0692, 0.0173, 0.0299, 0.0396, 0.0792],
       device='cuda:2'), in_proj_covar=tensor([0.0119, 0.0118, 0.0126, 0.0116, 0.0087, 0.0115, 0.0129, 0.0148],
       device='cuda:2'), out_proj_covar=tensor([0.0001, 0.0001, 0.0002, 0.0001, 0.0001, 0.0001, 0.0002, 0.0002],
       device='cuda:2')
2022-12-23 05:28:17,087 WARNING [train.py:1060] (2/4) Exclude cut with ID 7699-105389-0094-26379-0_sp0.9 from training. Duration: 26.6166875
2022-12-23 05:28:29,896 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([2.3899, 2.3536, 1.6111, 0.6006, 1.5882, 1.9413, 1.6820, 1.8368],
       device='cuda:2'), covar=tensor([0.0601, 0.0369, 0.1084, 0.1516, 0.1122, 0.1406, 0.1477, 0.0691],
       device='cuda:2'), in_proj_covar=tensor([0.0165, 0.0179, 0.0202, 0.0190, 0.0205, 0.0194, 0.0209, 0.0195],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-23 05:28:36,773 WARNING [train.py:1060] (2/4) Exclude cut with ID 2046-178027-0000-53705-0_sp0.9 from training. Duration: 20.3055625
2022-12-23 05:28:57,482 WARNING [train.py:1060] (2/4) Exclude cut with ID 7205-50138-0008-5373-0_sp0.9 from training. Duration: 20.7
2022-12-23 05:28:59,548 INFO [zipformer.py:660] (2/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=45117.0, num_to_drop=1, layers_to_drop={0}
2022-12-23 05:29:02,872 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([0.6937, 1.9411, 2.2506, 1.0004, 1.3682, 2.4964, 1.9271, 1.8794],
       device='cuda:2'), covar=tensor([0.0725, 0.0307, 0.0333, 0.0389, 0.0370, 0.0369, 0.0266, 0.0640],
       device='cuda:2'), in_proj_covar=tensor([0.0143, 0.0161, 0.0117, 0.0133, 0.0143, 0.0135, 0.0152, 0.0156],
       device='cuda:2'), out_proj_covar=tensor([1.1381e-04, 1.2954e-04, 9.2546e-05, 1.0363e-04, 1.1316e-04, 1.0931e-04,
        1.2319e-04, 1.2430e-04], device='cuda:2')
2022-12-23 05:29:11,120 INFO [train.py:894] (2/4) Epoch 13, batch 3050, loss[loss=0.2159, simple_loss=0.2949, pruned_loss=0.06844, over 18510.00 frames. ], tot_loss[loss=0.2075, simple_loss=0.2834, pruned_loss=0.0658, over 3714587.39 frames. ], batch size: 64, lr: 8.78e-03, grad_scale: 8.0
2022-12-23 05:29:18,119 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 3.268e+02 5.093e+02 6.049e+02 7.784e+02 1.480e+03, threshold=1.210e+03, percent-clipped=4.0
2022-12-23 05:29:36,881 INFO [zipformer.py:660] (2/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=45142.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 05:29:43,019 WARNING [train.py:1060] (2/4) Exclude cut with ID 6978-92210-0019-146985-0 from training. Duration: 22.48
2022-12-23 05:29:58,144 WARNING [train.py:1060] (2/4) Exclude cut with ID 7255-291500-0003-134302-0_sp0.9 from training. Duration: 29.816625
2022-12-23 05:30:17,665 WARNING [train.py:1060] (2/4) Exclude cut with ID 7255-291500-0005-134304-0_sp1.1 from training. Duration: 22.7590625
2022-12-23 05:30:23,366 WARNING [train.py:1060] (2/4) Exclude cut with ID 7357-94126-0026-15857-0 from training. Duration: 22.555
2022-12-23 05:30:26,298 INFO [train.py:894] (2/4) Epoch 13, batch 3100, loss[loss=0.1875, simple_loss=0.2671, pruned_loss=0.05393, over 18662.00 frames. ], tot_loss[loss=0.2072, simple_loss=0.2834, pruned_loss=0.06548, over 3714813.18 frames. ], batch size: 48, lr: 8.78e-03, grad_scale: 8.0
2022-12-23 05:30:39,076 INFO [zipformer.py:660] (2/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=45183.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 05:30:40,566 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.5601, 1.0240, 1.7558, 2.9068, 1.9284, 2.3651, 0.9100, 1.8894],
       device='cuda:2'), covar=tensor([0.2008, 0.2132, 0.1580, 0.0599, 0.1268, 0.1207, 0.2272, 0.1363],
       device='cuda:2'), in_proj_covar=tensor([0.0102, 0.0114, 0.0130, 0.0133, 0.0104, 0.0132, 0.0128, 0.0109],
       device='cuda:2'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0004, 0.0003, 0.0003],
       device='cuda:2')
2022-12-23 05:30:43,683 WARNING [train.py:1060] (2/4) Exclude cut with ID 1250-135782-0005-25975-0_sp0.9 from training. Duration: 21.688875
2022-12-23 05:30:59,132 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([2.5994, 2.3600, 1.8408, 0.8316, 1.8169, 2.0933, 1.7243, 1.9444],
       device='cuda:2'), covar=tensor([0.0642, 0.0504, 0.1278, 0.1566, 0.1304, 0.1377, 0.1597, 0.0858],
       device='cuda:2'), in_proj_covar=tensor([0.0166, 0.0180, 0.0203, 0.0192, 0.0206, 0.0195, 0.0209, 0.0196],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-23 05:31:14,918 INFO [zipformer.py:660] (2/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=45206.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 05:31:16,267 WARNING [train.py:1060] (2/4) Exclude cut with ID 3488-85273-0038-41224-0_sp0.9 from training. Duration: 22.6
2022-12-23 05:31:30,820 INFO [scaling.py:679] (2/4) Whitening: num_groups=8, num_channels=96, metric=1.35 vs. limit=2.0
2022-12-23 05:31:42,791 INFO [train.py:894] (2/4) Epoch 13, batch 3150, loss[loss=0.1989, simple_loss=0.283, pruned_loss=0.05745, over 18626.00 frames. ], tot_loss[loss=0.2078, simple_loss=0.2838, pruned_loss=0.06593, over 3715794.66 frames. ], batch size: 97, lr: 8.77e-03, grad_scale: 8.0
2022-12-23 05:31:50,719 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 2.862e+02 4.724e+02 5.980e+02 7.666e+02 1.502e+03, threshold=1.196e+03, percent-clipped=2.0
2022-12-23 05:31:51,709 INFO [scaling.py:679] (2/4) Whitening: num_groups=8, num_channels=96, metric=1.38 vs. limit=2.0
2022-12-23 05:31:52,617 WARNING [train.py:1060] (2/4) Exclude cut with ID 7357-94126-0009-15840-0 from training. Duration: 24.32
2022-12-23 05:31:52,713 INFO [zipformer.py:660] (2/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=45231.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 05:32:00,358 INFO [scaling.py:679] (2/4) Whitening: num_groups=8, num_channels=96, metric=1.54 vs. limit=2.0
2022-12-23 05:32:02,605 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.7019, 1.6333, 1.2747, 1.6665, 1.8229, 1.5973, 2.1148, 1.8328],
       device='cuda:2'), covar=tensor([0.0880, 0.1598, 0.2684, 0.1568, 0.1812, 0.0899, 0.1094, 0.1080],
       device='cuda:2'), in_proj_covar=tensor([0.0174, 0.0198, 0.0241, 0.0287, 0.0229, 0.0187, 0.0211, 0.0200],
       device='cuda:2'), out_proj_covar=tensor([0.0001, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-23 05:32:29,058 INFO [zipformer.py:660] (2/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=45254.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 05:32:34,856 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.5728, 1.5935, 1.6165, 1.5234, 1.1086, 3.5986, 1.5587, 1.9442],
       device='cuda:2'), covar=tensor([0.3238, 0.2006, 0.1972, 0.2126, 0.1509, 0.0167, 0.1577, 0.0879],
       device='cuda:2'), in_proj_covar=tensor([0.0137, 0.0117, 0.0128, 0.0121, 0.0104, 0.0099, 0.0094, 0.0091],
       device='cuda:2'), out_proj_covar=tensor([0.0005, 0.0005, 0.0005, 0.0005, 0.0004, 0.0003, 0.0004, 0.0004],
       device='cuda:2')
2022-12-23 05:32:53,379 WARNING [train.py:1060] (2/4) Exclude cut with ID 8291-276745-0093-13116-0_sp0.9 from training. Duration: 21.061125
2022-12-23 05:32:59,694 INFO [train.py:894] (2/4) Epoch 13, batch 3200, loss[loss=0.2145, simple_loss=0.2888, pruned_loss=0.0701, over 18551.00 frames. ], tot_loss[loss=0.2079, simple_loss=0.2838, pruned_loss=0.06598, over 3715740.45 frames. ], batch size: 77, lr: 8.77e-03, grad_scale: 8.0
2022-12-23 05:33:07,118 WARNING [train.py:1060] (2/4) Exclude cut with ID 7357-94126-0024-15855-0_sp0.9 from training. Duration: 20.32225
2022-12-23 05:33:22,371 WARNING [train.py:1060] (2/4) Exclude cut with ID 3033-130750-0096-55598-0_sp1.1 from training. Duration: 0.7545625
2022-12-23 05:33:37,176 WARNING [train.py:1060] (2/4) Exclude cut with ID 4295-39940-0007-92567-0_sp0.9 from training. Duration: 23.9333125
2022-12-23 05:34:04,659 INFO [scaling.py:679] (2/4) Whitening: num_groups=8, num_channels=96, metric=1.38 vs. limit=2.0
2022-12-23 05:34:09,845 WARNING [train.py:1060] (2/4) Exclude cut with ID 7255-291500-0008-134307-0_sp1.1 from training. Duration: 20.17275
2022-12-23 05:34:16,008 INFO [train.py:894] (2/4) Epoch 13, batch 3250, loss[loss=0.2299, simple_loss=0.3052, pruned_loss=0.07733, over 18613.00 frames. ], tot_loss[loss=0.2062, simple_loss=0.2822, pruned_loss=0.06511, over 3715049.19 frames. ], batch size: 78, lr: 8.76e-03, grad_scale: 8.0
2022-12-23 05:34:16,025 WARNING [train.py:1060] (2/4) Exclude cut with ID 6978-92210-0019-146985-0_sp1.1 from training. Duration: 20.436375
2022-12-23 05:34:24,014 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 3.138e+02 5.303e+02 6.543e+02 7.796e+02 1.452e+03, threshold=1.309e+03, percent-clipped=4.0
2022-12-23 05:34:44,007 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([2.7472, 2.1546, 1.5536, 2.5954, 1.9054, 1.9686, 2.0860, 2.7692],
       device='cuda:2'), covar=tensor([0.1756, 0.2897, 0.1851, 0.2691, 0.3207, 0.1096, 0.2780, 0.0706],
       device='cuda:2'), in_proj_covar=tensor([0.0280, 0.0273, 0.0233, 0.0344, 0.0255, 0.0217, 0.0268, 0.0194],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0003, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-23 05:34:47,062 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.8785, 1.9148, 1.3033, 2.0369, 2.0951, 1.7765, 2.5988, 2.0096],
       device='cuda:2'), covar=tensor([0.0898, 0.1583, 0.2833, 0.1662, 0.1732, 0.0913, 0.1034, 0.1152],
       device='cuda:2'), in_proj_covar=tensor([0.0174, 0.0198, 0.0241, 0.0288, 0.0229, 0.0187, 0.0211, 0.0200],
       device='cuda:2'), out_proj_covar=tensor([0.0001, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-23 05:34:51,978 INFO [zipformer.py:660] (2/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=45348.0, num_to_drop=1, layers_to_drop={1}
2022-12-23 05:35:32,625 INFO [train.py:894] (2/4) Epoch 13, batch 3300, loss[loss=0.197, simple_loss=0.264, pruned_loss=0.06503, over 18599.00 frames. ], tot_loss[loss=0.2053, simple_loss=0.2816, pruned_loss=0.06448, over 3714761.76 frames. ], batch size: 45, lr: 8.76e-03, grad_scale: 8.0
2022-12-23 05:35:38,572 WARNING [train.py:1060] (2/4) Exclude cut with ID 4234-40345-0022-142709-0_sp0.9 from training. Duration: 23.1055625
2022-12-23 05:35:39,304 INFO [scaling.py:679] (2/4) Whitening: num_groups=8, num_channels=96, metric=1.14 vs. limit=2.0
2022-12-23 05:35:40,137 WARNING [train.py:1060] (2/4) Exclude cut with ID 8291-282929-0007-12994-0_sp1.1 from training. Duration: 23.5
2022-12-23 05:35:52,653 WARNING [train.py:1060] (2/4) Exclude cut with ID 7255-291500-0009-134308-0_sp0.9 from training. Duration: 26.62775
2022-12-23 05:36:04,458 WARNING [train.py:1060] (2/4) Exclude cut with ID 6951-79737-0018-132285-0 from training. Duration: 21.105
2022-12-23 05:36:09,145 WARNING [train.py:1060] (2/4) Exclude cut with ID 4511-76322-0006-80011-0_sp0.9 from training. Duration: 24.411125
2022-12-23 05:36:24,783 INFO [zipformer.py:660] (2/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=45409.0, num_to_drop=1, layers_to_drop={0}
2022-12-23 05:36:29,104 INFO [zipformer.py:660] (2/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=45412.0, num_to_drop=1, layers_to_drop={3}
2022-12-23 05:36:35,331 WARNING [train.py:1060] (2/4) Exclude cut with ID 6758-72288-0033-108368-0_sp1.1 from training. Duration: 21.263625
2022-12-23 05:36:49,535 INFO [train.py:894] (2/4) Epoch 13, batch 3350, loss[loss=0.1934, simple_loss=0.2681, pruned_loss=0.05941, over 18404.00 frames. ], tot_loss[loss=0.2056, simple_loss=0.2822, pruned_loss=0.06453, over 3714278.37 frames. ], batch size: 42, lr: 8.75e-03, grad_scale: 8.0
2022-12-23 05:36:57,614 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 3.516e+02 4.658e+02 5.777e+02 7.131e+02 1.099e+03, threshold=1.155e+03, percent-clipped=0.0
2022-12-23 05:37:05,376 WARNING [train.py:1060] (2/4) Exclude cut with ID 4234-40345-0022-142709-0 from training. Duration: 20.795
2022-12-23 05:37:08,428 INFO [zipformer.py:660] (2/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=45437.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 05:37:16,386 WARNING [train.py:1060] (2/4) Exclude cut with ID 7357-94126-0021-15852-0 from training. Duration: 24.76
2022-12-23 05:37:17,974 WARNING [train.py:1060] (2/4) Exclude cut with ID 3867-173237-0077-144769-0_sp0.9 from training. Duration: 22.25
2022-12-23 05:37:40,177 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.9855, 1.5556, 1.9032, 2.5401, 2.0773, 4.6936, 1.5973, 2.0199],
       device='cuda:2'), covar=tensor([0.0766, 0.1820, 0.1092, 0.0813, 0.1324, 0.0182, 0.1348, 0.1438],
       device='cuda:2'), in_proj_covar=tensor([0.0070, 0.0080, 0.0072, 0.0073, 0.0089, 0.0071, 0.0083, 0.0076],
       device='cuda:2'), out_proj_covar=tensor([0.0003, 0.0004, 0.0003, 0.0003, 0.0004, 0.0003, 0.0004, 0.0004],
       device='cuda:2')
2022-12-23 05:37:42,588 WARNING [train.py:1060] (2/4) Exclude cut with ID 7357-94126-0026-15857-0_sp1.1 from training. Duration: 20.5045625
2022-12-23 05:38:05,793 INFO [train.py:894] (2/4) Epoch 13, batch 3400, loss[loss=0.2255, simple_loss=0.2924, pruned_loss=0.07933, over 18649.00 frames. ], tot_loss[loss=0.2058, simple_loss=0.2822, pruned_loss=0.06467, over 3714169.05 frames. ], batch size: 179, lr: 8.75e-03, grad_scale: 8.0
2022-12-23 05:39:16,509 INFO [train.py:894] (2/4) Epoch 13, batch 3450, loss[loss=0.2132, simple_loss=0.2951, pruned_loss=0.06566, over 18594.00 frames. ], tot_loss[loss=0.2067, simple_loss=0.283, pruned_loss=0.0652, over 3714063.30 frames. ], batch size: 51, lr: 8.74e-03, grad_scale: 8.0
2022-12-23 05:39:18,257 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.9297, 2.5331, 1.9001, 2.8475, 3.2141, 1.8887, 2.1582, 1.5209],
       device='cuda:2'), covar=tensor([0.1841, 0.1525, 0.1417, 0.0812, 0.1249, 0.1023, 0.1786, 0.1427],
       device='cuda:2'), in_proj_covar=tensor([0.0241, 0.0216, 0.0204, 0.0192, 0.0256, 0.0191, 0.0215, 0.0192],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-23 05:39:23,733 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 3.221e+02 5.288e+02 6.220e+02 8.428e+02 1.484e+03, threshold=1.244e+03, percent-clipped=4.0
2022-12-23 05:40:13,342 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.5853, 1.2865, 1.3191, 0.9181, 1.6962, 1.4166, 1.3915, 1.1335],
       device='cuda:2'), covar=tensor([0.0382, 0.0537, 0.0519, 0.0686, 0.0343, 0.0399, 0.0493, 0.1015],
       device='cuda:2'), in_proj_covar=tensor([0.0122, 0.0123, 0.0130, 0.0119, 0.0090, 0.0118, 0.0134, 0.0154],
       device='cuda:2'), out_proj_covar=tensor([0.0001, 0.0002, 0.0002, 0.0001, 0.0001, 0.0001, 0.0002, 0.0002],
       device='cuda:2')
2022-12-23 05:40:29,606 INFO [train.py:894] (2/4) Epoch 13, batch 3500, loss[loss=0.2258, simple_loss=0.3063, pruned_loss=0.07265, over 18604.00 frames. ], tot_loss[loss=0.2071, simple_loss=0.2834, pruned_loss=0.06545, over 3713861.76 frames. ], batch size: 98, lr: 8.74e-03, grad_scale: 8.0
2022-12-23 05:40:49,918 WARNING [train.py:1060] (2/4) Exclude cut with ID 7859-102521-0017-7548-0_sp1.1 from training. Duration: 22.2954375
2022-12-23 05:41:02,150 INFO [train.py:894] (2/4) Epoch 14, batch 0, loss[loss=0.2009, simple_loss=0.2795, pruned_loss=0.06111, over 18677.00 frames. ], tot_loss[loss=0.2009, simple_loss=0.2795, pruned_loss=0.06111, over 18677.00 frames. ], batch size: 48, lr: 8.42e-03, grad_scale: 8.0
2022-12-23 05:41:02,150 INFO [train.py:919] (2/4) Computing validation loss
2022-12-23 05:41:12,887 INFO [train.py:928] (2/4) Epoch 14, validation: loss=0.1689, simple_loss=0.2684, pruned_loss=0.03472, over 944034.00 frames. 
2022-12-23 05:41:12,888 INFO [train.py:929] (2/4) Maximum memory allocated so far is 24680MB
2022-12-23 05:41:33,698 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([2.4263, 1.9396, 1.9933, 2.2723, 2.3684, 2.2554, 2.3030, 1.6418],
       device='cuda:2'), covar=tensor([0.1843, 0.2822, 0.2136, 0.2427, 0.1471, 0.0778, 0.2506, 0.1082],
       device='cuda:2'), in_proj_covar=tensor([0.0258, 0.0292, 0.0266, 0.0301, 0.0288, 0.0241, 0.0316, 0.0228],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-23 05:41:41,478 INFO [scaling.py:679] (2/4) Whitening: num_groups=8, num_channels=96, metric=1.21 vs. limit=2.0
2022-12-23 05:42:09,106 WARNING [train.py:1060] (2/4) Exclude cut with ID 298-126791-0067-24026-0_sp0.9 from training. Duration: 21.438875
2022-12-23 05:42:13,480 WARNING [train.py:1060] (2/4) Exclude cut with ID 5652-39938-0025-23684-0_sp0.9 from training. Duration: 22.2055625
2022-12-23 05:42:23,910 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([2.6388, 2.4892, 1.8208, 1.3235, 3.0632, 2.7220, 2.1990, 1.8147],
       device='cuda:2'), covar=tensor([0.0359, 0.0371, 0.0565, 0.0753, 0.0165, 0.0303, 0.0450, 0.0794],
       device='cuda:2'), in_proj_covar=tensor([0.0123, 0.0123, 0.0130, 0.0120, 0.0090, 0.0119, 0.0134, 0.0154],
       device='cuda:2'), out_proj_covar=tensor([0.0001, 0.0002, 0.0002, 0.0001, 0.0001, 0.0001, 0.0002, 0.0002],
       device='cuda:2')
2022-12-23 05:42:26,606 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 2.512e+02 4.529e+02 5.962e+02 7.073e+02 1.880e+03, threshold=1.192e+03, percent-clipped=5.0
2022-12-23 05:42:28,168 INFO [train.py:894] (2/4) Epoch 14, batch 50, loss[loss=0.2402, simple_loss=0.3146, pruned_loss=0.08289, over 18589.00 frames. ], tot_loss[loss=0.1972, simple_loss=0.2808, pruned_loss=0.05681, over 837468.77 frames. ], batch size: 57, lr: 8.41e-03, grad_scale: 8.0
2022-12-23 05:42:38,195 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([2.6840, 2.5821, 1.9722, 1.4873, 3.0395, 2.7389, 2.2889, 1.8406],
       device='cuda:2'), covar=tensor([0.0341, 0.0324, 0.0519, 0.0706, 0.0170, 0.0309, 0.0449, 0.0785],
       device='cuda:2'), in_proj_covar=tensor([0.0122, 0.0123, 0.0130, 0.0119, 0.0090, 0.0118, 0.0134, 0.0154],
       device='cuda:2'), out_proj_covar=tensor([0.0001, 0.0002, 0.0002, 0.0001, 0.0001, 0.0001, 0.0002, 0.0002],
       device='cuda:2')
2022-12-23 05:43:43,695 INFO [train.py:894] (2/4) Epoch 14, batch 100, loss[loss=0.1768, simple_loss=0.2666, pruned_loss=0.04356, over 18577.00 frames. ], tot_loss[loss=0.1925, simple_loss=0.2783, pruned_loss=0.05336, over 1475516.11 frames. ], batch size: 49, lr: 8.41e-03, grad_scale: 8.0
2022-12-23 05:44:17,270 INFO [zipformer.py:660] (2/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=45702.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 05:44:17,320 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.4579, 2.8177, 3.4145, 1.0129, 2.7135, 3.5386, 2.4915, 2.9026],
       device='cuda:2'), covar=tensor([0.0657, 0.0323, 0.0285, 0.0464, 0.0346, 0.0311, 0.0322, 0.0529],
       device='cuda:2'), in_proj_covar=tensor([0.0142, 0.0160, 0.0115, 0.0133, 0.0141, 0.0134, 0.0150, 0.0156],
       device='cuda:2'), out_proj_covar=tensor([1.1273e-04, 1.2827e-04, 9.1078e-05, 1.0302e-04, 1.1112e-04, 1.0846e-04,
        1.2188e-04, 1.2417e-04], device='cuda:2')
2022-12-23 05:44:19,920 INFO [zipformer.py:660] (2/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=45704.0, num_to_drop=1, layers_to_drop={2}
2022-12-23 05:44:31,786 INFO [zipformer.py:660] (2/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=45712.0, num_to_drop=1, layers_to_drop={0}
2022-12-23 05:44:59,361 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 2.734e+02 3.533e+02 4.272e+02 5.188e+02 1.286e+03, threshold=8.545e+02, percent-clipped=1.0
2022-12-23 05:45:00,662 INFO [train.py:894] (2/4) Epoch 14, batch 150, loss[loss=0.1855, simple_loss=0.2739, pruned_loss=0.04859, over 18474.00 frames. ], tot_loss[loss=0.1893, simple_loss=0.2746, pruned_loss=0.052, over 1971757.03 frames. ], batch size: 77, lr: 8.40e-03, grad_scale: 8.0
2022-12-23 05:45:09,921 WARNING [train.py:1060] (2/4) Exclude cut with ID 7859-102521-0017-7548-0 from training. Duration: 24.525
2022-12-23 05:45:10,197 INFO [zipformer.py:660] (2/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=45737.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 05:45:43,332 WARNING [train.py:1060] (2/4) Exclude cut with ID 3699-47246-0007-3408-0_sp0.9 from training. Duration: 20.26675
2022-12-23 05:45:45,265 INFO [zipformer.py:660] (2/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=45760.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 05:45:49,737 INFO [zipformer.py:660] (2/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=45763.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 05:45:56,284 WARNING [train.py:1060] (2/4) Exclude cut with ID 7859-102521-0017-7548-0_sp0.9 from training. Duration: 27.25
2022-12-23 05:46:05,157 INFO [zipformer.py:660] (2/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=45774.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 05:46:14,108 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.3617, 1.9265, 1.3306, 2.0757, 2.7209, 1.4984, 1.5319, 1.1726],
       device='cuda:2'), covar=tensor([0.2056, 0.1715, 0.1663, 0.1031, 0.1134, 0.1128, 0.2010, 0.1564],
       device='cuda:2'), in_proj_covar=tensor([0.0239, 0.0215, 0.0204, 0.0191, 0.0254, 0.0190, 0.0214, 0.0191],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-23 05:46:15,002 INFO [train.py:894] (2/4) Epoch 14, batch 200, loss[loss=0.1844, simple_loss=0.2776, pruned_loss=0.04563, over 18571.00 frames. ], tot_loss[loss=0.1891, simple_loss=0.2742, pruned_loss=0.052, over 2358562.44 frames. ], batch size: 56, lr: 8.40e-03, grad_scale: 8.0
2022-12-23 05:46:25,674 INFO [zipformer.py:660] (2/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=45785.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 05:46:53,729 INFO [scaling.py:679] (2/4) Whitening: num_groups=1, num_channels=384, metric=4.68 vs. limit=5.0
2022-12-23 05:46:58,283 INFO [scaling.py:679] (2/4) Whitening: num_groups=8, num_channels=96, metric=1.18 vs. limit=2.0
2022-12-23 05:47:15,434 WARNING [train.py:1060] (2/4) Exclude cut with ID 6426-64292-0017-15984-0 from training. Duration: 21.68
2022-12-23 05:47:25,836 WARNING [train.py:1060] (2/4) Exclude cut with ID 4278-13270-0007-59342-0 from training. Duration: 21.6300625
2022-12-23 05:47:34,693 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 2.785e+02 3.863e+02 4.748e+02 5.932e+02 2.201e+03, threshold=9.496e+02, percent-clipped=7.0
2022-12-23 05:47:36,848 INFO [train.py:894] (2/4) Epoch 14, batch 250, loss[loss=0.174, simple_loss=0.2469, pruned_loss=0.05049, over 18585.00 frames. ], tot_loss[loss=0.1886, simple_loss=0.2737, pruned_loss=0.05175, over 2658892.98 frames. ], batch size: 41, lr: 8.40e-03, grad_scale: 8.0
2022-12-23 05:47:44,254 INFO [zipformer.py:660] (2/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=45835.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 05:47:51,518 WARNING [train.py:1060] (2/4) Exclude cut with ID 4278-13270-0007-59342-0_sp0.9 from training. Duration: 24.033375
2022-12-23 05:48:19,395 INFO [zipformer.py:660] (2/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=45859.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 05:48:46,428 WARNING [train.py:1060] (2/4) Exclude cut with ID 4278-13270-0009-59344-0 from training. Duration: 22.905
2022-12-23 05:48:48,384 WARNING [train.py:1060] (2/4) Exclude cut with ID 5622-44585-0006-90525-0_sp1.1 from training. Duration: 23.4318125
2022-12-23 05:48:51,791 INFO [train.py:894] (2/4) Epoch 14, batch 300, loss[loss=0.2162, simple_loss=0.2992, pruned_loss=0.06663, over 18561.00 frames. ], tot_loss[loss=0.1873, simple_loss=0.2726, pruned_loss=0.05101, over 2893020.67 frames. ], batch size: 78, lr: 8.39e-03, grad_scale: 8.0
2022-12-23 05:49:22,975 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([2.1819, 1.5806, 1.7507, 1.8705, 2.0936, 2.1221, 2.0214, 1.5809],
       device='cuda:2'), covar=tensor([0.1904, 0.2900, 0.2265, 0.2513, 0.1806, 0.0861, 0.2722, 0.1164],
       device='cuda:2'), in_proj_covar=tensor([0.0257, 0.0293, 0.0267, 0.0303, 0.0291, 0.0242, 0.0318, 0.0229],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-23 05:49:46,466 INFO [scaling.py:679] (2/4) Whitening: num_groups=1, num_channels=384, metric=4.06 vs. limit=5.0
2022-12-23 05:49:50,473 INFO [zipformer.py:660] (2/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=45920.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 05:50:05,347 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 2.191e+02 3.620e+02 4.473e+02 5.511e+02 1.268e+03, threshold=8.946e+02, percent-clipped=1.0
2022-12-23 05:50:06,952 INFO [train.py:894] (2/4) Epoch 14, batch 350, loss[loss=0.1709, simple_loss=0.2501, pruned_loss=0.04588, over 18403.00 frames. ], tot_loss[loss=0.187, simple_loss=0.272, pruned_loss=0.05099, over 3074759.58 frames. ], batch size: 42, lr: 8.39e-03, grad_scale: 8.0
2022-12-23 05:50:46,096 WARNING [train.py:1060] (2/4) Exclude cut with ID 4278-13270-0009-59344-0_sp1.1 from training. Duration: 20.82275
2022-12-23 05:50:47,648 WARNING [train.py:1060] (2/4) Exclude cut with ID 4278-13270-0009-59344-0_sp0.9 from training. Duration: 25.45
2022-12-23 05:51:24,538 INFO [train.py:894] (2/4) Epoch 14, batch 400, loss[loss=0.187, simple_loss=0.2718, pruned_loss=0.05103, over 18711.00 frames. ], tot_loss[loss=0.1886, simple_loss=0.2734, pruned_loss=0.0519, over 3216859.30 frames. ], batch size: 54, lr: 8.38e-03, grad_scale: 8.0
2022-12-23 05:51:48,662 WARNING [train.py:1060] (2/4) Exclude cut with ID 5622-44585-0006-90525-0 from training. Duration: 25.775
2022-12-23 05:52:01,557 INFO [zipformer.py:660] (2/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=46004.0, num_to_drop=1, layers_to_drop={1}
2022-12-23 05:52:13,161 WARNING [train.py:1060] (2/4) Exclude cut with ID 2929-85685-0071-62375-0_sp0.9 from training. Duration: 22.25
2022-12-23 05:52:40,420 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 2.282e+02 3.982e+02 5.008e+02 5.983e+02 8.979e+02, threshold=1.002e+03, percent-clipped=1.0
2022-12-23 05:52:40,441 WARNING [train.py:1060] (2/4) Exclude cut with ID 3972-170212-0014-23379-0 from training. Duration: 26.205
2022-12-23 05:52:41,920 INFO [train.py:894] (2/4) Epoch 14, batch 450, loss[loss=0.2046, simple_loss=0.2949, pruned_loss=0.0571, over 18596.00 frames. ], tot_loss[loss=0.1917, simple_loss=0.2763, pruned_loss=0.05356, over 3327294.41 frames. ], batch size: 69, lr: 8.38e-03, grad_scale: 8.0
2022-12-23 05:52:56,674 WARNING [train.py:1060] (2/4) Exclude cut with ID 5239-32139-0047-9341-0_sp0.9 from training. Duration: 30.1555625
2022-12-23 05:53:01,155 WARNING [train.py:1060] (2/4) Exclude cut with ID 1265-135635-0050-6781-0_sp0.9 from training. Duration: 21.8333125
2022-12-23 05:53:10,758 WARNING [train.py:1060] (2/4) Exclude cut with ID 1914-133440-0024-94914-0_sp1.1 from training. Duration: 20.6545625
2022-12-23 05:53:13,746 INFO [zipformer.py:660] (2/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=46052.0, num_to_drop=1, layers_to_drop={0}
2022-12-23 05:53:22,315 INFO [zipformer.py:660] (2/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=46058.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 05:53:34,162 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([3.7941, 3.2403, 2.7464, 1.2372, 2.5662, 2.9625, 2.4445, 2.7456],
       device='cuda:2'), covar=tensor([0.0575, 0.0609, 0.1366, 0.1965, 0.1680, 0.1093, 0.1325, 0.1015],
       device='cuda:2'), in_proj_covar=tensor([0.0167, 0.0182, 0.0202, 0.0194, 0.0209, 0.0194, 0.0210, 0.0194],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-23 05:53:43,250 INFO [scaling.py:679] (2/4) Whitening: num_groups=8, num_channels=192, metric=1.90 vs. limit=2.0
2022-12-23 05:53:53,846 WARNING [train.py:1060] (2/4) Exclude cut with ID 7395-89880-0045-39920-0_sp0.9 from training. Duration: 20.52225
2022-12-23 05:53:56,433 INFO [train.py:894] (2/4) Epoch 14, batch 500, loss[loss=0.1747, simple_loss=0.2584, pruned_loss=0.04548, over 18399.00 frames. ], tot_loss[loss=0.192, simple_loss=0.2764, pruned_loss=0.05382, over 3413405.13 frames. ], batch size: 46, lr: 8.37e-03, grad_scale: 8.0
2022-12-23 05:54:12,939 WARNING [train.py:1060] (2/4) Exclude cut with ID 3972-170212-0014-23379-0_sp0.9 from training. Duration: 29.1166875
2022-12-23 05:55:10,416 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 2.469e+02 4.016e+02 4.935e+02 6.120e+02 1.386e+03, threshold=9.871e+02, percent-clipped=2.0
2022-12-23 05:55:10,612 INFO [zipformer.py:660] (2/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=46130.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 05:55:12,151 INFO [train.py:894] (2/4) Epoch 14, batch 550, loss[loss=0.1963, simple_loss=0.2853, pruned_loss=0.05367, over 18512.00 frames. ], tot_loss[loss=0.1922, simple_loss=0.277, pruned_loss=0.05373, over 3480391.51 frames. ], batch size: 52, lr: 8.37e-03, grad_scale: 8.0
2022-12-23 05:55:15,265 WARNING [train.py:1060] (2/4) Exclude cut with ID 543-133211-0007-59831-0_sp0.9 from training. Duration: 21.388875
2022-12-23 05:55:50,812 WARNING [train.py:1060] (2/4) Exclude cut with ID 1914-133440-0024-94914-0 from training. Duration: 22.72
2022-12-23 05:55:52,346 WARNING [train.py:1060] (2/4) Exclude cut with ID 1914-133440-0031-94921-0_sp0.9 from training. Duration: 22.7444375
2022-12-23 05:56:27,832 INFO [train.py:894] (2/4) Epoch 14, batch 600, loss[loss=0.1706, simple_loss=0.2604, pruned_loss=0.04038, over 18696.00 frames. ], tot_loss[loss=0.193, simple_loss=0.2778, pruned_loss=0.05406, over 3532316.99 frames. ], batch size: 48, lr: 8.36e-03, grad_scale: 8.0
2022-12-23 05:56:32,120 WARNING [train.py:1060] (2/4) Exclude cut with ID 4133-6541-0027-40495-0_sp1.1 from training. Duration: 0.9681875
2022-12-23 05:56:35,042 WARNING [train.py:1060] (2/4) Exclude cut with ID 6330-62851-0022-91297-0_sp0.9 from training. Duration: 22.3166875
2022-12-23 05:56:41,096 WARNING [train.py:1060] (2/4) Exclude cut with ID 543-133212-0015-59917-0_sp0.9 from training. Duration: 21.8166875
2022-12-23 05:56:41,535 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([0.5444, 1.6583, 2.0988, 0.8791, 1.3132, 2.3194, 1.8264, 1.6188],
       device='cuda:2'), covar=tensor([0.0781, 0.0525, 0.0319, 0.0423, 0.0408, 0.0359, 0.0335, 0.0715],
       device='cuda:2'), in_proj_covar=tensor([0.0142, 0.0157, 0.0114, 0.0131, 0.0138, 0.0131, 0.0148, 0.0154],
       device='cuda:2'), out_proj_covar=tensor([1.1266e-04, 1.2593e-04, 9.0136e-05, 1.0161e-04, 1.0929e-04, 1.0569e-04,
        1.1975e-04, 1.2256e-04], device='cuda:2')
2022-12-23 05:57:18,794 INFO [zipformer.py:660] (2/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=46215.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 05:57:41,977 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 2.383e+02 4.078e+02 4.914e+02 6.557e+02 2.060e+03, threshold=9.827e+02, percent-clipped=9.0
2022-12-23 05:57:43,536 INFO [train.py:894] (2/4) Epoch 14, batch 650, loss[loss=0.1724, simple_loss=0.2683, pruned_loss=0.03824, over 18539.00 frames. ], tot_loss[loss=0.1932, simple_loss=0.2779, pruned_loss=0.05425, over 3571838.82 frames. ], batch size: 55, lr: 8.36e-03, grad_scale: 8.0
2022-12-23 05:57:59,873 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.3704, 2.6108, 2.9713, 1.1070, 2.3615, 3.3838, 2.3535, 2.8023],
       device='cuda:2'), covar=tensor([0.0810, 0.0345, 0.0383, 0.0468, 0.0437, 0.0311, 0.0367, 0.0550],
       device='cuda:2'), in_proj_covar=tensor([0.0142, 0.0157, 0.0114, 0.0130, 0.0139, 0.0131, 0.0148, 0.0154],
       device='cuda:2'), out_proj_covar=tensor([1.1252e-04, 1.2565e-04, 8.9643e-05, 1.0135e-04, 1.0963e-04, 1.0578e-04,
        1.1957e-04, 1.2236e-04], device='cuda:2')
2022-12-23 05:58:22,620 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.8418, 1.7870, 1.5312, 1.5246, 1.8568, 2.0061, 2.0933, 1.4622],
       device='cuda:2'), covar=tensor([0.0266, 0.0232, 0.0397, 0.0206, 0.0175, 0.0292, 0.0197, 0.0254],
       device='cuda:2'), in_proj_covar=tensor([0.0087, 0.0122, 0.0145, 0.0123, 0.0112, 0.0111, 0.0093, 0.0124],
       device='cuda:2'), out_proj_covar=tensor([7.1258e-05, 9.9402e-05, 1.2409e-04, 1.0083e-04, 9.3579e-05, 8.7079e-05,
        7.4771e-05, 1.0010e-04], device='cuda:2')
2022-12-23 05:58:28,386 WARNING [train.py:1060] (2/4) Exclude cut with ID 4957-30119-0041-23990-0_sp0.9 from training. Duration: 20.22775
2022-12-23 05:58:52,425 INFO [zipformer.py:660] (2/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=46276.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 05:58:59,233 INFO [train.py:894] (2/4) Epoch 14, batch 700, loss[loss=0.1998, simple_loss=0.2918, pruned_loss=0.05395, over 18667.00 frames. ], tot_loss[loss=0.1924, simple_loss=0.2772, pruned_loss=0.05381, over 3603672.27 frames. ], batch size: 60, lr: 8.36e-03, grad_scale: 8.0
2022-12-23 05:59:13,941 WARNING [train.py:1060] (2/4) Exclude cut with ID 5239-32139-0047-9341-0_sp1.1 from training. Duration: 24.67275
2022-12-23 05:59:43,724 WARNING [train.py:1060] (2/4) Exclude cut with ID 3082-165428-0081-50734-0_sp0.9 from training. Duration: 21.8055625
2022-12-23 06:00:13,188 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 2.178e+02 4.025e+02 4.936e+02 5.531e+02 1.060e+03, threshold=9.873e+02, percent-clipped=1.0
2022-12-23 06:00:14,543 INFO [train.py:894] (2/4) Epoch 14, batch 750, loss[loss=0.2277, simple_loss=0.3131, pruned_loss=0.07115, over 18710.00 frames. ], tot_loss[loss=0.192, simple_loss=0.2769, pruned_loss=0.05351, over 3628416.31 frames. ], batch size: 97, lr: 8.35e-03, grad_scale: 8.0
2022-12-23 06:00:21,718 WARNING [train.py:1060] (2/4) Exclude cut with ID 3340-169293-0054-76830-0_sp0.9 from training. Duration: 22.6666875
2022-12-23 06:00:23,626 INFO [zipformer.py:660] (2/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=46337.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 06:00:54,951 INFO [zipformer.py:660] (2/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=46358.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 06:00:55,346 INFO [scaling.py:679] (2/4) Whitening: num_groups=8, num_channels=96, metric=1.28 vs. limit=2.0
2022-12-23 06:01:23,579 WARNING [train.py:1060] (2/4) Exclude cut with ID 2411-132532-0017-82279-0_sp1.1 from training. Duration: 0.9681875
2022-12-23 06:01:29,301 INFO [train.py:894] (2/4) Epoch 14, batch 800, loss[loss=0.2097, simple_loss=0.2958, pruned_loss=0.06181, over 18521.00 frames. ], tot_loss[loss=0.1911, simple_loss=0.2762, pruned_loss=0.053, over 3647465.13 frames. ], batch size: 52, lr: 8.35e-03, grad_scale: 8.0
2022-12-23 06:01:50,096 WARNING [train.py:1060] (2/4) Exclude cut with ID 6330-62850-0007-91323-0 from training. Duration: 22.485
2022-12-23 06:02:06,129 INFO [zipformer.py:660] (2/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=46406.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 06:02:27,795 WARNING [train.py:1060] (2/4) Exclude cut with ID 3972-170212-0014-23379-0_sp1.1 from training. Duration: 23.82275
2022-12-23 06:02:40,426 WARNING [train.py:1060] (2/4) Exclude cut with ID 4860-13185-0032-76709-0 from training. Duration: 20.77
2022-12-23 06:02:42,763 INFO [scaling.py:679] (2/4) Whitening: num_groups=8, num_channels=192, metric=1.88 vs. limit=2.0
2022-12-23 06:02:43,273 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 2.502e+02 3.825e+02 4.476e+02 5.573e+02 1.333e+03, threshold=8.951e+02, percent-clipped=4.0
2022-12-23 06:02:43,516 INFO [zipformer.py:660] (2/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=46430.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 06:02:44,558 INFO [train.py:894] (2/4) Epoch 14, batch 850, loss[loss=0.1955, simple_loss=0.2754, pruned_loss=0.05781, over 18571.00 frames. ], tot_loss[loss=0.1916, simple_loss=0.2768, pruned_loss=0.05325, over 3662734.61 frames. ], batch size: 49, lr: 8.34e-03, grad_scale: 8.0
2022-12-23 06:02:47,781 WARNING [train.py:1060] (2/4) Exclude cut with ID 6426-64292-0017-15984-0_sp0.9 from training. Duration: 24.088875
2022-12-23 06:03:15,533 WARNING [train.py:1060] (2/4) Exclude cut with ID 6330-62850-0007-91323-0_sp1.1 from training. Duration: 20.4409375
2022-12-23 06:03:55,974 INFO [zipformer.py:660] (2/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=46478.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 06:04:00,227 INFO [train.py:894] (2/4) Epoch 14, batch 900, loss[loss=0.1889, simple_loss=0.2801, pruned_loss=0.04888, over 18541.00 frames. ], tot_loss[loss=0.1915, simple_loss=0.2769, pruned_loss=0.05301, over 3673298.91 frames. ], batch size: 97, lr: 8.34e-03, grad_scale: 8.0
2022-12-23 06:04:31,603 WARNING [train.py:1060] (2/4) Exclude cut with ID 2929-85685-0062-62366-0_sp0.9 from training. Duration: 22.511125
2022-12-23 06:04:31,626 WARNING [train.py:1060] (2/4) Exclude cut with ID 7395-89880-0031-39906-0 from training. Duration: 20.675
2022-12-23 06:04:45,669 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.9407, 1.9959, 2.2732, 1.1338, 2.3209, 2.3297, 1.6274, 2.6475],
       device='cuda:2'), covar=tensor([0.1134, 0.1577, 0.1133, 0.1842, 0.0637, 0.1018, 0.2066, 0.0506],
       device='cuda:2'), in_proj_covar=tensor([0.0200, 0.0201, 0.0206, 0.0194, 0.0176, 0.0214, 0.0208, 0.0184],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-23 06:04:52,115 INFO [zipformer.py:660] (2/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=46515.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 06:05:17,322 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 2.454e+02 3.881e+02 4.626e+02 5.706e+02 1.079e+03, threshold=9.251e+02, percent-clipped=7.0
2022-12-23 06:05:17,337 INFO [train.py:894] (2/4) Epoch 14, batch 950, loss[loss=0.1735, simple_loss=0.2538, pruned_loss=0.04663, over 18576.00 frames. ], tot_loss[loss=0.1906, simple_loss=0.276, pruned_loss=0.05263, over 3682336.52 frames. ], batch size: 49, lr: 8.33e-03, grad_scale: 8.0
2022-12-23 06:05:50,851 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([3.1804, 2.4947, 1.7320, 2.9978, 2.2308, 2.3235, 2.4510, 3.3886],
       device='cuda:2'), covar=tensor([0.1742, 0.3090, 0.1776, 0.2847, 0.3358, 0.0987, 0.2749, 0.0635],
       device='cuda:2'), in_proj_covar=tensor([0.0282, 0.0278, 0.0233, 0.0342, 0.0257, 0.0218, 0.0272, 0.0198],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0003, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-23 06:06:07,184 INFO [zipformer.py:660] (2/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=46563.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 06:06:11,695 WARNING [train.py:1060] (2/4) Exclude cut with ID 6330-62850-0007-91323-0_sp0.9 from training. Duration: 24.9833125
2022-12-23 06:06:33,808 INFO [train.py:894] (2/4) Epoch 14, batch 1000, loss[loss=0.1999, simple_loss=0.2747, pruned_loss=0.06261, over 18545.00 frames. ], tot_loss[loss=0.1907, simple_loss=0.2759, pruned_loss=0.05281, over 3688195.53 frames. ], batch size: 44, lr: 8.33e-03, grad_scale: 8.0
2022-12-23 06:06:42,923 WARNING [train.py:1060] (2/4) Exclude cut with ID 5239-32139-0047-9341-0 from training. Duration: 27.14
2022-12-23 06:06:55,873 INFO [zipformer.py:660] (2/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=46595.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 06:06:58,687 WARNING [train.py:1060] (2/4) Exclude cut with ID 2929-85685-0044-62348-0 from training. Duration: 22.44
2022-12-23 06:07:06,757 INFO [zipformer.py:660] (2/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=46602.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 06:07:50,415 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 2.396e+02 3.981e+02 4.528e+02 6.046e+02 1.105e+03, threshold=9.056e+02, percent-clipped=1.0
2022-12-23 06:07:50,430 INFO [train.py:894] (2/4) Epoch 14, batch 1050, loss[loss=0.2223, simple_loss=0.3094, pruned_loss=0.06757, over 18698.00 frames. ], tot_loss[loss=0.191, simple_loss=0.2759, pruned_loss=0.05301, over 3694700.45 frames. ], batch size: 78, lr: 8.32e-03, grad_scale: 8.0
2022-12-23 06:07:51,956 INFO [zipformer.py:660] (2/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=46632.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 06:08:12,633 WARNING [train.py:1060] (2/4) Exclude cut with ID 2929-85685-0060-62364-0_sp0.9 from training. Duration: 21.361125
2022-12-23 06:08:18,443 WARNING [train.py:1060] (2/4) Exclude cut with ID 2929-85685-0079-62383-0_sp1.1 from training. Duration: 27.0318125
2022-12-23 06:08:28,871 INFO [zipformer.py:660] (2/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=46656.0, num_to_drop=1, layers_to_drop={2}
2022-12-23 06:08:30,015 WARNING [train.py:1060] (2/4) Exclude cut with ID 5622-44585-0006-90525-0_sp0.9 from training. Duration: 28.638875
2022-12-23 06:08:39,725 INFO [zipformer.py:660] (2/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=46663.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 06:08:46,711 WARNING [train.py:1060] (2/4) Exclude cut with ID 3340-169293-0054-76830-0 from training. Duration: 20.4
2022-12-23 06:09:06,401 INFO [train.py:894] (2/4) Epoch 14, batch 1100, loss[loss=0.1704, simple_loss=0.2505, pruned_loss=0.04519, over 18383.00 frames. ], tot_loss[loss=0.1913, simple_loss=0.2764, pruned_loss=0.05303, over 3699294.80 frames. ], batch size: 46, lr: 8.32e-03, grad_scale: 8.0
2022-12-23 06:09:19,466 WARNING [train.py:1060] (2/4) Exclude cut with ID 2929-85685-0071-62375-0 from training. Duration: 20.025
2022-12-23 06:09:19,479 WARNING [train.py:1060] (2/4) Exclude cut with ID 2364-131735-0112-64612-0_sp0.9 from training. Duration: 20.488875
2022-12-23 06:09:23,683 WARNING [train.py:1060] (2/4) Exclude cut with ID 2929-85685-0079-62383-0 from training. Duration: 29.735
2022-12-23 06:09:33,754 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.6506, 1.5801, 1.3983, 1.4276, 1.8352, 1.7585, 1.8482, 1.2977],
       device='cuda:2'), covar=tensor([0.0268, 0.0195, 0.0413, 0.0181, 0.0148, 0.0278, 0.0213, 0.0245],
       device='cuda:2'), in_proj_covar=tensor([0.0086, 0.0119, 0.0144, 0.0122, 0.0110, 0.0109, 0.0093, 0.0121],
       device='cuda:2'), out_proj_covar=tensor([7.0666e-05, 9.6729e-05, 1.2239e-04, 9.9542e-05, 9.1870e-05, 8.6120e-05,
        7.4255e-05, 9.7401e-05], device='cuda:2')
2022-12-23 06:09:42,639 INFO [scaling.py:679] (2/4) Whitening: num_groups=1, num_channels=384, metric=3.63 vs. limit=5.0
2022-12-23 06:10:10,787 INFO [scaling.py:679] (2/4) Whitening: num_groups=8, num_channels=192, metric=1.91 vs. limit=2.0
2022-12-23 06:10:21,528 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 2.330e+02 4.103e+02 4.927e+02 5.923e+02 1.016e+03, threshold=9.854e+02, percent-clipped=2.0
2022-12-23 06:10:21,543 INFO [train.py:894] (2/4) Epoch 14, batch 1150, loss[loss=0.1936, simple_loss=0.2693, pruned_loss=0.059, over 18414.00 frames. ], tot_loss[loss=0.1906, simple_loss=0.2759, pruned_loss=0.05264, over 3702316.81 frames. ], batch size: 42, lr: 8.32e-03, grad_scale: 8.0
2022-12-23 06:10:46,278 WARNING [train.py:1060] (2/4) Exclude cut with ID 7276-92427-0014-12983-0_sp0.9 from training. Duration: 21.3055625
2022-12-23 06:10:48,322 WARNING [train.py:1060] (2/4) Exclude cut with ID 1025-75365-0008-79168-0_sp0.9 from training. Duration: 22.0666875
2022-12-23 06:10:54,576 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([2.7570, 2.2497, 1.6510, 2.5823, 1.9979, 2.0735, 2.1706, 2.7948],
       device='cuda:2'), covar=tensor([0.1779, 0.2908, 0.1785, 0.2541, 0.3455, 0.0976, 0.2743, 0.0729],
       device='cuda:2'), in_proj_covar=tensor([0.0284, 0.0278, 0.0233, 0.0342, 0.0258, 0.0218, 0.0272, 0.0200],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0003, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-23 06:11:36,878 INFO [train.py:894] (2/4) Epoch 14, batch 1200, loss[loss=0.1923, simple_loss=0.2803, pruned_loss=0.05218, over 18727.00 frames. ], tot_loss[loss=0.189, simple_loss=0.2746, pruned_loss=0.05175, over 3704597.01 frames. ], batch size: 54, lr: 8.31e-03, grad_scale: 8.0
2022-12-23 06:12:05,724 INFO [scaling.py:679] (2/4) Whitening: num_groups=1, num_channels=384, metric=4.01 vs. limit=5.0
2022-12-23 06:12:17,489 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([2.1470, 1.9998, 2.2870, 1.2000, 2.5365, 2.4768, 1.5243, 2.6088],
       device='cuda:2'), covar=tensor([0.1105, 0.1710, 0.1279, 0.2035, 0.0644, 0.1120, 0.2101, 0.0583],
       device='cuda:2'), in_proj_covar=tensor([0.0197, 0.0199, 0.0201, 0.0191, 0.0173, 0.0211, 0.0204, 0.0182],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-23 06:12:21,738 INFO [zipformer.py:660] (2/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=46810.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 06:12:40,010 WARNING [train.py:1060] (2/4) Exclude cut with ID 2929-85685-0062-62366-0 from training. Duration: 20.26
2022-12-23 06:12:52,910 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 2.710e+02 4.042e+02 4.765e+02 5.744e+02 1.398e+03, threshold=9.530e+02, percent-clipped=5.0
2022-12-23 06:12:52,926 INFO [train.py:894] (2/4) Epoch 14, batch 1250, loss[loss=0.1867, simple_loss=0.2754, pruned_loss=0.04898, over 18378.00 frames. ], tot_loss[loss=0.1887, simple_loss=0.2743, pruned_loss=0.05158, over 3705426.74 frames. ], batch size: 51, lr: 8.31e-03, grad_scale: 8.0
2022-12-23 06:12:52,990 WARNING [train.py:1060] (2/4) Exclude cut with ID 5239-32139-0030-9324-0_sp0.9 from training. Duration: 21.3444375
2022-12-23 06:13:53,595 WARNING [train.py:1060] (2/4) Exclude cut with ID 497-129325-0061-62254-0_sp1.1 from training. Duration: 0.97725
2022-12-23 06:13:54,006 INFO [zipformer.py:660] (2/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=46871.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 06:14:08,372 INFO [train.py:894] (2/4) Epoch 14, batch 1300, loss[loss=0.1486, simple_loss=0.2329, pruned_loss=0.03214, over 18485.00 frames. ], tot_loss[loss=0.188, simple_loss=0.274, pruned_loss=0.05099, over 3706500.15 frames. ], batch size: 43, lr: 8.30e-03, grad_scale: 8.0
2022-12-23 06:14:16,136 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.5392, 1.4641, 1.6560, 1.5503, 1.0726, 3.5841, 1.5186, 2.0773],
       device='cuda:2'), covar=tensor([0.3237, 0.2066, 0.1864, 0.1972, 0.1451, 0.0158, 0.1575, 0.0815],
       device='cuda:2'), in_proj_covar=tensor([0.0136, 0.0117, 0.0127, 0.0121, 0.0104, 0.0098, 0.0094, 0.0091],
       device='cuda:2'), out_proj_covar=tensor([0.0005, 0.0005, 0.0005, 0.0005, 0.0004, 0.0003, 0.0004, 0.0004],
       device='cuda:2')
2022-12-23 06:14:34,071 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([2.3471, 1.6936, 1.8909, 1.9829, 2.2184, 2.1810, 2.1785, 1.7744],
       device='cuda:2'), covar=tensor([0.1926, 0.2781, 0.2346, 0.2537, 0.1677, 0.0817, 0.2702, 0.1072],
       device='cuda:2'), in_proj_covar=tensor([0.0256, 0.0292, 0.0268, 0.0304, 0.0290, 0.0241, 0.0319, 0.0227],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-23 06:14:35,021 WARNING [train.py:1060] (2/4) Exclude cut with ID 7395-89880-0031-39906-0_sp0.9 from training. Duration: 22.97225
2022-12-23 06:15:04,731 WARNING [train.py:1060] (2/4) Exclude cut with ID 7395-89880-0047-39922-0_sp0.9 from training. Duration: 21.97775
2022-12-23 06:15:18,474 WARNING [train.py:1060] (2/4) Exclude cut with ID 1112-1043-0006-89194-0_sp0.9 from training. Duration: 21.8333125
2022-12-23 06:15:24,187 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 2.454e+02 4.083e+02 4.755e+02 5.873e+02 8.808e+02, threshold=9.511e+02, percent-clipped=0.0
2022-12-23 06:15:24,202 INFO [train.py:894] (2/4) Epoch 14, batch 1350, loss[loss=0.1791, simple_loss=0.2687, pruned_loss=0.04476, over 18504.00 frames. ], tot_loss[loss=0.1887, simple_loss=0.2747, pruned_loss=0.05131, over 3707694.35 frames. ], batch size: 52, lr: 8.30e-03, grad_scale: 8.0
2022-12-23 06:15:25,820 INFO [zipformer.py:660] (2/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=46932.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 06:15:28,541 WARNING [train.py:1060] (2/4) Exclude cut with ID 1914-133440-0031-94921-0 from training. Duration: 20.47
2022-12-23 06:15:43,773 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([2.5171, 2.1431, 1.5673, 2.3920, 1.8473, 2.0438, 1.9318, 2.5655],
       device='cuda:2'), covar=tensor([0.1795, 0.2781, 0.1744, 0.2281, 0.3037, 0.0934, 0.2680, 0.0746],
       device='cuda:2'), in_proj_covar=tensor([0.0285, 0.0279, 0.0234, 0.0342, 0.0259, 0.0218, 0.0272, 0.0200],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0003, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-23 06:15:46,328 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.8706, 1.7521, 1.5033, 1.2570, 2.0888, 1.9449, 1.7701, 1.4622],
       device='cuda:2'), covar=tensor([0.0325, 0.0342, 0.0430, 0.0566, 0.0232, 0.0289, 0.0375, 0.0640],
       device='cuda:2'), in_proj_covar=tensor([0.0119, 0.0119, 0.0125, 0.0116, 0.0088, 0.0117, 0.0132, 0.0150],
       device='cuda:2'), out_proj_covar=tensor([0.0001, 0.0001, 0.0002, 0.0001, 0.0001, 0.0001, 0.0002, 0.0002],
       device='cuda:2')
2022-12-23 06:15:54,701 INFO [zipformer.py:660] (2/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=46951.0, num_to_drop=1, layers_to_drop={2}
2022-12-23 06:16:05,717 INFO [zipformer.py:660] (2/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=46958.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 06:16:06,283 INFO [scaling.py:679] (2/4) Whitening: num_groups=8, num_channels=96, metric=1.17 vs. limit=2.0
2022-12-23 06:16:33,627 WARNING [train.py:1060] (2/4) Exclude cut with ID 7395-89880-0037-39912-0_sp0.9 from training. Duration: 20.67225
2022-12-23 06:16:38,022 INFO [zipformer.py:660] (2/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=46980.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 06:16:39,249 INFO [train.py:894] (2/4) Epoch 14, batch 1400, loss[loss=0.1809, simple_loss=0.2641, pruned_loss=0.0488, over 18532.00 frames. ], tot_loss[loss=0.1899, simple_loss=0.2758, pruned_loss=0.05202, over 3708112.82 frames. ], batch size: 47, lr: 8.29e-03, grad_scale: 8.0
2022-12-23 06:16:50,075 WARNING [train.py:1060] (2/4) Exclude cut with ID 1914-133440-0024-94914-0_sp0.9 from training. Duration: 25.2444375
2022-12-23 06:17:14,843 WARNING [train.py:1060] (2/4) Exclude cut with ID 3340-169293-0021-76797-0_sp0.9 from training. Duration: 21.1445
2022-12-23 06:17:44,220 INFO [scaling.py:679] (2/4) Whitening: num_groups=8, num_channels=192, metric=1.72 vs. limit=2.0
2022-12-23 06:17:56,990 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 2.138e+02 3.877e+02 4.597e+02 5.782e+02 1.688e+03, threshold=9.194e+02, percent-clipped=3.0
2022-12-23 06:17:57,006 INFO [train.py:894] (2/4) Epoch 14, batch 1450, loss[loss=0.1883, simple_loss=0.2812, pruned_loss=0.04772, over 18711.00 frames. ], tot_loss[loss=0.1892, simple_loss=0.2753, pruned_loss=0.05151, over 3709372.24 frames. ], batch size: 60, lr: 8.29e-03, grad_scale: 8.0
2022-12-23 06:18:29,686 WARNING [train.py:1060] (2/4) Exclude cut with ID 2929-85685-0079-62383-0_sp0.9 from training. Duration: 33.038875
2022-12-23 06:18:42,770 INFO [scaling.py:679] (2/4) Whitening: num_groups=8, num_channels=96, metric=1.37 vs. limit=2.0
2022-12-23 06:19:06,805 WARNING [train.py:1060] (2/4) Exclude cut with ID 6426-64291-0000-16059-0_sp0.9 from training. Duration: 20.0944375
2022-12-23 06:19:12,445 INFO [train.py:894] (2/4) Epoch 14, batch 1500, loss[loss=0.1714, simple_loss=0.2511, pruned_loss=0.04582, over 18489.00 frames. ], tot_loss[loss=0.1883, simple_loss=0.2745, pruned_loss=0.05103, over 3710040.76 frames. ], batch size: 43, lr: 8.28e-03, grad_scale: 8.0
2022-12-23 06:19:19,904 WARNING [train.py:1060] (2/4) Exclude cut with ID 2929-85685-0044-62348-0_sp1.1 from training. Duration: 20.4
2022-12-23 06:19:28,844 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.8991, 1.6487, 1.6583, 2.4439, 2.2291, 4.4871, 1.5716, 1.9702],
       device='cuda:2'), covar=tensor([0.0785, 0.1683, 0.1035, 0.0864, 0.1332, 0.0180, 0.1336, 0.1449],
       device='cuda:2'), in_proj_covar=tensor([0.0073, 0.0082, 0.0073, 0.0075, 0.0091, 0.0074, 0.0084, 0.0078],
       device='cuda:2'), out_proj_covar=tensor([0.0003, 0.0004, 0.0004, 0.0003, 0.0004, 0.0003, 0.0004, 0.0004],
       device='cuda:2')
2022-12-23 06:19:29,870 WARNING [train.py:1060] (2/4) Exclude cut with ID 6330-62851-0022-91297-0 from training. Duration: 20.085
2022-12-23 06:19:40,554 WARNING [train.py:1060] (2/4) Exclude cut with ID 4860-13185-0032-76709-0_sp0.9 from training. Duration: 23.07775
2022-12-23 06:20:25,452 WARNING [train.py:1060] (2/4) Exclude cut with ID 2929-85685-0044-62348-0_sp0.9 from training. Duration: 24.9333125
2022-12-23 06:20:28,369 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 2.178e+02 3.794e+02 4.635e+02 5.636e+02 1.005e+03, threshold=9.269e+02, percent-clipped=3.0
2022-12-23 06:20:28,384 INFO [train.py:894] (2/4) Epoch 14, batch 1550, loss[loss=0.2014, simple_loss=0.2937, pruned_loss=0.05459, over 18544.00 frames. ], tot_loss[loss=0.1876, simple_loss=0.2738, pruned_loss=0.05066, over 3710744.52 frames. ], batch size: 55, lr: 8.28e-03, grad_scale: 8.0
2022-12-23 06:21:10,323 WARNING [train.py:1060] (2/4) Exclude cut with ID 5118-111612-0016-124680-0_sp0.9 from training. Duration: 20.388875
2022-12-23 06:21:17,568 WARNING [train.py:1060] (2/4) Exclude cut with ID 432-122774-0017-62487-0_sp1.1 from training. Duration: 20.3590625
2022-12-23 06:21:21,954 INFO [zipformer.py:660] (2/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=47166.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 06:21:44,086 INFO [train.py:894] (2/4) Epoch 14, batch 1600, loss[loss=0.1805, simple_loss=0.2653, pruned_loss=0.04786, over 18415.00 frames. ], tot_loss[loss=0.187, simple_loss=0.2732, pruned_loss=0.05039, over 3710631.77 frames. ], batch size: 48, lr: 8.28e-03, grad_scale: 8.0
2022-12-23 06:22:25,296 WARNING [train.py:1060] (2/4) Exclude cut with ID 3557-8342-0013-54691-0_sp1.1 from training. Duration: 0.836375
2022-12-23 06:22:58,396 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 2.803e+02 4.204e+02 4.894e+02 6.203e+02 1.673e+03, threshold=9.789e+02, percent-clipped=2.0
2022-12-23 06:22:58,411 INFO [train.py:894] (2/4) Epoch 14, batch 1650, loss[loss=0.2155, simple_loss=0.3004, pruned_loss=0.06534, over 18711.00 frames. ], tot_loss[loss=0.1889, simple_loss=0.2744, pruned_loss=0.05167, over 3711049.08 frames. ], batch size: 52, lr: 8.27e-03, grad_scale: 8.0
2022-12-23 06:22:59,457 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([2.1674, 2.0965, 1.6056, 1.0428, 2.7113, 2.2925, 2.0231, 1.5067],
       device='cuda:2'), covar=tensor([0.0360, 0.0369, 0.0490, 0.0734, 0.0209, 0.0339, 0.0406, 0.0872],
       device='cuda:2'), in_proj_covar=tensor([0.0121, 0.0120, 0.0127, 0.0117, 0.0090, 0.0119, 0.0134, 0.0153],
       device='cuda:2'), out_proj_covar=tensor([0.0001, 0.0001, 0.0002, 0.0001, 0.0001, 0.0001, 0.0002, 0.0002],
       device='cuda:2')
2022-12-23 06:23:09,759 WARNING [train.py:1060] (2/4) Exclude cut with ID 8565-290391-0049-67394-0_sp0.9 from training. Duration: 21.3166875
2022-12-23 06:23:14,525 INFO [zipformer.py:660] (2/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=47241.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 06:23:29,866 INFO [zipformer.py:660] (2/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=47251.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 06:23:39,895 INFO [zipformer.py:660] (2/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=47258.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 06:23:42,522 WARNING [train.py:1060] (2/4) Exclude cut with ID 6533-399-0029-104863-0_sp0.9 from training. Duration: 22.1055625
2022-12-23 06:23:53,893 WARNING [train.py:1060] (2/4) Exclude cut with ID 7699-105389-0094-26379-0_sp1.1 from training. Duration: 21.77725
2022-12-23 06:24:12,227 WARNING [train.py:1060] (2/4) Exclude cut with ID 7255-291500-0005-134304-0_sp0.9 from training. Duration: 27.8166875
2022-12-23 06:24:13,402 INFO [train.py:894] (2/4) Epoch 14, batch 1700, loss[loss=0.2232, simple_loss=0.3028, pruned_loss=0.07179, over 18646.00 frames. ], tot_loss[loss=0.1921, simple_loss=0.2762, pruned_loss=0.054, over 3711382.07 frames. ], batch size: 62, lr: 8.27e-03, grad_scale: 8.0
2022-12-23 06:24:38,740 WARNING [train.py:1060] (2/4) Exclude cut with ID 7357-94126-0021-15852-0_sp1.1 from training. Duration: 22.5090625
2022-12-23 06:24:41,900 INFO [zipformer.py:660] (2/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=47299.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 06:24:44,452 WARNING [train.py:1060] (2/4) Exclude cut with ID 7255-291500-0005-134304-0 from training. Duration: 25.035
2022-12-23 06:24:46,194 INFO [zipformer.py:660] (2/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=47302.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 06:24:51,935 INFO [zipformer.py:660] (2/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=47306.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 06:25:03,573 WARNING [train.py:1060] (2/4) Exclude cut with ID 774-127930-0014-10412-0_sp1.1 from training. Duration: 0.95
2022-12-23 06:25:14,021 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.6397, 2.3595, 1.6678, 2.7413, 2.8717, 1.7197, 2.0509, 1.3139],
       device='cuda:2'), covar=tensor([0.1765, 0.1381, 0.1338, 0.0786, 0.1236, 0.1033, 0.1620, 0.1374],
       device='cuda:2'), in_proj_covar=tensor([0.0239, 0.0216, 0.0206, 0.0189, 0.0254, 0.0191, 0.0214, 0.0191],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-23 06:25:21,380 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.4638, 1.3901, 1.2903, 0.6934, 1.7121, 1.4383, 1.3402, 1.1120],
       device='cuda:2'), covar=tensor([0.0356, 0.0404, 0.0431, 0.0664, 0.0304, 0.0341, 0.0415, 0.0867],
       device='cuda:2'), in_proj_covar=tensor([0.0120, 0.0119, 0.0126, 0.0116, 0.0089, 0.0117, 0.0133, 0.0152],
       device='cuda:2'), out_proj_covar=tensor([0.0001, 0.0001, 0.0002, 0.0001, 0.0001, 0.0001, 0.0002, 0.0002],
       device='cuda:2')
2022-12-23 06:25:22,538 WARNING [train.py:1060] (2/4) Exclude cut with ID 3033-130750-0096-55598-0_sp0.9 from training. Duration: 0.92225
2022-12-23 06:25:28,998 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 2.980e+02 4.300e+02 5.386e+02 7.051e+02 1.470e+03, threshold=1.077e+03, percent-clipped=6.0
2022-12-23 06:25:29,014 INFO [train.py:894] (2/4) Epoch 14, batch 1750, loss[loss=0.1904, simple_loss=0.2756, pruned_loss=0.05258, over 18510.00 frames. ], tot_loss[loss=0.1948, simple_loss=0.2776, pruned_loss=0.05602, over 3711542.24 frames. ], batch size: 58, lr: 8.26e-03, grad_scale: 8.0
2022-12-23 06:25:48,234 WARNING [train.py:1060] (2/4) Exclude cut with ID 4511-76322-0006-80011-0 from training. Duration: 21.97
2022-12-23 06:25:56,315 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.7745, 1.1840, 0.8417, 1.3512, 2.0890, 1.1574, 1.4053, 1.6745],
       device='cuda:2'), covar=tensor([0.1589, 0.2117, 0.2391, 0.1568, 0.1699, 0.1727, 0.1461, 0.1608],
       device='cuda:2'), in_proj_covar=tensor([0.0092, 0.0098, 0.0118, 0.0096, 0.0113, 0.0091, 0.0098, 0.0093],
       device='cuda:2'), out_proj_covar=tensor([0.0003, 0.0003, 0.0004, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003],
       device='cuda:2')
2022-12-23 06:26:05,037 WARNING [train.py:1060] (2/4) Exclude cut with ID 7492-105653-0055-62765-0_sp0.9 from training. Duration: 21.97225
2022-12-23 06:26:06,473 WARNING [train.py:1060] (2/4) Exclude cut with ID 453-131332-0000-47844-0_sp0.9 from training. Duration: 25.3333125
2022-12-23 06:26:13,038 INFO [scaling.py:679] (2/4) Whitening: num_groups=8, num_channels=96, metric=1.37 vs. limit=2.0
2022-12-23 06:26:16,809 WARNING [train.py:1060] (2/4) Exclude cut with ID 5172-29468-0015-19128-0_sp0.9 from training. Duration: 21.5055625
2022-12-23 06:26:24,055 WARNING [train.py:1060] (2/4) Exclude cut with ID 453-131332-0000-47844-0_sp1.1 from training. Duration: 20.72725
2022-12-23 06:26:43,334 INFO [zipformer.py:660] (2/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=47379.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 06:26:46,087 INFO [train.py:894] (2/4) Epoch 14, batch 1800, loss[loss=0.2031, simple_loss=0.2745, pruned_loss=0.06585, over 18632.00 frames. ], tot_loss[loss=0.1973, simple_loss=0.2788, pruned_loss=0.05792, over 3711584.05 frames. ], batch size: 41, lr: 8.26e-03, grad_scale: 8.0
2022-12-23 06:26:58,916 WARNING [train.py:1060] (2/4) Exclude cut with ID 8631-249866-0030-130156-0_sp0.9 from training. Duration: 26.32775
2022-12-23 06:27:28,714 WARNING [train.py:1060] (2/4) Exclude cut with ID 3867-173237-0077-144769-0 from training. Duration: 20.025
2022-12-23 06:27:33,034 WARNING [train.py:1060] (2/4) Exclude cut with ID 6709-74022-0004-86860-0_sp1.1 from training. Duration: 0.9409375
2022-12-23 06:27:33,041 WARNING [train.py:1060] (2/4) Exclude cut with ID 4757-1811-0023-62229-0_sp0.9 from training. Duration: 21.37775
2022-12-23 06:27:55,179 WARNING [train.py:1060] (2/4) Exclude cut with ID 1250-135782-0004-25974-0_sp0.9 from training. Duration: 21.17225
2022-12-23 06:27:55,186 WARNING [train.py:1060] (2/4) Exclude cut with ID 7357-94126-0021-15852-0_sp0.9 from training. Duration: 27.511125
2022-12-23 06:28:04,823 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 3.184e+02 5.129e+02 6.200e+02 7.319e+02 2.210e+03, threshold=1.240e+03, percent-clipped=6.0
2022-12-23 06:28:04,838 INFO [train.py:894] (2/4) Epoch 14, batch 1850, loss[loss=0.2498, simple_loss=0.3184, pruned_loss=0.09055, over 18651.00 frames. ], tot_loss[loss=0.1997, simple_loss=0.28, pruned_loss=0.05971, over 3712681.88 frames. ], batch size: 176, lr: 8.25e-03, grad_scale: 8.0
2022-12-23 06:28:19,009 INFO [zipformer.py:660] (2/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=47440.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 06:28:28,766 WARNING [train.py:1060] (2/4) Exclude cut with ID 453-131332-0000-47844-0 from training. Duration: 22.8
2022-12-23 06:28:31,627 WARNING [train.py:1060] (2/4) Exclude cut with ID 4964-30587-0040-44509-0 from training. Duration: 22.585
2022-12-23 06:28:56,950 INFO [zipformer.py:660] (2/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=47466.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 06:29:03,590 WARNING [train.py:1060] (2/4) Exclude cut with ID 6978-92210-0001-146967-0_sp0.9 from training. Duration: 22.0166875
2022-12-23 06:29:06,085 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.5065, 1.3447, 1.3211, 1.8938, 1.6494, 3.1842, 1.3447, 1.5759],
       device='cuda:2'), covar=tensor([0.0923, 0.1924, 0.1095, 0.0895, 0.1447, 0.0282, 0.1521, 0.1582],
       device='cuda:2'), in_proj_covar=tensor([0.0073, 0.0082, 0.0073, 0.0075, 0.0090, 0.0074, 0.0083, 0.0077],
       device='cuda:2'), out_proj_covar=tensor([0.0003, 0.0004, 0.0004, 0.0003, 0.0004, 0.0003, 0.0004, 0.0004],
       device='cuda:2')
2022-12-23 06:29:19,648 INFO [train.py:894] (2/4) Epoch 14, batch 1900, loss[loss=0.2301, simple_loss=0.2999, pruned_loss=0.08021, over 18586.00 frames. ], tot_loss[loss=0.201, simple_loss=0.2807, pruned_loss=0.06065, over 3712913.94 frames. ], batch size: 49, lr: 8.25e-03, grad_scale: 8.0
2022-12-23 06:29:19,682 WARNING [train.py:1060] (2/4) Exclude cut with ID 7255-291500-0003-134302-0_sp1.1 from training. Duration: 24.395375
2022-12-23 06:29:27,131 WARNING [train.py:1060] (2/4) Exclude cut with ID 3488-85273-0017-41203-0_sp0.9 from training. Duration: 27.47775
2022-12-23 06:29:30,090 WARNING [train.py:1060] (2/4) Exclude cut with ID 432-122774-0017-62487-0_sp0.9 from training. Duration: 24.8833125
2022-12-23 06:29:32,957 WARNING [train.py:1060] (2/4) Exclude cut with ID 6758-72288-0033-108368-0 from training. Duration: 23.39
2022-12-23 06:29:33,808 INFO [scaling.py:679] (2/4) Whitening: num_groups=1, num_channels=384, metric=3.83 vs. limit=5.0
2022-12-23 06:29:38,559 WARNING [train.py:1060] (2/4) Exclude cut with ID 8291-282929-0007-12994-0_sp0.9 from training. Duration: 28.72225
2022-12-23 06:29:48,362 WARNING [train.py:1060] (2/4) Exclude cut with ID 585-294811-0110-133686-0_sp0.9 from training. Duration: 20.8944375
2022-12-23 06:30:02,740 WARNING [train.py:1060] (2/4) Exclude cut with ID 5796-66357-0007-116447-0_sp0.9 from training. Duration: 23.8444375
2022-12-23 06:30:07,855 INFO [zipformer.py:660] (2/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=47514.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 06:30:17,376 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([2.5936, 1.5808, 0.8403, 1.7828, 2.6670, 1.7579, 2.3994, 2.4652],
       device='cuda:2'), covar=tensor([0.1240, 0.1819, 0.2335, 0.1375, 0.1411, 0.1522, 0.1230, 0.1316],
       device='cuda:2'), in_proj_covar=tensor([0.0092, 0.0098, 0.0117, 0.0096, 0.0113, 0.0091, 0.0097, 0.0093],
       device='cuda:2'), out_proj_covar=tensor([0.0003, 0.0003, 0.0004, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003],
       device='cuda:2')
2022-12-23 06:30:28,538 WARNING [train.py:1060] (2/4) Exclude cut with ID 8291-282929-0007-12994-0 from training. Duration: 25.85
2022-12-23 06:30:28,548 WARNING [train.py:1060] (2/4) Exclude cut with ID 8291-282929-0023-13010-0 from training. Duration: 21.39
2022-12-23 06:30:34,534 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 2.979e+02 5.048e+02 5.917e+02 7.311e+02 1.442e+03, threshold=1.183e+03, percent-clipped=2.0
2022-12-23 06:30:34,549 INFO [train.py:894] (2/4) Epoch 14, batch 1950, loss[loss=0.1931, simple_loss=0.2576, pruned_loss=0.06431, over 18636.00 frames. ], tot_loss[loss=0.2034, simple_loss=0.282, pruned_loss=0.06236, over 3714109.87 frames. ], batch size: 41, lr: 8.25e-03, grad_scale: 8.0
2022-12-23 06:30:39,024 WARNING [train.py:1060] (2/4) Exclude cut with ID 7357-94126-0014-15845-0 from training. Duration: 27.92
2022-12-23 06:30:39,304 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.2346, 1.0628, 0.8895, 1.3637, 1.5400, 2.4018, 1.0747, 1.3756],
       device='cuda:2'), covar=tensor([0.0919, 0.1925, 0.1089, 0.0867, 0.1372, 0.0354, 0.1591, 0.1564],
       device='cuda:2'), in_proj_covar=tensor([0.0073, 0.0081, 0.0073, 0.0074, 0.0090, 0.0074, 0.0083, 0.0077],
       device='cuda:2'), out_proj_covar=tensor([0.0003, 0.0004, 0.0004, 0.0003, 0.0004, 0.0003, 0.0004, 0.0004],
       device='cuda:2')
2022-12-23 06:31:05,752 WARNING [train.py:1060] (2/4) Exclude cut with ID 8631-249866-0039-130165-0_sp0.9 from training. Duration: 20.661125
2022-12-23 06:31:30,310 WARNING [train.py:1060] (2/4) Exclude cut with ID 7357-94126-0043-15874-0_sp0.9 from training. Duration: 20.07225
2022-12-23 06:31:39,728 WARNING [train.py:1060] (2/4) Exclude cut with ID 1085-156170-0017-128270-0 from training. Duration: 21.01
2022-12-23 06:31:48,671 INFO [train.py:894] (2/4) Epoch 14, batch 2000, loss[loss=0.1689, simple_loss=0.2425, pruned_loss=0.04764, over 18608.00 frames. ], tot_loss[loss=0.2046, simple_loss=0.2828, pruned_loss=0.06326, over 3714866.18 frames. ], batch size: 45, lr: 8.24e-03, grad_scale: 8.0
2022-12-23 06:32:13,300 INFO [zipformer.py:660] (2/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=47597.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 06:32:27,592 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([2.5570, 2.3883, 1.9547, 0.9524, 1.9104, 2.0662, 1.6700, 1.8716],
       device='cuda:2'), covar=tensor([0.0621, 0.0473, 0.1206, 0.1608, 0.1188, 0.1418, 0.1596, 0.0826],
       device='cuda:2'), in_proj_covar=tensor([0.0167, 0.0181, 0.0205, 0.0194, 0.0208, 0.0194, 0.0209, 0.0197],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-23 06:32:43,690 WARNING [train.py:1060] (2/4) Exclude cut with ID 2195-150901-0045-59933-0 from training. Duration: 20.65
2022-12-23 06:32:51,898 WARNING [train.py:1060] (2/4) Exclude cut with ID 5796-66357-0007-116447-0 from training. Duration: 21.46
2022-12-23 06:33:05,405 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 2.860e+02 5.137e+02 6.273e+02 7.724e+02 1.894e+03, threshold=1.255e+03, percent-clipped=3.0
2022-12-23 06:33:05,420 INFO [train.py:894] (2/4) Epoch 14, batch 2050, loss[loss=0.1859, simple_loss=0.2722, pruned_loss=0.04977, over 18547.00 frames. ], tot_loss[loss=0.2049, simple_loss=0.2828, pruned_loss=0.0635, over 3714450.39 frames. ], batch size: 55, lr: 8.24e-03, grad_scale: 8.0
2022-12-23 06:33:08,987 INFO [scaling.py:679] (2/4) Whitening: num_groups=8, num_channels=96, metric=1.19 vs. limit=2.0
2022-12-23 06:33:36,863 WARNING [train.py:1060] (2/4) Exclude cut with ID 3557-8342-0013-54691-0 from training. Duration: 0.92
2022-12-23 06:33:42,947 WARNING [train.py:1060] (2/4) Exclude cut with ID 8291-282929-0023-13010-0_sp0.9 from training. Duration: 23.7666875
2022-12-23 06:34:20,885 INFO [train.py:894] (2/4) Epoch 14, batch 2100, loss[loss=0.1892, simple_loss=0.2601, pruned_loss=0.05915, over 18556.00 frames. ], tot_loss[loss=0.2049, simple_loss=0.2823, pruned_loss=0.06376, over 3714255.08 frames. ], batch size: 44, lr: 8.23e-03, grad_scale: 8.0
2022-12-23 06:34:20,931 WARNING [train.py:1060] (2/4) Exclude cut with ID 8544-281189-0060-101339-0_sp0.9 from training. Duration: 20.861125
2022-12-23 06:34:31,827 WARNING [train.py:1060] (2/4) Exclude cut with ID 3488-65654-0031-41259-0_sp0.9 from training. Duration: 22.711125
2022-12-23 06:34:35,736 INFO [scaling.py:679] (2/4) Whitening: num_groups=1, num_channels=384, metric=3.79 vs. limit=5.0
2022-12-23 06:35:16,045 WARNING [train.py:1060] (2/4) Exclude cut with ID 6951-79737-0043-132310-0_sp1.1 from training. Duration: 22.986375
2022-12-23 06:35:32,225 WARNING [train.py:1060] (2/4) Exclude cut with ID 8040-260924-0003-80960-0_sp0.9 from training. Duration: 22.07225
2022-12-23 06:35:38,042 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 3.231e+02 4.876e+02 6.072e+02 7.489e+02 2.027e+03, threshold=1.214e+03, percent-clipped=4.0
2022-12-23 06:35:38,058 INFO [train.py:894] (2/4) Epoch 14, batch 2150, loss[loss=0.199, simple_loss=0.2784, pruned_loss=0.05984, over 18462.00 frames. ], tot_loss[loss=0.2056, simple_loss=0.2828, pruned_loss=0.06418, over 3714054.30 frames. ], batch size: 50, lr: 8.23e-03, grad_scale: 8.0
2022-12-23 06:35:38,074 WARNING [train.py:1060] (2/4) Exclude cut with ID 7699-105389-0045-26330-0_sp0.9 from training. Duration: 20.3055625
2022-12-23 06:35:41,140 WARNING [train.py:1060] (2/4) Exclude cut with ID 6356-271890-0060-94317-0_sp0.9 from training. Duration: 20.72225
2022-12-23 06:35:44,072 INFO [zipformer.py:660] (2/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=47735.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 06:35:48,566 INFO [zipformer.py:660] (2/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=47738.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 06:36:00,233 WARNING [train.py:1060] (2/4) Exclude cut with ID 3488-85273-0017-41203-0_sp1.1 from training. Duration: 22.4818125
2022-12-23 06:36:16,507 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([4.7158, 4.0125, 3.9822, 4.5682, 4.2770, 4.1276, 4.8111, 1.4264],
       device='cuda:2'), covar=tensor([0.0714, 0.0672, 0.0595, 0.0809, 0.1481, 0.1159, 0.0572, 0.4914],
       device='cuda:2'), in_proj_covar=tensor([0.0320, 0.0212, 0.0218, 0.0241, 0.0301, 0.0252, 0.0263, 0.0271],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0001, 0.0001, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-23 06:36:28,891 WARNING [train.py:1060] (2/4) Exclude cut with ID 4964-30587-0040-44509-0_sp0.9 from training. Duration: 25.0944375
2022-12-23 06:36:33,955 WARNING [train.py:1060] (2/4) Exclude cut with ID 6533-399-0047-104881-0 from training. Duration: 21.515
2022-12-23 06:36:40,222 WARNING [train.py:1060] (2/4) Exclude cut with ID 7357-94126-0009-15840-0_sp0.9 from training. Duration: 27.02225
2022-12-23 06:36:46,017 WARNING [train.py:1060] (2/4) Exclude cut with ID 432-122774-0010-62480-0_sp0.9 from training. Duration: 22.22225
2022-12-23 06:36:53,119 WARNING [train.py:1060] (2/4) Exclude cut with ID 4964-30587-0085-44554-0_sp0.9 from training. Duration: 20.85
2022-12-23 06:36:54,592 INFO [train.py:894] (2/4) Epoch 14, batch 2200, loss[loss=0.1801, simple_loss=0.2502, pruned_loss=0.05502, over 18480.00 frames. ], tot_loss[loss=0.2042, simple_loss=0.2813, pruned_loss=0.06358, over 3713694.26 frames. ], batch size: 43, lr: 8.22e-03, grad_scale: 8.0
2022-12-23 06:37:05,286 INFO [zipformer.py:660] (2/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=47788.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 06:37:21,368 INFO [zipformer.py:660] (2/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=47799.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 06:37:25,994 WARNING [train.py:1060] (2/4) Exclude cut with ID 4295-39940-0007-92567-0 from training. Duration: 21.54
2022-12-23 06:37:29,641 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.9147, 1.6690, 1.7230, 1.1441, 2.3698, 2.0201, 1.9077, 1.0548],
       device='cuda:2'), covar=tensor([0.0419, 0.0520, 0.0465, 0.0761, 0.0255, 0.0405, 0.0487, 0.1594],
       device='cuda:2'), in_proj_covar=tensor([0.0121, 0.0121, 0.0127, 0.0118, 0.0091, 0.0119, 0.0134, 0.0153],
       device='cuda:2'), out_proj_covar=tensor([0.0001, 0.0001, 0.0002, 0.0001, 0.0001, 0.0001, 0.0002, 0.0002],
       device='cuda:2')
2022-12-23 06:37:32,024 WARNING [train.py:1060] (2/4) Exclude cut with ID 4964-30587-0040-44509-0_sp1.1 from training. Duration: 20.5318125
2022-12-23 06:37:37,483 INFO [zipformer.py:660] (2/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=47809.0, num_to_drop=1, layers_to_drop={1}
2022-12-23 06:37:40,252 WARNING [train.py:1060] (2/4) Exclude cut with ID 7255-291500-0012-134311-0_sp0.9 from training. Duration: 21.9333125
2022-12-23 06:38:09,844 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 2.708e+02 5.175e+02 6.084e+02 7.017e+02 2.042e+03, threshold=1.217e+03, percent-clipped=4.0
2022-12-23 06:38:09,860 INFO [train.py:894] (2/4) Epoch 14, batch 2250, loss[loss=0.2184, simple_loss=0.2817, pruned_loss=0.07757, over 18603.00 frames. ], tot_loss[loss=0.2047, simple_loss=0.2816, pruned_loss=0.06396, over 3714666.10 frames. ], batch size: 45, lr: 8.22e-03, grad_scale: 8.0
2022-12-23 06:38:27,199 WARNING [train.py:1060] (2/4) Exclude cut with ID 8631-249866-0025-130151-0_sp0.9 from training. Duration: 21.7944375
2022-12-23 06:38:28,701 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([3.6941, 3.1912, 3.2164, 3.6266, 3.3560, 3.3314, 3.8052, 1.4208],
       device='cuda:2'), covar=tensor([0.0776, 0.0670, 0.0625, 0.0874, 0.1347, 0.1061, 0.0663, 0.4169],
       device='cuda:2'), in_proj_covar=tensor([0.0319, 0.0211, 0.0216, 0.0239, 0.0300, 0.0253, 0.0261, 0.0268],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0001, 0.0001, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-23 06:38:36,735 INFO [zipformer.py:660] (2/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=47849.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 06:38:41,616 WARNING [train.py:1060] (2/4) Exclude cut with ID 8291-282929-0002-12989-0_sp0.9 from training. Duration: 22.4666875
2022-12-23 06:38:47,455 WARNING [train.py:1060] (2/4) Exclude cut with ID 6121-9014-0076-24124-0 from training. Duration: 21.635
2022-12-23 06:38:53,162 WARNING [train.py:1060] (2/4) Exclude cut with ID 6121-9014-0076-24124-0_sp0.9 from training. Duration: 24.038875
2022-12-23 06:38:57,352 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.6952, 2.1626, 1.5123, 2.4731, 2.9828, 1.6714, 1.9860, 1.4435],
       device='cuda:2'), covar=tensor([0.1930, 0.1760, 0.1631, 0.0981, 0.1444, 0.1221, 0.1919, 0.1538],
       device='cuda:2'), in_proj_covar=tensor([0.0239, 0.0216, 0.0204, 0.0191, 0.0256, 0.0191, 0.0216, 0.0192],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-23 06:39:08,973 INFO [zipformer.py:660] (2/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=47870.0, num_to_drop=1, layers_to_drop={1}
2022-12-23 06:39:25,657 INFO [train.py:894] (2/4) Epoch 14, batch 2300, loss[loss=0.2334, simple_loss=0.2989, pruned_loss=0.08395, over 18624.00 frames. ], tot_loss[loss=0.2046, simple_loss=0.2814, pruned_loss=0.06395, over 3715161.68 frames. ], batch size: 177, lr: 8.22e-03, grad_scale: 8.0
2022-12-23 06:39:35,763 WARNING [train.py:1060] (2/4) Exclude cut with ID 7255-291500-0009-134308-0_sp1.1 from training. Duration: 21.786375
2022-12-23 06:39:48,687 WARNING [train.py:1060] (2/4) Exclude cut with ID 8291-282929-0002-12989-0 from training. Duration: 20.22
2022-12-23 06:39:50,409 INFO [zipformer.py:660] (2/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=47897.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 06:40:41,724 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 3.478e+02 5.323e+02 6.580e+02 8.724e+02 2.507e+03, threshold=1.316e+03, percent-clipped=6.0
2022-12-23 06:40:41,740 INFO [train.py:894] (2/4) Epoch 14, batch 2350, loss[loss=0.1901, simple_loss=0.2721, pruned_loss=0.05409, over 18625.00 frames. ], tot_loss[loss=0.2053, simple_loss=0.2816, pruned_loss=0.06447, over 3715532.18 frames. ], batch size: 53, lr: 8.21e-03, grad_scale: 8.0
2022-12-23 06:41:02,301 INFO [zipformer.py:660] (2/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=47945.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 06:41:52,652 WARNING [train.py:1060] (2/4) Exclude cut with ID 6951-79737-0043-132310-0 from training. Duration: 25.285
2022-12-23 06:41:57,057 INFO [train.py:894] (2/4) Epoch 14, batch 2400, loss[loss=0.2014, simple_loss=0.2831, pruned_loss=0.05982, over 18616.00 frames. ], tot_loss[loss=0.2048, simple_loss=0.2814, pruned_loss=0.06407, over 3715806.64 frames. ], batch size: 98, lr: 8.21e-03, grad_scale: 8.0
2022-12-23 06:42:48,785 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.7596, 1.1016, 0.8638, 1.2705, 2.0319, 0.9119, 1.3516, 1.6354],
       device='cuda:2'), covar=tensor([0.1504, 0.2223, 0.2274, 0.1648, 0.1761, 0.1867, 0.1538, 0.1608],
       device='cuda:2'), in_proj_covar=tensor([0.0092, 0.0098, 0.0118, 0.0096, 0.0113, 0.0092, 0.0097, 0.0093],
       device='cuda:2'), out_proj_covar=tensor([0.0003, 0.0003, 0.0004, 0.0003, 0.0004, 0.0003, 0.0003, 0.0003],
       device='cuda:2')
2022-12-23 06:43:00,378 WARNING [train.py:1060] (2/4) Exclude cut with ID 811-130148-0001-63453-0_sp0.9 from training. Duration: 20.861125
2022-12-23 06:43:16,634 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 2.720e+02 4.531e+02 5.657e+02 7.183e+02 1.580e+03, threshold=1.131e+03, percent-clipped=2.0
2022-12-23 06:43:16,649 INFO [train.py:894] (2/4) Epoch 14, batch 2450, loss[loss=0.1921, simple_loss=0.2769, pruned_loss=0.0536, over 18499.00 frames. ], tot_loss[loss=0.2054, simple_loss=0.2822, pruned_loss=0.0643, over 3714863.63 frames. ], batch size: 52, lr: 8.20e-03, grad_scale: 8.0
2022-12-23 06:43:20,786 WARNING [train.py:1060] (2/4) Exclude cut with ID 6010-56788-0055-90261-0 from training. Duration: 20.88
2022-12-23 06:43:23,116 INFO [zipformer.py:660] (2/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=48035.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 06:43:57,571 WARNING [train.py:1060] (2/4) Exclude cut with ID 7357-94126-0045-15876-0_sp0.9 from training. Duration: 23.4166875
2022-12-23 06:44:11,283 INFO [scaling.py:679] (2/4) Whitening: num_groups=1, num_channels=384, metric=3.00 vs. limit=5.0
2022-12-23 06:44:18,181 INFO [zipformer.py:660] (2/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=48072.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 06:44:30,887 INFO [train.py:894] (2/4) Epoch 14, batch 2500, loss[loss=0.2078, simple_loss=0.2908, pruned_loss=0.06236, over 18565.00 frames. ], tot_loss[loss=0.2039, simple_loss=0.281, pruned_loss=0.06336, over 3714484.99 frames. ], batch size: 57, lr: 8.20e-03, grad_scale: 8.0
2022-12-23 06:44:34,713 INFO [zipformer.py:660] (2/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=48083.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 06:44:52,062 INFO [zipformer.py:660] (2/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=48094.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 06:45:16,272 WARNING [train.py:1060] (2/4) Exclude cut with ID 7255-291500-0006-134305-0 from training. Duration: 21.24
2022-12-23 06:45:16,286 WARNING [train.py:1060] (2/4) Exclude cut with ID 6533-399-0047-104881-0_sp0.9 from training. Duration: 23.9055625
2022-12-23 06:45:21,553 INFO [scaling.py:679] (2/4) Whitening: num_groups=8, num_channels=192, metric=1.89 vs. limit=2.0
2022-12-23 06:45:31,603 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([6.0557, 5.0957, 5.4079, 6.0261, 5.5725, 5.4453, 6.0200, 1.5767],
       device='cuda:2'), covar=tensor([0.0581, 0.0568, 0.0478, 0.0654, 0.1226, 0.1128, 0.0399, 0.5431],
       device='cuda:2'), in_proj_covar=tensor([0.0319, 0.0211, 0.0219, 0.0239, 0.0302, 0.0256, 0.0263, 0.0269],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0001, 0.0001, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-23 06:45:40,664 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([2.6360, 2.7116, 1.7778, 3.1291, 2.8520, 2.3326, 3.8123, 2.5657],
       device='cuda:2'), covar=tensor([0.0761, 0.1651, 0.2711, 0.1827, 0.1552, 0.0917, 0.0843, 0.1191],
       device='cuda:2'), in_proj_covar=tensor([0.0173, 0.0197, 0.0240, 0.0285, 0.0228, 0.0183, 0.0206, 0.0198],
       device='cuda:2'), out_proj_covar=tensor([0.0001, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-23 06:45:48,282 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 3.059e+02 4.733e+02 5.770e+02 7.680e+02 1.345e+03, threshold=1.154e+03, percent-clipped=2.0
2022-12-23 06:45:48,300 INFO [train.py:894] (2/4) Epoch 14, batch 2550, loss[loss=0.1815, simple_loss=0.2557, pruned_loss=0.0536, over 18411.00 frames. ], tot_loss[loss=0.2031, simple_loss=0.2804, pruned_loss=0.0629, over 3713897.10 frames. ], batch size: 42, lr: 8.19e-03, grad_scale: 8.0
2022-12-23 06:45:48,336 WARNING [train.py:1060] (2/4) Exclude cut with ID 6758-72288-0033-108368-0_sp0.9 from training. Duration: 25.988875
2022-12-23 06:45:51,952 INFO [zipformer.py:660] (2/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=48133.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 06:45:58,488 WARNING [train.py:1060] (2/4) Exclude cut with ID 7255-291500-0001-134300-0_sp0.9 from training. Duration: 20.67225
2022-12-23 06:46:09,285 INFO [zipformer.py:660] (2/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=48144.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 06:46:16,175 INFO [scaling.py:679] (2/4) Whitening: num_groups=8, num_channels=96, metric=1.62 vs. limit=2.0
2022-12-23 06:46:40,959 INFO [zipformer.py:660] (2/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=48165.0, num_to_drop=1, layers_to_drop={3}
2022-12-23 06:46:45,136 WARNING [train.py:1060] (2/4) Exclude cut with ID 3488-85273-0038-41224-0 from training. Duration: 20.34
2022-12-23 06:47:05,104 INFO [train.py:894] (2/4) Epoch 14, batch 2600, loss[loss=0.1933, simple_loss=0.2784, pruned_loss=0.05404, over 18599.00 frames. ], tot_loss[loss=0.2027, simple_loss=0.2802, pruned_loss=0.0626, over 3713850.35 frames. ], batch size: 99, lr: 8.19e-03, grad_scale: 8.0
2022-12-23 06:47:47,273 INFO [scaling.py:679] (2/4) Whitening: num_groups=8, num_channels=96, metric=1.37 vs. limit=2.0
2022-12-23 06:47:52,370 INFO [zipformer.py:660] (2/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=48212.0, num_to_drop=1, layers_to_drop={1}
2022-12-23 06:47:54,966 WARNING [train.py:1060] (2/4) Exclude cut with ID 7357-94126-0026-15857-0_sp0.9 from training. Duration: 25.061125
2022-12-23 06:48:07,518 WARNING [train.py:1060] (2/4) Exclude cut with ID 3033-130750-0096-55598-0 from training. Duration: 0.83
2022-12-23 06:48:22,224 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 3.675e+02 5.046e+02 6.298e+02 8.011e+02 1.904e+03, threshold=1.260e+03, percent-clipped=4.0
2022-12-23 06:48:22,241 INFO [train.py:894] (2/4) Epoch 14, batch 2650, loss[loss=0.2111, simple_loss=0.2793, pruned_loss=0.0715, over 18525.00 frames. ], tot_loss[loss=0.2013, simple_loss=0.2791, pruned_loss=0.06173, over 3713886.55 frames. ], batch size: 47, lr: 8.19e-03, grad_scale: 8.0
2022-12-23 06:48:34,082 WARNING [train.py:1060] (2/4) Exclude cut with ID 3488-85273-0017-41203-0 from training. Duration: 24.73
2022-12-23 06:48:45,848 WARNING [train.py:1060] (2/4) Exclude cut with ID 7255-291500-0009-134308-0 from training. Duration: 23.965
2022-12-23 06:48:55,959 WARNING [train.py:1060] (2/4) Exclude cut with ID 6978-92210-0030-146996-0_sp0.9 from training. Duration: 22.088875
2022-12-23 06:49:04,363 INFO [scaling.py:679] (2/4) Whitening: num_groups=8, num_channels=96, metric=1.43 vs. limit=2.0
2022-12-23 06:49:10,102 WARNING [train.py:1060] (2/4) Exclude cut with ID 7255-291500-0006-134305-0_sp0.9 from training. Duration: 23.6
2022-12-23 06:49:26,283 INFO [zipformer.py:660] (2/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=48273.0, num_to_drop=1, layers_to_drop={2}
2022-12-23 06:49:38,603 INFO [train.py:894] (2/4) Epoch 14, batch 2700, loss[loss=0.2032, simple_loss=0.2857, pruned_loss=0.06033, over 18685.00 frames. ], tot_loss[loss=0.2015, simple_loss=0.2792, pruned_loss=0.06193, over 3713611.50 frames. ], batch size: 46, lr: 8.18e-03, grad_scale: 8.0
2022-12-23 06:50:44,328 INFO [zipformer.py:660] (2/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=48324.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 06:50:54,771 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 3.448e+02 4.967e+02 5.745e+02 6.581e+02 2.319e+03, threshold=1.149e+03, percent-clipped=2.0
2022-12-23 06:50:54,787 INFO [train.py:894] (2/4) Epoch 14, batch 2750, loss[loss=0.1726, simple_loss=0.2473, pruned_loss=0.04896, over 18404.00 frames. ], tot_loss[loss=0.2012, simple_loss=0.279, pruned_loss=0.06168, over 3713692.68 frames. ], batch size: 42, lr: 8.18e-03, grad_scale: 8.0
2022-12-23 06:50:54,842 WARNING [train.py:1060] (2/4) Exclude cut with ID 8291-282929-0024-13011-0 from training. Duration: 23.795
2022-12-23 06:51:10,192 WARNING [train.py:1060] (2/4) Exclude cut with ID 8631-249866-0030-130156-0_sp1.1 from training. Duration: 21.5409375
2022-12-23 06:51:14,663 WARNING [train.py:1060] (2/4) Exclude cut with ID 6978-92210-0019-146985-0_sp0.9 from training. Duration: 24.97775
2022-12-23 06:51:22,825 INFO [scaling.py:679] (2/4) Whitening: num_groups=1, num_channels=384, metric=4.74 vs. limit=5.0
2022-12-23 06:51:24,777 WARNING [train.py:1060] (2/4) Exclude cut with ID 1085-156170-0017-128270-0_sp0.9 from training. Duration: 23.3444375
2022-12-23 06:51:28,086 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.7762, 1.2670, 1.4220, 2.0044, 1.7704, 3.5315, 1.2967, 1.5039],
       device='cuda:2'), covar=tensor([0.1071, 0.2520, 0.1339, 0.1074, 0.1666, 0.0262, 0.2004, 0.2077],
       device='cuda:2'), in_proj_covar=tensor([0.0073, 0.0081, 0.0073, 0.0073, 0.0089, 0.0073, 0.0083, 0.0076],
       device='cuda:2'), out_proj_covar=tensor([0.0003, 0.0004, 0.0004, 0.0003, 0.0004, 0.0003, 0.0004, 0.0004],
       device='cuda:2')
2022-12-23 06:51:53,295 WARNING [train.py:1060] (2/4) Exclude cut with ID 6010-56788-0055-90261-0_sp0.9 from training. Duration: 23.2
2022-12-23 06:51:59,875 WARNING [train.py:1060] (2/4) Exclude cut with ID 5653-46179-0060-117930-0_sp0.9 from training. Duration: 21.17225
2022-12-23 06:52:11,702 INFO [train.py:894] (2/4) Epoch 14, batch 2800, loss[loss=0.2384, simple_loss=0.3076, pruned_loss=0.08463, over 18709.00 frames. ], tot_loss[loss=0.2018, simple_loss=0.2793, pruned_loss=0.06214, over 3713428.52 frames. ], batch size: 52, lr: 8.17e-03, grad_scale: 8.0
2022-12-23 06:52:18,324 INFO [zipformer.py:660] (2/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=48385.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 06:52:19,448 WARNING [train.py:1060] (2/4) Exclude cut with ID 7255-291500-0008-134307-0_sp0.9 from training. Duration: 24.6555625
2022-12-23 06:52:31,331 INFO [zipformer.py:660] (2/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=48394.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 06:52:35,619 INFO [zipformer.py:660] (2/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=48397.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 06:52:54,478 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([2.6778, 2.7146, 2.1134, 2.0736, 3.0881, 2.9963, 2.6499, 2.2361],
       device='cuda:2'), covar=tensor([0.0340, 0.0281, 0.0485, 0.0527, 0.0202, 0.0267, 0.0372, 0.0623],
       device='cuda:2'), in_proj_covar=tensor([0.0123, 0.0122, 0.0129, 0.0119, 0.0092, 0.0119, 0.0136, 0.0154],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0001, 0.0002, 0.0001, 0.0001, 0.0001, 0.0002, 0.0002],
       device='cuda:2')
2022-12-23 06:53:15,106 WARNING [train.py:1060] (2/4) Exclude cut with ID 3488-65654-0031-41259-0 from training. Duration: 20.44
2022-12-23 06:53:17,258 INFO [zipformer.py:660] (2/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=48423.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 06:53:24,379 INFO [zipformer.py:660] (2/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=48428.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 06:53:28,734 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 3.239e+02 4.838e+02 5.602e+02 6.913e+02 2.084e+03, threshold=1.120e+03, percent-clipped=3.0
2022-12-23 06:53:28,749 INFO [train.py:894] (2/4) Epoch 14, batch 2850, loss[loss=0.2228, simple_loss=0.2934, pruned_loss=0.07614, over 18501.00 frames. ], tot_loss[loss=0.2014, simple_loss=0.2788, pruned_loss=0.06194, over 3712936.92 frames. ], batch size: 52, lr: 8.17e-03, grad_scale: 8.0
2022-12-23 06:53:30,255 WARNING [train.py:1060] (2/4) Exclude cut with ID 6951-79737-0018-132285-0_sp0.9 from training. Duration: 23.45
2022-12-23 06:53:44,637 INFO [zipformer.py:660] (2/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=48442.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 06:53:47,603 INFO [zipformer.py:660] (2/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=48444.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 06:54:01,297 WARNING [train.py:1060] (2/4) Exclude cut with ID 6945-60535-0076-12784-0_sp0.9 from training. Duration: 20.52225
2022-12-23 06:54:08,780 WARNING [train.py:1060] (2/4) Exclude cut with ID 7255-291500-0008-134307-0 from training. Duration: 22.19
2022-12-23 06:54:09,105 INFO [zipformer.py:660] (2/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=48458.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 06:54:20,262 WARNING [train.py:1060] (2/4) Exclude cut with ID 7357-94126-0014-15845-0_sp1.1 from training. Duration: 25.3818125
2022-12-23 06:54:20,531 INFO [zipformer.py:660] (2/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=48465.0, num_to_drop=1, layers_to_drop={0}
2022-12-23 06:54:38,045 WARNING [train.py:1060] (2/4) Exclude cut with ID 6951-79737-0043-132310-0_sp0.9 from training. Duration: 28.0944375
2022-12-23 06:54:42,569 WARNING [train.py:1060] (2/4) Exclude cut with ID 2195-150901-0045-59933-0_sp0.9 from training. Duration: 22.9444375
2022-12-23 06:54:44,030 INFO [train.py:894] (2/4) Epoch 14, batch 2900, loss[loss=0.162, simple_loss=0.2368, pruned_loss=0.0436, over 18396.00 frames. ], tot_loss[loss=0.2017, simple_loss=0.2791, pruned_loss=0.06219, over 3712388.41 frames. ], batch size: 46, lr: 8.17e-03, grad_scale: 8.0
2022-12-23 06:54:44,571 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.4610, 2.0013, 1.4323, 2.2731, 2.4327, 1.5173, 1.6132, 1.2336],
       device='cuda:2'), covar=tensor([0.1966, 0.1606, 0.1595, 0.0960, 0.1383, 0.1150, 0.2008, 0.1554],
       device='cuda:2'), in_proj_covar=tensor([0.0240, 0.0217, 0.0206, 0.0191, 0.0257, 0.0193, 0.0217, 0.0194],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-23 06:54:49,068 INFO [zipformer.py:660] (2/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=48484.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 06:54:50,280 WARNING [train.py:1060] (2/4) Exclude cut with ID 8291-282929-0024-13011-0_sp1.1 from training. Duration: 21.6318125
2022-12-23 06:55:00,608 INFO [zipformer.py:660] (2/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=48492.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 06:55:08,048 WARNING [train.py:1060] (2/4) Exclude cut with ID 8631-249866-0030-130156-0 from training. Duration: 23.695
2022-12-23 06:55:33,864 INFO [zipformer.py:660] (2/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=48513.0, num_to_drop=1, layers_to_drop={0}
2022-12-23 06:55:37,882 WARNING [train.py:1060] (2/4) Exclude cut with ID 7699-105389-0094-26379-0 from training. Duration: 23.955
2022-12-23 06:56:00,358 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 3.111e+02 4.661e+02 5.748e+02 7.181e+02 1.318e+03, threshold=1.150e+03, percent-clipped=2.0
2022-12-23 06:56:00,373 INFO [train.py:894] (2/4) Epoch 14, batch 2950, loss[loss=0.2116, simple_loss=0.2859, pruned_loss=0.06868, over 18631.00 frames. ], tot_loss[loss=0.2015, simple_loss=0.279, pruned_loss=0.06198, over 3712883.43 frames. ], batch size: 97, lr: 8.16e-03, grad_scale: 16.0
2022-12-23 06:56:09,267 WARNING [train.py:1060] (2/4) Exclude cut with ID 8291-282929-0024-13011-0_sp0.9 from training. Duration: 26.438875
2022-12-23 06:56:09,592 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.7283, 1.7246, 2.0694, 1.1128, 1.8882, 2.0746, 1.5427, 2.4716],
       device='cuda:2'), covar=tensor([0.1099, 0.1660, 0.1044, 0.1692, 0.0740, 0.1033, 0.2057, 0.0453],
       device='cuda:2'), in_proj_covar=tensor([0.0197, 0.0203, 0.0203, 0.0192, 0.0175, 0.0212, 0.0207, 0.0184],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-23 06:56:55,648 WARNING [train.py:1060] (2/4) Exclude cut with ID 7699-105389-0021-26306-0_sp0.9 from training. Duration: 21.2444375
2022-12-23 06:56:57,088 WARNING [train.py:1060] (2/4) Exclude cut with ID 7357-94126-0014-15845-0_sp0.9 from training. Duration: 31.02225
2022-12-23 06:56:57,386 INFO [zipformer.py:660] (2/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=48568.0, num_to_drop=1, layers_to_drop={0}
2022-12-23 06:57:07,203 WARNING [train.py:1060] (2/4) Exclude cut with ID 432-122774-0017-62487-0 from training. Duration: 22.395
2022-12-23 06:57:16,283 INFO [train.py:894] (2/4) Epoch 14, batch 3000, loss[loss=0.2024, simple_loss=0.2679, pruned_loss=0.06844, over 18534.00 frames. ], tot_loss[loss=0.2009, simple_loss=0.2785, pruned_loss=0.0616, over 3713246.88 frames. ], batch size: 44, lr: 8.16e-03, grad_scale: 16.0
2022-12-23 06:57:16,283 INFO [train.py:919] (2/4) Computing validation loss
2022-12-23 06:57:27,212 INFO [train.py:928] (2/4) Epoch 14, validation: loss=0.1688, simple_loss=0.2675, pruned_loss=0.03502, over 944034.00 frames. 
2022-12-23 06:57:27,212 INFO [train.py:929] (2/4) Maximum memory allocated so far is 24680MB
2022-12-23 06:57:33,073 WARNING [train.py:1060] (2/4) Exclude cut with ID 7357-94126-0045-15876-0 from training. Duration: 21.075
2022-12-23 06:57:39,768 WARNING [train.py:1060] (2/4) Exclude cut with ID 6482-98857-0025-147532-0_sp0.9 from training. Duration: 20.0055625
2022-12-23 06:57:39,775 WARNING [train.py:1060] (2/4) Exclude cut with ID 6951-79737-0037-132304-0_sp0.9 from training. Duration: 22.05
2022-12-23 06:57:39,782 WARNING [train.py:1060] (2/4) Exclude cut with ID 7255-291500-0003-134302-0 from training. Duration: 26.8349375
2022-12-23 06:57:43,306 WARNING [train.py:1060] (2/4) Exclude cut with ID 7357-94126-0009-15840-0_sp1.1 from training. Duration: 22.1090625
2022-12-23 06:57:50,572 WARNING [train.py:1060] (2/4) Exclude cut with ID 7699-105389-0094-26379-0_sp0.9 from training. Duration: 26.6166875
2022-12-23 06:58:07,588 WARNING [train.py:1060] (2/4) Exclude cut with ID 2046-178027-0000-53705-0_sp0.9 from training. Duration: 20.3055625
2022-12-23 06:58:28,576 WARNING [train.py:1060] (2/4) Exclude cut with ID 7205-50138-0008-5373-0_sp0.9 from training. Duration: 20.7
2022-12-23 06:58:43,454 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 3.119e+02 4.638e+02 5.784e+02 7.167e+02 1.598e+03, threshold=1.157e+03, percent-clipped=4.0
2022-12-23 06:58:43,470 INFO [train.py:894] (2/4) Epoch 14, batch 3050, loss[loss=0.1544, simple_loss=0.2324, pruned_loss=0.03825, over 18595.00 frames. ], tot_loss[loss=0.2021, simple_loss=0.2798, pruned_loss=0.06218, over 3713187.96 frames. ], batch size: 45, lr: 8.15e-03, grad_scale: 16.0
2022-12-23 06:58:49,404 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([5.6441, 4.8176, 5.0270, 5.6086, 5.0827, 5.0229, 5.6940, 1.6441],
       device='cuda:2'), covar=tensor([0.0586, 0.0594, 0.0456, 0.0621, 0.1236, 0.1026, 0.0425, 0.4998],
       device='cuda:2'), in_proj_covar=tensor([0.0322, 0.0214, 0.0222, 0.0242, 0.0305, 0.0255, 0.0269, 0.0270],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0001, 0.0001, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-23 06:59:15,014 WARNING [train.py:1060] (2/4) Exclude cut with ID 6978-92210-0019-146985-0 from training. Duration: 22.48
2022-12-23 06:59:15,493 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([2.5232, 2.4750, 1.6600, 3.1541, 2.8958, 2.3372, 3.9686, 2.3495],
       device='cuda:2'), covar=tensor([0.0773, 0.1824, 0.2865, 0.1819, 0.1531, 0.0929, 0.0780, 0.1160],
       device='cuda:2'), in_proj_covar=tensor([0.0174, 0.0199, 0.0242, 0.0287, 0.0228, 0.0186, 0.0207, 0.0198],
       device='cuda:2'), out_proj_covar=tensor([0.0001, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-23 06:59:30,155 WARNING [train.py:1060] (2/4) Exclude cut with ID 7255-291500-0003-134302-0_sp0.9 from training. Duration: 29.816625
2022-12-23 06:59:50,810 WARNING [train.py:1060] (2/4) Exclude cut with ID 7255-291500-0005-134304-0_sp1.1 from training. Duration: 22.7590625
2022-12-23 06:59:55,547 WARNING [train.py:1060] (2/4) Exclude cut with ID 7357-94126-0026-15857-0 from training. Duration: 22.555
2022-12-23 06:59:58,798 INFO [zipformer.py:660] (2/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=48680.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 07:00:00,080 INFO [train.py:894] (2/4) Epoch 14, batch 3100, loss[loss=0.1931, simple_loss=0.2783, pruned_loss=0.05391, over 18605.00 frames. ], tot_loss[loss=0.2014, simple_loss=0.279, pruned_loss=0.06185, over 3713708.10 frames. ], batch size: 51, lr: 8.15e-03, grad_scale: 16.0
2022-12-23 07:00:17,787 WARNING [train.py:1060] (2/4) Exclude cut with ID 1250-135782-0005-25975-0_sp0.9 from training. Duration: 21.688875
2022-12-23 07:00:43,933 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([2.0873, 2.0493, 1.5641, 2.2649, 2.2882, 1.9514, 2.9124, 2.1376],
       device='cuda:2'), covar=tensor([0.0848, 0.1549, 0.2737, 0.1747, 0.1618, 0.0906, 0.0885, 0.1155],
       device='cuda:2'), in_proj_covar=tensor([0.0174, 0.0199, 0.0242, 0.0286, 0.0228, 0.0185, 0.0206, 0.0198],
       device='cuda:2'), out_proj_covar=tensor([0.0001, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-23 07:00:50,871 WARNING [train.py:1060] (2/4) Exclude cut with ID 3488-85273-0038-41224-0_sp0.9 from training. Duration: 22.6
2022-12-23 07:01:14,132 INFO [zipformer.py:660] (2/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=48728.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 07:01:18,804 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 2.420e+02 4.890e+02 5.907e+02 7.278e+02 1.415e+03, threshold=1.181e+03, percent-clipped=1.0
2022-12-23 07:01:18,827 INFO [train.py:894] (2/4) Epoch 14, batch 3150, loss[loss=0.222, simple_loss=0.2956, pruned_loss=0.07416, over 18514.00 frames. ], tot_loss[loss=0.201, simple_loss=0.2789, pruned_loss=0.06157, over 3713266.50 frames. ], batch size: 77, lr: 8.14e-03, grad_scale: 16.0
2022-12-23 07:01:29,640 WARNING [train.py:1060] (2/4) Exclude cut with ID 7357-94126-0009-15840-0 from training. Duration: 24.32
2022-12-23 07:01:51,789 INFO [zipformer.py:660] (2/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=48753.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 07:02:26,834 WARNING [train.py:1060] (2/4) Exclude cut with ID 8291-276745-0093-13116-0_sp0.9 from training. Duration: 21.061125
2022-12-23 07:02:28,925 INFO [zipformer.py:660] (2/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=48776.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 07:02:33,312 INFO [zipformer.py:660] (2/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=48779.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 07:02:36,151 INFO [train.py:894] (2/4) Epoch 14, batch 3200, loss[loss=0.2259, simple_loss=0.3032, pruned_loss=0.07426, over 18565.00 frames. ], tot_loss[loss=0.2003, simple_loss=0.2781, pruned_loss=0.06124, over 3712558.68 frames. ], batch size: 57, lr: 8.14e-03, grad_scale: 16.0
2022-12-23 07:02:42,160 WARNING [train.py:1060] (2/4) Exclude cut with ID 7357-94126-0024-15855-0_sp0.9 from training. Duration: 20.32225
2022-12-23 07:02:56,287 WARNING [train.py:1060] (2/4) Exclude cut with ID 3033-130750-0096-55598-0_sp1.1 from training. Duration: 0.7545625
2022-12-23 07:03:12,361 WARNING [train.py:1060] (2/4) Exclude cut with ID 4295-39940-0007-92567-0_sp0.9 from training. Duration: 23.9333125
2022-12-23 07:03:35,179 INFO [zipformer.py:660] (2/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=48821.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 07:03:42,775 INFO [scaling.py:679] (2/4) Whitening: num_groups=8, num_channels=96, metric=1.35 vs. limit=2.0
2022-12-23 07:03:45,346 WARNING [train.py:1060] (2/4) Exclude cut with ID 7255-291500-0008-134307-0_sp1.1 from training. Duration: 20.17275
2022-12-23 07:03:51,307 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 3.097e+02 4.705e+02 5.372e+02 6.667e+02 1.296e+03, threshold=1.074e+03, percent-clipped=3.0
2022-12-23 07:03:51,323 INFO [train.py:894] (2/4) Epoch 14, batch 3250, loss[loss=0.1997, simple_loss=0.2825, pruned_loss=0.05849, over 18498.00 frames. ], tot_loss[loss=0.2012, simple_loss=0.2791, pruned_loss=0.06165, over 3713681.62 frames. ], batch size: 52, lr: 8.14e-03, grad_scale: 16.0
2022-12-23 07:03:52,757 WARNING [train.py:1060] (2/4) Exclude cut with ID 6978-92210-0019-146985-0_sp1.1 from training. Duration: 20.436375
2022-12-23 07:04:02,065 INFO [zipformer.py:660] (2/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=48838.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 07:04:15,082 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([2.0454, 2.0941, 1.4126, 2.2837, 2.2098, 1.9056, 2.9298, 2.0994],
       device='cuda:2'), covar=tensor([0.0847, 0.1655, 0.2859, 0.1904, 0.1754, 0.0922, 0.0952, 0.1100],
       device='cuda:2'), in_proj_covar=tensor([0.0175, 0.0200, 0.0243, 0.0286, 0.0230, 0.0186, 0.0207, 0.0198],
       device='cuda:2'), out_proj_covar=tensor([0.0001, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-23 07:04:24,524 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.6008, 1.1576, 1.7828, 3.0440, 2.1105, 2.3459, 1.1374, 2.2662],
       device='cuda:2'), covar=tensor([0.2013, 0.2230, 0.1813, 0.0782, 0.1222, 0.1440, 0.2293, 0.1257],
       device='cuda:2'), in_proj_covar=tensor([0.0101, 0.0116, 0.0130, 0.0137, 0.0105, 0.0135, 0.0129, 0.0110],
       device='cuda:2'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0004, 0.0003, 0.0004, 0.0003, 0.0003],
       device='cuda:2')
2022-12-23 07:04:41,851 INFO [scaling.py:679] (2/4) Whitening: num_groups=8, num_channels=96, metric=1.68 vs. limit=2.0
2022-12-23 07:04:51,956 INFO [zipformer.py:660] (2/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=48868.0, num_to_drop=1, layers_to_drop={0}
2022-12-23 07:05:12,149 INFO [train.py:894] (2/4) Epoch 14, batch 3300, loss[loss=0.1737, simple_loss=0.2542, pruned_loss=0.04663, over 18541.00 frames. ], tot_loss[loss=0.2015, simple_loss=0.2792, pruned_loss=0.06186, over 3714191.12 frames. ], batch size: 47, lr: 8.13e-03, grad_scale: 16.0
2022-12-23 07:05:14,085 INFO [zipformer.py:660] (2/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=48882.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 07:05:15,241 WARNING [train.py:1060] (2/4) Exclude cut with ID 4234-40345-0022-142709-0_sp0.9 from training. Duration: 23.1055625
2022-12-23 07:05:16,807 WARNING [train.py:1060] (2/4) Exclude cut with ID 8291-282929-0007-12994-0_sp1.1 from training. Duration: 23.5
2022-12-23 07:05:27,153 WARNING [train.py:1060] (2/4) Exclude cut with ID 7255-291500-0009-134308-0_sp0.9 from training. Duration: 26.62775
2022-12-23 07:05:27,453 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.4880, 1.4938, 1.4972, 1.5396, 1.1308, 2.9868, 1.2663, 1.7456],
       device='cuda:2'), covar=tensor([0.3977, 0.2403, 0.2237, 0.2364, 0.1597, 0.0280, 0.1673, 0.0973],
       device='cuda:2'), in_proj_covar=tensor([0.0135, 0.0116, 0.0126, 0.0120, 0.0103, 0.0098, 0.0093, 0.0089],
       device='cuda:2'), out_proj_covar=tensor([0.0005, 0.0005, 0.0005, 0.0005, 0.0004, 0.0003, 0.0004, 0.0004],
       device='cuda:2')
2022-12-23 07:05:38,848 INFO [zipformer.py:660] (2/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=48899.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 07:05:39,967 WARNING [train.py:1060] (2/4) Exclude cut with ID 6951-79737-0018-132285-0 from training. Duration: 21.105
2022-12-23 07:05:44,444 WARNING [train.py:1060] (2/4) Exclude cut with ID 4511-76322-0006-80011-0_sp0.9 from training. Duration: 24.411125
2022-12-23 07:06:04,730 INFO [zipformer.py:660] (2/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=48916.0, num_to_drop=1, layers_to_drop={0}
2022-12-23 07:06:11,770 WARNING [train.py:1060] (2/4) Exclude cut with ID 6758-72288-0033-108368-0_sp1.1 from training. Duration: 21.263625
2022-12-23 07:06:26,996 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 2.866e+02 5.170e+02 6.260e+02 9.545e+02 2.280e+03, threshold=1.252e+03, percent-clipped=16.0
2022-12-23 07:06:27,011 INFO [train.py:894] (2/4) Epoch 14, batch 3350, loss[loss=0.1936, simple_loss=0.2679, pruned_loss=0.0597, over 18529.00 frames. ], tot_loss[loss=0.2025, simple_loss=0.2798, pruned_loss=0.06259, over 3714956.43 frames. ], batch size: 47, lr: 8.13e-03, grad_scale: 16.0
2022-12-23 07:06:44,256 WARNING [train.py:1060] (2/4) Exclude cut with ID 4234-40345-0022-142709-0 from training. Duration: 20.795
2022-12-23 07:06:54,544 WARNING [train.py:1060] (2/4) Exclude cut with ID 7357-94126-0021-15852-0 from training. Duration: 24.76
2022-12-23 07:06:54,558 WARNING [train.py:1060] (2/4) Exclude cut with ID 3867-173237-0077-144769-0_sp0.9 from training. Duration: 22.25
2022-12-23 07:07:21,938 WARNING [train.py:1060] (2/4) Exclude cut with ID 7357-94126-0026-15857-0_sp1.1 from training. Duration: 20.5045625
2022-12-23 07:07:41,923 INFO [zipformer.py:660] (2/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=48980.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 07:07:43,306 INFO [train.py:894] (2/4) Epoch 14, batch 3400, loss[loss=0.2229, simple_loss=0.2964, pruned_loss=0.0747, over 18697.00 frames. ], tot_loss[loss=0.2019, simple_loss=0.2791, pruned_loss=0.06235, over 3713753.64 frames. ], batch size: 78, lr: 8.12e-03, grad_scale: 16.0
2022-12-23 07:08:51,317 INFO [zipformer.py:660] (2/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=49028.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 07:08:55,425 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 3.249e+02 4.856e+02 6.127e+02 7.528e+02 2.019e+03, threshold=1.225e+03, percent-clipped=5.0
2022-12-23 07:08:55,441 INFO [train.py:894] (2/4) Epoch 14, batch 3450, loss[loss=0.197, simple_loss=0.2792, pruned_loss=0.05733, over 18666.00 frames. ], tot_loss[loss=0.2034, simple_loss=0.2803, pruned_loss=0.06323, over 3713724.86 frames. ], batch size: 62, lr: 8.12e-03, grad_scale: 16.0
2022-12-23 07:09:27,233 INFO [zipformer.py:660] (2/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=49053.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 07:10:08,286 INFO [zipformer.py:660] (2/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=49079.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 07:10:11,319 INFO [train.py:894] (2/4) Epoch 14, batch 3500, loss[loss=0.2362, simple_loss=0.3055, pruned_loss=0.08346, over 18678.00 frames. ], tot_loss[loss=0.2029, simple_loss=0.2796, pruned_loss=0.06311, over 3713412.60 frames. ], batch size: 188, lr: 8.12e-03, grad_scale: 16.0
2022-12-23 07:10:31,227 WARNING [train.py:1060] (2/4) Exclude cut with ID 7859-102521-0017-7548-0_sp1.1 from training. Duration: 22.2954375
2022-12-23 07:10:41,352 INFO [train.py:894] (2/4) Epoch 15, batch 0, loss[loss=0.185, simple_loss=0.2673, pruned_loss=0.05138, over 18519.00 frames. ], tot_loss[loss=0.185, simple_loss=0.2673, pruned_loss=0.05138, over 18519.00 frames. ], batch size: 47, lr: 7.84e-03, grad_scale: 16.0
2022-12-23 07:10:41,352 INFO [train.py:919] (2/4) Computing validation loss
2022-12-23 07:10:52,028 INFO [train.py:928] (2/4) Epoch 15, validation: loss=0.1674, simple_loss=0.2666, pruned_loss=0.03411, over 944034.00 frames. 
2022-12-23 07:10:52,029 INFO [train.py:929] (2/4) Maximum memory allocated so far is 24680MB
2022-12-23 07:10:59,296 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.9861, 1.8373, 1.8770, 2.1254, 2.0813, 5.2597, 2.4539, 2.6971],
       device='cuda:2'), covar=tensor([0.3044, 0.2015, 0.1837, 0.1854, 0.1168, 0.0075, 0.1212, 0.0774],
       device='cuda:2'), in_proj_covar=tensor([0.0136, 0.0117, 0.0127, 0.0121, 0.0104, 0.0099, 0.0094, 0.0090],
       device='cuda:2'), out_proj_covar=tensor([0.0005, 0.0005, 0.0005, 0.0005, 0.0004, 0.0003, 0.0004, 0.0004],
       device='cuda:2')
2022-12-23 07:11:13,616 INFO [zipformer.py:660] (2/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=49101.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 07:11:45,173 WARNING [train.py:1060] (2/4) Exclude cut with ID 298-126791-0067-24026-0_sp0.9 from training. Duration: 21.438875
2022-12-23 07:11:48,152 WARNING [train.py:1060] (2/4) Exclude cut with ID 5652-39938-0025-23684-0_sp0.9 from training. Duration: 22.2055625
2022-12-23 07:11:50,614 INFO [scaling.py:679] (2/4) Whitening: num_groups=1, num_channels=384, metric=3.81 vs. limit=5.0
2022-12-23 07:11:52,703 INFO [zipformer.py:660] (2/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=49127.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 07:11:58,176 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 2.381e+02 4.087e+02 5.188e+02 6.525e+02 1.846e+03, threshold=1.038e+03, percent-clipped=3.0
2022-12-23 07:12:06,980 INFO [train.py:894] (2/4) Epoch 15, batch 50, loss[loss=0.1888, simple_loss=0.2721, pruned_loss=0.05274, over 18692.00 frames. ], tot_loss[loss=0.1918, simple_loss=0.2767, pruned_loss=0.05348, over 837886.21 frames. ], batch size: 50, lr: 7.83e-03, grad_scale: 16.0
2022-12-23 07:12:51,569 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.8211, 1.7098, 1.5914, 1.4733, 1.8889, 2.0387, 2.1269, 1.4969],
       device='cuda:2'), covar=tensor([0.0319, 0.0297, 0.0459, 0.0245, 0.0195, 0.0380, 0.0259, 0.0296],
       device='cuda:2'), in_proj_covar=tensor([0.0089, 0.0124, 0.0151, 0.0125, 0.0113, 0.0114, 0.0096, 0.0124],
       device='cuda:2'), out_proj_covar=tensor([7.2953e-05, 1.0012e-04, 1.2820e-04, 1.0149e-04, 9.3995e-05, 8.9491e-05,
        7.6752e-05, 9.9520e-05], device='cuda:2')
2022-12-23 07:13:09,578 INFO [zipformer.py:660] (2/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=49177.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 07:13:24,206 INFO [train.py:894] (2/4) Epoch 15, batch 100, loss[loss=0.2029, simple_loss=0.2899, pruned_loss=0.05796, over 18511.00 frames. ], tot_loss[loss=0.1882, simple_loss=0.2731, pruned_loss=0.05162, over 1475352.30 frames. ], batch size: 58, lr: 7.83e-03, grad_scale: 16.0
2022-12-23 07:13:35,584 INFO [zipformer.py:660] (2/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=49194.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 07:14:30,647 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 1.987e+02 3.186e+02 4.193e+02 4.956e+02 1.857e+03, threshold=8.385e+02, percent-clipped=1.0
2022-12-23 07:14:38,215 INFO [zipformer.py:660] (2/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=49236.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 07:14:39,180 INFO [train.py:894] (2/4) Epoch 15, batch 150, loss[loss=0.2009, simple_loss=0.2845, pruned_loss=0.05867, over 18727.00 frames. ], tot_loss[loss=0.1856, simple_loss=0.2701, pruned_loss=0.05057, over 1971221.58 frames. ], batch size: 54, lr: 7.83e-03, grad_scale: 16.0
2022-12-23 07:14:50,218 WARNING [train.py:1060] (2/4) Exclude cut with ID 7859-102521-0017-7548-0 from training. Duration: 24.525
2022-12-23 07:15:22,318 WARNING [train.py:1060] (2/4) Exclude cut with ID 3699-47246-0007-3408-0_sp0.9 from training. Duration: 20.26675
2022-12-23 07:15:37,462 WARNING [train.py:1060] (2/4) Exclude cut with ID 7859-102521-0017-7548-0_sp0.9 from training. Duration: 27.25
2022-12-23 07:15:49,520 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([2.1345, 1.5526, 2.3012, 4.4025, 3.2865, 2.8439, 0.8770, 3.1082],
       device='cuda:2'), covar=tensor([0.1785, 0.1777, 0.1541, 0.0372, 0.1000, 0.1115, 0.2484, 0.0898],
       device='cuda:2'), in_proj_covar=tensor([0.0099, 0.0115, 0.0128, 0.0135, 0.0103, 0.0133, 0.0127, 0.0109],
       device='cuda:2'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0004, 0.0003, 0.0003],
       device='cuda:2')
2022-12-23 07:15:55,577 INFO [train.py:894] (2/4) Epoch 15, batch 200, loss[loss=0.1916, simple_loss=0.2697, pruned_loss=0.05669, over 18403.00 frames. ], tot_loss[loss=0.185, simple_loss=0.27, pruned_loss=0.05004, over 2357849.04 frames. ], batch size: 48, lr: 7.82e-03, grad_scale: 16.0
2022-12-23 07:16:13,310 INFO [zipformer.py:660] (2/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=49297.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 07:16:48,460 WARNING [train.py:1060] (2/4) Exclude cut with ID 6426-64292-0017-15984-0 from training. Duration: 21.68
2022-12-23 07:17:01,138 WARNING [train.py:1060] (2/4) Exclude cut with ID 4278-13270-0007-59342-0 from training. Duration: 21.6300625
2022-12-23 07:17:02,600 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 2.256e+02 3.589e+02 4.467e+02 5.180e+02 1.013e+03, threshold=8.934e+02, percent-clipped=1.0
2022-12-23 07:17:11,972 INFO [train.py:894] (2/4) Epoch 15, batch 250, loss[loss=0.1705, simple_loss=0.2458, pruned_loss=0.04762, over 18585.00 frames. ], tot_loss[loss=0.1841, simple_loss=0.269, pruned_loss=0.0496, over 2658075.76 frames. ], batch size: 41, lr: 7.82e-03, grad_scale: 16.0
2022-12-23 07:17:26,515 WARNING [train.py:1060] (2/4) Exclude cut with ID 4278-13270-0007-59342-0_sp0.9 from training. Duration: 24.033375
2022-12-23 07:17:52,497 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([2.1005, 1.8634, 2.2087, 1.2742, 2.3253, 2.3541, 1.6662, 2.6472],
       device='cuda:2'), covar=tensor([0.1007, 0.1684, 0.1189, 0.1864, 0.0640, 0.0935, 0.2090, 0.0491],
       device='cuda:2'), in_proj_covar=tensor([0.0197, 0.0200, 0.0199, 0.0190, 0.0172, 0.0207, 0.0205, 0.0183],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-23 07:18:15,688 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.5253, 1.4213, 1.4903, 1.4580, 1.0267, 3.0493, 1.2941, 1.8305],
       device='cuda:2'), covar=tensor([0.3279, 0.2172, 0.1982, 0.2121, 0.1465, 0.0216, 0.1719, 0.0853],
       device='cuda:2'), in_proj_covar=tensor([0.0135, 0.0115, 0.0126, 0.0120, 0.0103, 0.0098, 0.0093, 0.0089],
       device='cuda:2'), out_proj_covar=tensor([0.0005, 0.0005, 0.0005, 0.0005, 0.0004, 0.0003, 0.0004, 0.0004],
       device='cuda:2')
2022-12-23 07:18:22,715 WARNING [train.py:1060] (2/4) Exclude cut with ID 4278-13270-0009-59344-0 from training. Duration: 22.905
2022-12-23 07:18:24,108 WARNING [train.py:1060] (2/4) Exclude cut with ID 5622-44585-0006-90525-0_sp1.1 from training. Duration: 23.4318125
2022-12-23 07:18:26,183 INFO [train.py:894] (2/4) Epoch 15, batch 300, loss[loss=0.1871, simple_loss=0.2776, pruned_loss=0.04828, over 18628.00 frames. ], tot_loss[loss=0.1832, simple_loss=0.2684, pruned_loss=0.04899, over 2892113.94 frames. ], batch size: 53, lr: 7.82e-03, grad_scale: 16.0
2022-12-23 07:18:48,917 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([2.5340, 2.2508, 1.6630, 0.7824, 1.6783, 2.1985, 1.7455, 1.9311],
       device='cuda:2'), covar=tensor([0.0603, 0.0498, 0.1173, 0.1543, 0.1193, 0.1303, 0.1501, 0.0750],
       device='cuda:2'), in_proj_covar=tensor([0.0168, 0.0184, 0.0203, 0.0194, 0.0210, 0.0197, 0.0210, 0.0196],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-23 07:19:19,074 INFO [zipformer.py:660] (2/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=49422.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 07:19:27,204 INFO [scaling.py:679] (2/4) Whitening: num_groups=1, num_channels=384, metric=2.62 vs. limit=5.0
2022-12-23 07:19:32,365 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 2.246e+02 3.698e+02 4.337e+02 5.424e+02 1.306e+03, threshold=8.675e+02, percent-clipped=3.0
2022-12-23 07:19:34,175 INFO [zipformer.py:660] (2/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=49432.0, num_to_drop=1, layers_to_drop={0}
2022-12-23 07:19:41,985 INFO [train.py:894] (2/4) Epoch 15, batch 350, loss[loss=0.183, simple_loss=0.2714, pruned_loss=0.04728, over 18525.00 frames. ], tot_loss[loss=0.1845, simple_loss=0.27, pruned_loss=0.0495, over 3074094.99 frames. ], batch size: 58, lr: 7.81e-03, grad_scale: 16.0
2022-12-23 07:20:20,395 WARNING [train.py:1060] (2/4) Exclude cut with ID 4278-13270-0009-59344-0_sp1.1 from training. Duration: 20.82275
2022-12-23 07:20:20,442 WARNING [train.py:1060] (2/4) Exclude cut with ID 4278-13270-0009-59344-0_sp0.9 from training. Duration: 25.45
2022-12-23 07:20:41,201 INFO [zipformer.py:660] (2/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=49477.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 07:20:50,844 INFO [zipformer.py:660] (2/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=49483.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 07:20:56,137 INFO [train.py:894] (2/4) Epoch 15, batch 400, loss[loss=0.1993, simple_loss=0.2836, pruned_loss=0.05747, over 18389.00 frames. ], tot_loss[loss=0.1857, simple_loss=0.2711, pruned_loss=0.05017, over 3215709.99 frames. ], batch size: 53, lr: 7.81e-03, grad_scale: 16.0
2022-12-23 07:21:07,194 INFO [zipformer.py:660] (2/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=49493.0, num_to_drop=1, layers_to_drop={0}
2022-12-23 07:21:08,289 INFO [zipformer.py:660] (2/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=49494.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 07:21:16,992 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.6651, 1.6149, 1.9251, 0.9839, 1.9781, 1.9123, 1.5190, 2.1995],
       device='cuda:2'), covar=tensor([0.1003, 0.1716, 0.1049, 0.1605, 0.0624, 0.0928, 0.2041, 0.0499],
       device='cuda:2'), in_proj_covar=tensor([0.0198, 0.0201, 0.0202, 0.0191, 0.0173, 0.0210, 0.0205, 0.0184],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-23 07:21:23,805 WARNING [train.py:1060] (2/4) Exclude cut with ID 5622-44585-0006-90525-0 from training. Duration: 25.775
2022-12-23 07:21:46,663 WARNING [train.py:1060] (2/4) Exclude cut with ID 2929-85685-0071-62375-0_sp0.9 from training. Duration: 22.25
2022-12-23 07:21:52,753 INFO [zipformer.py:660] (2/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=49525.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 07:22:01,187 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 2.732e+02 3.654e+02 4.284e+02 4.993e+02 1.074e+03, threshold=8.568e+02, percent-clipped=3.0
2022-12-23 07:22:10,269 INFO [train.py:894] (2/4) Epoch 15, batch 450, loss[loss=0.1736, simple_loss=0.2575, pruned_loss=0.04488, over 18554.00 frames. ], tot_loss[loss=0.1868, simple_loss=0.2723, pruned_loss=0.05068, over 3326250.41 frames. ], batch size: 45, lr: 7.80e-03, grad_scale: 16.0
2022-12-23 07:22:13,510 WARNING [train.py:1060] (2/4) Exclude cut with ID 3972-170212-0014-23379-0 from training. Duration: 26.205
2022-12-23 07:22:18,503 INFO [zipformer.py:660] (2/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=49542.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 07:22:28,614 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([2.7432, 4.1187, 3.9166, 1.5677, 4.1586, 3.1099, 0.7477, 2.9146],
       device='cuda:2'), covar=tensor([0.1997, 0.0937, 0.1390, 0.3918, 0.0758, 0.0878, 0.5243, 0.1367],
       device='cuda:2'), in_proj_covar=tensor([0.0139, 0.0130, 0.0151, 0.0122, 0.0132, 0.0107, 0.0141, 0.0109],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0002, 0.0003, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-23 07:22:29,919 WARNING [train.py:1060] (2/4) Exclude cut with ID 5239-32139-0047-9341-0_sp0.9 from training. Duration: 30.1555625
2022-12-23 07:22:35,800 WARNING [train.py:1060] (2/4) Exclude cut with ID 1265-135635-0050-6781-0_sp0.9 from training. Duration: 21.8333125
2022-12-23 07:22:39,322 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.4332, 1.6280, 1.3729, 2.0800, 2.1229, 1.5514, 1.2169, 1.3020],
       device='cuda:2'), covar=tensor([0.1857, 0.1812, 0.1526, 0.0992, 0.1202, 0.1119, 0.2052, 0.1493],
       device='cuda:2'), in_proj_covar=tensor([0.0239, 0.0216, 0.0205, 0.0191, 0.0254, 0.0190, 0.0214, 0.0192],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-23 07:22:45,958 WARNING [train.py:1060] (2/4) Exclude cut with ID 1914-133440-0024-94914-0_sp1.1 from training. Duration: 20.6545625
2022-12-23 07:22:46,406 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.4520, 2.4806, 2.7023, 1.1964, 2.4150, 2.9584, 2.3285, 2.3952],
       device='cuda:2'), covar=tensor([0.0724, 0.0335, 0.0361, 0.0468, 0.0327, 0.0368, 0.0320, 0.0579],
       device='cuda:2'), in_proj_covar=tensor([0.0147, 0.0163, 0.0118, 0.0136, 0.0146, 0.0138, 0.0152, 0.0158],
       device='cuda:2'), out_proj_covar=tensor([1.1543e-04, 1.2955e-04, 9.1755e-05, 1.0507e-04, 1.1422e-04, 1.1011e-04,
        1.2197e-04, 1.2476e-04], device='cuda:2')
2022-12-23 07:23:10,103 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([2.6709, 2.1428, 1.6130, 2.4426, 1.9876, 2.0915, 2.0493, 2.5586],
       device='cuda:2'), covar=tensor([0.1864, 0.3106, 0.1772, 0.2747, 0.3444, 0.0961, 0.2891, 0.0869],
       device='cuda:2'), in_proj_covar=tensor([0.0285, 0.0281, 0.0237, 0.0349, 0.0261, 0.0217, 0.0276, 0.0202],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0003, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-23 07:23:26,572 INFO [train.py:894] (2/4) Epoch 15, batch 500, loss[loss=0.1892, simple_loss=0.2807, pruned_loss=0.04884, over 18388.00 frames. ], tot_loss[loss=0.1892, simple_loss=0.2744, pruned_loss=0.05199, over 3412122.36 frames. ], batch size: 53, lr: 7.80e-03, grad_scale: 16.0
2022-12-23 07:23:26,616 WARNING [train.py:1060] (2/4) Exclude cut with ID 7395-89880-0045-39920-0_sp0.9 from training. Duration: 20.52225
2022-12-23 07:23:34,401 INFO [zipformer.py:660] (2/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=49592.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 07:23:40,580 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([2.3413, 1.6587, 0.8746, 1.7174, 2.5954, 1.6056, 2.4963, 2.2841],
       device='cuda:2'), covar=tensor([0.1532, 0.2008, 0.2462, 0.1564, 0.1544, 0.1657, 0.1314, 0.1586],
       device='cuda:2'), in_proj_covar=tensor([0.0093, 0.0097, 0.0117, 0.0095, 0.0112, 0.0091, 0.0097, 0.0093],
       device='cuda:2'), out_proj_covar=tensor([0.0003, 0.0003, 0.0004, 0.0003, 0.0004, 0.0003, 0.0003, 0.0003],
       device='cuda:2')
2022-12-23 07:23:44,706 WARNING [train.py:1060] (2/4) Exclude cut with ID 3972-170212-0014-23379-0_sp0.9 from training. Duration: 29.1166875
2022-12-23 07:24:02,209 INFO [scaling.py:679] (2/4) Whitening: num_groups=8, num_channels=96, metric=1.35 vs. limit=2.0
2022-12-23 07:24:32,845 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 2.334e+02 4.035e+02 5.056e+02 6.562e+02 1.635e+03, threshold=1.011e+03, percent-clipped=6.0
2022-12-23 07:24:43,080 INFO [train.py:894] (2/4) Epoch 15, batch 550, loss[loss=0.1965, simple_loss=0.2896, pruned_loss=0.05168, over 18701.00 frames. ], tot_loss[loss=0.1884, simple_loss=0.2739, pruned_loss=0.05146, over 3478590.77 frames. ], batch size: 69, lr: 7.80e-03, grad_scale: 16.0
2022-12-23 07:24:47,890 WARNING [train.py:1060] (2/4) Exclude cut with ID 543-133211-0007-59831-0_sp0.9 from training. Duration: 21.388875
2022-12-23 07:24:50,925 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.9966, 1.9783, 1.4192, 2.2822, 2.1406, 2.0077, 2.7905, 2.0550],
       device='cuda:2'), covar=tensor([0.0843, 0.1588, 0.2739, 0.1706, 0.1722, 0.0828, 0.0873, 0.1189],
       device='cuda:2'), in_proj_covar=tensor([0.0175, 0.0201, 0.0244, 0.0286, 0.0230, 0.0186, 0.0208, 0.0201],
       device='cuda:2'), out_proj_covar=tensor([0.0001, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-23 07:25:08,574 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.8989, 2.0021, 1.1833, 2.5796, 2.0146, 1.7519, 3.0476, 1.9463],
       device='cuda:2'), covar=tensor([0.1040, 0.1727, 0.3232, 0.1950, 0.1922, 0.1138, 0.0941, 0.1395],
       device='cuda:2'), in_proj_covar=tensor([0.0175, 0.0201, 0.0244, 0.0286, 0.0230, 0.0186, 0.0208, 0.0201],
       device='cuda:2'), out_proj_covar=tensor([0.0001, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-23 07:25:21,222 WARNING [train.py:1060] (2/4) Exclude cut with ID 1914-133440-0024-94914-0 from training. Duration: 22.72
2022-12-23 07:25:21,260 WARNING [train.py:1060] (2/4) Exclude cut with ID 1914-133440-0031-94921-0_sp0.9 from training. Duration: 22.7444375
2022-12-23 07:25:34,394 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([3.2705, 2.5590, 1.9829, 3.0697, 2.4161, 2.5212, 2.5000, 3.5196],
       device='cuda:2'), covar=tensor([0.1634, 0.2840, 0.1635, 0.2486, 0.3342, 0.0859, 0.2891, 0.0610],
       device='cuda:2'), in_proj_covar=tensor([0.0283, 0.0279, 0.0236, 0.0347, 0.0259, 0.0216, 0.0273, 0.0201],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0003, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-23 07:25:57,313 INFO [train.py:894] (2/4) Epoch 15, batch 600, loss[loss=0.1803, simple_loss=0.275, pruned_loss=0.04281, over 18581.00 frames. ], tot_loss[loss=0.1902, simple_loss=0.2757, pruned_loss=0.05231, over 3531734.30 frames. ], batch size: 56, lr: 7.79e-03, grad_scale: 16.0
2022-12-23 07:26:04,574 WARNING [train.py:1060] (2/4) Exclude cut with ID 4133-6541-0027-40495-0_sp1.1 from training. Duration: 0.9681875
2022-12-23 07:26:07,451 WARNING [train.py:1060] (2/4) Exclude cut with ID 6330-62851-0022-91297-0_sp0.9 from training. Duration: 22.3166875
2022-12-23 07:26:12,171 WARNING [train.py:1060] (2/4) Exclude cut with ID 543-133212-0015-59917-0_sp0.9 from training. Duration: 21.8166875
2022-12-23 07:26:18,925 INFO [scaling.py:679] (2/4) Whitening: num_groups=1, num_channels=384, metric=2.62 vs. limit=5.0
2022-12-23 07:27:03,252 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 2.339e+02 3.610e+02 4.307e+02 5.287e+02 1.018e+03, threshold=8.614e+02, percent-clipped=1.0
2022-12-23 07:27:13,047 INFO [train.py:894] (2/4) Epoch 15, batch 650, loss[loss=0.178, simple_loss=0.2625, pruned_loss=0.04675, over 18705.00 frames. ], tot_loss[loss=0.1906, simple_loss=0.2758, pruned_loss=0.05267, over 3572848.86 frames. ], batch size: 46, lr: 7.79e-03, grad_scale: 16.0
2022-12-23 07:27:56,400 INFO [zipformer.py:660] (2/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=49766.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 07:27:57,467 WARNING [train.py:1060] (2/4) Exclude cut with ID 4957-30119-0041-23990-0_sp0.9 from training. Duration: 20.22775
2022-12-23 07:28:14,608 INFO [zipformer.py:660] (2/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=49778.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 07:28:28,631 INFO [train.py:894] (2/4) Epoch 15, batch 700, loss[loss=0.1632, simple_loss=0.2473, pruned_loss=0.03957, over 18668.00 frames. ], tot_loss[loss=0.1904, simple_loss=0.2761, pruned_loss=0.05233, over 3603813.76 frames. ], batch size: 48, lr: 7.78e-03, grad_scale: 16.0
2022-12-23 07:28:30,303 INFO [zipformer.py:660] (2/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=49788.0, num_to_drop=1, layers_to_drop={3}
2022-12-23 07:28:32,385 INFO [scaling.py:679] (2/4) Whitening: num_groups=8, num_channels=192, metric=1.98 vs. limit=2.0
2022-12-23 07:28:41,852 WARNING [train.py:1060] (2/4) Exclude cut with ID 5239-32139-0047-9341-0_sp1.1 from training. Duration: 24.67275
2022-12-23 07:29:08,690 WARNING [train.py:1060] (2/4) Exclude cut with ID 3082-165428-0081-50734-0_sp0.9 from training. Duration: 21.8055625
2022-12-23 07:29:28,883 INFO [zipformer.py:660] (2/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=49827.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 07:29:34,936 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 2.205e+02 3.616e+02 4.297e+02 5.395e+02 9.947e+02, threshold=8.594e+02, percent-clipped=4.0
2022-12-23 07:29:44,198 INFO [train.py:894] (2/4) Epoch 15, batch 750, loss[loss=0.179, simple_loss=0.2679, pruned_loss=0.04501, over 18362.00 frames. ], tot_loss[loss=0.1891, simple_loss=0.2744, pruned_loss=0.05187, over 3626684.60 frames. ], batch size: 46, lr: 7.78e-03, grad_scale: 16.0
2022-12-23 07:29:49,037 WARNING [train.py:1060] (2/4) Exclude cut with ID 3340-169293-0054-76830-0_sp0.9 from training. Duration: 22.6666875
2022-12-23 07:30:49,920 WARNING [train.py:1060] (2/4) Exclude cut with ID 2411-132532-0017-82279-0_sp1.1 from training. Duration: 0.9681875
2022-12-23 07:30:58,945 INFO [train.py:894] (2/4) Epoch 15, batch 800, loss[loss=0.1531, simple_loss=0.2358, pruned_loss=0.03519, over 18620.00 frames. ], tot_loss[loss=0.1887, simple_loss=0.2743, pruned_loss=0.05154, over 3646522.12 frames. ], batch size: 45, lr: 7.78e-03, grad_scale: 8.0
2022-12-23 07:31:06,278 INFO [zipformer.py:660] (2/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=49892.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 07:31:16,085 WARNING [train.py:1060] (2/4) Exclude cut with ID 6330-62850-0007-91323-0 from training. Duration: 22.485
2022-12-23 07:31:51,867 WARNING [train.py:1060] (2/4) Exclude cut with ID 3972-170212-0014-23379-0_sp1.1 from training. Duration: 23.82275
2022-12-23 07:32:06,627 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 2.483e+02 3.966e+02 4.783e+02 5.574e+02 1.191e+03, threshold=9.565e+02, percent-clipped=3.0
2022-12-23 07:32:07,920 WARNING [train.py:1060] (2/4) Exclude cut with ID 4860-13185-0032-76709-0 from training. Duration: 20.77
2022-12-23 07:32:13,604 INFO [train.py:894] (2/4) Epoch 15, batch 850, loss[loss=0.175, simple_loss=0.2676, pruned_loss=0.04125, over 18629.00 frames. ], tot_loss[loss=0.1888, simple_loss=0.2747, pruned_loss=0.05141, over 3661294.53 frames. ], batch size: 53, lr: 7.77e-03, grad_scale: 8.0
2022-12-23 07:32:13,661 WARNING [train.py:1060] (2/4) Exclude cut with ID 6426-64292-0017-15984-0_sp0.9 from training. Duration: 24.088875
2022-12-23 07:32:18,207 INFO [zipformer.py:660] (2/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=49940.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 07:32:41,697 WARNING [train.py:1060] (2/4) Exclude cut with ID 6330-62850-0007-91323-0_sp1.1 from training. Duration: 20.4409375
2022-12-23 07:33:20,587 INFO [scaling.py:679] (2/4) Whitening: num_groups=8, num_channels=96, metric=1.29 vs. limit=2.0
2022-12-23 07:33:29,290 INFO [train.py:894] (2/4) Epoch 15, batch 900, loss[loss=0.2062, simple_loss=0.2945, pruned_loss=0.05894, over 18717.00 frames. ], tot_loss[loss=0.1882, simple_loss=0.2745, pruned_loss=0.05095, over 3673146.66 frames. ], batch size: 98, lr: 7.77e-03, grad_scale: 8.0
2022-12-23 07:33:39,264 INFO [scaling.py:679] (2/4) Whitening: num_groups=1, num_channels=384, metric=4.21 vs. limit=5.0
2022-12-23 07:34:01,021 WARNING [train.py:1060] (2/4) Exclude cut with ID 2929-85685-0062-62366-0_sp0.9 from training. Duration: 22.511125
2022-12-23 07:34:02,342 WARNING [train.py:1060] (2/4) Exclude cut with ID 7395-89880-0031-39906-0 from training. Duration: 20.675
2022-12-23 07:34:40,524 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 2.188e+02 3.776e+02 4.591e+02 6.047e+02 1.033e+03, threshold=9.181e+02, percent-clipped=2.0
2022-12-23 07:34:47,833 INFO [train.py:894] (2/4) Epoch 15, batch 950, loss[loss=0.2046, simple_loss=0.2996, pruned_loss=0.0548, over 18406.00 frames. ], tot_loss[loss=0.1888, simple_loss=0.2753, pruned_loss=0.05119, over 3682552.26 frames. ], batch size: 53, lr: 7.76e-03, grad_scale: 8.0
2022-12-23 07:35:38,971 WARNING [train.py:1060] (2/4) Exclude cut with ID 6330-62850-0007-91323-0_sp0.9 from training. Duration: 24.9833125
2022-12-23 07:35:51,319 INFO [zipformer.py:660] (2/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=50078.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 07:36:03,967 INFO [train.py:894] (2/4) Epoch 15, batch 1000, loss[loss=0.1828, simple_loss=0.2679, pruned_loss=0.04881, over 18456.00 frames. ], tot_loss[loss=0.189, simple_loss=0.2749, pruned_loss=0.05162, over 3688992.67 frames. ], batch size: 50, lr: 7.76e-03, grad_scale: 8.0
2022-12-23 07:36:05,737 INFO [zipformer.py:660] (2/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=50088.0, num_to_drop=1, layers_to_drop={2}
2022-12-23 07:36:10,015 WARNING [train.py:1060] (2/4) Exclude cut with ID 5239-32139-0047-9341-0 from training. Duration: 27.14
2022-12-23 07:36:24,699 WARNING [train.py:1060] (2/4) Exclude cut with ID 2929-85685-0044-62348-0 from training. Duration: 22.44
2022-12-23 07:36:57,537 INFO [zipformer.py:660] (2/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=50122.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 07:37:03,199 INFO [zipformer.py:660] (2/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=50126.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 07:37:06,259 INFO [zipformer.py:660] (2/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=50128.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 07:37:11,411 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 2.398e+02 3.589e+02 4.498e+02 6.076e+02 1.358e+03, threshold=8.996e+02, percent-clipped=1.0
2022-12-23 07:37:17,321 INFO [zipformer.py:660] (2/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=50136.0, num_to_drop=1, layers_to_drop={0}
2022-12-23 07:37:18,478 INFO [train.py:894] (2/4) Epoch 15, batch 1050, loss[loss=0.183, simple_loss=0.2748, pruned_loss=0.0456, over 18470.00 frames. ], tot_loss[loss=0.1888, simple_loss=0.2748, pruned_loss=0.05139, over 3694071.17 frames. ], batch size: 54, lr: 7.76e-03, grad_scale: 8.0
2022-12-23 07:37:46,425 WARNING [train.py:1060] (2/4) Exclude cut with ID 2929-85685-0060-62364-0_sp0.9 from training. Duration: 21.361125
2022-12-23 07:37:51,901 WARNING [train.py:1060] (2/4) Exclude cut with ID 2929-85685-0079-62383-0_sp1.1 from training. Duration: 27.0318125
2022-12-23 07:38:03,164 WARNING [train.py:1060] (2/4) Exclude cut with ID 5622-44585-0006-90525-0_sp0.9 from training. Duration: 28.638875
2022-12-23 07:38:05,175 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.3304, 1.3952, 1.2414, 1.7986, 1.6824, 1.3866, 0.9350, 1.1901],
       device='cuda:2'), covar=tensor([0.1876, 0.1899, 0.1600, 0.1015, 0.1206, 0.1134, 0.2158, 0.1475],
       device='cuda:2'), in_proj_covar=tensor([0.0240, 0.0216, 0.0206, 0.0192, 0.0256, 0.0191, 0.0216, 0.0192],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-23 07:38:19,566 WARNING [train.py:1060] (2/4) Exclude cut with ID 3340-169293-0054-76830-0 from training. Duration: 20.4
2022-12-23 07:38:32,489 INFO [train.py:894] (2/4) Epoch 15, batch 1100, loss[loss=0.1589, simple_loss=0.2408, pruned_loss=0.03855, over 18560.00 frames. ], tot_loss[loss=0.1885, simple_loss=0.2744, pruned_loss=0.05132, over 3698315.09 frames. ], batch size: 41, lr: 7.75e-03, grad_scale: 8.0
2022-12-23 07:38:35,615 INFO [zipformer.py:660] (2/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=50189.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 07:38:46,629 INFO [zipformer.py:660] (2/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=50197.0, num_to_drop=1, layers_to_drop={1}
2022-12-23 07:38:50,691 WARNING [train.py:1060] (2/4) Exclude cut with ID 2929-85685-0071-62375-0 from training. Duration: 20.025
2022-12-23 07:38:50,703 WARNING [train.py:1060] (2/4) Exclude cut with ID 2364-131735-0112-64612-0_sp0.9 from training. Duration: 20.488875
2022-12-23 07:38:57,831 WARNING [train.py:1060] (2/4) Exclude cut with ID 2929-85685-0079-62383-0 from training. Duration: 29.735
2022-12-23 07:39:39,224 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 2.273e+02 3.557e+02 4.399e+02 5.755e+02 2.138e+03, threshold=8.797e+02, percent-clipped=6.0
2022-12-23 07:39:43,136 INFO [scaling.py:679] (2/4) Whitening: num_groups=1, num_channels=384, metric=4.25 vs. limit=5.0
2022-12-23 07:39:46,188 INFO [train.py:894] (2/4) Epoch 15, batch 1150, loss[loss=0.199, simple_loss=0.2877, pruned_loss=0.05516, over 18696.00 frames. ], tot_loss[loss=0.1886, simple_loss=0.2745, pruned_loss=0.05131, over 3701282.44 frames. ], batch size: 60, lr: 7.75e-03, grad_scale: 8.0
2022-12-23 07:40:14,536 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([0.4203, 1.7900, 2.0997, 1.2426, 1.2768, 2.2448, 2.0318, 1.7368],
       device='cuda:2'), covar=tensor([0.0783, 0.0300, 0.0301, 0.0320, 0.0386, 0.0353, 0.0199, 0.0607],
       device='cuda:2'), in_proj_covar=tensor([0.0149, 0.0164, 0.0120, 0.0138, 0.0148, 0.0138, 0.0155, 0.0161],
       device='cuda:2'), out_proj_covar=tensor([1.1684e-04, 1.3027e-04, 9.3500e-05, 1.0638e-04, 1.1599e-04, 1.1031e-04,
        1.2413e-04, 1.2668e-04], device='cuda:2')
2022-12-23 07:40:17,416 INFO [zipformer.py:660] (2/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=50258.0, num_to_drop=1, layers_to_drop={3}
2022-12-23 07:40:18,503 WARNING [train.py:1060] (2/4) Exclude cut with ID 7276-92427-0014-12983-0_sp0.9 from training. Duration: 21.3055625
2022-12-23 07:40:19,933 WARNING [train.py:1060] (2/4) Exclude cut with ID 1025-75365-0008-79168-0_sp0.9 from training. Duration: 22.0666875
2022-12-23 07:41:01,622 INFO [train.py:894] (2/4) Epoch 15, batch 1200, loss[loss=0.2114, simple_loss=0.2934, pruned_loss=0.06473, over 18592.00 frames. ], tot_loss[loss=0.1883, simple_loss=0.2742, pruned_loss=0.05122, over 3703870.46 frames. ], batch size: 57, lr: 7.75e-03, grad_scale: 8.0
2022-12-23 07:41:58,484 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([2.1926, 1.6358, 2.4931, 4.3584, 3.1221, 2.7578, 1.3497, 3.1388],
       device='cuda:2'), covar=tensor([0.1846, 0.1828, 0.1564, 0.0631, 0.0936, 0.1162, 0.2099, 0.0962],
       device='cuda:2'), in_proj_covar=tensor([0.0102, 0.0116, 0.0132, 0.0138, 0.0103, 0.0135, 0.0129, 0.0110],
       device='cuda:2'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0004, 0.0003, 0.0004, 0.0003, 0.0003],
       device='cuda:2')
2022-12-23 07:42:09,331 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 2.658e+02 3.741e+02 4.483e+02 5.877e+02 1.567e+03, threshold=8.966e+02, percent-clipped=2.0
2022-12-23 07:42:09,366 WARNING [train.py:1060] (2/4) Exclude cut with ID 2929-85685-0062-62366-0 from training. Duration: 20.26
2022-12-23 07:42:16,726 INFO [train.py:894] (2/4) Epoch 15, batch 1250, loss[loss=0.1545, simple_loss=0.2342, pruned_loss=0.0374, over 18415.00 frames. ], tot_loss[loss=0.1871, simple_loss=0.2727, pruned_loss=0.05073, over 3706071.05 frames. ], batch size: 42, lr: 7.74e-03, grad_scale: 8.0
2022-12-23 07:42:23,663 WARNING [train.py:1060] (2/4) Exclude cut with ID 5239-32139-0030-9324-0_sp0.9 from training. Duration: 21.3444375
2022-12-23 07:43:18,910 WARNING [train.py:1060] (2/4) Exclude cut with ID 497-129325-0061-62254-0_sp1.1 from training. Duration: 0.97725
2022-12-23 07:43:31,603 INFO [train.py:894] (2/4) Epoch 15, batch 1300, loss[loss=0.2088, simple_loss=0.2944, pruned_loss=0.06166, over 18493.00 frames. ], tot_loss[loss=0.1867, simple_loss=0.2727, pruned_loss=0.05028, over 3707946.76 frames. ], batch size: 64, lr: 7.74e-03, grad_scale: 8.0
2022-12-23 07:44:02,811 WARNING [train.py:1060] (2/4) Exclude cut with ID 7395-89880-0031-39906-0_sp0.9 from training. Duration: 22.97225
2022-12-23 07:44:24,683 INFO [zipformer.py:660] (2/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=50422.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 07:44:33,457 WARNING [train.py:1060] (2/4) Exclude cut with ID 7395-89880-0047-39922-0_sp0.9 from training. Duration: 21.97775
2022-12-23 07:44:38,887 INFO [scaling.py:679] (2/4) Whitening: num_groups=8, num_channels=192, metric=2.03 vs. limit=2.0
2022-12-23 07:44:39,441 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 2.804e+02 4.009e+02 4.601e+02 5.649e+02 1.183e+03, threshold=9.201e+02, percent-clipped=3.0
2022-12-23 07:44:46,983 INFO [train.py:894] (2/4) Epoch 15, batch 1350, loss[loss=0.1965, simple_loss=0.2783, pruned_loss=0.05733, over 18589.00 frames. ], tot_loss[loss=0.1861, simple_loss=0.2721, pruned_loss=0.0501, over 3708827.47 frames. ], batch size: 51, lr: 7.73e-03, grad_scale: 8.0
2022-12-23 07:44:47,001 WARNING [train.py:1060] (2/4) Exclude cut with ID 1112-1043-0006-89194-0_sp0.9 from training. Duration: 21.8333125
2022-12-23 07:44:57,165 WARNING [train.py:1060] (2/4) Exclude cut with ID 1914-133440-0031-94921-0 from training. Duration: 20.47
2022-12-23 07:45:36,654 INFO [zipformer.py:660] (2/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=50470.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 07:45:48,115 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([2.4959, 4.0668, 3.8776, 1.6104, 4.1366, 2.9052, 0.6241, 2.6423],
       device='cuda:2'), covar=tensor([0.2093, 0.0965, 0.1153, 0.3617, 0.0686, 0.1025, 0.5076, 0.1524],
       device='cuda:2'), in_proj_covar=tensor([0.0136, 0.0126, 0.0148, 0.0119, 0.0130, 0.0106, 0.0139, 0.0108],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-23 07:45:56,904 INFO [zipformer.py:660] (2/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=50484.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 07:46:00,938 INFO [train.py:894] (2/4) Epoch 15, batch 1400, loss[loss=0.2382, simple_loss=0.3091, pruned_loss=0.08363, over 18526.00 frames. ], tot_loss[loss=0.1869, simple_loss=0.273, pruned_loss=0.0504, over 3710283.39 frames. ], batch size: 55, lr: 7.73e-03, grad_scale: 8.0
2022-12-23 07:46:02,469 WARNING [train.py:1060] (2/4) Exclude cut with ID 7395-89880-0037-39912-0_sp0.9 from training. Duration: 20.67225
2022-12-23 07:46:20,661 WARNING [train.py:1060] (2/4) Exclude cut with ID 1914-133440-0024-94914-0_sp0.9 from training. Duration: 25.2444375
2022-12-23 07:46:27,556 INFO [scaling.py:679] (2/4) Whitening: num_groups=1, num_channels=384, metric=3.80 vs. limit=5.0
2022-12-23 07:46:30,145 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([2.2233, 1.8850, 1.5110, 2.0921, 1.8342, 1.8578, 1.8036, 2.2012],
       device='cuda:2'), covar=tensor([0.2416, 0.3627, 0.2350, 0.2852, 0.3553, 0.1338, 0.3533, 0.0983],
       device='cuda:2'), in_proj_covar=tensor([0.0286, 0.0281, 0.0237, 0.0349, 0.0261, 0.0219, 0.0278, 0.0203],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0003, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-23 07:46:42,450 WARNING [train.py:1060] (2/4) Exclude cut with ID 3340-169293-0021-76797-0_sp0.9 from training. Duration: 21.1445
2022-12-23 07:47:08,330 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 2.114e+02 3.582e+02 4.419e+02 5.521e+02 1.088e+03, threshold=8.838e+02, percent-clipped=3.0
2022-12-23 07:47:08,713 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.5954, 1.5567, 1.6582, 1.6113, 1.0596, 3.3678, 1.5165, 1.9191],
       device='cuda:2'), covar=tensor([0.3076, 0.2037, 0.1949, 0.1970, 0.1501, 0.0188, 0.1635, 0.0886],
       device='cuda:2'), in_proj_covar=tensor([0.0135, 0.0118, 0.0128, 0.0121, 0.0104, 0.0098, 0.0094, 0.0091],
       device='cuda:2'), out_proj_covar=tensor([0.0005, 0.0005, 0.0005, 0.0005, 0.0004, 0.0003, 0.0004, 0.0004],
       device='cuda:2')
2022-12-23 07:47:15,709 INFO [train.py:894] (2/4) Epoch 15, batch 1450, loss[loss=0.1686, simple_loss=0.2589, pruned_loss=0.03916, over 18702.00 frames. ], tot_loss[loss=0.1873, simple_loss=0.2735, pruned_loss=0.05051, over 3712836.29 frames. ], batch size: 46, lr: 7.73e-03, grad_scale: 8.0
2022-12-23 07:47:29,672 INFO [zipformer.py:660] (2/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=50546.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 07:47:40,094 INFO [zipformer.py:660] (2/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=50553.0, num_to_drop=1, layers_to_drop={3}
2022-12-23 07:47:57,217 WARNING [train.py:1060] (2/4) Exclude cut with ID 2929-85685-0079-62383-0_sp0.9 from training. Duration: 33.038875
2022-12-23 07:48:31,373 INFO [train.py:894] (2/4) Epoch 15, batch 1500, loss[loss=0.1772, simple_loss=0.2605, pruned_loss=0.04697, over 18668.00 frames. ], tot_loss[loss=0.187, simple_loss=0.2732, pruned_loss=0.0504, over 3712951.27 frames. ], batch size: 46, lr: 7.72e-03, grad_scale: 8.0
2022-12-23 07:48:31,908 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([2.0564, 2.0264, 1.5164, 2.3971, 2.3251, 1.9747, 2.9109, 2.1405],
       device='cuda:2'), covar=tensor([0.0836, 0.1673, 0.2734, 0.1643, 0.1595, 0.0835, 0.0804, 0.1140],
       device='cuda:2'), in_proj_covar=tensor([0.0174, 0.0199, 0.0242, 0.0284, 0.0230, 0.0185, 0.0206, 0.0200],
       device='cuda:2'), out_proj_covar=tensor([0.0001, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-23 07:48:34,373 WARNING [train.py:1060] (2/4) Exclude cut with ID 6426-64291-0000-16059-0_sp0.9 from training. Duration: 20.0944375
2022-12-23 07:48:51,398 WARNING [train.py:1060] (2/4) Exclude cut with ID 2929-85685-0044-62348-0_sp1.1 from training. Duration: 20.4
2022-12-23 07:48:58,503 WARNING [train.py:1060] (2/4) Exclude cut with ID 6330-62851-0022-91297-0 from training. Duration: 20.085
2022-12-23 07:49:02,153 INFO [zipformer.py:660] (2/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=50607.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 07:49:11,604 WARNING [train.py:1060] (2/4) Exclude cut with ID 4860-13185-0032-76709-0_sp0.9 from training. Duration: 23.07775
2022-12-23 07:49:39,105 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([2.0313, 2.1629, 1.4997, 2.3829, 2.1414, 1.9698, 2.8200, 2.1528],
       device='cuda:2'), covar=tensor([0.0876, 0.1607, 0.2669, 0.1777, 0.1759, 0.0879, 0.0907, 0.1208],
       device='cuda:2'), in_proj_covar=tensor([0.0174, 0.0199, 0.0241, 0.0284, 0.0229, 0.0185, 0.0205, 0.0200],
       device='cuda:2'), out_proj_covar=tensor([0.0001, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-23 07:49:40,184 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 1.951e+02 3.779e+02 4.506e+02 5.486e+02 9.567e+02, threshold=9.012e+02, percent-clipped=3.0
2022-12-23 07:49:47,217 INFO [train.py:894] (2/4) Epoch 15, batch 1550, loss[loss=0.204, simple_loss=0.2869, pruned_loss=0.06059, over 18550.00 frames. ], tot_loss[loss=0.1867, simple_loss=0.273, pruned_loss=0.05023, over 3713126.89 frames. ], batch size: 171, lr: 7.72e-03, grad_scale: 8.0
2022-12-23 07:49:56,272 WARNING [train.py:1060] (2/4) Exclude cut with ID 2929-85685-0044-62348-0_sp0.9 from training. Duration: 24.9333125
2022-12-23 07:50:15,034 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.7779, 2.1962, 1.6350, 2.3341, 2.6593, 1.7093, 1.7874, 1.4985],
       device='cuda:2'), covar=tensor([0.1668, 0.1414, 0.1438, 0.0876, 0.1181, 0.1032, 0.1885, 0.1364],
       device='cuda:2'), in_proj_covar=tensor([0.0239, 0.0215, 0.0205, 0.0191, 0.0253, 0.0190, 0.0213, 0.0192],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-23 07:50:38,025 INFO [scaling.py:679] (2/4) Whitening: num_groups=8, num_channels=96, metric=1.48 vs. limit=2.0
2022-12-23 07:50:41,851 WARNING [train.py:1060] (2/4) Exclude cut with ID 5118-111612-0016-124680-0_sp0.9 from training. Duration: 20.388875
2022-12-23 07:50:49,342 WARNING [train.py:1060] (2/4) Exclude cut with ID 432-122774-0017-62487-0_sp1.1 from training. Duration: 20.3590625
2022-12-23 07:51:02,344 INFO [train.py:894] (2/4) Epoch 15, batch 1600, loss[loss=0.1948, simple_loss=0.2727, pruned_loss=0.05846, over 18524.00 frames. ], tot_loss[loss=0.1862, simple_loss=0.2724, pruned_loss=0.05005, over 3713282.79 frames. ], batch size: 47, lr: 7.72e-03, grad_scale: 8.0
2022-12-23 07:51:13,213 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([2.4558, 2.1648, 1.7861, 1.2063, 2.7352, 2.5250, 2.1641, 1.7327],
       device='cuda:2'), covar=tensor([0.0390, 0.0405, 0.0504, 0.0764, 0.0222, 0.0332, 0.0444, 0.0801],
       device='cuda:2'), in_proj_covar=tensor([0.0122, 0.0122, 0.0128, 0.0120, 0.0091, 0.0120, 0.0134, 0.0153],
       device='cuda:2'), out_proj_covar=tensor([0.0001, 0.0001, 0.0002, 0.0001, 0.0001, 0.0001, 0.0002, 0.0002],
       device='cuda:2')
2022-12-23 07:51:19,450 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.9308, 1.2312, 0.6795, 1.2752, 2.2295, 1.2346, 1.7694, 1.9761],
       device='cuda:2'), covar=tensor([0.1493, 0.2101, 0.2439, 0.1573, 0.1522, 0.1684, 0.1390, 0.1511],
       device='cuda:2'), in_proj_covar=tensor([0.0093, 0.0098, 0.0118, 0.0096, 0.0113, 0.0091, 0.0099, 0.0095],
       device='cuda:2'), out_proj_covar=tensor([0.0003, 0.0003, 0.0004, 0.0003, 0.0004, 0.0003, 0.0003, 0.0003],
       device='cuda:2')
2022-12-23 07:51:56,706 WARNING [train.py:1060] (2/4) Exclude cut with ID 3557-8342-0013-54691-0_sp1.1 from training. Duration: 0.836375
2022-12-23 07:52:09,725 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 2.709e+02 4.126e+02 5.150e+02 6.276e+02 1.539e+03, threshold=1.030e+03, percent-clipped=5.0
2022-12-23 07:52:17,180 INFO [train.py:894] (2/4) Epoch 15, batch 1650, loss[loss=0.2415, simple_loss=0.3231, pruned_loss=0.07998, over 18518.00 frames. ], tot_loss[loss=0.1884, simple_loss=0.274, pruned_loss=0.05141, over 3712211.73 frames. ], batch size: 58, lr: 7.71e-03, grad_scale: 8.0
2022-12-23 07:52:37,951 WARNING [train.py:1060] (2/4) Exclude cut with ID 8565-290391-0049-67394-0_sp0.9 from training. Duration: 21.3166875
2022-12-23 07:53:09,208 WARNING [train.py:1060] (2/4) Exclude cut with ID 6533-399-0029-104863-0_sp0.9 from training. Duration: 22.1055625
2022-12-23 07:53:20,908 WARNING [train.py:1060] (2/4) Exclude cut with ID 7699-105389-0094-26379-0_sp1.1 from training. Duration: 21.77725
2022-12-23 07:53:24,701 INFO [zipformer.py:660] (2/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=50782.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 07:53:27,462 INFO [zipformer.py:660] (2/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=50784.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 07:53:31,803 INFO [train.py:894] (2/4) Epoch 15, batch 1700, loss[loss=0.1652, simple_loss=0.246, pruned_loss=0.04219, over 18598.00 frames. ], tot_loss[loss=0.1916, simple_loss=0.2762, pruned_loss=0.05351, over 3712702.91 frames. ], batch size: 41, lr: 7.71e-03, grad_scale: 8.0
2022-12-23 07:53:38,140 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([2.4655, 1.9173, 2.0294, 2.2216, 2.2678, 2.2518, 2.3306, 1.7779],
       device='cuda:2'), covar=tensor([0.1716, 0.2732, 0.2068, 0.2363, 0.1632, 0.0820, 0.2457, 0.1050],
       device='cuda:2'), in_proj_covar=tensor([0.0258, 0.0291, 0.0267, 0.0303, 0.0290, 0.0242, 0.0323, 0.0229],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-23 07:53:40,241 WARNING [train.py:1060] (2/4) Exclude cut with ID 7255-291500-0005-134304-0_sp0.9 from training. Duration: 27.8166875
2022-12-23 07:54:04,750 WARNING [train.py:1060] (2/4) Exclude cut with ID 7357-94126-0021-15852-0_sp1.1 from training. Duration: 22.5090625
2022-12-23 07:54:05,091 INFO [zipformer.py:660] (2/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=50808.0, num_to_drop=1, layers_to_drop={1}
2022-12-23 07:54:10,680 WARNING [train.py:1060] (2/4) Exclude cut with ID 7255-291500-0005-134304-0 from training. Duration: 25.035
2022-12-23 07:54:27,976 WARNING [train.py:1060] (2/4) Exclude cut with ID 774-127930-0014-10412-0_sp1.1 from training. Duration: 0.95
2022-12-23 07:54:39,921 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 2.931e+02 4.643e+02 5.788e+02 7.435e+02 4.288e+03, threshold=1.158e+03, percent-clipped=5.0
2022-12-23 07:54:40,098 INFO [zipformer.py:660] (2/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=50832.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 07:54:45,941 WARNING [train.py:1060] (2/4) Exclude cut with ID 3033-130750-0096-55598-0_sp0.9 from training. Duration: 0.92225
2022-12-23 07:54:47,262 INFO [train.py:894] (2/4) Epoch 15, batch 1750, loss[loss=0.187, simple_loss=0.2772, pruned_loss=0.04838, over 18472.00 frames. ], tot_loss[loss=0.1955, simple_loss=0.2785, pruned_loss=0.0562, over 3714024.40 frames. ], batch size: 54, lr: 7.70e-03, grad_scale: 8.0
2022-12-23 07:54:57,619 INFO [zipformer.py:660] (2/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=50843.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 07:55:12,328 WARNING [train.py:1060] (2/4) Exclude cut with ID 4511-76322-0006-80011-0 from training. Duration: 21.97
2022-12-23 07:55:13,260 INFO [zipformer.py:660] (2/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=50853.0, num_to_drop=1, layers_to_drop={1}
2022-12-23 07:55:32,001 WARNING [train.py:1060] (2/4) Exclude cut with ID 7492-105653-0055-62765-0_sp0.9 from training. Duration: 21.97225
2022-12-23 07:55:32,039 WARNING [train.py:1060] (2/4) Exclude cut with ID 453-131332-0000-47844-0_sp0.9 from training. Duration: 25.3333125
2022-12-23 07:55:36,759 INFO [zipformer.py:660] (2/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=50869.0, num_to_drop=1, layers_to_drop={1}
2022-12-23 07:55:42,140 WARNING [train.py:1060] (2/4) Exclude cut with ID 5172-29468-0015-19128-0_sp0.9 from training. Duration: 21.5055625
2022-12-23 07:55:52,927 WARNING [train.py:1060] (2/4) Exclude cut with ID 453-131332-0000-47844-0_sp1.1 from training. Duration: 20.72725
2022-12-23 07:56:03,524 INFO [train.py:894] (2/4) Epoch 15, batch 1800, loss[loss=0.1605, simple_loss=0.2339, pruned_loss=0.04352, over 18462.00 frames. ], tot_loss[loss=0.1975, simple_loss=0.2796, pruned_loss=0.05766, over 3713766.11 frames. ], batch size: 43, lr: 7.70e-03, grad_scale: 8.0
2022-12-23 07:56:25,073 WARNING [train.py:1060] (2/4) Exclude cut with ID 8631-249866-0030-130156-0_sp0.9 from training. Duration: 26.32775
2022-12-23 07:56:25,229 INFO [zipformer.py:660] (2/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=50901.0, num_to_drop=1, layers_to_drop={1}
2022-12-23 07:56:27,011 INFO [zipformer.py:660] (2/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=50902.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 07:56:56,011 WARNING [train.py:1060] (2/4) Exclude cut with ID 3867-173237-0077-144769-0 from training. Duration: 20.025
2022-12-23 07:57:00,728 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([0.1191, 0.9486, 1.1708, 0.5447, 0.5837, 1.1824, 1.2472, 1.1254],
       device='cuda:2'), covar=tensor([0.0737, 0.0328, 0.0324, 0.0424, 0.0438, 0.0519, 0.0262, 0.0575],
       device='cuda:2'), in_proj_covar=tensor([0.0149, 0.0165, 0.0120, 0.0138, 0.0148, 0.0140, 0.0155, 0.0162],
       device='cuda:2'), out_proj_covar=tensor([1.1617e-04, 1.3045e-04, 9.3796e-05, 1.0640e-04, 1.1548e-04, 1.1140e-04,
        1.2346e-04, 1.2734e-04], device='cuda:2')
2022-12-23 07:57:01,850 WARNING [train.py:1060] (2/4) Exclude cut with ID 6709-74022-0004-86860-0_sp1.1 from training. Duration: 0.9409375
2022-12-23 07:57:01,857 WARNING [train.py:1060] (2/4) Exclude cut with ID 4757-1811-0023-62229-0_sp0.9 from training. Duration: 21.37775
2022-12-23 07:57:08,455 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.3955, 1.7714, 1.3689, 1.9936, 2.3569, 1.4761, 1.2969, 1.2094],
       device='cuda:2'), covar=tensor([0.1854, 0.1704, 0.1522, 0.0987, 0.1160, 0.1080, 0.2033, 0.1474],
       device='cuda:2'), in_proj_covar=tensor([0.0240, 0.0215, 0.0206, 0.0192, 0.0256, 0.0191, 0.0215, 0.0193],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-23 07:57:10,950 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 3.207e+02 5.141e+02 6.153e+02 7.821e+02 1.885e+03, threshold=1.231e+03, percent-clipped=4.0
2022-12-23 07:57:16,417 INFO [scaling.py:679] (2/4) Whitening: num_groups=8, num_channels=96, metric=1.34 vs. limit=2.0
2022-12-23 07:57:18,334 INFO [train.py:894] (2/4) Epoch 15, batch 1850, loss[loss=0.2277, simple_loss=0.2988, pruned_loss=0.07825, over 18523.00 frames. ], tot_loss[loss=0.1997, simple_loss=0.281, pruned_loss=0.0592, over 3713976.80 frames. ], batch size: 77, lr: 7.70e-03, grad_scale: 8.0
2022-12-23 07:57:18,594 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([2.5122, 3.2286, 3.1500, 1.1997, 3.3272, 2.5110, 0.5610, 2.1726],
       device='cuda:2'), covar=tensor([0.2002, 0.1319, 0.1655, 0.3805, 0.1131, 0.1086, 0.5145, 0.1767],
       device='cuda:2'), in_proj_covar=tensor([0.0139, 0.0130, 0.0151, 0.0121, 0.0133, 0.0108, 0.0143, 0.0111],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0002, 0.0003, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-23 07:57:22,688 WARNING [train.py:1060] (2/4) Exclude cut with ID 1250-135782-0004-25974-0_sp0.9 from training. Duration: 21.17225
2022-12-23 07:57:22,698 WARNING [train.py:1060] (2/4) Exclude cut with ID 7357-94126-0021-15852-0_sp0.9 from training. Duration: 27.511125
2022-12-23 07:57:32,309 INFO [zipformer.py:660] (2/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=50946.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 07:57:45,979 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([2.5354, 2.3277, 1.9359, 1.0657, 1.8575, 2.0916, 1.8442, 2.0930],
       device='cuda:2'), covar=tensor([0.0564, 0.0473, 0.1040, 0.1371, 0.1092, 0.1098, 0.1186, 0.0633],
       device='cuda:2'), in_proj_covar=tensor([0.0170, 0.0184, 0.0205, 0.0193, 0.0208, 0.0198, 0.0214, 0.0198],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-23 07:57:56,225 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([2.2738, 2.0889, 1.8545, 1.2467, 2.6335, 2.3215, 1.9448, 1.6052],
       device='cuda:2'), covar=tensor([0.0384, 0.0374, 0.0432, 0.0668, 0.0212, 0.0322, 0.0443, 0.0861],
       device='cuda:2'), in_proj_covar=tensor([0.0125, 0.0124, 0.0131, 0.0122, 0.0093, 0.0123, 0.0138, 0.0155],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0001, 0.0001, 0.0001, 0.0002, 0.0002],
       device='cuda:2')
2022-12-23 07:57:57,394 WARNING [train.py:1060] (2/4) Exclude cut with ID 453-131332-0000-47844-0 from training. Duration: 22.8
2022-12-23 07:58:00,163 WARNING [train.py:1060] (2/4) Exclude cut with ID 4964-30587-0040-44509-0 from training. Duration: 22.585
2022-12-23 07:58:31,862 WARNING [train.py:1060] (2/4) Exclude cut with ID 6978-92210-0001-146967-0_sp0.9 from training. Duration: 22.0166875
2022-12-23 07:58:33,304 INFO [train.py:894] (2/4) Epoch 15, batch 1900, loss[loss=0.2055, simple_loss=0.2879, pruned_loss=0.06149, over 18731.00 frames. ], tot_loss[loss=0.2012, simple_loss=0.2814, pruned_loss=0.06051, over 3714527.25 frames. ], batch size: 52, lr: 7.69e-03, grad_scale: 8.0
2022-12-23 07:58:49,222 WARNING [train.py:1060] (2/4) Exclude cut with ID 7255-291500-0003-134302-0_sp1.1 from training. Duration: 24.395375
2022-12-23 07:58:56,666 WARNING [train.py:1060] (2/4) Exclude cut with ID 3488-85273-0017-41203-0_sp0.9 from training. Duration: 27.47775
2022-12-23 07:58:59,619 WARNING [train.py:1060] (2/4) Exclude cut with ID 432-122774-0017-62487-0_sp0.9 from training. Duration: 24.8833125
2022-12-23 07:59:02,534 WARNING [train.py:1060] (2/4) Exclude cut with ID 6758-72288-0033-108368-0 from training. Duration: 23.39
2022-12-23 07:59:04,358 INFO [zipformer.py:660] (2/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=51007.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 07:59:08,707 WARNING [train.py:1060] (2/4) Exclude cut with ID 8291-282929-0007-12994-0_sp0.9 from training. Duration: 28.72225
2022-12-23 07:59:19,769 WARNING [train.py:1060] (2/4) Exclude cut with ID 585-294811-0110-133686-0_sp0.9 from training. Duration: 20.8944375
2022-12-23 07:59:34,344 WARNING [train.py:1060] (2/4) Exclude cut with ID 5796-66357-0007-116447-0_sp0.9 from training. Duration: 23.8444375
2022-12-23 07:59:40,935 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.9799, 2.1486, 1.4183, 2.4954, 2.2819, 1.9674, 2.9545, 2.1302],
       device='cuda:2'), covar=tensor([0.0900, 0.1514, 0.2748, 0.1607, 0.1649, 0.0917, 0.0889, 0.1088],
       device='cuda:2'), in_proj_covar=tensor([0.0172, 0.0197, 0.0238, 0.0280, 0.0225, 0.0183, 0.0203, 0.0196],
       device='cuda:2'), out_proj_covar=tensor([0.0001, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-23 07:59:41,796 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 3.113e+02 4.916e+02 5.880e+02 7.888e+02 2.150e+03, threshold=1.176e+03, percent-clipped=4.0
2022-12-23 07:59:49,404 INFO [train.py:894] (2/4) Epoch 15, batch 1950, loss[loss=0.2025, simple_loss=0.2716, pruned_loss=0.06665, over 18451.00 frames. ], tot_loss[loss=0.2007, simple_loss=0.2803, pruned_loss=0.06058, over 3714925.34 frames. ], batch size: 50, lr: 7.69e-03, grad_scale: 8.0
2022-12-23 08:00:00,220 WARNING [train.py:1060] (2/4) Exclude cut with ID 8291-282929-0007-12994-0 from training. Duration: 25.85
2022-12-23 08:00:00,228 WARNING [train.py:1060] (2/4) Exclude cut with ID 8291-282929-0023-13010-0 from training. Duration: 21.39
2022-12-23 08:00:12,184 WARNING [train.py:1060] (2/4) Exclude cut with ID 7357-94126-0014-15845-0 from training. Duration: 27.92
2022-12-23 08:00:27,298 INFO [zipformer.py:660] (2/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=51062.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 08:00:40,065 WARNING [train.py:1060] (2/4) Exclude cut with ID 8631-249866-0039-130165-0_sp0.9 from training. Duration: 20.661125
2022-12-23 08:00:59,957 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([2.1089, 2.1436, 1.5479, 2.1748, 2.1870, 1.9627, 2.8750, 2.1918],
       device='cuda:2'), covar=tensor([0.0828, 0.1324, 0.2546, 0.1658, 0.1580, 0.0818, 0.0841, 0.1004],
       device='cuda:2'), in_proj_covar=tensor([0.0171, 0.0197, 0.0237, 0.0280, 0.0224, 0.0182, 0.0203, 0.0195],
       device='cuda:2'), out_proj_covar=tensor([0.0001, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-23 08:01:03,631 INFO [train.py:894] (2/4) Epoch 15, batch 2000, loss[loss=0.1657, simple_loss=0.2382, pruned_loss=0.04657, over 18459.00 frames. ], tot_loss[loss=0.2011, simple_loss=0.2799, pruned_loss=0.06113, over 3714388.72 frames. ], batch size: 43, lr: 7.69e-03, grad_scale: 8.0
2022-12-23 08:01:03,729 WARNING [train.py:1060] (2/4) Exclude cut with ID 7357-94126-0043-15874-0_sp0.9 from training. Duration: 20.07225
2022-12-23 08:01:11,440 WARNING [train.py:1060] (2/4) Exclude cut with ID 1085-156170-0017-128270-0 from training. Duration: 21.01
2022-12-23 08:01:38,759 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([4.6724, 3.9664, 4.0555, 4.5767, 4.1545, 4.0813, 4.7487, 1.5721],
       device='cuda:2'), covar=tensor([0.0623, 0.0647, 0.0553, 0.0701, 0.1443, 0.1162, 0.0578, 0.4930],
       device='cuda:2'), in_proj_covar=tensor([0.0318, 0.0213, 0.0220, 0.0244, 0.0301, 0.0253, 0.0267, 0.0269],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0001, 0.0001, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-23 08:01:58,696 INFO [zipformer.py:660] (2/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=51123.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 08:02:11,369 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 3.251e+02 4.739e+02 5.620e+02 7.194e+02 1.175e+03, threshold=1.124e+03, percent-clipped=0.0
2022-12-23 08:02:17,159 WARNING [train.py:1060] (2/4) Exclude cut with ID 2195-150901-0045-59933-0 from training. Duration: 20.65
2022-12-23 08:02:19,375 INFO [train.py:894] (2/4) Epoch 15, batch 2050, loss[loss=0.2532, simple_loss=0.32, pruned_loss=0.09325, over 18598.00 frames. ], tot_loss[loss=0.2012, simple_loss=0.2796, pruned_loss=0.06144, over 3714141.01 frames. ], batch size: 174, lr: 7.68e-03, grad_scale: 8.0
2022-12-23 08:02:21,417 INFO [zipformer.py:660] (2/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=51138.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 08:02:25,609 WARNING [train.py:1060] (2/4) Exclude cut with ID 5796-66357-0007-116447-0 from training. Duration: 21.46
2022-12-23 08:02:59,884 INFO [zipformer.py:660] (2/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=51164.0, num_to_drop=1, layers_to_drop={0}
2022-12-23 08:03:10,537 WARNING [train.py:1060] (2/4) Exclude cut with ID 3557-8342-0013-54691-0 from training. Duration: 0.92
2022-12-23 08:03:12,410 INFO [zipformer.py:660] (2/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=51172.0, num_to_drop=1, layers_to_drop={1}
2022-12-23 08:03:14,685 WARNING [train.py:1060] (2/4) Exclude cut with ID 8291-282929-0023-13010-0_sp0.9 from training. Duration: 23.7666875
2022-12-23 08:03:35,198 INFO [train.py:894] (2/4) Epoch 15, batch 2100, loss[loss=0.194, simple_loss=0.2753, pruned_loss=0.05637, over 18726.00 frames. ], tot_loss[loss=0.2039, simple_loss=0.2818, pruned_loss=0.06299, over 3714658.96 frames. ], batch size: 54, lr: 7.68e-03, grad_scale: 8.0
2022-12-23 08:03:45,850 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([2.2723, 1.5619, 1.8830, 1.9475, 2.1893, 2.1052, 2.0829, 1.6496],
       device='cuda:2'), covar=tensor([0.1878, 0.2940, 0.2130, 0.2427, 0.1616, 0.0819, 0.2540, 0.1106],
       device='cuda:2'), in_proj_covar=tensor([0.0259, 0.0292, 0.0269, 0.0304, 0.0290, 0.0243, 0.0321, 0.0231],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-23 08:03:54,283 WARNING [train.py:1060] (2/4) Exclude cut with ID 8544-281189-0060-101339-0_sp0.9 from training. Duration: 20.861125
2022-12-23 08:03:57,775 INFO [zipformer.py:660] (2/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=51202.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 08:04:03,194 WARNING [train.py:1060] (2/4) Exclude cut with ID 3488-65654-0031-41259-0_sp0.9 from training. Duration: 22.711125
2022-12-23 08:04:05,308 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([3.0640, 2.6959, 2.7944, 2.8357, 2.5304, 2.5184, 3.1989, 1.0776],
       device='cuda:2'), covar=tensor([0.1845, 0.1740, 0.1456, 0.2103, 0.3213, 0.2465, 0.1762, 0.6698],
       device='cuda:2'), in_proj_covar=tensor([0.0323, 0.0216, 0.0224, 0.0247, 0.0305, 0.0257, 0.0270, 0.0272],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0001, 0.0001, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-23 08:04:44,104 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 2.951e+02 4.953e+02 6.008e+02 8.155e+02 2.142e+03, threshold=1.202e+03, percent-clipped=7.0
2022-12-23 08:04:45,700 WARNING [train.py:1060] (2/4) Exclude cut with ID 6951-79737-0043-132310-0_sp1.1 from training. Duration: 22.986375
2022-12-23 08:04:46,030 INFO [zipformer.py:660] (2/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=51233.0, num_to_drop=1, layers_to_drop={2}
2022-12-23 08:04:51,358 INFO [train.py:894] (2/4) Epoch 15, batch 2150, loss[loss=0.2213, simple_loss=0.2947, pruned_loss=0.07393, over 18667.00 frames. ], tot_loss[loss=0.2025, simple_loss=0.2802, pruned_loss=0.06237, over 3714063.41 frames. ], batch size: 79, lr: 7.67e-03, grad_scale: 8.0
2022-12-23 08:05:01,837 WARNING [train.py:1060] (2/4) Exclude cut with ID 8040-260924-0003-80960-0_sp0.9 from training. Duration: 22.07225
2022-12-23 08:05:06,186 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.8195, 1.7960, 1.6317, 1.5563, 1.6779, 2.0538, 2.1057, 1.4200],
       device='cuda:2'), covar=tensor([0.0300, 0.0211, 0.0370, 0.0210, 0.0209, 0.0300, 0.0183, 0.0258],
       device='cuda:2'), in_proj_covar=tensor([0.0091, 0.0124, 0.0150, 0.0125, 0.0114, 0.0116, 0.0095, 0.0125],
       device='cuda:2'), out_proj_covar=tensor([7.4441e-05, 1.0005e-04, 1.2671e-04, 1.0170e-04, 9.5076e-05, 9.1505e-05,
        7.5384e-05, 1.0013e-04], device='cuda:2')
2022-12-23 08:05:07,118 WARNING [train.py:1060] (2/4) Exclude cut with ID 7699-105389-0045-26330-0_sp0.9 from training. Duration: 20.3055625
2022-12-23 08:05:08,585 WARNING [train.py:1060] (2/4) Exclude cut with ID 6356-271890-0060-94317-0_sp0.9 from training. Duration: 20.72225
2022-12-23 08:05:10,310 INFO [zipformer.py:660] (2/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=51250.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 08:05:16,023 INFO [scaling.py:679] (2/4) Whitening: num_groups=1, num_channels=384, metric=4.03 vs. limit=5.0
2022-12-23 08:05:28,915 WARNING [train.py:1060] (2/4) Exclude cut with ID 3488-85273-0017-41203-0_sp1.1 from training. Duration: 22.4818125
2022-12-23 08:05:56,349 WARNING [train.py:1060] (2/4) Exclude cut with ID 4964-30587-0040-44509-0_sp0.9 from training. Duration: 25.0944375
2022-12-23 08:05:59,343 WARNING [train.py:1060] (2/4) Exclude cut with ID 6533-399-0047-104881-0 from training. Duration: 21.515
2022-12-23 08:06:06,753 WARNING [train.py:1060] (2/4) Exclude cut with ID 7357-94126-0009-15840-0_sp0.9 from training. Duration: 27.02225
2022-12-23 08:06:08,200 INFO [train.py:894] (2/4) Epoch 15, batch 2200, loss[loss=0.2041, simple_loss=0.286, pruned_loss=0.06116, over 18520.00 frames. ], tot_loss[loss=0.2024, simple_loss=0.2799, pruned_loss=0.06242, over 3713895.34 frames. ], batch size: 58, lr: 7.67e-03, grad_scale: 8.0
2022-12-23 08:06:11,124 WARNING [train.py:1060] (2/4) Exclude cut with ID 432-122774-0010-62480-0_sp0.9 from training. Duration: 22.22225
2022-12-23 08:06:18,430 WARNING [train.py:1060] (2/4) Exclude cut with ID 4964-30587-0085-44554-0_sp0.9 from training. Duration: 20.85
2022-12-23 08:06:20,136 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([2.7006, 3.7043, 3.7797, 1.5796, 3.9072, 3.0552, 0.8170, 2.7281],
       device='cuda:2'), covar=tensor([0.2012, 0.1298, 0.1509, 0.3888, 0.0946, 0.0936, 0.5009, 0.1542],
       device='cuda:2'), in_proj_covar=tensor([0.0141, 0.0133, 0.0155, 0.0123, 0.0134, 0.0110, 0.0144, 0.0112],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0002, 0.0003, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-23 08:06:30,879 INFO [zipformer.py:660] (2/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=51302.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 08:06:51,617 WARNING [train.py:1060] (2/4) Exclude cut with ID 4295-39940-0007-92567-0 from training. Duration: 21.54
2022-12-23 08:06:57,204 WARNING [train.py:1060] (2/4) Exclude cut with ID 4964-30587-0040-44509-0_sp1.1 from training. Duration: 20.5318125
2022-12-23 08:07:06,004 WARNING [train.py:1060] (2/4) Exclude cut with ID 7255-291500-0012-134311-0_sp0.9 from training. Duration: 21.9333125
2022-12-23 08:07:16,952 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 3.356e+02 4.912e+02 6.307e+02 7.418e+02 1.261e+03, threshold=1.261e+03, percent-clipped=3.0
2022-12-23 08:07:24,362 INFO [train.py:894] (2/4) Epoch 15, batch 2250, loss[loss=0.1655, simple_loss=0.2355, pruned_loss=0.04777, over 18519.00 frames. ], tot_loss[loss=0.2021, simple_loss=0.2797, pruned_loss=0.0622, over 3713339.41 frames. ], batch size: 41, lr: 7.67e-03, grad_scale: 8.0
2022-12-23 08:07:26,008 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([2.5545, 3.2131, 3.1963, 1.2490, 3.3620, 2.6156, 0.7162, 2.2503],
       device='cuda:2'), covar=tensor([0.2031, 0.1372, 0.1591, 0.3992, 0.1075, 0.1096, 0.5040, 0.1681],
       device='cuda:2'), in_proj_covar=tensor([0.0141, 0.0133, 0.0155, 0.0123, 0.0134, 0.0110, 0.0144, 0.0113],
       device='cuda:2'), out_proj_covar=tensor([0.0003, 0.0002, 0.0003, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-23 08:07:33,288 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([2.6897, 4.0218, 3.8613, 1.5286, 4.0867, 3.1205, 0.5950, 2.6788],
       device='cuda:2'), covar=tensor([0.1976, 0.1085, 0.1416, 0.3703, 0.0820, 0.0971, 0.5345, 0.1462],
       device='cuda:2'), in_proj_covar=tensor([0.0142, 0.0133, 0.0156, 0.0123, 0.0134, 0.0111, 0.0144, 0.0113],
       device='cuda:2'), out_proj_covar=tensor([0.0003, 0.0002, 0.0003, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-23 08:07:52,010 INFO [scaling.py:679] (2/4) Whitening: num_groups=8, num_channels=96, metric=1.21 vs. limit=2.0
2022-12-23 08:07:52,798 WARNING [train.py:1060] (2/4) Exclude cut with ID 8631-249866-0025-130151-0_sp0.9 from training. Duration: 21.7944375
2022-12-23 08:07:53,172 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([2.3085, 1.9599, 1.3454, 0.4974, 1.5140, 1.9908, 1.5192, 1.7966],
       device='cuda:2'), covar=tensor([0.0535, 0.0459, 0.0964, 0.1404, 0.0910, 0.1184, 0.1493, 0.0543],
       device='cuda:2'), in_proj_covar=tensor([0.0170, 0.0183, 0.0205, 0.0192, 0.0208, 0.0197, 0.0212, 0.0198],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-23 08:08:07,555 WARNING [train.py:1060] (2/4) Exclude cut with ID 8291-282929-0002-12989-0_sp0.9 from training. Duration: 22.4666875
2022-12-23 08:08:12,711 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.5473, 1.8871, 1.5170, 2.2396, 2.5076, 1.5975, 1.4106, 1.3219],
       device='cuda:2'), covar=tensor([0.1842, 0.1588, 0.1431, 0.0856, 0.1085, 0.1073, 0.1978, 0.1468],
       device='cuda:2'), in_proj_covar=tensor([0.0242, 0.0217, 0.0206, 0.0191, 0.0256, 0.0191, 0.0215, 0.0192],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-23 08:08:13,800 WARNING [train.py:1060] (2/4) Exclude cut with ID 6121-9014-0076-24124-0 from training. Duration: 21.635
2022-12-23 08:08:20,459 WARNING [train.py:1060] (2/4) Exclude cut with ID 6121-9014-0076-24124-0_sp0.9 from training. Duration: 24.038875
2022-12-23 08:08:32,034 INFO [zipformer.py:660] (2/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=51382.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 08:08:39,393 INFO [train.py:894] (2/4) Epoch 15, batch 2300, loss[loss=0.1973, simple_loss=0.2773, pruned_loss=0.05867, over 18583.00 frames. ], tot_loss[loss=0.2025, simple_loss=0.2798, pruned_loss=0.06254, over 3713880.96 frames. ], batch size: 56, lr: 7.66e-03, grad_scale: 8.0
2022-12-23 08:08:47,480 INFO [zipformer.py:660] (2/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=51392.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 08:08:59,362 WARNING [train.py:1060] (2/4) Exclude cut with ID 7255-291500-0009-134308-0_sp1.1 from training. Duration: 21.786375
2022-12-23 08:09:06,487 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([2.9844, 3.2989, 2.0476, 1.4849, 3.5617, 3.5498, 2.5993, 2.3044],
       device='cuda:2'), covar=tensor([0.0448, 0.0299, 0.0613, 0.0816, 0.0179, 0.0345, 0.0601, 0.0759],
       device='cuda:2'), in_proj_covar=tensor([0.0125, 0.0123, 0.0129, 0.0121, 0.0092, 0.0122, 0.0136, 0.0153],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0001, 0.0001, 0.0001, 0.0002, 0.0002],
       device='cuda:2')
2022-12-23 08:09:07,076 INFO [scaling.py:679] (2/4) Whitening: num_groups=1, num_channels=384, metric=4.81 vs. limit=5.0
2022-12-23 08:09:12,240 WARNING [train.py:1060] (2/4) Exclude cut with ID 8291-282929-0002-12989-0 from training. Duration: 20.22
2022-12-23 08:09:27,841 INFO [zipformer.py:660] (2/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=51418.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 08:09:47,809 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 3.635e+02 5.135e+02 6.604e+02 7.825e+02 1.376e+03, threshold=1.321e+03, percent-clipped=3.0
2022-12-23 08:09:55,127 INFO [train.py:894] (2/4) Epoch 15, batch 2350, loss[loss=0.2329, simple_loss=0.303, pruned_loss=0.08139, over 18560.00 frames. ], tot_loss[loss=0.2038, simple_loss=0.2811, pruned_loss=0.06326, over 3714542.58 frames. ], batch size: 98, lr: 7.66e-03, grad_scale: 8.0
2022-12-23 08:09:56,970 INFO [zipformer.py:660] (2/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=51438.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 08:10:04,861 INFO [zipformer.py:660] (2/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=51443.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 08:10:19,424 INFO [zipformer.py:660] (2/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=51453.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 08:10:35,842 INFO [zipformer.py:660] (2/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=51464.0, num_to_drop=1, layers_to_drop={1}
2022-12-23 08:10:38,597 INFO [scaling.py:679] (2/4) Whitening: num_groups=8, num_channels=192, metric=1.94 vs. limit=2.0
2022-12-23 08:11:05,659 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([3.1181, 2.3994, 1.7118, 2.9706, 2.1126, 2.2774, 2.3318, 3.4012],
       device='cuda:2'), covar=tensor([0.1713, 0.3071, 0.1829, 0.2821, 0.3637, 0.1004, 0.2831, 0.0658],
       device='cuda:2'), in_proj_covar=tensor([0.0289, 0.0282, 0.0237, 0.0352, 0.0263, 0.0219, 0.0276, 0.0204],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0003, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-23 08:11:08,174 INFO [zipformer.py:660] (2/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=51486.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 08:11:09,475 INFO [train.py:894] (2/4) Epoch 15, batch 2400, loss[loss=0.2171, simple_loss=0.291, pruned_loss=0.07161, over 18432.00 frames. ], tot_loss[loss=0.2032, simple_loss=0.2808, pruned_loss=0.06287, over 3715255.29 frames. ], batch size: 48, lr: 7.66e-03, grad_scale: 8.0
2022-12-23 08:11:09,482 WARNING [train.py:1060] (2/4) Exclude cut with ID 6951-79737-0043-132310-0 from training. Duration: 25.285
2022-12-23 08:11:22,362 INFO [zipformer.py:660] (2/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=51495.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 08:11:48,460 INFO [zipformer.py:660] (2/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=51512.0, num_to_drop=1, layers_to_drop={1}
2022-12-23 08:12:11,648 INFO [zipformer.py:660] (2/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=51528.0, num_to_drop=1, layers_to_drop={0}
2022-12-23 08:12:14,808 WARNING [train.py:1060] (2/4) Exclude cut with ID 811-130148-0001-63453-0_sp0.9 from training. Duration: 20.861125
2022-12-23 08:12:17,682 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 3.080e+02 5.000e+02 6.055e+02 7.391e+02 1.821e+03, threshold=1.211e+03, percent-clipped=2.0
2022-12-23 08:12:25,894 INFO [train.py:894] (2/4) Epoch 15, batch 2450, loss[loss=0.2098, simple_loss=0.2917, pruned_loss=0.06397, over 18500.00 frames. ], tot_loss[loss=0.2022, simple_loss=0.2802, pruned_loss=0.06211, over 3715310.00 frames. ], batch size: 52, lr: 7.65e-03, grad_scale: 8.0
2022-12-23 08:12:38,269 WARNING [train.py:1060] (2/4) Exclude cut with ID 6010-56788-0055-90261-0 from training. Duration: 20.88
2022-12-23 08:12:55,433 INFO [zipformer.py:660] (2/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=51556.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 08:13:08,694 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.2553, 1.3862, 1.1193, 1.6710, 1.6108, 1.3329, 0.9340, 1.1391],
       device='cuda:2'), covar=tensor([0.2027, 0.1915, 0.1701, 0.1151, 0.1256, 0.1134, 0.2199, 0.1596],
       device='cuda:2'), in_proj_covar=tensor([0.0242, 0.0217, 0.0206, 0.0192, 0.0257, 0.0190, 0.0216, 0.0193],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-23 08:13:09,633 WARNING [train.py:1060] (2/4) Exclude cut with ID 7357-94126-0045-15876-0_sp0.9 from training. Duration: 23.4166875
2022-12-23 08:13:40,945 INFO [train.py:894] (2/4) Epoch 15, batch 2500, loss[loss=0.2422, simple_loss=0.3106, pruned_loss=0.08686, over 18569.00 frames. ], tot_loss[loss=0.2009, simple_loss=0.2786, pruned_loss=0.0616, over 3713879.10 frames. ], batch size: 77, lr: 7.65e-03, grad_scale: 8.0
2022-12-23 08:13:47,450 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([2.9336, 5.2201, 4.7106, 2.6330, 5.2017, 3.8712, 0.7432, 3.3388],
       device='cuda:2'), covar=tensor([0.1891, 0.0756, 0.1279, 0.2830, 0.0635, 0.0834, 0.5281, 0.1345],
       device='cuda:2'), in_proj_covar=tensor([0.0139, 0.0131, 0.0154, 0.0122, 0.0133, 0.0109, 0.0142, 0.0112],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0002, 0.0003, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-23 08:14:04,565 INFO [zipformer.py:660] (2/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=51602.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 08:14:28,588 WARNING [train.py:1060] (2/4) Exclude cut with ID 7255-291500-0006-134305-0 from training. Duration: 21.24
2022-12-23 08:14:28,605 WARNING [train.py:1060] (2/4) Exclude cut with ID 6533-399-0047-104881-0_sp0.9 from training. Duration: 23.9055625
2022-12-23 08:14:49,858 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 2.827e+02 4.732e+02 5.703e+02 7.714e+02 1.744e+03, threshold=1.141e+03, percent-clipped=4.0
2022-12-23 08:14:58,035 INFO [train.py:894] (2/4) Epoch 15, batch 2550, loss[loss=0.2321, simple_loss=0.3028, pruned_loss=0.08069, over 18679.00 frames. ], tot_loss[loss=0.2003, simple_loss=0.2782, pruned_loss=0.0612, over 3714622.31 frames. ], batch size: 62, lr: 7.64e-03, grad_scale: 8.0
2022-12-23 08:15:01,003 WARNING [train.py:1060] (2/4) Exclude cut with ID 6758-72288-0033-108368-0_sp0.9 from training. Duration: 25.988875
2022-12-23 08:15:10,450 WARNING [train.py:1060] (2/4) Exclude cut with ID 7255-291500-0001-134300-0_sp0.9 from training. Duration: 20.67225
2022-12-23 08:15:19,046 INFO [zipformer.py:660] (2/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=51650.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 08:15:57,974 WARNING [train.py:1060] (2/4) Exclude cut with ID 3488-85273-0038-41224-0 from training. Duration: 20.34
2022-12-23 08:16:15,222 INFO [train.py:894] (2/4) Epoch 15, batch 2600, loss[loss=0.1975, simple_loss=0.2845, pruned_loss=0.0552, over 18456.00 frames. ], tot_loss[loss=0.1995, simple_loss=0.2774, pruned_loss=0.06077, over 3714170.17 frames. ], batch size: 54, lr: 7.64e-03, grad_scale: 8.0
2022-12-23 08:16:44,079 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.3615, 2.7896, 3.3164, 0.9945, 2.7903, 3.5236, 2.4520, 2.9141],
       device='cuda:2'), covar=tensor([0.0864, 0.0345, 0.0306, 0.0532, 0.0366, 0.0403, 0.0403, 0.0600],
       device='cuda:2'), in_proj_covar=tensor([0.0147, 0.0164, 0.0120, 0.0138, 0.0146, 0.0139, 0.0155, 0.0162],
       device='cuda:2'), out_proj_covar=tensor([1.1504e-04, 1.2944e-04, 9.3230e-05, 1.0623e-04, 1.1383e-04, 1.1058e-04,
        1.2346e-04, 1.2678e-04], device='cuda:2')
2022-12-23 08:17:03,410 INFO [zipformer.py:660] (2/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=51718.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 08:17:07,880 WARNING [train.py:1060] (2/4) Exclude cut with ID 7357-94126-0026-15857-0_sp0.9 from training. Duration: 25.061125
2022-12-23 08:17:20,455 WARNING [train.py:1060] (2/4) Exclude cut with ID 3033-130750-0096-55598-0 from training. Duration: 0.83
2022-12-23 08:17:24,696 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 3.474e+02 4.798e+02 5.976e+02 6.876e+02 1.380e+03, threshold=1.195e+03, percent-clipped=6.0
2022-12-23 08:17:32,247 INFO [train.py:894] (2/4) Epoch 15, batch 2650, loss[loss=0.2053, simple_loss=0.2817, pruned_loss=0.06446, over 18702.00 frames. ], tot_loss[loss=0.1993, simple_loss=0.2775, pruned_loss=0.06056, over 3714023.15 frames. ], batch size: 50, lr: 7.64e-03, grad_scale: 8.0
2022-12-23 08:17:33,890 INFO [zipformer.py:660] (2/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=51738.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 08:17:46,845 WARNING [train.py:1060] (2/4) Exclude cut with ID 3488-85273-0017-41203-0 from training. Duration: 24.73
2022-12-23 08:17:48,646 INFO [zipformer.py:660] (2/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=51747.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 08:17:50,012 INFO [zipformer.py:660] (2/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=51748.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 08:18:00,246 WARNING [train.py:1060] (2/4) Exclude cut with ID 7255-291500-0009-134308-0 from training. Duration: 23.965
2022-12-23 08:18:07,239 WARNING [train.py:1060] (2/4) Exclude cut with ID 6978-92210-0030-146996-0_sp0.9 from training. Duration: 22.088875
2022-12-23 08:18:16,971 INFO [zipformer.py:660] (2/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=51766.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 08:18:24,751 WARNING [train.py:1060] (2/4) Exclude cut with ID 7255-291500-0006-134305-0_sp0.9 from training. Duration: 23.6
2022-12-23 08:18:48,946 INFO [train.py:894] (2/4) Epoch 15, batch 2700, loss[loss=0.1939, simple_loss=0.2779, pruned_loss=0.05489, over 18625.00 frames. ], tot_loss[loss=0.1986, simple_loss=0.2769, pruned_loss=0.06013, over 3714252.01 frames. ], batch size: 53, lr: 7.63e-03, grad_scale: 8.0
2022-12-23 08:19:20,150 INFO [zipformer.py:660] (2/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=51808.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 08:19:50,323 INFO [zipformer.py:660] (2/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=51828.0, num_to_drop=1, layers_to_drop={2}
2022-12-23 08:19:55,520 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 3.453e+02 5.078e+02 6.132e+02 7.520e+02 1.991e+03, threshold=1.226e+03, percent-clipped=4.0
2022-12-23 08:20:03,362 INFO [train.py:894] (2/4) Epoch 15, batch 2750, loss[loss=0.1909, simple_loss=0.2721, pruned_loss=0.05483, over 18630.00 frames. ], tot_loss[loss=0.1985, simple_loss=0.2767, pruned_loss=0.06013, over 3713632.64 frames. ], batch size: 53, lr: 7.63e-03, grad_scale: 8.0
2022-12-23 08:20:09,956 WARNING [train.py:1060] (2/4) Exclude cut with ID 8291-282929-0024-13011-0 from training. Duration: 23.795
2022-12-23 08:20:23,793 WARNING [train.py:1060] (2/4) Exclude cut with ID 8631-249866-0030-130156-0_sp1.1 from training. Duration: 21.5409375
2022-12-23 08:20:25,585 INFO [zipformer.py:660] (2/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=51851.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 08:20:26,730 WARNING [train.py:1060] (2/4) Exclude cut with ID 6978-92210-0019-146985-0_sp0.9 from training. Duration: 24.97775
2022-12-23 08:20:39,226 WARNING [train.py:1060] (2/4) Exclude cut with ID 1085-156170-0017-128270-0_sp0.9 from training. Duration: 23.3444375
2022-12-23 08:20:41,454 INFO [scaling.py:679] (2/4) Whitening: num_groups=8, num_channels=192, metric=1.97 vs. limit=2.0
2022-12-23 08:21:02,801 INFO [zipformer.py:660] (2/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=51876.0, num_to_drop=1, layers_to_drop={0}
2022-12-23 08:21:06,789 WARNING [train.py:1060] (2/4) Exclude cut with ID 6010-56788-0055-90261-0_sp0.9 from training. Duration: 23.2
2022-12-23 08:21:13,977 WARNING [train.py:1060] (2/4) Exclude cut with ID 5653-46179-0060-117930-0_sp0.9 from training. Duration: 21.17225
2022-12-23 08:21:20,032 INFO [train.py:894] (2/4) Epoch 15, batch 2800, loss[loss=0.1893, simple_loss=0.2673, pruned_loss=0.05564, over 18686.00 frames. ], tot_loss[loss=0.1971, simple_loss=0.2757, pruned_loss=0.05928, over 3714234.20 frames. ], batch size: 48, lr: 7.63e-03, grad_scale: 16.0
2022-12-23 08:21:32,257 WARNING [train.py:1060] (2/4) Exclude cut with ID 7255-291500-0008-134307-0_sp0.9 from training. Duration: 24.6555625
2022-12-23 08:21:43,338 INFO [scaling.py:679] (2/4) Whitening: num_groups=1, num_channels=384, metric=3.43 vs. limit=5.0
2022-12-23 08:22:28,075 WARNING [train.py:1060] (2/4) Exclude cut with ID 3488-65654-0031-41259-0 from training. Duration: 20.44
2022-12-23 08:22:29,758 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 2.494e+02 4.534e+02 5.436e+02 6.882e+02 1.456e+03, threshold=1.087e+03, percent-clipped=1.0
2022-12-23 08:22:35,860 INFO [train.py:894] (2/4) Epoch 15, batch 2850, loss[loss=0.1856, simple_loss=0.2696, pruned_loss=0.05077, over 18716.00 frames. ], tot_loss[loss=0.1977, simple_loss=0.2758, pruned_loss=0.05983, over 3714496.93 frames. ], batch size: 78, lr: 7.62e-03, grad_scale: 8.0
2022-12-23 08:22:43,574 WARNING [train.py:1060] (2/4) Exclude cut with ID 6951-79737-0018-132285-0_sp0.9 from training. Duration: 23.45
2022-12-23 08:23:12,459 WARNING [train.py:1060] (2/4) Exclude cut with ID 6945-60535-0076-12784-0_sp0.9 from training. Duration: 20.52225
2022-12-23 08:23:21,349 WARNING [train.py:1060] (2/4) Exclude cut with ID 7255-291500-0008-134307-0 from training. Duration: 22.19
2022-12-23 08:23:30,701 WARNING [train.py:1060] (2/4) Exclude cut with ID 7357-94126-0014-15845-0_sp1.1 from training. Duration: 25.3818125
2022-12-23 08:23:48,166 WARNING [train.py:1060] (2/4) Exclude cut with ID 6951-79737-0043-132310-0_sp0.9 from training. Duration: 28.0944375
2022-12-23 08:23:52,678 INFO [train.py:894] (2/4) Epoch 15, batch 2900, loss[loss=0.2071, simple_loss=0.2811, pruned_loss=0.06655, over 18714.00 frames. ], tot_loss[loss=0.1974, simple_loss=0.2755, pruned_loss=0.05964, over 3715379.87 frames. ], batch size: 52, lr: 7.62e-03, grad_scale: 8.0
2022-12-23 08:23:54,180 WARNING [train.py:1060] (2/4) Exclude cut with ID 2195-150901-0045-59933-0_sp0.9 from training. Duration: 22.9444375
2022-12-23 08:24:01,707 WARNING [train.py:1060] (2/4) Exclude cut with ID 8291-282929-0024-13011-0_sp1.1 from training. Duration: 21.6318125
2022-12-23 08:24:23,335 WARNING [train.py:1060] (2/4) Exclude cut with ID 8631-249866-0030-130156-0 from training. Duration: 23.695
2022-12-23 08:24:47,557 WARNING [train.py:1060] (2/4) Exclude cut with ID 7699-105389-0094-26379-0 from training. Duration: 23.955
2022-12-23 08:24:54,002 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.4397, 1.6959, 1.3566, 2.0462, 2.2364, 1.4906, 1.1662, 1.3017],
       device='cuda:2'), covar=tensor([0.1959, 0.1769, 0.1588, 0.0988, 0.1162, 0.1186, 0.2277, 0.1475],
       device='cuda:2'), in_proj_covar=tensor([0.0243, 0.0219, 0.0207, 0.0193, 0.0259, 0.0193, 0.0219, 0.0194],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-23 08:25:04,484 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([2.4573, 2.6425, 2.1600, 3.0590, 2.7163, 2.3132, 3.2477, 2.6948],
       device='cuda:2'), covar=tensor([0.0714, 0.1475, 0.1991, 0.1425, 0.1273, 0.0770, 0.0896, 0.0849],
       device='cuda:2'), in_proj_covar=tensor([0.0175, 0.0200, 0.0241, 0.0286, 0.0228, 0.0186, 0.0207, 0.0199],
       device='cuda:2'), out_proj_covar=tensor([0.0001, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-23 08:25:07,019 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 3.195e+02 4.890e+02 5.936e+02 7.455e+02 1.185e+03, threshold=1.187e+03, percent-clipped=3.0
2022-12-23 08:25:13,458 INFO [train.py:894] (2/4) Epoch 15, batch 2950, loss[loss=0.1765, simple_loss=0.2674, pruned_loss=0.04283, over 18504.00 frames. ], tot_loss[loss=0.1982, simple_loss=0.2759, pruned_loss=0.0602, over 3714983.25 frames. ], batch size: 58, lr: 7.62e-03, grad_scale: 8.0
2022-12-23 08:25:15,333 INFO [zipformer.py:660] (2/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=52038.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 08:25:18,152 INFO [zipformer.py:660] (2/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=52040.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 08:25:22,691 WARNING [train.py:1060] (2/4) Exclude cut with ID 8291-282929-0024-13011-0_sp0.9 from training. Duration: 26.438875
2022-12-23 08:25:29,934 INFO [zipformer.py:660] (2/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=52048.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 08:26:06,653 WARNING [train.py:1060] (2/4) Exclude cut with ID 7699-105389-0021-26306-0_sp0.9 from training. Duration: 21.2444375
2022-12-23 08:26:06,674 WARNING [train.py:1060] (2/4) Exclude cut with ID 7357-94126-0014-15845-0_sp0.9 from training. Duration: 31.02225
2022-12-23 08:26:08,622 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([2.3298, 2.6257, 1.7076, 2.9211, 2.7521, 2.1829, 3.6695, 2.3494],
       device='cuda:2'), covar=tensor([0.0825, 0.1654, 0.2609, 0.1884, 0.1560, 0.0883, 0.0899, 0.1100],
       device='cuda:2'), in_proj_covar=tensor([0.0175, 0.0201, 0.0242, 0.0288, 0.0228, 0.0187, 0.0207, 0.0199],
       device='cuda:2'), out_proj_covar=tensor([0.0001, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-23 08:26:17,391 WARNING [train.py:1060] (2/4) Exclude cut with ID 432-122774-0017-62487-0 from training. Duration: 22.395
2022-12-23 08:26:21,038 INFO [scaling.py:679] (2/4) Whitening: num_groups=1, num_channels=384, metric=3.73 vs. limit=5.0
2022-12-23 08:26:27,558 INFO [zipformer.py:660] (2/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=52086.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 08:26:28,966 INFO [train.py:894] (2/4) Epoch 15, batch 3000, loss[loss=0.1651, simple_loss=0.2379, pruned_loss=0.04616, over 18518.00 frames. ], tot_loss[loss=0.1988, simple_loss=0.2762, pruned_loss=0.0607, over 3713686.53 frames. ], batch size: 41, lr: 7.61e-03, grad_scale: 8.0
2022-12-23 08:26:28,966 INFO [train.py:919] (2/4) Computing validation loss
2022-12-23 08:26:35,047 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.6040, 1.5989, 1.7921, 1.6663, 1.1728, 3.2413, 1.5667, 1.9202],
       device='cuda:2'), covar=tensor([0.3187, 0.2093, 0.1752, 0.1942, 0.1429, 0.0228, 0.1580, 0.0854],
       device='cuda:2'), in_proj_covar=tensor([0.0135, 0.0115, 0.0126, 0.0119, 0.0102, 0.0098, 0.0092, 0.0089],
       device='cuda:2'), out_proj_covar=tensor([0.0005, 0.0005, 0.0005, 0.0005, 0.0004, 0.0003, 0.0004, 0.0004],
       device='cuda:2')
2022-12-23 08:26:39,939 INFO [train.py:928] (2/4) Epoch 15, validation: loss=0.1652, simple_loss=0.2646, pruned_loss=0.03286, over 944034.00 frames. 
2022-12-23 08:26:39,940 INFO [train.py:929] (2/4) Maximum memory allocated so far is 24680MB
2022-12-23 08:26:44,119 WARNING [train.py:1060] (2/4) Exclude cut with ID 7357-94126-0045-15876-0 from training. Duration: 21.075
2022-12-23 08:26:50,637 WARNING [train.py:1060] (2/4) Exclude cut with ID 6482-98857-0025-147532-0_sp0.9 from training. Duration: 20.0055625
2022-12-23 08:26:50,649 WARNING [train.py:1060] (2/4) Exclude cut with ID 6951-79737-0037-132304-0_sp0.9 from training. Duration: 22.05
2022-12-23 08:26:50,662 WARNING [train.py:1060] (2/4) Exclude cut with ID 7255-291500-0003-134302-0 from training. Duration: 26.8349375
2022-12-23 08:26:54,496 WARNING [train.py:1060] (2/4) Exclude cut with ID 7357-94126-0009-15840-0_sp1.1 from training. Duration: 22.1090625
2022-12-23 08:26:54,623 INFO [zipformer.py:660] (2/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=52096.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 08:27:02,111 WARNING [train.py:1060] (2/4) Exclude cut with ID 7699-105389-0094-26379-0_sp0.9 from training. Duration: 26.6166875
2022-12-23 08:27:02,477 INFO [zipformer.py:660] (2/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=52101.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 08:27:05,019 INFO [zipformer.py:660] (2/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=52103.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 08:27:19,165 WARNING [train.py:1060] (2/4) Exclude cut with ID 2046-178027-0000-53705-0_sp0.9 from training. Duration: 20.3055625
2022-12-23 08:27:31,618 INFO [scaling.py:679] (2/4) Whitening: num_groups=8, num_channels=192, metric=1.85 vs. limit=2.0
2022-12-23 08:27:35,679 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([3.8735, 3.3414, 3.3837, 3.8414, 3.5262, 3.4030, 3.9678, 1.1971],
       device='cuda:2'), covar=tensor([0.0753, 0.0714, 0.0686, 0.0792, 0.1387, 0.1148, 0.0695, 0.4994],
       device='cuda:2'), in_proj_covar=tensor([0.0331, 0.0218, 0.0227, 0.0254, 0.0309, 0.0262, 0.0278, 0.0275],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0001, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-23 08:27:41,403 WARNING [train.py:1060] (2/4) Exclude cut with ID 7205-50138-0008-5373-0_sp0.9 from training. Duration: 20.7
2022-12-23 08:27:49,518 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 3.226e+02 4.827e+02 5.658e+02 7.296e+02 1.550e+03, threshold=1.132e+03, percent-clipped=2.0
2022-12-23 08:27:56,244 INFO [train.py:894] (2/4) Epoch 15, batch 3050, loss[loss=0.2252, simple_loss=0.3068, pruned_loss=0.07182, over 18575.00 frames. ], tot_loss[loss=0.1994, simple_loss=0.2773, pruned_loss=0.06076, over 3713548.88 frames. ], batch size: 57, lr: 7.61e-03, grad_scale: 8.0
2022-12-23 08:28:15,133 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.6961, 1.4891, 1.0937, 0.1925, 1.2262, 1.5975, 1.4295, 1.4875],
       device='cuda:2'), covar=tensor([0.0590, 0.0526, 0.0959, 0.1619, 0.0990, 0.1600, 0.1617, 0.0661],
       device='cuda:2'), in_proj_covar=tensor([0.0168, 0.0182, 0.0204, 0.0192, 0.0207, 0.0197, 0.0212, 0.0199],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-23 08:28:17,818 INFO [zipformer.py:660] (2/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=52151.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 08:28:24,950 WARNING [train.py:1060] (2/4) Exclude cut with ID 6978-92210-0019-146985-0 from training. Duration: 22.48
2022-12-23 08:28:39,795 WARNING [train.py:1060] (2/4) Exclude cut with ID 7255-291500-0003-134302-0_sp0.9 from training. Duration: 29.816625
2022-12-23 08:28:44,453 INFO [scaling.py:679] (2/4) Whitening: num_groups=8, num_channels=192, metric=1.86 vs. limit=2.0
2022-12-23 08:29:00,311 WARNING [train.py:1060] (2/4) Exclude cut with ID 7255-291500-0005-134304-0_sp1.1 from training. Duration: 22.7590625
2022-12-23 08:29:05,889 WARNING [train.py:1060] (2/4) Exclude cut with ID 7357-94126-0026-15857-0 from training. Duration: 22.555
2022-12-23 08:29:11,485 INFO [scaling.py:679] (2/4) Whitening: num_groups=8, num_channels=96, metric=1.21 vs. limit=2.0
2022-12-23 08:29:11,935 INFO [train.py:894] (2/4) Epoch 15, batch 3100, loss[loss=0.1773, simple_loss=0.2519, pruned_loss=0.05134, over 18545.00 frames. ], tot_loss[loss=0.199, simple_loss=0.277, pruned_loss=0.06051, over 3713280.02 frames. ], batch size: 44, lr: 7.60e-03, grad_scale: 8.0
2022-12-23 08:29:21,832 INFO [zipformer.py:660] (2/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=52193.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 08:29:24,474 WARNING [train.py:1060] (2/4) Exclude cut with ID 1250-135782-0005-25975-0_sp0.9 from training. Duration: 21.688875
2022-12-23 08:29:30,476 INFO [zipformer.py:660] (2/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=52199.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 08:29:33,215 INFO [scaling.py:679] (2/4) Whitening: num_groups=8, num_channels=96, metric=1.38 vs. limit=2.0
2022-12-23 08:29:39,867 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([2.0767, 1.8492, 1.7005, 1.0864, 2.3202, 2.1005, 1.8853, 1.4230],
       device='cuda:2'), covar=tensor([0.0339, 0.0387, 0.0442, 0.0725, 0.0273, 0.0321, 0.0399, 0.0834],
       device='cuda:2'), in_proj_covar=tensor([0.0122, 0.0124, 0.0129, 0.0121, 0.0092, 0.0121, 0.0134, 0.0153],
       device='cuda:2'), out_proj_covar=tensor([0.0001, 0.0002, 0.0002, 0.0001, 0.0001, 0.0001, 0.0002, 0.0002],
       device='cuda:2')
2022-12-23 08:30:00,874 WARNING [train.py:1060] (2/4) Exclude cut with ID 3488-85273-0038-41224-0_sp0.9 from training. Duration: 22.6
2022-12-23 08:30:22,175 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 2.672e+02 4.895e+02 6.130e+02 7.505e+02 1.385e+03, threshold=1.226e+03, percent-clipped=5.0
2022-12-23 08:30:27,881 INFO [train.py:894] (2/4) Epoch 15, batch 3150, loss[loss=0.1764, simple_loss=0.2516, pruned_loss=0.05061, over 18672.00 frames. ], tot_loss[loss=0.1988, simple_loss=0.2769, pruned_loss=0.06036, over 3713799.85 frames. ], batch size: 48, lr: 7.60e-03, grad_scale: 8.0
2022-12-23 08:30:36,482 INFO [scaling.py:679] (2/4) Whitening: num_groups=8, num_channels=96, metric=1.21 vs. limit=2.0
2022-12-23 08:30:39,761 WARNING [train.py:1060] (2/4) Exclude cut with ID 7357-94126-0009-15840-0 from training. Duration: 24.32
2022-12-23 08:30:53,357 INFO [zipformer.py:660] (2/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=52254.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 08:31:32,759 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.5468, 1.6482, 1.4463, 2.0414, 2.0150, 1.5471, 1.2396, 1.3463],
       device='cuda:2'), covar=tensor([0.1764, 0.1729, 0.1441, 0.0951, 0.1216, 0.1035, 0.2136, 0.1444],
       device='cuda:2'), in_proj_covar=tensor([0.0243, 0.0219, 0.0207, 0.0192, 0.0258, 0.0191, 0.0217, 0.0192],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-23 08:31:38,082 WARNING [train.py:1060] (2/4) Exclude cut with ID 8291-276745-0093-13116-0_sp0.9 from training. Duration: 21.061125
2022-12-23 08:31:42,986 INFO [train.py:894] (2/4) Epoch 15, batch 3200, loss[loss=0.1871, simple_loss=0.2756, pruned_loss=0.04932, over 18514.00 frames. ], tot_loss[loss=0.1976, simple_loss=0.2762, pruned_loss=0.05948, over 3715153.22 frames. ], batch size: 58, lr: 7.60e-03, grad_scale: 8.0
2022-12-23 08:31:51,489 WARNING [train.py:1060] (2/4) Exclude cut with ID 7357-94126-0024-15855-0_sp0.9 from training. Duration: 20.32225
2022-12-23 08:32:03,708 WARNING [train.py:1060] (2/4) Exclude cut with ID 3033-130750-0096-55598-0_sp1.1 from training. Duration: 0.7545625
2022-12-23 08:32:19,472 WARNING [train.py:1060] (2/4) Exclude cut with ID 4295-39940-0007-92567-0_sp0.9 from training. Duration: 23.9333125
2022-12-23 08:32:21,258 INFO [zipformer.py:660] (2/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=52312.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 08:32:51,689 WARNING [train.py:1060] (2/4) Exclude cut with ID 7255-291500-0008-134307-0_sp1.1 from training. Duration: 20.17275
2022-12-23 08:32:53,102 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 3.115e+02 4.671e+02 5.458e+02 7.644e+02 1.526e+03, threshold=1.092e+03, percent-clipped=4.0
2022-12-23 08:32:57,365 WARNING [train.py:1060] (2/4) Exclude cut with ID 6978-92210-0019-146985-0_sp1.1 from training. Duration: 20.436375
2022-12-23 08:32:58,899 INFO [train.py:894] (2/4) Epoch 15, batch 3250, loss[loss=0.1971, simple_loss=0.2733, pruned_loss=0.06044, over 18447.00 frames. ], tot_loss[loss=0.1965, simple_loss=0.2755, pruned_loss=0.05876, over 3715000.94 frames. ], batch size: 50, lr: 7.59e-03, grad_scale: 8.0
2022-12-23 08:33:09,306 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([2.5423, 3.8436, 3.7227, 1.6030, 3.7781, 2.8025, 0.6534, 2.6520],
       device='cuda:2'), covar=tensor([0.2268, 0.1078, 0.1433, 0.3778, 0.0947, 0.1060, 0.5460, 0.1546],
       device='cuda:2'), in_proj_covar=tensor([0.0140, 0.0133, 0.0155, 0.0123, 0.0135, 0.0111, 0.0142, 0.0112],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0002, 0.0003, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-23 08:33:44,945 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([2.5524, 2.2777, 1.6909, 0.7439, 1.5608, 2.0449, 1.7233, 1.9108],
       device='cuda:2'), covar=tensor([0.0530, 0.0443, 0.1096, 0.1476, 0.1098, 0.1247, 0.1344, 0.0652],
       device='cuda:2'), in_proj_covar=tensor([0.0166, 0.0180, 0.0203, 0.0189, 0.0204, 0.0195, 0.0210, 0.0196],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-23 08:33:54,357 INFO [zipformer.py:660] (2/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=52373.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 08:34:15,744 INFO [train.py:894] (2/4) Epoch 15, batch 3300, loss[loss=0.1824, simple_loss=0.2602, pruned_loss=0.05226, over 18661.00 frames. ], tot_loss[loss=0.1975, simple_loss=0.2761, pruned_loss=0.0594, over 3715718.45 frames. ], batch size: 48, lr: 7.59e-03, grad_scale: 8.0
2022-12-23 08:34:20,212 WARNING [train.py:1060] (2/4) Exclude cut with ID 4234-40345-0022-142709-0_sp0.9 from training. Duration: 23.1055625
2022-12-23 08:34:20,264 WARNING [train.py:1060] (2/4) Exclude cut with ID 8291-282929-0007-12994-0_sp1.1 from training. Duration: 23.5
2022-12-23 08:34:25,343 INFO [scaling.py:679] (2/4) Whitening: num_groups=1, num_channels=384, metric=3.28 vs. limit=5.0
2022-12-23 08:34:29,072 INFO [zipformer.py:660] (2/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=52396.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 08:34:31,924 WARNING [train.py:1060] (2/4) Exclude cut with ID 7255-291500-0009-134308-0_sp0.9 from training. Duration: 26.62775
2022-12-23 08:34:40,054 INFO [zipformer.py:660] (2/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=52403.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 08:34:44,914 WARNING [train.py:1060] (2/4) Exclude cut with ID 6951-79737-0018-132285-0 from training. Duration: 21.105
2022-12-23 08:34:45,152 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([3.0383, 2.6833, 2.6829, 3.0092, 2.7847, 2.7673, 3.1172, 1.4072],
       device='cuda:2'), covar=tensor([0.0668, 0.0538, 0.0606, 0.0718, 0.1074, 0.0927, 0.0746, 0.3355],
       device='cuda:2'), in_proj_covar=tensor([0.0328, 0.0216, 0.0226, 0.0253, 0.0307, 0.0259, 0.0275, 0.0274],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0001, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-23 08:34:48,168 WARNING [train.py:1060] (2/4) Exclude cut with ID 4511-76322-0006-80011-0_sp0.9 from training. Duration: 24.411125
2022-12-23 08:35:17,036 WARNING [train.py:1060] (2/4) Exclude cut with ID 6758-72288-0033-108368-0_sp1.1 from training. Duration: 21.263625
2022-12-23 08:35:27,536 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 2.913e+02 4.685e+02 5.736e+02 7.121e+02 1.192e+03, threshold=1.147e+03, percent-clipped=2.0
2022-12-23 08:35:33,512 INFO [train.py:894] (2/4) Epoch 15, batch 3350, loss[loss=0.1584, simple_loss=0.239, pruned_loss=0.03895, over 18537.00 frames. ], tot_loss[loss=0.1973, simple_loss=0.2758, pruned_loss=0.05942, over 3715638.55 frames. ], batch size: 47, lr: 7.59e-03, grad_scale: 8.0
2022-12-23 08:35:49,081 WARNING [train.py:1060] (2/4) Exclude cut with ID 4234-40345-0022-142709-0 from training. Duration: 20.795
2022-12-23 08:35:55,729 INFO [zipformer.py:660] (2/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=52451.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 08:35:59,957 WARNING [train.py:1060] (2/4) Exclude cut with ID 7357-94126-0021-15852-0 from training. Duration: 24.76
2022-12-23 08:35:59,975 WARNING [train.py:1060] (2/4) Exclude cut with ID 3867-173237-0077-144769-0_sp0.9 from training. Duration: 22.25
2022-12-23 08:36:27,574 WARNING [train.py:1060] (2/4) Exclude cut with ID 7357-94126-0026-15857-0_sp1.1 from training. Duration: 20.5045625
2022-12-23 08:36:49,785 INFO [train.py:894] (2/4) Epoch 15, batch 3400, loss[loss=0.1564, simple_loss=0.2326, pruned_loss=0.04009, over 18476.00 frames. ], tot_loss[loss=0.1966, simple_loss=0.275, pruned_loss=0.05908, over 3715804.42 frames. ], batch size: 43, lr: 7.58e-03, grad_scale: 8.0
2022-12-23 08:37:57,115 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 3.148e+02 4.332e+02 5.107e+02 6.639e+02 1.156e+03, threshold=1.021e+03, percent-clipped=1.0
2022-12-23 08:38:02,848 INFO [train.py:894] (2/4) Epoch 15, batch 3450, loss[loss=0.1939, simple_loss=0.2768, pruned_loss=0.05552, over 18453.00 frames. ], tot_loss[loss=0.1971, simple_loss=0.2757, pruned_loss=0.0593, over 3715292.89 frames. ], batch size: 50, lr: 7.58e-03, grad_scale: 8.0
2022-12-23 08:38:20,360 INFO [zipformer.py:660] (2/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=52549.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 08:38:28,986 INFO [zipformer.py:660] (2/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=52555.0, num_to_drop=1, layers_to_drop={1}
2022-12-23 08:39:01,849 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.7146, 1.2287, 1.8958, 3.3107, 2.3417, 2.5481, 0.6181, 2.3734],
       device='cuda:2'), covar=tensor([0.1745, 0.1867, 0.1501, 0.0582, 0.1053, 0.1149, 0.2410, 0.1096],
       device='cuda:2'), in_proj_covar=tensor([0.0103, 0.0116, 0.0132, 0.0140, 0.0106, 0.0136, 0.0129, 0.0112],
       device='cuda:2'), out_proj_covar=tensor([0.0003, 0.0003, 0.0004, 0.0004, 0.0003, 0.0004, 0.0003, 0.0003],
       device='cuda:2')
2022-12-23 08:39:16,290 INFO [train.py:894] (2/4) Epoch 15, batch 3500, loss[loss=0.2253, simple_loss=0.3008, pruned_loss=0.07491, over 18628.00 frames. ], tot_loss[loss=0.198, simple_loss=0.2764, pruned_loss=0.05976, over 3715281.17 frames. ], batch size: 175, lr: 7.58e-03, grad_scale: 8.0
2022-12-23 08:39:21,531 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([2.2789, 1.9589, 2.2410, 2.4872, 2.3229, 3.2874, 1.9566, 2.0580],
       device='cuda:2'), covar=tensor([0.0799, 0.1476, 0.1151, 0.0778, 0.1113, 0.0355, 0.1175, 0.1224],
       device='cuda:2'), in_proj_covar=tensor([0.0074, 0.0082, 0.0074, 0.0074, 0.0091, 0.0075, 0.0085, 0.0078],
       device='cuda:2'), out_proj_covar=tensor([0.0003, 0.0004, 0.0004, 0.0003, 0.0004, 0.0003, 0.0004, 0.0004],
       device='cuda:2')
2022-12-23 08:39:37,367 WARNING [train.py:1060] (2/4) Exclude cut with ID 7859-102521-0017-7548-0_sp1.1 from training. Duration: 22.2954375
2022-12-23 08:39:47,481 INFO [train.py:894] (2/4) Epoch 16, batch 0, loss[loss=0.1889, simple_loss=0.2729, pruned_loss=0.05243, over 18581.00 frames. ], tot_loss[loss=0.1889, simple_loss=0.2729, pruned_loss=0.05243, over 18581.00 frames. ], batch size: 51, lr: 7.33e-03, grad_scale: 8.0
2022-12-23 08:39:47,481 INFO [train.py:919] (2/4) Computing validation loss
2022-12-23 08:39:58,260 INFO [train.py:928] (2/4) Epoch 16, validation: loss=0.1687, simple_loss=0.2674, pruned_loss=0.03497, over 944034.00 frames. 
2022-12-23 08:39:58,260 INFO [train.py:929] (2/4) Maximum memory allocated so far is 24680MB
2022-12-23 08:40:03,352 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.9220, 2.0910, 1.5320, 2.2432, 2.9189, 1.6528, 1.9328, 1.4936],
       device='cuda:2'), covar=tensor([0.1919, 0.1759, 0.1654, 0.1135, 0.1506, 0.1154, 0.1934, 0.1534],
       device='cuda:2'), in_proj_covar=tensor([0.0246, 0.0221, 0.0209, 0.0195, 0.0261, 0.0194, 0.0219, 0.0195],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-23 08:40:34,386 INFO [zipformer.py:660] (2/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=52616.0, num_to_drop=1, layers_to_drop={1}
2022-12-23 08:40:49,780 WARNING [train.py:1060] (2/4) Exclude cut with ID 298-126791-0067-24026-0_sp0.9 from training. Duration: 21.438875
2022-12-23 08:40:54,037 WARNING [train.py:1060] (2/4) Exclude cut with ID 5652-39938-0025-23684-0_sp0.9 from training. Duration: 22.2055625
2022-12-23 08:40:58,104 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 2.574e+02 3.834e+02 5.457e+02 6.710e+02 1.332e+03, threshold=1.091e+03, percent-clipped=4.0
2022-12-23 08:41:14,214 INFO [train.py:894] (2/4) Epoch 16, batch 50, loss[loss=0.1953, simple_loss=0.2855, pruned_loss=0.05255, over 18722.00 frames. ], tot_loss[loss=0.1865, simple_loss=0.2727, pruned_loss=0.05016, over 837272.01 frames. ], batch size: 54, lr: 7.33e-03, grad_scale: 8.0
2022-12-23 08:41:52,364 INFO [zipformer.py:660] (2/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=52668.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 08:42:29,861 INFO [train.py:894] (2/4) Epoch 16, batch 100, loss[loss=0.234, simple_loss=0.3131, pruned_loss=0.07747, over 18664.00 frames. ], tot_loss[loss=0.1829, simple_loss=0.2695, pruned_loss=0.04813, over 1475194.84 frames. ], batch size: 60, lr: 7.33e-03, grad_scale: 8.0
2022-12-23 08:42:34,309 INFO [zipformer.py:660] (2/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=52696.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 08:43:14,316 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.7579, 1.6337, 1.4491, 1.6986, 1.6545, 1.8632, 1.9698, 1.3957],
       device='cuda:2'), covar=tensor([0.0302, 0.0236, 0.0437, 0.0201, 0.0206, 0.0317, 0.0210, 0.0278],
       device='cuda:2'), in_proj_covar=tensor([0.0092, 0.0124, 0.0153, 0.0127, 0.0116, 0.0117, 0.0095, 0.0126],
       device='cuda:2'), out_proj_covar=tensor([7.4982e-05, 1.0037e-04, 1.2900e-04, 1.0263e-04, 9.5691e-05, 9.1853e-05,
        7.5210e-05, 1.0110e-04], device='cuda:2')
2022-12-23 08:43:29,093 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 2.318e+02 3.254e+02 4.108e+02 5.312e+02 8.636e+02, threshold=8.216e+02, percent-clipped=0.0
2022-12-23 08:43:35,879 INFO [scaling.py:679] (2/4) Whitening: num_groups=8, num_channels=192, metric=1.90 vs. limit=2.0
2022-12-23 08:43:43,853 INFO [train.py:894] (2/4) Epoch 16, batch 150, loss[loss=0.1793, simple_loss=0.269, pruned_loss=0.04477, over 18472.00 frames. ], tot_loss[loss=0.1819, simple_loss=0.2685, pruned_loss=0.04759, over 1971096.95 frames. ], batch size: 64, lr: 7.32e-03, grad_scale: 8.0
2022-12-23 08:43:45,433 INFO [zipformer.py:660] (2/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=52744.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 08:43:57,770 INFO [zipformer.py:660] (2/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=52752.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 08:43:58,904 WARNING [train.py:1060] (2/4) Exclude cut with ID 7859-102521-0017-7548-0 from training. Duration: 24.525
2022-12-23 08:44:32,750 WARNING [train.py:1060] (2/4) Exclude cut with ID 3699-47246-0007-3408-0_sp0.9 from training. Duration: 20.26675
2022-12-23 08:44:45,463 WARNING [train.py:1060] (2/4) Exclude cut with ID 7859-102521-0017-7548-0_sp0.9 from training. Duration: 27.25
2022-12-23 08:44:59,482 INFO [train.py:894] (2/4) Epoch 16, batch 200, loss[loss=0.1819, simple_loss=0.2767, pruned_loss=0.04358, over 18732.00 frames. ], tot_loss[loss=0.1822, simple_loss=0.2687, pruned_loss=0.04784, over 2357496.94 frames. ], batch size: 54, lr: 7.32e-03, grad_scale: 8.0
2022-12-23 08:45:31,786 INFO [zipformer.py:660] (2/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=52813.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 08:45:59,683 WARNING [train.py:1060] (2/4) Exclude cut with ID 6426-64292-0017-15984-0 from training. Duration: 21.68
2022-12-23 08:46:00,911 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 2.133e+02 3.642e+02 4.189e+02 4.973e+02 1.073e+03, threshold=8.378e+02, percent-clipped=4.0
2022-12-23 08:46:13,095 WARNING [train.py:1060] (2/4) Exclude cut with ID 4278-13270-0007-59342-0 from training. Duration: 21.6300625
2022-12-23 08:46:15,852 INFO [train.py:894] (2/4) Epoch 16, batch 250, loss[loss=0.1928, simple_loss=0.2822, pruned_loss=0.05172, over 18589.00 frames. ], tot_loss[loss=0.1817, simple_loss=0.2687, pruned_loss=0.04733, over 2659798.04 frames. ], batch size: 57, lr: 7.32e-03, grad_scale: 8.0
2022-12-23 08:46:24,618 INFO [scaling.py:679] (2/4) Whitening: num_groups=1, num_channels=384, metric=2.98 vs. limit=5.0
2022-12-23 08:46:25,650 INFO [zipformer.py:660] (2/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=52849.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 08:46:36,033 WARNING [train.py:1060] (2/4) Exclude cut with ID 4278-13270-0007-59342-0_sp0.9 from training. Duration: 24.033375
2022-12-23 08:47:17,613 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([0.5878, 1.9053, 2.1893, 1.0266, 1.4153, 2.3963, 2.0578, 1.8404],
       device='cuda:2'), covar=tensor([0.0748, 0.0344, 0.0314, 0.0400, 0.0362, 0.0393, 0.0227, 0.0654],
       device='cuda:2'), in_proj_covar=tensor([0.0149, 0.0164, 0.0121, 0.0138, 0.0146, 0.0140, 0.0156, 0.0164],
       device='cuda:2'), out_proj_covar=tensor([1.1571e-04, 1.2966e-04, 9.3993e-05, 1.0577e-04, 1.1311e-04, 1.1060e-04,
        1.2382e-04, 1.2821e-04], device='cuda:2')
2022-12-23 08:47:32,158 INFO [train.py:894] (2/4) Epoch 16, batch 300, loss[loss=0.1741, simple_loss=0.2762, pruned_loss=0.036, over 18392.00 frames. ], tot_loss[loss=0.1816, simple_loss=0.2683, pruned_loss=0.04744, over 2893436.24 frames. ], batch size: 53, lr: 7.31e-03, grad_scale: 8.0
2022-12-23 08:47:33,976 WARNING [train.py:1060] (2/4) Exclude cut with ID 4278-13270-0009-59344-0 from training. Duration: 22.905
2022-12-23 08:47:34,019 WARNING [train.py:1060] (2/4) Exclude cut with ID 5622-44585-0006-90525-0_sp1.1 from training. Duration: 23.4318125
2022-12-23 08:47:39,102 INFO [zipformer.py:660] (2/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=52897.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 08:47:46,650 INFO [scaling.py:679] (2/4) Whitening: num_groups=8, num_channels=96, metric=1.28 vs. limit=2.0
2022-12-23 08:48:00,880 INFO [zipformer.py:660] (2/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=52911.0, num_to_drop=1, layers_to_drop={1}
2022-12-23 08:48:32,000 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 2.517e+02 3.926e+02 4.602e+02 5.842e+02 9.674e+02, threshold=9.204e+02, percent-clipped=4.0
2022-12-23 08:48:38,530 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([2.4028, 2.5599, 2.8179, 1.4369, 3.0480, 2.8714, 1.9604, 3.4140],
       device='cuda:2'), covar=tensor([0.1347, 0.1652, 0.1563, 0.2385, 0.0759, 0.1321, 0.2199, 0.0502],
       device='cuda:2'), in_proj_covar=tensor([0.0196, 0.0203, 0.0206, 0.0191, 0.0174, 0.0213, 0.0212, 0.0187],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-23 08:48:47,088 INFO [train.py:894] (2/4) Epoch 16, batch 350, loss[loss=0.1604, simple_loss=0.2366, pruned_loss=0.04205, over 18700.00 frames. ], tot_loss[loss=0.1824, simple_loss=0.2689, pruned_loss=0.04799, over 3075754.30 frames. ], batch size: 46, lr: 7.31e-03, grad_scale: 8.0
2022-12-23 08:49:24,342 INFO [zipformer.py:660] (2/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=52968.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 08:49:29,922 WARNING [train.py:1060] (2/4) Exclude cut with ID 4278-13270-0009-59344-0_sp1.1 from training. Duration: 20.82275
2022-12-23 08:49:31,387 WARNING [train.py:1060] (2/4) Exclude cut with ID 4278-13270-0009-59344-0_sp0.9 from training. Duration: 25.45
2022-12-23 08:50:01,796 INFO [train.py:894] (2/4) Epoch 16, batch 400, loss[loss=0.207, simple_loss=0.3026, pruned_loss=0.05569, over 18649.00 frames. ], tot_loss[loss=0.1847, simple_loss=0.2713, pruned_loss=0.04909, over 3216860.39 frames. ], batch size: 53, lr: 7.30e-03, grad_scale: 8.0
2022-12-23 08:50:02,886 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([0.2126, 1.3374, 1.6313, 0.8046, 0.9633, 1.6907, 1.6746, 1.4176],
       device='cuda:2'), covar=tensor([0.0659, 0.0326, 0.0332, 0.0318, 0.0380, 0.0419, 0.0219, 0.0609],
       device='cuda:2'), in_proj_covar=tensor([0.0148, 0.0164, 0.0121, 0.0138, 0.0146, 0.0140, 0.0156, 0.0163],
       device='cuda:2'), out_proj_covar=tensor([1.1552e-04, 1.2936e-04, 9.3792e-05, 1.0549e-04, 1.1279e-04, 1.1072e-04,
        1.2387e-04, 1.2802e-04], device='cuda:2')
2022-12-23 08:50:24,441 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.9018, 1.8831, 2.2426, 1.1189, 2.3419, 2.1779, 1.6279, 2.6594],
       device='cuda:2'), covar=tensor([0.1266, 0.1629, 0.1292, 0.1944, 0.0648, 0.1177, 0.1977, 0.0442],
       device='cuda:2'), in_proj_covar=tensor([0.0194, 0.0202, 0.0204, 0.0189, 0.0172, 0.0211, 0.0210, 0.0185],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-23 08:50:32,604 WARNING [train.py:1060] (2/4) Exclude cut with ID 5622-44585-0006-90525-0 from training. Duration: 25.775
2022-12-23 08:50:38,330 INFO [zipformer.py:660] (2/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=53016.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 08:50:55,804 WARNING [train.py:1060] (2/4) Exclude cut with ID 2929-85685-0071-62375-0_sp0.9 from training. Duration: 22.25
2022-12-23 08:51:03,193 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 1.768e+02 3.604e+02 4.680e+02 6.020e+02 1.897e+03, threshold=9.359e+02, percent-clipped=5.0
2022-12-23 08:51:17,416 INFO [train.py:894] (2/4) Epoch 16, batch 450, loss[loss=0.1987, simple_loss=0.2895, pruned_loss=0.05395, over 18612.00 frames. ], tot_loss[loss=0.1851, simple_loss=0.2721, pruned_loss=0.049, over 3327706.68 frames. ], batch size: 98, lr: 7.30e-03, grad_scale: 8.0
2022-12-23 08:51:21,207 WARNING [train.py:1060] (2/4) Exclude cut with ID 3972-170212-0014-23379-0 from training. Duration: 26.205
2022-12-23 08:51:38,098 WARNING [train.py:1060] (2/4) Exclude cut with ID 5239-32139-0047-9341-0_sp0.9 from training. Duration: 30.1555625
2022-12-23 08:51:38,407 INFO [zipformer.py:660] (2/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=53056.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 08:51:42,461 WARNING [train.py:1060] (2/4) Exclude cut with ID 1265-135635-0050-6781-0_sp0.9 from training. Duration: 21.8333125
2022-12-23 08:51:52,830 WARNING [train.py:1060] (2/4) Exclude cut with ID 1914-133440-0024-94914-0_sp1.1 from training. Duration: 20.6545625
2022-12-23 08:52:32,573 WARNING [train.py:1060] (2/4) Exclude cut with ID 7395-89880-0045-39920-0_sp0.9 from training. Duration: 20.52225
2022-12-23 08:52:35,690 INFO [train.py:894] (2/4) Epoch 16, batch 500, loss[loss=0.1656, simple_loss=0.2575, pruned_loss=0.03688, over 18529.00 frames. ], tot_loss[loss=0.1852, simple_loss=0.2724, pruned_loss=0.04905, over 3412531.15 frames. ], batch size: 55, lr: 7.30e-03, grad_scale: 8.0
2022-12-23 08:52:39,144 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.9096, 1.4240, 1.4471, 1.8885, 1.5792, 3.4333, 1.2688, 1.3923],
       device='cuda:2'), covar=tensor([0.0804, 0.1878, 0.1175, 0.0935, 0.1524, 0.0234, 0.1468, 0.1615],
       device='cuda:2'), in_proj_covar=tensor([0.0073, 0.0082, 0.0073, 0.0073, 0.0090, 0.0075, 0.0084, 0.0077],
       device='cuda:2'), out_proj_covar=tensor([0.0003, 0.0004, 0.0004, 0.0003, 0.0004, 0.0003, 0.0004, 0.0004],
       device='cuda:2')
2022-12-23 08:52:52,269 WARNING [train.py:1060] (2/4) Exclude cut with ID 3972-170212-0014-23379-0_sp0.9 from training. Duration: 29.1166875
2022-12-23 08:52:57,938 INFO [zipformer.py:660] (2/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=53108.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 08:53:11,390 INFO [zipformer.py:660] (2/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=53117.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 08:53:36,074 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 2.439e+02 3.611e+02 4.573e+02 5.633e+02 1.501e+03, threshold=9.146e+02, percent-clipped=5.0
2022-12-23 08:53:41,768 INFO [scaling.py:679] (2/4) Whitening: num_groups=8, num_channels=96, metric=1.31 vs. limit=2.0
2022-12-23 08:53:52,522 INFO [train.py:894] (2/4) Epoch 16, batch 550, loss[loss=0.2028, simple_loss=0.2949, pruned_loss=0.05534, over 18675.00 frames. ], tot_loss[loss=0.185, simple_loss=0.2716, pruned_loss=0.04916, over 3478965.76 frames. ], batch size: 60, lr: 7.29e-03, grad_scale: 8.0
2022-12-23 08:53:55,460 WARNING [train.py:1060] (2/4) Exclude cut with ID 543-133211-0007-59831-0_sp0.9 from training. Duration: 21.388875
2022-12-23 08:54:32,132 WARNING [train.py:1060] (2/4) Exclude cut with ID 1914-133440-0024-94914-0 from training. Duration: 22.72
2022-12-23 08:54:32,182 WARNING [train.py:1060] (2/4) Exclude cut with ID 1914-133440-0031-94921-0_sp0.9 from training. Duration: 22.7444375
2022-12-23 08:55:08,111 INFO [train.py:894] (2/4) Epoch 16, batch 600, loss[loss=0.1966, simple_loss=0.2892, pruned_loss=0.05199, over 18492.00 frames. ], tot_loss[loss=0.1856, simple_loss=0.2723, pruned_loss=0.04945, over 3529572.60 frames. ], batch size: 52, lr: 7.29e-03, grad_scale: 8.0
2022-12-23 08:55:15,480 WARNING [train.py:1060] (2/4) Exclude cut with ID 4133-6541-0027-40495-0_sp1.1 from training. Duration: 0.9681875
2022-12-23 08:55:18,574 WARNING [train.py:1060] (2/4) Exclude cut with ID 6330-62851-0022-91297-0_sp0.9 from training. Duration: 22.3166875
2022-12-23 08:55:24,776 WARNING [train.py:1060] (2/4) Exclude cut with ID 543-133212-0015-59917-0_sp0.9 from training. Duration: 21.8166875
2022-12-23 08:55:34,989 INFO [zipformer.py:660] (2/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=53211.0, num_to_drop=1, layers_to_drop={1}
2022-12-23 08:55:41,392 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.7115, 1.6865, 1.2118, 1.5829, 1.7857, 1.5478, 2.1128, 1.7655],
       device='cuda:2'), covar=tensor([0.0959, 0.1768, 0.2924, 0.1885, 0.1885, 0.0994, 0.1065, 0.1266],
       device='cuda:2'), in_proj_covar=tensor([0.0174, 0.0201, 0.0241, 0.0285, 0.0227, 0.0185, 0.0204, 0.0198],
       device='cuda:2'), out_proj_covar=tensor([0.0001, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-23 08:55:50,203 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([2.3278, 1.9875, 1.9007, 1.1741, 2.5540, 2.3039, 2.2474, 1.6025],
       device='cuda:2'), covar=tensor([0.0343, 0.0393, 0.0477, 0.0721, 0.0231, 0.0350, 0.0367, 0.0777],
       device='cuda:2'), in_proj_covar=tensor([0.0123, 0.0123, 0.0128, 0.0118, 0.0092, 0.0121, 0.0133, 0.0152],
       device='cuda:2'), out_proj_covar=tensor([0.0001, 0.0002, 0.0002, 0.0001, 0.0001, 0.0001, 0.0002, 0.0002],
       device='cuda:2')
2022-12-23 08:56:07,402 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 1.754e+02 3.712e+02 4.400e+02 5.395e+02 1.040e+03, threshold=8.800e+02, percent-clipped=2.0
2022-12-23 08:56:07,828 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.5713, 1.4453, 1.4300, 0.8086, 1.8071, 1.5167, 1.5262, 1.2886],
       device='cuda:2'), covar=tensor([0.0353, 0.0447, 0.0459, 0.0706, 0.0335, 0.0368, 0.0420, 0.0851],
       device='cuda:2'), in_proj_covar=tensor([0.0123, 0.0123, 0.0128, 0.0119, 0.0092, 0.0121, 0.0133, 0.0152],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0001, 0.0001, 0.0001, 0.0002, 0.0002],
       device='cuda:2')
2022-12-23 08:56:16,514 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.5768, 1.1267, 0.7231, 1.1590, 2.1118, 0.6219, 1.3062, 1.4624],
       device='cuda:2'), covar=tensor([0.1598, 0.2052, 0.2115, 0.1609, 0.1600, 0.1806, 0.1424, 0.1675],
       device='cuda:2'), in_proj_covar=tensor([0.0093, 0.0098, 0.0118, 0.0096, 0.0114, 0.0091, 0.0098, 0.0094],
       device='cuda:2'), out_proj_covar=tensor([0.0003, 0.0003, 0.0004, 0.0003, 0.0004, 0.0003, 0.0003, 0.0003],
       device='cuda:2')
2022-12-23 08:56:18,668 INFO [scaling.py:679] (2/4) Whitening: num_groups=8, num_channels=96, metric=1.29 vs. limit=2.0
2022-12-23 08:56:21,011 INFO [zipformer.py:660] (2/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=53241.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 08:56:23,459 INFO [train.py:894] (2/4) Epoch 16, batch 650, loss[loss=0.1675, simple_loss=0.2499, pruned_loss=0.04254, over 18444.00 frames. ], tot_loss[loss=0.1858, simple_loss=0.2723, pruned_loss=0.0496, over 3571060.32 frames. ], batch size: 43, lr: 7.29e-03, grad_scale: 8.0
2022-12-23 08:56:46,960 INFO [zipformer.py:660] (2/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=53259.0, num_to_drop=1, layers_to_drop={0}
2022-12-23 08:57:07,898 WARNING [train.py:1060] (2/4) Exclude cut with ID 4957-30119-0041-23990-0_sp0.9 from training. Duration: 20.22775
2022-12-23 08:57:38,501 INFO [train.py:894] (2/4) Epoch 16, batch 700, loss[loss=0.1508, simple_loss=0.2309, pruned_loss=0.03536, over 18478.00 frames. ], tot_loss[loss=0.1849, simple_loss=0.2714, pruned_loss=0.04922, over 3602541.17 frames. ], batch size: 43, lr: 7.28e-03, grad_scale: 8.0
2022-12-23 08:57:44,582 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([2.2719, 2.5572, 2.7766, 1.2638, 2.9823, 2.8778, 2.0873, 3.4945],
       device='cuda:2'), covar=tensor([0.1357, 0.1674, 0.1706, 0.2364, 0.0817, 0.1269, 0.2157, 0.0563],
       device='cuda:2'), in_proj_covar=tensor([0.0196, 0.0204, 0.0205, 0.0190, 0.0174, 0.0214, 0.0211, 0.0186],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-23 08:57:52,122 INFO [zipformer.py:660] (2/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=53302.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 08:57:53,240 WARNING [train.py:1060] (2/4) Exclude cut with ID 5239-32139-0047-9341-0_sp1.1 from training. Duration: 24.67275
2022-12-23 08:57:54,959 INFO [zipformer.py:660] (2/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=53304.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 08:58:12,002 INFO [zipformer.py:660] (2/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=53315.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 08:58:21,004 WARNING [train.py:1060] (2/4) Exclude cut with ID 3082-165428-0081-50734-0_sp0.9 from training. Duration: 21.8055625
2022-12-23 08:58:39,335 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 2.032e+02 3.701e+02 4.471e+02 5.554e+02 1.139e+03, threshold=8.941e+02, percent-clipped=3.0
2022-12-23 08:58:54,396 INFO [train.py:894] (2/4) Epoch 16, batch 750, loss[loss=0.1721, simple_loss=0.2611, pruned_loss=0.04152, over 18479.00 frames. ], tot_loss[loss=0.1855, simple_loss=0.2723, pruned_loss=0.04937, over 3627902.36 frames. ], batch size: 50, lr: 7.28e-03, grad_scale: 8.0
2022-12-23 08:59:00,117 WARNING [train.py:1060] (2/4) Exclude cut with ID 3340-169293-0054-76830-0_sp0.9 from training. Duration: 22.6666875
2022-12-23 08:59:26,718 INFO [zipformer.py:660] (2/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=53365.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 08:59:29,841 INFO [zipformer.py:660] (2/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=53367.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 08:59:43,818 INFO [zipformer.py:660] (2/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=53376.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 09:00:01,935 WARNING [train.py:1060] (2/4) Exclude cut with ID 2411-132532-0017-82279-0_sp1.1 from training. Duration: 0.9681875
2022-12-23 09:00:05,303 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([2.8460, 2.2971, 1.7585, 2.6633, 2.0830, 2.1723, 2.1422, 2.9382],
       device='cuda:2'), covar=tensor([0.1707, 0.3005, 0.1812, 0.2548, 0.3437, 0.0975, 0.2884, 0.0671],
       device='cuda:2'), in_proj_covar=tensor([0.0286, 0.0276, 0.0233, 0.0340, 0.0256, 0.0216, 0.0271, 0.0200],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0003, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-23 09:00:09,100 INFO [train.py:894] (2/4) Epoch 16, batch 800, loss[loss=0.2086, simple_loss=0.293, pruned_loss=0.06209, over 18490.00 frames. ], tot_loss[loss=0.1858, simple_loss=0.2726, pruned_loss=0.04948, over 3646810.95 frames. ], batch size: 52, lr: 7.28e-03, grad_scale: 8.0
2022-12-23 09:00:27,205 WARNING [train.py:1060] (2/4) Exclude cut with ID 6330-62850-0007-91323-0 from training. Duration: 22.485
2022-12-23 09:00:32,064 INFO [zipformer.py:660] (2/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=53408.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 09:00:38,168 INFO [zipformer.py:660] (2/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=53412.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 09:01:02,649 INFO [zipformer.py:660] (2/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=53428.0, num_to_drop=1, layers_to_drop={0}
2022-12-23 09:01:05,573 WARNING [train.py:1060] (2/4) Exclude cut with ID 3972-170212-0014-23379-0_sp1.1 from training. Duration: 23.82275
2022-12-23 09:01:10,796 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 2.669e+02 3.732e+02 4.519e+02 5.573e+02 1.568e+03, threshold=9.039e+02, percent-clipped=4.0
2022-12-23 09:01:18,547 WARNING [train.py:1060] (2/4) Exclude cut with ID 4860-13185-0032-76709-0 from training. Duration: 20.77
2022-12-23 09:01:25,724 INFO [train.py:894] (2/4) Epoch 16, batch 850, loss[loss=0.1805, simple_loss=0.2721, pruned_loss=0.0445, over 18620.00 frames. ], tot_loss[loss=0.1854, simple_loss=0.2723, pruned_loss=0.0492, over 3661169.15 frames. ], batch size: 53, lr: 7.27e-03, grad_scale: 8.0
2022-12-23 09:01:27,228 WARNING [train.py:1060] (2/4) Exclude cut with ID 6426-64292-0017-15984-0_sp0.9 from training. Duration: 24.088875
2022-12-23 09:01:44,879 INFO [zipformer.py:660] (2/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=53456.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 09:01:55,771 WARNING [train.py:1060] (2/4) Exclude cut with ID 6330-62850-0007-91323-0_sp1.1 from training. Duration: 20.4409375
2022-12-23 09:02:41,478 INFO [train.py:894] (2/4) Epoch 16, batch 900, loss[loss=0.1649, simple_loss=0.2494, pruned_loss=0.0402, over 18394.00 frames. ], tot_loss[loss=0.1863, simple_loss=0.273, pruned_loss=0.04977, over 3673029.15 frames. ], batch size: 46, lr: 7.27e-03, grad_scale: 8.0
2022-12-23 09:03:12,654 WARNING [train.py:1060] (2/4) Exclude cut with ID 2929-85685-0062-62366-0_sp0.9 from training. Duration: 22.511125
2022-12-23 09:03:12,680 WARNING [train.py:1060] (2/4) Exclude cut with ID 7395-89880-0031-39906-0 from training. Duration: 20.675
2022-12-23 09:03:32,869 INFO [scaling.py:679] (2/4) Whitening: num_groups=8, num_channels=192, metric=1.70 vs. limit=2.0
2022-12-23 09:03:42,940 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 2.459e+02 3.846e+02 4.469e+02 5.428e+02 1.357e+03, threshold=8.938e+02, percent-clipped=3.0
2022-12-23 09:03:54,212 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.7652, 0.6612, 1.5252, 1.3911, 1.7548, 1.8150, 1.4233, 1.4422],
       device='cuda:2'), covar=tensor([0.1979, 0.2997, 0.2516, 0.2394, 0.1895, 0.0995, 0.2681, 0.1292],
       device='cuda:2'), in_proj_covar=tensor([0.0258, 0.0289, 0.0268, 0.0301, 0.0291, 0.0243, 0.0323, 0.0230],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-23 09:03:57,835 INFO [train.py:894] (2/4) Epoch 16, batch 950, loss[loss=0.2059, simple_loss=0.2937, pruned_loss=0.05903, over 18522.00 frames. ], tot_loss[loss=0.1858, simple_loss=0.2724, pruned_loss=0.04962, over 3681623.01 frames. ], batch size: 58, lr: 7.27e-03, grad_scale: 8.0
2022-12-23 09:04:47,701 INFO [scaling.py:679] (2/4) Whitening: num_groups=8, num_channels=192, metric=1.84 vs. limit=2.0
2022-12-23 09:04:50,913 WARNING [train.py:1060] (2/4) Exclude cut with ID 6330-62850-0007-91323-0_sp0.9 from training. Duration: 24.9833125
2022-12-23 09:05:12,798 INFO [train.py:894] (2/4) Epoch 16, batch 1000, loss[loss=0.1688, simple_loss=0.2607, pruned_loss=0.03847, over 18493.00 frames. ], tot_loss[loss=0.1855, simple_loss=0.2722, pruned_loss=0.0494, over 3689252.20 frames. ], batch size: 54, lr: 7.26e-03, grad_scale: 8.0
2022-12-23 09:05:18,867 INFO [zipformer.py:660] (2/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=53597.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 09:05:21,558 WARNING [train.py:1060] (2/4) Exclude cut with ID 5239-32139-0047-9341-0 from training. Duration: 27.14
2022-12-23 09:05:39,095 WARNING [train.py:1060] (2/4) Exclude cut with ID 2929-85685-0044-62348-0 from training. Duration: 22.44
2022-12-23 09:06:11,973 INFO [scaling.py:679] (2/4) Whitening: num_groups=8, num_channels=96, metric=1.23 vs. limit=2.0
2022-12-23 09:06:14,007 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 2.372e+02 3.452e+02 4.205e+02 5.209e+02 9.505e+02, threshold=8.411e+02, percent-clipped=1.0
2022-12-23 09:06:28,586 INFO [train.py:894] (2/4) Epoch 16, batch 1050, loss[loss=0.1777, simple_loss=0.2559, pruned_loss=0.04969, over 18399.00 frames. ], tot_loss[loss=0.1852, simple_loss=0.272, pruned_loss=0.04925, over 3694647.19 frames. ], batch size: 46, lr: 7.26e-03, grad_scale: 8.0
2022-12-23 09:06:54,323 INFO [scaling.py:679] (2/4) Whitening: num_groups=8, num_channels=192, metric=1.87 vs. limit=2.0
2022-12-23 09:06:55,014 INFO [zipformer.py:660] (2/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=53660.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 09:07:00,259 WARNING [train.py:1060] (2/4) Exclude cut with ID 2929-85685-0060-62364-0_sp0.9 from training. Duration: 21.361125
2022-12-23 09:07:06,373 WARNING [train.py:1060] (2/4) Exclude cut with ID 2929-85685-0079-62383-0_sp1.1 from training. Duration: 27.0318125
2022-12-23 09:07:11,335 INFO [zipformer.py:660] (2/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=53671.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 09:07:12,961 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.6286, 1.5160, 1.6217, 1.6338, 1.2532, 3.6181, 1.5399, 1.9664],
       device='cuda:2'), covar=tensor([0.3213, 0.2142, 0.1937, 0.2011, 0.1445, 0.0174, 0.1631, 0.0905],
       device='cuda:2'), in_proj_covar=tensor([0.0135, 0.0117, 0.0126, 0.0121, 0.0103, 0.0097, 0.0092, 0.0090],
       device='cuda:2'), out_proj_covar=tensor([0.0005, 0.0005, 0.0005, 0.0005, 0.0004, 0.0003, 0.0004, 0.0004],
       device='cuda:2')
2022-12-23 09:07:16,718 WARNING [train.py:1060] (2/4) Exclude cut with ID 5622-44585-0006-90525-0_sp0.9 from training. Duration: 28.638875
2022-12-23 09:07:29,325 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.8551, 1.6308, 1.6536, 2.2779, 2.0172, 4.4296, 1.5822, 1.8598],
       device='cuda:2'), covar=tensor([0.0781, 0.1686, 0.1164, 0.0872, 0.1353, 0.0167, 0.1242, 0.1395],
       device='cuda:2'), in_proj_covar=tensor([0.0072, 0.0081, 0.0072, 0.0073, 0.0089, 0.0074, 0.0083, 0.0076],
       device='cuda:2'), out_proj_covar=tensor([0.0003, 0.0004, 0.0004, 0.0003, 0.0004, 0.0003, 0.0004, 0.0004],
       device='cuda:2')
2022-12-23 09:07:33,437 WARNING [train.py:1060] (2/4) Exclude cut with ID 3340-169293-0054-76830-0 from training. Duration: 20.4
2022-12-23 09:07:43,949 INFO [train.py:894] (2/4) Epoch 16, batch 1100, loss[loss=0.2078, simple_loss=0.2917, pruned_loss=0.06199, over 18587.00 frames. ], tot_loss[loss=0.1851, simple_loss=0.2717, pruned_loss=0.04919, over 3698405.28 frames. ], batch size: 172, lr: 7.26e-03, grad_scale: 8.0
2022-12-23 09:08:05,612 WARNING [train.py:1060] (2/4) Exclude cut with ID 2929-85685-0071-62375-0 from training. Duration: 20.025
2022-12-23 09:08:05,622 WARNING [train.py:1060] (2/4) Exclude cut with ID 2364-131735-0112-64612-0_sp0.9 from training. Duration: 20.488875
2022-12-23 09:08:11,765 WARNING [train.py:1060] (2/4) Exclude cut with ID 2929-85685-0079-62383-0 from training. Duration: 29.735
2022-12-23 09:08:13,738 INFO [zipformer.py:660] (2/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=53712.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 09:08:30,228 INFO [zipformer.py:660] (2/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=53723.0, num_to_drop=1, layers_to_drop={1}
2022-12-23 09:08:45,994 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 2.409e+02 3.829e+02 4.383e+02 5.525e+02 1.392e+03, threshold=8.765e+02, percent-clipped=4.0
2022-12-23 09:08:56,984 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([2.0877, 2.0156, 2.2175, 2.0875, 1.8068, 3.5731, 2.1071, 2.5706],
       device='cuda:2'), covar=tensor([0.2500, 0.1696, 0.1404, 0.1651, 0.1120, 0.0182, 0.1605, 0.0679],
       device='cuda:2'), in_proj_covar=tensor([0.0134, 0.0116, 0.0125, 0.0120, 0.0103, 0.0097, 0.0092, 0.0089],
       device='cuda:2'), out_proj_covar=tensor([0.0005, 0.0005, 0.0005, 0.0005, 0.0004, 0.0003, 0.0004, 0.0004],
       device='cuda:2')
2022-12-23 09:08:57,041 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([2.3129, 2.0242, 2.4429, 1.7624, 2.4896, 2.5035, 1.9143, 2.6977],
       device='cuda:2'), covar=tensor([0.0858, 0.1318, 0.1102, 0.1478, 0.0520, 0.0790, 0.1671, 0.0385],
       device='cuda:2'), in_proj_covar=tensor([0.0197, 0.0203, 0.0203, 0.0192, 0.0175, 0.0213, 0.0212, 0.0187],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-23 09:09:00,846 INFO [train.py:894] (2/4) Epoch 16, batch 1150, loss[loss=0.1904, simple_loss=0.288, pruned_loss=0.04646, over 18615.00 frames. ], tot_loss[loss=0.1859, simple_loss=0.2726, pruned_loss=0.04958, over 3701491.64 frames. ], batch size: 56, lr: 7.25e-03, grad_scale: 8.0
2022-12-23 09:09:26,181 INFO [scaling.py:679] (2/4) Whitening: num_groups=8, num_channels=96, metric=1.34 vs. limit=2.0
2022-12-23 09:09:27,021 INFO [zipformer.py:660] (2/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=53760.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 09:09:31,756 WARNING [train.py:1060] (2/4) Exclude cut with ID 7276-92427-0014-12983-0_sp0.9 from training. Duration: 21.3055625
2022-12-23 09:09:34,709 WARNING [train.py:1060] (2/4) Exclude cut with ID 1025-75365-0008-79168-0_sp0.9 from training. Duration: 22.0666875
2022-12-23 09:10:16,301 INFO [train.py:894] (2/4) Epoch 16, batch 1200, loss[loss=0.1811, simple_loss=0.2666, pruned_loss=0.04784, over 18605.00 frames. ], tot_loss[loss=0.1849, simple_loss=0.2719, pruned_loss=0.04901, over 3705486.01 frames. ], batch size: 45, lr: 7.25e-03, grad_scale: 8.0
2022-12-23 09:10:37,543 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.5915, 1.8834, 1.5257, 2.2578, 2.4527, 1.5244, 1.3231, 1.3657],
       device='cuda:2'), covar=tensor([0.1853, 0.1700, 0.1523, 0.0924, 0.1137, 0.1172, 0.2139, 0.1491],
       device='cuda:2'), in_proj_covar=tensor([0.0241, 0.0218, 0.0207, 0.0193, 0.0254, 0.0193, 0.0216, 0.0194],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-23 09:10:49,373 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([3.0887, 5.3938, 4.7350, 2.3367, 5.3973, 4.0483, 0.8660, 3.1266],
       device='cuda:2'), covar=tensor([0.1857, 0.0836, 0.1110, 0.3222, 0.0541, 0.0756, 0.5144, 0.1477],
       device='cuda:2'), in_proj_covar=tensor([0.0138, 0.0131, 0.0152, 0.0121, 0.0133, 0.0109, 0.0142, 0.0112],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0002, 0.0003, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-23 09:11:16,021 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 2.339e+02 3.426e+02 4.196e+02 5.027e+02 8.544e+02, threshold=8.392e+02, percent-clipped=0.0
2022-12-23 09:11:25,034 WARNING [train.py:1060] (2/4) Exclude cut with ID 2929-85685-0062-62366-0 from training. Duration: 20.26
2022-12-23 09:11:31,014 INFO [train.py:894] (2/4) Epoch 16, batch 1250, loss[loss=0.179, simple_loss=0.2653, pruned_loss=0.04642, over 18437.00 frames. ], tot_loss[loss=0.1833, simple_loss=0.2704, pruned_loss=0.04804, over 3707294.56 frames. ], batch size: 50, lr: 7.25e-03, grad_scale: 8.0
2022-12-23 09:11:37,613 WARNING [train.py:1060] (2/4) Exclude cut with ID 5239-32139-0030-9324-0_sp0.9 from training. Duration: 21.3444375
2022-12-23 09:11:54,206 INFO [scaling.py:679] (2/4) Whitening: num_groups=8, num_channels=96, metric=1.43 vs. limit=2.0
2022-12-23 09:12:33,854 WARNING [train.py:1060] (2/4) Exclude cut with ID 497-129325-0061-62254-0_sp1.1 from training. Duration: 0.97725
2022-12-23 09:12:49,449 INFO [train.py:894] (2/4) Epoch 16, batch 1300, loss[loss=0.1844, simple_loss=0.2796, pruned_loss=0.04461, over 18628.00 frames. ], tot_loss[loss=0.1837, simple_loss=0.271, pruned_loss=0.04824, over 3709266.22 frames. ], batch size: 69, lr: 7.24e-03, grad_scale: 8.0
2022-12-23 09:12:55,855 INFO [zipformer.py:660] (2/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=53897.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 09:13:14,674 WARNING [train.py:1060] (2/4) Exclude cut with ID 7395-89880-0031-39906-0_sp0.9 from training. Duration: 22.97225
2022-12-23 09:13:43,602 WARNING [train.py:1060] (2/4) Exclude cut with ID 7395-89880-0047-39922-0_sp0.9 from training. Duration: 21.97775
2022-12-23 09:13:48,129 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 1.926e+02 3.573e+02 4.453e+02 5.683e+02 9.534e+02, threshold=8.906e+02, percent-clipped=1.0
2022-12-23 09:13:48,546 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([2.6338, 2.3232, 1.7897, 0.6990, 1.6330, 2.0572, 1.6845, 1.9980],
       device='cuda:2'), covar=tensor([0.0608, 0.0529, 0.1249, 0.1661, 0.1343, 0.1428, 0.1549, 0.0810],
       device='cuda:2'), in_proj_covar=tensor([0.0170, 0.0184, 0.0204, 0.0191, 0.0209, 0.0198, 0.0212, 0.0201],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-23 09:13:57,553 WARNING [train.py:1060] (2/4) Exclude cut with ID 1112-1043-0006-89194-0_sp0.9 from training. Duration: 21.8333125
2022-12-23 09:14:04,500 INFO [train.py:894] (2/4) Epoch 16, batch 1350, loss[loss=0.2095, simple_loss=0.2994, pruned_loss=0.05981, over 18617.00 frames. ], tot_loss[loss=0.1842, simple_loss=0.2715, pruned_loss=0.04841, over 3711083.79 frames. ], batch size: 53, lr: 7.24e-03, grad_scale: 16.0
2022-12-23 09:14:07,595 INFO [zipformer.py:660] (2/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=53945.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 09:14:07,857 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.9169, 1.8112, 2.0317, 1.1687, 2.2400, 2.1398, 1.4363, 2.4981],
       device='cuda:2'), covar=tensor([0.1158, 0.1769, 0.1336, 0.1922, 0.0670, 0.1085, 0.2395, 0.0589],
       device='cuda:2'), in_proj_covar=tensor([0.0195, 0.0201, 0.0202, 0.0189, 0.0173, 0.0212, 0.0210, 0.0186],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-23 09:14:09,047 WARNING [train.py:1060] (2/4) Exclude cut with ID 1914-133440-0031-94921-0 from training. Duration: 20.47
2022-12-23 09:14:30,042 INFO [zipformer.py:660] (2/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=53960.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 09:14:46,769 INFO [zipformer.py:660] (2/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=53971.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 09:14:56,796 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.6009, 1.0470, 0.6772, 1.1000, 2.0252, 0.5576, 1.1692, 1.3939],
       device='cuda:2'), covar=tensor([0.1382, 0.2094, 0.2062, 0.1580, 0.1605, 0.1728, 0.1520, 0.1600],
       device='cuda:2'), in_proj_covar=tensor([0.0092, 0.0098, 0.0117, 0.0096, 0.0115, 0.0091, 0.0098, 0.0094],
       device='cuda:2'), out_proj_covar=tensor([0.0003, 0.0003, 0.0004, 0.0003, 0.0004, 0.0003, 0.0003, 0.0003],
       device='cuda:2')
2022-12-23 09:15:16,445 WARNING [train.py:1060] (2/4) Exclude cut with ID 7395-89880-0037-39912-0_sp0.9 from training. Duration: 20.67225
2022-12-23 09:15:19,737 INFO [train.py:894] (2/4) Epoch 16, batch 1400, loss[loss=0.1943, simple_loss=0.2831, pruned_loss=0.05273, over 18714.00 frames. ], tot_loss[loss=0.1841, simple_loss=0.2711, pruned_loss=0.04851, over 3711073.74 frames. ], batch size: 52, lr: 7.24e-03, grad_scale: 16.0
2022-12-23 09:15:39,434 WARNING [train.py:1060] (2/4) Exclude cut with ID 1914-133440-0024-94914-0_sp0.9 from training. Duration: 25.2444375
2022-12-23 09:15:45,864 INFO [zipformer.py:660] (2/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=54008.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 09:16:01,811 INFO [zipformer.py:660] (2/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=54019.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 09:16:03,166 WARNING [train.py:1060] (2/4) Exclude cut with ID 3340-169293-0021-76797-0_sp0.9 from training. Duration: 21.1445
2022-12-23 09:16:08,382 INFO [zipformer.py:660] (2/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=54023.0, num_to_drop=1, layers_to_drop={0}
2022-12-23 09:16:22,367 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 2.191e+02 3.783e+02 4.326e+02 5.287e+02 1.206e+03, threshold=8.652e+02, percent-clipped=5.0
2022-12-23 09:16:28,611 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([2.7323, 2.2776, 1.6907, 2.4872, 1.9878, 2.1021, 2.1438, 2.6783],
       device='cuda:2'), covar=tensor([0.1752, 0.2839, 0.1756, 0.2575, 0.3308, 0.0953, 0.2572, 0.0792],
       device='cuda:2'), in_proj_covar=tensor([0.0288, 0.0281, 0.0236, 0.0347, 0.0260, 0.0221, 0.0278, 0.0203],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0003, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-23 09:16:38,044 INFO [train.py:894] (2/4) Epoch 16, batch 1450, loss[loss=0.199, simple_loss=0.2858, pruned_loss=0.0561, over 18718.00 frames. ], tot_loss[loss=0.1846, simple_loss=0.2714, pruned_loss=0.04889, over 3712403.94 frames. ], batch size: 54, lr: 7.23e-03, grad_scale: 16.0
2022-12-23 09:16:55,574 INFO [scaling.py:679] (2/4) Whitening: num_groups=8, num_channels=96, metric=1.34 vs. limit=2.0
2022-12-23 09:17:13,731 WARNING [train.py:1060] (2/4) Exclude cut with ID 2929-85685-0079-62383-0_sp0.9 from training. Duration: 33.038875
2022-12-23 09:17:20,269 INFO [zipformer.py:660] (2/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=54071.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 09:17:36,107 INFO [scaling.py:679] (2/4) Whitening: num_groups=8, num_channels=192, metric=1.90 vs. limit=2.0
2022-12-23 09:17:52,841 INFO [train.py:894] (2/4) Epoch 16, batch 1500, loss[loss=0.1853, simple_loss=0.2788, pruned_loss=0.04594, over 18723.00 frames. ], tot_loss[loss=0.1851, simple_loss=0.2719, pruned_loss=0.0491, over 3713718.59 frames. ], batch size: 54, lr: 7.23e-03, grad_scale: 16.0
2022-12-23 09:17:52,958 WARNING [train.py:1060] (2/4) Exclude cut with ID 6426-64291-0000-16059-0_sp0.9 from training. Duration: 20.0944375
2022-12-23 09:18:07,812 WARNING [train.py:1060] (2/4) Exclude cut with ID 2929-85685-0044-62348-0_sp1.1 from training. Duration: 20.4
2022-12-23 09:18:15,988 WARNING [train.py:1060] (2/4) Exclude cut with ID 6330-62851-0022-91297-0 from training. Duration: 20.085
2022-12-23 09:18:20,729 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([2.2273, 1.8708, 1.8745, 1.3497, 2.5183, 2.2906, 2.1158, 1.1964],
       device='cuda:2'), covar=tensor([0.0391, 0.0574, 0.0551, 0.0872, 0.0273, 0.0435, 0.0515, 0.1507],
       device='cuda:2'), in_proj_covar=tensor([0.0125, 0.0127, 0.0131, 0.0122, 0.0096, 0.0125, 0.0136, 0.0157],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0001, 0.0001, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-23 09:18:26,436 WARNING [train.py:1060] (2/4) Exclude cut with ID 4860-13185-0032-76709-0_sp0.9 from training. Duration: 23.07775
2022-12-23 09:18:51,413 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 2.264e+02 4.118e+02 4.740e+02 5.968e+02 1.195e+03, threshold=9.480e+02, percent-clipped=4.0
2022-12-23 09:19:06,980 INFO [train.py:894] (2/4) Epoch 16, batch 1550, loss[loss=0.1974, simple_loss=0.2861, pruned_loss=0.05442, over 18633.00 frames. ], tot_loss[loss=0.185, simple_loss=0.2722, pruned_loss=0.04889, over 3713602.74 frames. ], batch size: 53, lr: 7.23e-03, grad_scale: 16.0
2022-12-23 09:19:13,414 WARNING [train.py:1060] (2/4) Exclude cut with ID 2929-85685-0044-62348-0_sp0.9 from training. Duration: 24.9333125
2022-12-23 09:19:58,598 WARNING [train.py:1060] (2/4) Exclude cut with ID 5118-111612-0016-124680-0_sp0.9 from training. Duration: 20.388875
2022-12-23 09:20:05,877 WARNING [train.py:1060] (2/4) Exclude cut with ID 432-122774-0017-62487-0_sp1.1 from training. Duration: 20.3590625
2022-12-23 09:20:15,468 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.8702, 1.7554, 2.2274, 1.3029, 2.2586, 2.1485, 1.5367, 2.4976],
       device='cuda:2'), covar=tensor([0.1128, 0.1782, 0.1209, 0.1742, 0.0677, 0.1087, 0.2204, 0.0551],
       device='cuda:2'), in_proj_covar=tensor([0.0197, 0.0204, 0.0206, 0.0192, 0.0175, 0.0214, 0.0213, 0.0188],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-23 09:20:22,361 INFO [train.py:894] (2/4) Epoch 16, batch 1600, loss[loss=0.2007, simple_loss=0.297, pruned_loss=0.05221, over 18713.00 frames. ], tot_loss[loss=0.1859, simple_loss=0.2729, pruned_loss=0.04947, over 3714435.26 frames. ], batch size: 60, lr: 7.22e-03, grad_scale: 16.0
2022-12-23 09:20:39,344 INFO [scaling.py:679] (2/4) Whitening: num_groups=8, num_channels=96, metric=1.44 vs. limit=2.0
2022-12-23 09:21:13,982 WARNING [train.py:1060] (2/4) Exclude cut with ID 3557-8342-0013-54691-0_sp1.1 from training. Duration: 0.836375
2022-12-23 09:21:19,009 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([2.4148, 2.1130, 1.6915, 2.3084, 1.8884, 2.0594, 1.9639, 2.4304],
       device='cuda:2'), covar=tensor([0.1816, 0.2730, 0.1695, 0.2428, 0.2986, 0.0999, 0.2548, 0.0847],
       device='cuda:2'), in_proj_covar=tensor([0.0289, 0.0282, 0.0236, 0.0347, 0.0261, 0.0221, 0.0279, 0.0202],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0003, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-23 09:21:23,104 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 2.634e+02 3.712e+02 4.661e+02 5.717e+02 1.218e+03, threshold=9.323e+02, percent-clipped=2.0
2022-12-23 09:21:38,403 INFO [train.py:894] (2/4) Epoch 16, batch 1650, loss[loss=0.2325, simple_loss=0.3036, pruned_loss=0.08069, over 18467.00 frames. ], tot_loss[loss=0.187, simple_loss=0.2733, pruned_loss=0.0503, over 3713993.70 frames. ], batch size: 54, lr: 7.22e-03, grad_scale: 16.0
2022-12-23 09:21:58,400 WARNING [train.py:1060] (2/4) Exclude cut with ID 8565-290391-0049-67394-0_sp0.9 from training. Duration: 21.3166875
2022-12-23 09:22:27,303 WARNING [train.py:1060] (2/4) Exclude cut with ID 6533-399-0029-104863-0_sp0.9 from training. Duration: 22.1055625
2022-12-23 09:22:40,096 WARNING [train.py:1060] (2/4) Exclude cut with ID 7699-105389-0094-26379-0_sp1.1 from training. Duration: 21.77725
2022-12-23 09:22:55,259 INFO [train.py:894] (2/4) Epoch 16, batch 1700, loss[loss=0.2187, simple_loss=0.2997, pruned_loss=0.06883, over 18593.00 frames. ], tot_loss[loss=0.1891, simple_loss=0.2746, pruned_loss=0.05178, over 3713956.85 frames. ], batch size: 69, lr: 7.22e-03, grad_scale: 16.0
2022-12-23 09:22:59,884 WARNING [train.py:1060] (2/4) Exclude cut with ID 7255-291500-0005-134304-0_sp0.9 from training. Duration: 27.8166875
2022-12-23 09:23:25,008 WARNING [train.py:1060] (2/4) Exclude cut with ID 7357-94126-0021-15852-0_sp1.1 from training. Duration: 22.5090625
2022-12-23 09:23:25,377 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.6295, 1.5455, 1.6469, 1.5399, 1.0703, 3.0562, 1.1415, 1.7629],
       device='cuda:2'), covar=tensor([0.3051, 0.1964, 0.1858, 0.1995, 0.1424, 0.0219, 0.1595, 0.0802],
       device='cuda:2'), in_proj_covar=tensor([0.0135, 0.0117, 0.0127, 0.0121, 0.0104, 0.0098, 0.0093, 0.0090],
       device='cuda:2'), out_proj_covar=tensor([0.0005, 0.0005, 0.0005, 0.0005, 0.0004, 0.0003, 0.0004, 0.0004],
       device='cuda:2')
2022-12-23 09:23:30,730 WARNING [train.py:1060] (2/4) Exclude cut with ID 7255-291500-0005-134304-0 from training. Duration: 25.035
2022-12-23 09:23:49,360 WARNING [train.py:1060] (2/4) Exclude cut with ID 774-127930-0014-10412-0_sp1.1 from training. Duration: 0.95
2022-12-23 09:23:55,487 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 2.933e+02 4.291e+02 5.331e+02 6.399e+02 1.011e+03, threshold=1.066e+03, percent-clipped=3.0
2022-12-23 09:23:57,785 INFO [scaling.py:679] (2/4) Whitening: num_groups=8, num_channels=96, metric=1.21 vs. limit=2.0
2022-12-23 09:24:10,669 INFO [train.py:894] (2/4) Epoch 16, batch 1750, loss[loss=0.1891, simple_loss=0.2664, pruned_loss=0.05591, over 18589.00 frames. ], tot_loss[loss=0.1905, simple_loss=0.2748, pruned_loss=0.05317, over 3713890.45 frames. ], batch size: 51, lr: 7.21e-03, grad_scale: 16.0
2022-12-23 09:24:10,686 WARNING [train.py:1060] (2/4) Exclude cut with ID 3033-130750-0096-55598-0_sp0.9 from training. Duration: 0.92225
2022-12-23 09:24:34,879 WARNING [train.py:1060] (2/4) Exclude cut with ID 4511-76322-0006-80011-0 from training. Duration: 21.97
2022-12-23 09:24:55,964 WARNING [train.py:1060] (2/4) Exclude cut with ID 7492-105653-0055-62765-0_sp0.9 from training. Duration: 21.97225
2022-12-23 09:24:57,458 WARNING [train.py:1060] (2/4) Exclude cut with ID 453-131332-0000-47844-0_sp0.9 from training. Duration: 25.3333125
2022-12-23 09:25:02,444 INFO [scaling.py:679] (2/4) Whitening: num_groups=1, num_channels=384, metric=3.64 vs. limit=5.0
2022-12-23 09:25:08,352 WARNING [train.py:1060] (2/4) Exclude cut with ID 5172-29468-0015-19128-0_sp0.9 from training. Duration: 21.5055625
2022-12-23 09:25:18,446 WARNING [train.py:1060] (2/4) Exclude cut with ID 453-131332-0000-47844-0_sp1.1 from training. Duration: 20.72725
2022-12-23 09:25:24,685 INFO [train.py:894] (2/4) Epoch 16, batch 1800, loss[loss=0.1879, simple_loss=0.2591, pruned_loss=0.05834, over 18591.00 frames. ], tot_loss[loss=0.193, simple_loss=0.2758, pruned_loss=0.05505, over 3713504.62 frames. ], batch size: 45, lr: 7.21e-03, grad_scale: 16.0
2022-12-23 09:25:52,600 WARNING [train.py:1060] (2/4) Exclude cut with ID 8631-249866-0030-130156-0_sp0.9 from training. Duration: 26.32775
2022-12-23 09:26:23,446 WARNING [train.py:1060] (2/4) Exclude cut with ID 3867-173237-0077-144769-0 from training. Duration: 20.025
2022-12-23 09:26:24,903 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 3.008e+02 5.203e+02 6.457e+02 7.874e+02 1.566e+03, threshold=1.291e+03, percent-clipped=4.0
2022-12-23 09:26:27,883 WARNING [train.py:1060] (2/4) Exclude cut with ID 6709-74022-0004-86860-0_sp1.1 from training. Duration: 0.9409375
2022-12-23 09:26:27,897 WARNING [train.py:1060] (2/4) Exclude cut with ID 4757-1811-0023-62229-0_sp0.9 from training. Duration: 21.37775
2022-12-23 09:26:40,526 INFO [train.py:894] (2/4) Epoch 16, batch 1850, loss[loss=0.1836, simple_loss=0.2644, pruned_loss=0.05143, over 18568.00 frames. ], tot_loss[loss=0.1955, simple_loss=0.277, pruned_loss=0.057, over 3714059.12 frames. ], batch size: 49, lr: 7.21e-03, grad_scale: 8.0
2022-12-23 09:26:49,216 WARNING [train.py:1060] (2/4) Exclude cut with ID 1250-135782-0004-25974-0_sp0.9 from training. Duration: 21.17225
2022-12-23 09:26:49,229 WARNING [train.py:1060] (2/4) Exclude cut with ID 7357-94126-0021-15852-0_sp0.9 from training. Duration: 27.511125
2022-12-23 09:27:20,778 WARNING [train.py:1060] (2/4) Exclude cut with ID 453-131332-0000-47844-0 from training. Duration: 22.8
2022-12-23 09:27:25,270 WARNING [train.py:1060] (2/4) Exclude cut with ID 4964-30587-0040-44509-0 from training. Duration: 22.585
2022-12-23 09:27:56,002 INFO [train.py:894] (2/4) Epoch 16, batch 1900, loss[loss=0.2102, simple_loss=0.2707, pruned_loss=0.07485, over 18601.00 frames. ], tot_loss[loss=0.1969, simple_loss=0.2775, pruned_loss=0.05817, over 3713687.88 frames. ], batch size: 45, lr: 7.20e-03, grad_scale: 8.0
2022-12-23 09:27:56,037 WARNING [train.py:1060] (2/4) Exclude cut with ID 6978-92210-0001-146967-0_sp0.9 from training. Duration: 22.0166875
2022-12-23 09:28:13,961 WARNING [train.py:1060] (2/4) Exclude cut with ID 7255-291500-0003-134302-0_sp1.1 from training. Duration: 24.395375
2022-12-23 09:28:19,761 WARNING [train.py:1060] (2/4) Exclude cut with ID 3488-85273-0017-41203-0_sp0.9 from training. Duration: 27.47775
2022-12-23 09:28:24,197 WARNING [train.py:1060] (2/4) Exclude cut with ID 432-122774-0017-62487-0_sp0.9 from training. Duration: 24.8833125
2022-12-23 09:28:26,887 WARNING [train.py:1060] (2/4) Exclude cut with ID 6758-72288-0033-108368-0 from training. Duration: 23.39
2022-12-23 09:28:32,888 WARNING [train.py:1060] (2/4) Exclude cut with ID 8291-282929-0007-12994-0_sp0.9 from training. Duration: 28.72225
2022-12-23 09:28:42,671 WARNING [train.py:1060] (2/4) Exclude cut with ID 585-294811-0110-133686-0_sp0.9 from training. Duration: 20.8944375
2022-12-23 09:28:44,591 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([2.9058, 2.6168, 2.2140, 1.0479, 2.0063, 2.3775, 1.8906, 2.0454],
       device='cuda:2'), covar=tensor([0.0628, 0.0530, 0.1267, 0.1678, 0.1337, 0.1316, 0.1604, 0.0921],
       device='cuda:2'), in_proj_covar=tensor([0.0169, 0.0183, 0.0206, 0.0192, 0.0210, 0.0199, 0.0214, 0.0201],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-23 09:28:58,202 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 3.390e+02 4.851e+02 5.792e+02 6.940e+02 1.812e+03, threshold=1.158e+03, percent-clipped=4.0
2022-12-23 09:28:58,259 WARNING [train.py:1060] (2/4) Exclude cut with ID 5796-66357-0007-116447-0_sp0.9 from training. Duration: 23.8444375
2022-12-23 09:29:11,813 INFO [train.py:894] (2/4) Epoch 16, batch 1950, loss[loss=0.1693, simple_loss=0.2433, pruned_loss=0.04761, over 18551.00 frames. ], tot_loss[loss=0.1971, simple_loss=0.2771, pruned_loss=0.05854, over 3713578.62 frames. ], batch size: 41, lr: 7.20e-03, grad_scale: 8.0
2022-12-23 09:29:22,598 WARNING [train.py:1060] (2/4) Exclude cut with ID 8291-282929-0007-12994-0 from training. Duration: 25.85
2022-12-23 09:29:22,609 WARNING [train.py:1060] (2/4) Exclude cut with ID 8291-282929-0023-13010-0 from training. Duration: 21.39
2022-12-23 09:29:26,363 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([3.0654, 5.1862, 4.7695, 2.6551, 5.3001, 4.0782, 0.7097, 3.4071],
       device='cuda:2'), covar=tensor([0.1741, 0.0968, 0.1088, 0.2978, 0.0686, 0.0749, 0.5120, 0.1319],
       device='cuda:2'), in_proj_covar=tensor([0.0140, 0.0133, 0.0154, 0.0123, 0.0135, 0.0109, 0.0144, 0.0113],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0002, 0.0003, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-23 09:29:34,255 WARNING [train.py:1060] (2/4) Exclude cut with ID 7357-94126-0014-15845-0 from training. Duration: 27.92
2022-12-23 09:29:58,201 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([2.1278, 1.3834, 1.7100, 1.7833, 2.0672, 2.0021, 1.9487, 1.6505],
       device='cuda:2'), covar=tensor([0.1891, 0.2855, 0.2261, 0.2489, 0.1681, 0.0862, 0.2589, 0.1108],
       device='cuda:2'), in_proj_covar=tensor([0.0262, 0.0292, 0.0272, 0.0306, 0.0294, 0.0246, 0.0326, 0.0232],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-23 09:30:02,623 WARNING [train.py:1060] (2/4) Exclude cut with ID 8631-249866-0039-130165-0_sp0.9 from training. Duration: 20.661125
2022-12-23 09:30:03,570 INFO [scaling.py:679] (2/4) Whitening: num_groups=8, num_channels=192, metric=2.06 vs. limit=2.0
2022-12-23 09:30:26,419 WARNING [train.py:1060] (2/4) Exclude cut with ID 7357-94126-0043-15874-0_sp0.9 from training. Duration: 20.07225
2022-12-23 09:30:27,844 INFO [train.py:894] (2/4) Epoch 16, batch 2000, loss[loss=0.2301, simple_loss=0.3037, pruned_loss=0.0783, over 18732.00 frames. ], tot_loss[loss=0.1974, simple_loss=0.277, pruned_loss=0.05889, over 3713693.96 frames. ], batch size: 54, lr: 7.20e-03, grad_scale: 8.0
2022-12-23 09:30:35,025 WARNING [train.py:1060] (2/4) Exclude cut with ID 1085-156170-0017-128270-0 from training. Duration: 21.01
2022-12-23 09:31:28,777 INFO [scaling.py:679] (2/4) Whitening: num_groups=8, num_channels=96, metric=1.68 vs. limit=2.0
2022-12-23 09:31:30,804 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 3.471e+02 4.848e+02 5.721e+02 6.996e+02 1.149e+03, threshold=1.144e+03, percent-clipped=0.0
2022-12-23 09:31:42,492 WARNING [train.py:1060] (2/4) Exclude cut with ID 2195-150901-0045-59933-0 from training. Duration: 20.65
2022-12-23 09:31:43,881 INFO [train.py:894] (2/4) Epoch 16, batch 2050, loss[loss=0.2317, simple_loss=0.3025, pruned_loss=0.08051, over 18698.00 frames. ], tot_loss[loss=0.1978, simple_loss=0.2771, pruned_loss=0.05927, over 3714271.59 frames. ], batch size: 99, lr: 7.19e-03, grad_scale: 8.0
2022-12-23 09:31:49,215 WARNING [train.py:1060] (2/4) Exclude cut with ID 5796-66357-0007-116447-0 from training. Duration: 21.46
2022-12-23 09:32:18,366 INFO [zipformer.py:660] (2/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=54664.0, num_to_drop=1, layers_to_drop={0}
2022-12-23 09:32:19,912 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([3.0327, 2.3475, 1.7280, 2.9267, 2.0331, 2.1565, 2.3409, 3.4269],
       device='cuda:2'), covar=tensor([0.1994, 0.3174, 0.1818, 0.3040, 0.3908, 0.1004, 0.3130, 0.0663],
       device='cuda:2'), in_proj_covar=tensor([0.0287, 0.0279, 0.0234, 0.0346, 0.0259, 0.0218, 0.0275, 0.0200],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0003, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-23 09:32:31,875 WARNING [train.py:1060] (2/4) Exclude cut with ID 3557-8342-0013-54691-0 from training. Duration: 0.92
2022-12-23 09:32:39,547 WARNING [train.py:1060] (2/4) Exclude cut with ID 8291-282929-0023-13010-0_sp0.9 from training. Duration: 23.7666875
2022-12-23 09:33:02,490 INFO [train.py:894] (2/4) Epoch 16, batch 2100, loss[loss=0.1903, simple_loss=0.2763, pruned_loss=0.05217, over 18388.00 frames. ], tot_loss[loss=0.1989, simple_loss=0.2777, pruned_loss=0.06009, over 3714477.09 frames. ], batch size: 53, lr: 7.19e-03, grad_scale: 8.0
2022-12-23 09:33:17,097 WARNING [train.py:1060] (2/4) Exclude cut with ID 8544-281189-0060-101339-0_sp0.9 from training. Duration: 20.861125
2022-12-23 09:33:28,159 WARNING [train.py:1060] (2/4) Exclude cut with ID 3488-65654-0031-41259-0_sp0.9 from training. Duration: 22.711125
2022-12-23 09:33:39,765 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([2.2781, 2.0236, 1.5530, 2.0193, 1.7823, 1.8810, 1.8474, 2.1713],
       device='cuda:2'), covar=tensor([0.2003, 0.2666, 0.1750, 0.2554, 0.3074, 0.0972, 0.2650, 0.0884],
       device='cuda:2'), in_proj_covar=tensor([0.0285, 0.0277, 0.0233, 0.0344, 0.0257, 0.0217, 0.0273, 0.0199],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0003, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-23 09:33:51,575 INFO [zipformer.py:660] (2/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=54725.0, num_to_drop=1, layers_to_drop={3}
2022-12-23 09:34:04,006 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 2.766e+02 4.934e+02 6.087e+02 7.585e+02 2.101e+03, threshold=1.217e+03, percent-clipped=2.0
2022-12-23 09:34:09,256 WARNING [train.py:1060] (2/4) Exclude cut with ID 6951-79737-0043-132310-0_sp1.1 from training. Duration: 22.986375
2022-12-23 09:34:18,059 INFO [train.py:894] (2/4) Epoch 16, batch 2150, loss[loss=0.2294, simple_loss=0.3055, pruned_loss=0.07668, over 18652.00 frames. ], tot_loss[loss=0.1991, simple_loss=0.2774, pruned_loss=0.06037, over 3714264.34 frames. ], batch size: 60, lr: 7.19e-03, grad_scale: 8.0
2022-12-23 09:34:25,393 WARNING [train.py:1060] (2/4) Exclude cut with ID 8040-260924-0003-80960-0_sp0.9 from training. Duration: 22.07225
2022-12-23 09:34:29,564 WARNING [train.py:1060] (2/4) Exclude cut with ID 7699-105389-0045-26330-0_sp0.9 from training. Duration: 20.3055625
2022-12-23 09:34:32,456 WARNING [train.py:1060] (2/4) Exclude cut with ID 6356-271890-0060-94317-0_sp0.9 from training. Duration: 20.72225
2022-12-23 09:34:46,982 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([2.7626, 4.1314, 3.9952, 1.8187, 4.2281, 3.1852, 0.8980, 2.9307],
       device='cuda:2'), covar=tensor([0.1722, 0.0935, 0.1350, 0.3535, 0.0855, 0.0888, 0.4932, 0.1401],
       device='cuda:2'), in_proj_covar=tensor([0.0138, 0.0132, 0.0153, 0.0122, 0.0135, 0.0108, 0.0144, 0.0112],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0002, 0.0003, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-23 09:34:51,128 WARNING [train.py:1060] (2/4) Exclude cut with ID 3488-85273-0017-41203-0_sp1.1 from training. Duration: 22.4818125
2022-12-23 09:35:16,529 WARNING [train.py:1060] (2/4) Exclude cut with ID 4964-30587-0040-44509-0_sp0.9 from training. Duration: 25.0944375
2022-12-23 09:35:18,578 WARNING [train.py:1060] (2/4) Exclude cut with ID 6533-399-0047-104881-0 from training. Duration: 21.515
2022-12-23 09:35:26,159 WARNING [train.py:1060] (2/4) Exclude cut with ID 7357-94126-0009-15840-0_sp0.9 from training. Duration: 27.02225
2022-12-23 09:35:30,726 WARNING [train.py:1060] (2/4) Exclude cut with ID 432-122774-0010-62480-0_sp0.9 from training. Duration: 22.22225
2022-12-23 09:35:33,177 INFO [train.py:894] (2/4) Epoch 16, batch 2200, loss[loss=0.2267, simple_loss=0.2989, pruned_loss=0.07728, over 18525.00 frames. ], tot_loss[loss=0.1992, simple_loss=0.2769, pruned_loss=0.0607, over 3713431.00 frames. ], batch size: 58, lr: 7.18e-03, grad_scale: 8.0
2022-12-23 09:35:37,685 WARNING [train.py:1060] (2/4) Exclude cut with ID 4964-30587-0085-44554-0_sp0.9 from training. Duration: 20.85
2022-12-23 09:36:11,279 WARNING [train.py:1060] (2/4) Exclude cut with ID 4295-39940-0007-92567-0 from training. Duration: 21.54
2022-12-23 09:36:18,938 WARNING [train.py:1060] (2/4) Exclude cut with ID 4964-30587-0040-44509-0_sp1.1 from training. Duration: 20.5318125
2022-12-23 09:36:29,024 WARNING [train.py:1060] (2/4) Exclude cut with ID 7255-291500-0012-134311-0_sp0.9 from training. Duration: 21.9333125
2022-12-23 09:36:38,351 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 2.727e+02 4.896e+02 6.033e+02 8.075e+02 1.619e+03, threshold=1.207e+03, percent-clipped=7.0
2022-12-23 09:36:51,273 INFO [train.py:894] (2/4) Epoch 16, batch 2250, loss[loss=0.1892, simple_loss=0.2822, pruned_loss=0.04812, over 18458.00 frames. ], tot_loss[loss=0.199, simple_loss=0.2773, pruned_loss=0.06033, over 3713513.89 frames. ], batch size: 54, lr: 7.18e-03, grad_scale: 8.0
2022-12-23 09:37:20,766 WARNING [train.py:1060] (2/4) Exclude cut with ID 8631-249866-0025-130151-0_sp0.9 from training. Duration: 21.7944375
2022-12-23 09:37:21,192 INFO [zipformer.py:660] (2/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=54862.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 09:37:33,864 WARNING [train.py:1060] (2/4) Exclude cut with ID 8291-282929-0002-12989-0_sp0.9 from training. Duration: 22.4666875
2022-12-23 09:37:40,996 WARNING [train.py:1060] (2/4) Exclude cut with ID 6121-9014-0076-24124-0 from training. Duration: 21.635
2022-12-23 09:37:46,783 WARNING [train.py:1060] (2/4) Exclude cut with ID 6121-9014-0076-24124-0_sp0.9 from training. Duration: 24.038875
2022-12-23 09:38:05,991 INFO [train.py:894] (2/4) Epoch 16, batch 2300, loss[loss=0.2173, simple_loss=0.2846, pruned_loss=0.07499, over 18721.00 frames. ], tot_loss[loss=0.1984, simple_loss=0.2769, pruned_loss=0.05993, over 3714924.81 frames. ], batch size: 52, lr: 7.18e-03, grad_scale: 8.0
2022-12-23 09:38:21,849 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([2.1997, 2.1688, 1.4587, 2.4257, 2.1968, 1.9965, 2.8637, 2.1977],
       device='cuda:2'), covar=tensor([0.0894, 0.1739, 0.2953, 0.1977, 0.1910, 0.0941, 0.1051, 0.1203],
       device='cuda:2'), in_proj_covar=tensor([0.0177, 0.0204, 0.0245, 0.0286, 0.0230, 0.0186, 0.0208, 0.0200],
       device='cuda:2'), out_proj_covar=tensor([0.0001, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-23 09:38:30,436 WARNING [train.py:1060] (2/4) Exclude cut with ID 7255-291500-0009-134308-0_sp1.1 from training. Duration: 21.786375
2022-12-23 09:38:40,753 WARNING [train.py:1060] (2/4) Exclude cut with ID 8291-282929-0002-12989-0 from training. Duration: 20.22
2022-12-23 09:38:51,805 INFO [zipformer.py:660] (2/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=54923.0, num_to_drop=1, layers_to_drop={0}
2022-12-23 09:39:08,575 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 2.948e+02 4.616e+02 5.200e+02 6.876e+02 1.499e+03, threshold=1.040e+03, percent-clipped=1.0
2022-12-23 09:39:22,500 INFO [train.py:894] (2/4) Epoch 16, batch 2350, loss[loss=0.221, simple_loss=0.2982, pruned_loss=0.07188, over 18476.00 frames. ], tot_loss[loss=0.1981, simple_loss=0.2767, pruned_loss=0.05975, over 3715235.14 frames. ], batch size: 54, lr: 7.18e-03, grad_scale: 8.0
2022-12-23 09:39:58,029 INFO [scaling.py:679] (2/4) Whitening: num_groups=1, num_channels=384, metric=2.73 vs. limit=5.0
2022-12-23 09:40:39,437 INFO [train.py:894] (2/4) Epoch 16, batch 2400, loss[loss=0.1744, simple_loss=0.2504, pruned_loss=0.0492, over 18549.00 frames. ], tot_loss[loss=0.1981, simple_loss=0.2768, pruned_loss=0.05974, over 3715219.58 frames. ], batch size: 47, lr: 7.17e-03, grad_scale: 8.0
2022-12-23 09:40:39,510 WARNING [train.py:1060] (2/4) Exclude cut with ID 6951-79737-0043-132310-0 from training. Duration: 25.285
2022-12-23 09:41:20,334 INFO [zipformer.py:660] (2/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=55020.0, num_to_drop=1, layers_to_drop={1}
2022-12-23 09:41:42,166 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 2.922e+02 4.598e+02 5.703e+02 7.103e+02 1.452e+03, threshold=1.141e+03, percent-clipped=8.0
2022-12-23 09:41:45,453 WARNING [train.py:1060] (2/4) Exclude cut with ID 811-130148-0001-63453-0_sp0.9 from training. Duration: 20.861125
2022-12-23 09:41:56,542 INFO [train.py:894] (2/4) Epoch 16, batch 2450, loss[loss=0.1521, simple_loss=0.2295, pruned_loss=0.03733, over 18435.00 frames. ], tot_loss[loss=0.1987, simple_loss=0.2772, pruned_loss=0.06004, over 3714338.59 frames. ], batch size: 42, lr: 7.17e-03, grad_scale: 8.0
2022-12-23 09:42:07,306 WARNING [train.py:1060] (2/4) Exclude cut with ID 6010-56788-0055-90261-0 from training. Duration: 20.88
2022-12-23 09:42:41,436 WARNING [train.py:1060] (2/4) Exclude cut with ID 7357-94126-0045-15876-0_sp0.9 from training. Duration: 23.4166875
2022-12-23 09:43:13,753 INFO [train.py:894] (2/4) Epoch 16, batch 2500, loss[loss=0.2261, simple_loss=0.2922, pruned_loss=0.07999, over 18451.00 frames. ], tot_loss[loss=0.1979, simple_loss=0.2764, pruned_loss=0.05975, over 3714835.87 frames. ], batch size: 50, lr: 7.17e-03, grad_scale: 8.0
2022-12-23 09:43:58,815 WARNING [train.py:1060] (2/4) Exclude cut with ID 7255-291500-0006-134305-0 from training. Duration: 21.24
2022-12-23 09:43:58,830 WARNING [train.py:1060] (2/4) Exclude cut with ID 6533-399-0047-104881-0_sp0.9 from training. Duration: 23.9055625
2022-12-23 09:44:14,934 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 2.619e+02 4.431e+02 5.249e+02 6.433e+02 1.075e+03, threshold=1.050e+03, percent-clipped=0.0
2022-12-23 09:44:28,869 INFO [train.py:894] (2/4) Epoch 16, batch 2550, loss[loss=0.191, simple_loss=0.27, pruned_loss=0.056, over 18425.00 frames. ], tot_loss[loss=0.1983, simple_loss=0.2769, pruned_loss=0.05988, over 3715058.78 frames. ], batch size: 48, lr: 7.16e-03, grad_scale: 8.0
2022-12-23 09:44:30,540 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([2.5799, 4.1391, 3.9898, 1.7587, 4.1791, 3.2105, 0.6206, 2.6316],
       device='cuda:2'), covar=tensor([0.2074, 0.0893, 0.1299, 0.3419, 0.0839, 0.0859, 0.5223, 0.1475],
       device='cuda:2'), in_proj_covar=tensor([0.0141, 0.0133, 0.0155, 0.0123, 0.0136, 0.0109, 0.0146, 0.0113],
       device='cuda:2'), out_proj_covar=tensor([0.0003, 0.0002, 0.0003, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-23 09:44:31,678 WARNING [train.py:1060] (2/4) Exclude cut with ID 6758-72288-0033-108368-0_sp0.9 from training. Duration: 25.988875
2022-12-23 09:44:36,221 INFO [zipformer.py:660] (2/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=55148.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 09:44:40,330 WARNING [train.py:1060] (2/4) Exclude cut with ID 7255-291500-0001-134300-0_sp0.9 from training. Duration: 20.67225
2022-12-23 09:45:32,469 WARNING [train.py:1060] (2/4) Exclude cut with ID 3488-85273-0038-41224-0 from training. Duration: 20.34
2022-12-23 09:45:44,367 INFO [train.py:894] (2/4) Epoch 16, batch 2600, loss[loss=0.1562, simple_loss=0.2332, pruned_loss=0.0396, over 18542.00 frames. ], tot_loss[loss=0.1977, simple_loss=0.2761, pruned_loss=0.05963, over 3713875.32 frames. ], batch size: 44, lr: 7.16e-03, grad_scale: 8.0
2022-12-23 09:46:09,150 INFO [zipformer.py:660] (2/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=55209.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 09:46:12,313 INFO [zipformer.py:660] (2/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=55211.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 09:46:22,187 INFO [zipformer.py:660] (2/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=55218.0, num_to_drop=1, layers_to_drop={0}
2022-12-23 09:46:42,915 WARNING [train.py:1060] (2/4) Exclude cut with ID 7357-94126-0026-15857-0_sp0.9 from training. Duration: 25.061125
2022-12-23 09:46:47,525 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 2.767e+02 4.537e+02 5.819e+02 6.990e+02 1.532e+03, threshold=1.164e+03, percent-clipped=6.0
2022-12-23 09:46:53,342 WARNING [train.py:1060] (2/4) Exclude cut with ID 3033-130750-0096-55598-0 from training. Duration: 0.83
2022-12-23 09:47:01,080 INFO [train.py:894] (2/4) Epoch 16, batch 2650, loss[loss=0.1732, simple_loss=0.2466, pruned_loss=0.04994, over 18404.00 frames. ], tot_loss[loss=0.1993, simple_loss=0.2774, pruned_loss=0.06054, over 3713883.00 frames. ], batch size: 42, lr: 7.16e-03, grad_scale: 8.0
2022-12-23 09:47:17,787 WARNING [train.py:1060] (2/4) Exclude cut with ID 3488-85273-0017-41203-0 from training. Duration: 24.73
2022-12-23 09:47:30,107 WARNING [train.py:1060] (2/4) Exclude cut with ID 7255-291500-0009-134308-0 from training. Duration: 23.965
2022-12-23 09:47:39,825 WARNING [train.py:1060] (2/4) Exclude cut with ID 6978-92210-0030-146996-0_sp0.9 from training. Duration: 22.088875
2022-12-23 09:47:46,664 INFO [zipformer.py:660] (2/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=55272.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 09:47:54,207 WARNING [train.py:1060] (2/4) Exclude cut with ID 7255-291500-0006-134305-0_sp0.9 from training. Duration: 23.6
2022-12-23 09:48:11,885 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([2.7112, 5.4129, 4.8753, 2.6560, 5.1602, 3.8873, 1.0947, 3.6799],
       device='cuda:2'), covar=tensor([0.2109, 0.0832, 0.1168, 0.3184, 0.0738, 0.0815, 0.4964, 0.1351],
       device='cuda:2'), in_proj_covar=tensor([0.0141, 0.0134, 0.0154, 0.0123, 0.0137, 0.0110, 0.0146, 0.0113],
       device='cuda:2'), out_proj_covar=tensor([0.0003, 0.0002, 0.0003, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-23 09:48:17,953 INFO [train.py:894] (2/4) Epoch 16, batch 2700, loss[loss=0.2057, simple_loss=0.2859, pruned_loss=0.06276, over 18691.00 frames. ], tot_loss[loss=0.1981, simple_loss=0.2765, pruned_loss=0.05988, over 3713994.01 frames. ], batch size: 62, lr: 7.15e-03, grad_scale: 8.0
2022-12-23 09:49:00,576 INFO [zipformer.py:660] (2/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=55320.0, num_to_drop=1, layers_to_drop={1}
2022-12-23 09:49:02,127 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.8774, 1.6927, 1.5139, 1.4407, 1.8240, 1.9475, 2.0413, 1.4744],
       device='cuda:2'), covar=tensor([0.0259, 0.0261, 0.0417, 0.0231, 0.0190, 0.0356, 0.0263, 0.0272],
       device='cuda:2'), in_proj_covar=tensor([0.0090, 0.0121, 0.0147, 0.0122, 0.0113, 0.0114, 0.0094, 0.0123],
       device='cuda:2'), out_proj_covar=tensor([7.2398e-05, 9.6885e-05, 1.2364e-04, 9.8302e-05, 9.3243e-05, 8.8452e-05,
        7.4392e-05, 9.8025e-05], device='cuda:2')
2022-12-23 09:49:05,048 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([2.7167, 3.9960, 3.8193, 2.0223, 3.8797, 2.9973, 0.7218, 2.8239],
       device='cuda:2'), covar=tensor([0.2066, 0.1190, 0.1553, 0.3390, 0.1126, 0.1023, 0.5377, 0.1428],
       device='cuda:2'), in_proj_covar=tensor([0.0142, 0.0135, 0.0155, 0.0123, 0.0137, 0.0110, 0.0146, 0.0114],
       device='cuda:2'), out_proj_covar=tensor([0.0003, 0.0002, 0.0003, 0.0002, 0.0002, 0.0002, 0.0003, 0.0002],
       device='cuda:2')
2022-12-23 09:49:20,787 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 2.908e+02 4.633e+02 5.450e+02 6.979e+02 1.469e+03, threshold=1.090e+03, percent-clipped=1.0
2022-12-23 09:49:33,916 INFO [train.py:894] (2/4) Epoch 16, batch 2750, loss[loss=0.2064, simple_loss=0.2848, pruned_loss=0.06399, over 18629.00 frames. ], tot_loss[loss=0.1985, simple_loss=0.2767, pruned_loss=0.06017, over 3713591.50 frames. ], batch size: 69, lr: 7.15e-03, grad_scale: 8.0
2022-12-23 09:49:36,946 WARNING [train.py:1060] (2/4) Exclude cut with ID 8291-282929-0024-13011-0 from training. Duration: 23.795
2022-12-23 09:49:53,665 WARNING [train.py:1060] (2/4) Exclude cut with ID 8631-249866-0030-130156-0_sp1.1 from training. Duration: 21.5409375
2022-12-23 09:49:57,184 WARNING [train.py:1060] (2/4) Exclude cut with ID 6978-92210-0019-146985-0_sp0.9 from training. Duration: 24.97775
2022-12-23 09:50:07,411 WARNING [train.py:1060] (2/4) Exclude cut with ID 1085-156170-0017-128270-0_sp0.9 from training. Duration: 23.3444375
2022-12-23 09:50:13,983 INFO [zipformer.py:660] (2/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=55368.0, num_to_drop=1, layers_to_drop={0}
2022-12-23 09:50:33,591 WARNING [train.py:1060] (2/4) Exclude cut with ID 6010-56788-0055-90261-0_sp0.9 from training. Duration: 23.2
2022-12-23 09:50:41,383 WARNING [train.py:1060] (2/4) Exclude cut with ID 5653-46179-0060-117930-0_sp0.9 from training. Duration: 21.17225
2022-12-23 09:50:51,456 INFO [train.py:894] (2/4) Epoch 16, batch 2800, loss[loss=0.2408, simple_loss=0.3059, pruned_loss=0.08788, over 18678.00 frames. ], tot_loss[loss=0.197, simple_loss=0.2754, pruned_loss=0.05928, over 3714843.39 frames. ], batch size: 179, lr: 7.15e-03, grad_scale: 8.0
2022-12-23 09:50:59,426 WARNING [train.py:1060] (2/4) Exclude cut with ID 7255-291500-0008-134307-0_sp0.9 from training. Duration: 24.6555625
2022-12-23 09:51:24,577 INFO [scaling.py:679] (2/4) Whitening: num_groups=8, num_channels=96, metric=1.18 vs. limit=2.0
2022-12-23 09:51:55,062 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 2.964e+02 4.626e+02 5.479e+02 6.826e+02 2.106e+03, threshold=1.096e+03, percent-clipped=3.0
2022-12-23 09:51:58,249 WARNING [train.py:1060] (2/4) Exclude cut with ID 3488-65654-0031-41259-0 from training. Duration: 20.44
2022-12-23 09:52:08,914 INFO [train.py:894] (2/4) Epoch 16, batch 2850, loss[loss=0.1765, simple_loss=0.2528, pruned_loss=0.05009, over 18524.00 frames. ], tot_loss[loss=0.1956, simple_loss=0.2741, pruned_loss=0.05852, over 3714789.32 frames. ], batch size: 47, lr: 7.14e-03, grad_scale: 8.0
2022-12-23 09:52:13,450 WARNING [train.py:1060] (2/4) Exclude cut with ID 6951-79737-0018-132285-0_sp0.9 from training. Duration: 23.45
2022-12-23 09:52:41,959 WARNING [train.py:1060] (2/4) Exclude cut with ID 6945-60535-0076-12784-0_sp0.9 from training. Duration: 20.52225
2022-12-23 09:52:49,409 WARNING [train.py:1060] (2/4) Exclude cut with ID 7255-291500-0008-134307-0 from training. Duration: 22.19
2022-12-23 09:52:58,314 WARNING [train.py:1060] (2/4) Exclude cut with ID 7357-94126-0014-15845-0_sp1.1 from training. Duration: 25.3818125
2022-12-23 09:53:15,940 WARNING [train.py:1060] (2/4) Exclude cut with ID 6951-79737-0043-132310-0_sp0.9 from training. Duration: 28.0944375
2022-12-23 09:53:24,573 INFO [train.py:894] (2/4) Epoch 16, batch 2900, loss[loss=0.2431, simple_loss=0.3028, pruned_loss=0.09172, over 18610.00 frames. ], tot_loss[loss=0.1962, simple_loss=0.275, pruned_loss=0.05874, over 3715279.35 frames. ], batch size: 180, lr: 7.14e-03, grad_scale: 8.0
2022-12-23 09:53:24,607 WARNING [train.py:1060] (2/4) Exclude cut with ID 2195-150901-0045-59933-0_sp0.9 from training. Duration: 22.9444375
2022-12-23 09:53:31,853 WARNING [train.py:1060] (2/4) Exclude cut with ID 8291-282929-0024-13011-0_sp1.1 from training. Duration: 21.6318125
2022-12-23 09:53:41,104 INFO [zipformer.py:660] (2/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=55504.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 09:53:47,526 WARNING [train.py:1060] (2/4) Exclude cut with ID 8631-249866-0030-130156-0 from training. Duration: 23.695
2022-12-23 09:54:02,827 INFO [zipformer.py:660] (2/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=55518.0, num_to_drop=1, layers_to_drop={0}
2022-12-23 09:54:14,153 WARNING [train.py:1060] (2/4) Exclude cut with ID 7699-105389-0094-26379-0 from training. Duration: 23.955
2022-12-23 09:54:26,967 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 2.388e+02 4.402e+02 5.592e+02 6.941e+02 1.570e+03, threshold=1.118e+03, percent-clipped=4.0
2022-12-23 09:54:40,716 INFO [train.py:894] (2/4) Epoch 16, batch 2950, loss[loss=0.1895, simple_loss=0.2717, pruned_loss=0.05362, over 18590.00 frames. ], tot_loss[loss=0.1963, simple_loss=0.2748, pruned_loss=0.05889, over 3714372.29 frames. ], batch size: 51, lr: 7.14e-03, grad_scale: 8.0
2022-12-23 09:54:46,610 WARNING [train.py:1060] (2/4) Exclude cut with ID 8291-282929-0024-13011-0_sp0.9 from training. Duration: 26.438875
2022-12-23 09:55:15,018 INFO [zipformer.py:660] (2/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=55566.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 09:55:16,493 INFO [zipformer.py:660] (2/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=55567.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 09:55:32,390 WARNING [train.py:1060] (2/4) Exclude cut with ID 7699-105389-0021-26306-0_sp0.9 from training. Duration: 21.2444375
2022-12-23 09:55:33,567 WARNING [train.py:1060] (2/4) Exclude cut with ID 7357-94126-0014-15845-0_sp0.9 from training. Duration: 31.02225
2022-12-23 09:55:44,227 WARNING [train.py:1060] (2/4) Exclude cut with ID 432-122774-0017-62487-0 from training. Duration: 22.395
2022-12-23 09:55:55,249 INFO [train.py:894] (2/4) Epoch 16, batch 3000, loss[loss=0.2511, simple_loss=0.3094, pruned_loss=0.09639, over 18569.00 frames. ], tot_loss[loss=0.1957, simple_loss=0.274, pruned_loss=0.05874, over 3714265.74 frames. ], batch size: 186, lr: 7.13e-03, grad_scale: 8.0
2022-12-23 09:55:55,250 INFO [train.py:919] (2/4) Computing validation loss
2022-12-23 09:56:00,802 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([2.6524, 2.4224, 2.0117, 1.5887, 2.9486, 2.6884, 2.5177, 1.9999],
       device='cuda:2'), covar=tensor([0.0307, 0.0338, 0.0543, 0.0740, 0.0214, 0.0297, 0.0387, 0.0755],
       device='cuda:2'), in_proj_covar=tensor([0.0121, 0.0125, 0.0126, 0.0119, 0.0093, 0.0120, 0.0134, 0.0154],
       device='cuda:2'), out_proj_covar=tensor([0.0001, 0.0002, 0.0002, 0.0001, 0.0001, 0.0001, 0.0002, 0.0002],
       device='cuda:2')
2022-12-23 09:56:06,242 INFO [train.py:928] (2/4) Epoch 16, validation: loss=0.1656, simple_loss=0.2645, pruned_loss=0.03333, over 944034.00 frames. 
2022-12-23 09:56:06,243 INFO [train.py:929] (2/4) Maximum memory allocated so far is 24680MB
2022-12-23 09:56:09,386 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.7815, 1.1633, 0.4977, 1.4213, 2.2456, 1.3622, 1.4155, 1.7096],
       device='cuda:2'), covar=tensor([0.2184, 0.3152, 0.3183, 0.2099, 0.1888, 0.2193, 0.2191, 0.2474],
       device='cuda:2'), in_proj_covar=tensor([0.0092, 0.0097, 0.0117, 0.0095, 0.0115, 0.0090, 0.0097, 0.0093],
       device='cuda:2'), out_proj_covar=tensor([0.0003, 0.0003, 0.0004, 0.0003, 0.0004, 0.0003, 0.0003, 0.0003],
       device='cuda:2')
2022-12-23 09:56:11,934 WARNING [train.py:1060] (2/4) Exclude cut with ID 7357-94126-0045-15876-0 from training. Duration: 21.075
2022-12-23 09:56:16,274 WARNING [train.py:1060] (2/4) Exclude cut with ID 6482-98857-0025-147532-0_sp0.9 from training. Duration: 20.0055625
2022-12-23 09:56:16,281 WARNING [train.py:1060] (2/4) Exclude cut with ID 6951-79737-0037-132304-0_sp0.9 from training. Duration: 22.05
2022-12-23 09:56:16,288 WARNING [train.py:1060] (2/4) Exclude cut with ID 7255-291500-0003-134302-0 from training. Duration: 26.8349375
2022-12-23 09:56:19,666 WARNING [train.py:1060] (2/4) Exclude cut with ID 7357-94126-0009-15840-0_sp1.1 from training. Duration: 22.1090625
2022-12-23 09:56:26,246 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([2.1586, 1.8841, 1.7616, 1.3697, 2.4515, 2.2257, 2.1231, 1.5190],
       device='cuda:2'), covar=tensor([0.0372, 0.0453, 0.0546, 0.0701, 0.0281, 0.0355, 0.0410, 0.0972],
       device='cuda:2'), in_proj_covar=tensor([0.0121, 0.0125, 0.0127, 0.0120, 0.0094, 0.0120, 0.0134, 0.0154],
       device='cuda:2'), out_proj_covar=tensor([0.0001, 0.0002, 0.0002, 0.0001, 0.0001, 0.0001, 0.0002, 0.0002],
       device='cuda:2')
2022-12-23 09:56:27,294 WARNING [train.py:1060] (2/4) Exclude cut with ID 7699-105389-0094-26379-0_sp0.9 from training. Duration: 26.6166875
2022-12-23 09:56:47,096 WARNING [train.py:1060] (2/4) Exclude cut with ID 2046-178027-0000-53705-0_sp0.9 from training. Duration: 20.3055625
2022-12-23 09:57:07,822 WARNING [train.py:1060] (2/4) Exclude cut with ID 7205-50138-0008-5373-0_sp0.9 from training. Duration: 20.7
2022-12-23 09:57:09,330 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 3.134e+02 4.889e+02 6.104e+02 7.743e+02 1.774e+03, threshold=1.221e+03, percent-clipped=6.0
2022-12-23 09:57:23,356 INFO [train.py:894] (2/4) Epoch 16, batch 3050, loss[loss=0.1884, simple_loss=0.2684, pruned_loss=0.05421, over 18507.00 frames. ], tot_loss[loss=0.1959, simple_loss=0.2743, pruned_loss=0.05877, over 3714222.45 frames. ], batch size: 52, lr: 7.13e-03, grad_scale: 8.0
2022-12-23 09:57:39,033 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([2.5618, 2.2067, 0.6772, 1.8113, 2.8076, 1.7644, 2.4420, 2.7043],
       device='cuda:2'), covar=tensor([0.1436, 0.1701, 0.2599, 0.1486, 0.1416, 0.1489, 0.1283, 0.1432],
       device='cuda:2'), in_proj_covar=tensor([0.0092, 0.0097, 0.0118, 0.0096, 0.0115, 0.0091, 0.0097, 0.0094],
       device='cuda:2'), out_proj_covar=tensor([0.0003, 0.0003, 0.0004, 0.0003, 0.0004, 0.0003, 0.0003, 0.0003],
       device='cuda:2')
2022-12-23 09:57:54,146 WARNING [train.py:1060] (2/4) Exclude cut with ID 6978-92210-0019-146985-0 from training. Duration: 22.48
2022-12-23 09:58:11,471 WARNING [train.py:1060] (2/4) Exclude cut with ID 7255-291500-0003-134302-0_sp0.9 from training. Duration: 29.816625
2022-12-23 09:58:26,016 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([2.0250, 1.7611, 2.3693, 2.8929, 2.2679, 2.8115, 1.6785, 2.4212],
       device='cuda:2'), covar=tensor([0.1727, 0.1988, 0.1380, 0.1150, 0.1231, 0.2198, 0.1977, 0.1339],
       device='cuda:2'), in_proj_covar=tensor([0.0101, 0.0115, 0.0130, 0.0142, 0.0105, 0.0136, 0.0128, 0.0111],
       device='cuda:2'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0004, 0.0003, 0.0004, 0.0003, 0.0003],
       device='cuda:2')
2022-12-23 09:58:31,705 WARNING [train.py:1060] (2/4) Exclude cut with ID 7255-291500-0005-134304-0_sp1.1 from training. Duration: 22.7590625
2022-12-23 09:58:36,101 WARNING [train.py:1060] (2/4) Exclude cut with ID 7357-94126-0026-15857-0 from training. Duration: 22.555
2022-12-23 09:58:39,599 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([3.7188, 2.8763, 2.7693, 1.2373, 2.5570, 2.6297, 2.2955, 2.4770],
       device='cuda:2'), covar=tensor([0.0625, 0.0723, 0.1602, 0.2003, 0.1666, 0.1384, 0.1531, 0.1166],
       device='cuda:2'), in_proj_covar=tensor([0.0170, 0.0185, 0.0205, 0.0191, 0.0209, 0.0200, 0.0214, 0.0200],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-23 09:58:40,498 INFO [train.py:894] (2/4) Epoch 16, batch 3100, loss[loss=0.1846, simple_loss=0.2675, pruned_loss=0.05084, over 18648.00 frames. ], tot_loss[loss=0.1947, simple_loss=0.2737, pruned_loss=0.05788, over 3714508.48 frames. ], batch size: 48, lr: 7.13e-03, grad_scale: 8.0
2022-12-23 09:58:56,791 WARNING [train.py:1060] (2/4) Exclude cut with ID 1250-135782-0005-25975-0_sp0.9 from training. Duration: 21.688875
2022-12-23 09:59:01,012 INFO [scaling.py:679] (2/4) Whitening: num_groups=8, num_channels=96, metric=1.60 vs. limit=2.0
2022-12-23 09:59:17,787 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.8442, 1.4449, 1.4211, 1.9829, 1.7290, 3.4959, 1.3963, 1.4587],
       device='cuda:2'), covar=tensor([0.0832, 0.1991, 0.1275, 0.0984, 0.1481, 0.0256, 0.1511, 0.1667],
       device='cuda:2'), in_proj_covar=tensor([0.0072, 0.0082, 0.0073, 0.0074, 0.0090, 0.0074, 0.0083, 0.0077],
       device='cuda:2'), out_proj_covar=tensor([0.0003, 0.0004, 0.0004, 0.0003, 0.0004, 0.0003, 0.0004, 0.0004],
       device='cuda:2')
2022-12-23 09:59:33,210 WARNING [train.py:1060] (2/4) Exclude cut with ID 3488-85273-0038-41224-0_sp0.9 from training. Duration: 22.6
2022-12-23 09:59:43,768 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 2.787e+02 4.394e+02 5.359e+02 6.672e+02 1.594e+03, threshold=1.072e+03, percent-clipped=1.0
2022-12-23 09:59:56,466 INFO [train.py:894] (2/4) Epoch 16, batch 3150, loss[loss=0.2371, simple_loss=0.3001, pruned_loss=0.08704, over 18564.00 frames. ], tot_loss[loss=0.1946, simple_loss=0.2735, pruned_loss=0.05785, over 3713615.77 frames. ], batch size: 49, lr: 7.12e-03, grad_scale: 8.0
2022-12-23 10:00:04,210 INFO [zipformer.py:660] (2/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=55748.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 10:00:10,614 WARNING [train.py:1060] (2/4) Exclude cut with ID 7357-94126-0009-15840-0 from training. Duration: 24.32
2022-12-23 10:00:43,406 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.3530, 1.0164, 1.4571, 2.2222, 1.5823, 2.1779, 0.6865, 1.6604],
       device='cuda:2'), covar=tensor([0.1775, 0.1823, 0.1246, 0.0781, 0.1246, 0.0909, 0.1912, 0.1252],
       device='cuda:2'), in_proj_covar=tensor([0.0100, 0.0114, 0.0129, 0.0140, 0.0105, 0.0135, 0.0127, 0.0110],
       device='cuda:2'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0004, 0.0003, 0.0004, 0.0003, 0.0003],
       device='cuda:2')
2022-12-23 10:01:08,948 WARNING [train.py:1060] (2/4) Exclude cut with ID 8291-276745-0093-13116-0_sp0.9 from training. Duration: 21.061125
2022-12-23 10:01:10,546 INFO [train.py:894] (2/4) Epoch 16, batch 3200, loss[loss=0.1749, simple_loss=0.2663, pruned_loss=0.04173, over 18707.00 frames. ], tot_loss[loss=0.1954, simple_loss=0.2742, pruned_loss=0.05829, over 3714367.00 frames. ], batch size: 50, lr: 7.12e-03, grad_scale: 8.0
2022-12-23 10:01:20,436 WARNING [train.py:1060] (2/4) Exclude cut with ID 7357-94126-0024-15855-0_sp0.9 from training. Duration: 20.32225
2022-12-23 10:01:27,848 INFO [zipformer.py:660] (2/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=55804.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 10:01:34,049 WARNING [train.py:1060] (2/4) Exclude cut with ID 3033-130750-0096-55598-0_sp1.1 from training. Duration: 0.7545625
2022-12-23 10:01:36,287 INFO [zipformer.py:660] (2/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=55809.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 10:01:50,449 WARNING [train.py:1060] (2/4) Exclude cut with ID 4295-39940-0007-92567-0_sp0.9 from training. Duration: 23.9333125
2022-12-23 10:02:03,025 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.9832, 1.8859, 1.7422, 1.0985, 2.3188, 2.0228, 1.9374, 1.5447],
       device='cuda:2'), covar=tensor([0.0370, 0.0420, 0.0452, 0.0732, 0.0274, 0.0366, 0.0427, 0.0871],
       device='cuda:2'), in_proj_covar=tensor([0.0123, 0.0126, 0.0127, 0.0121, 0.0095, 0.0121, 0.0136, 0.0155],
       device='cuda:2'), out_proj_covar=tensor([0.0001, 0.0002, 0.0002, 0.0001, 0.0001, 0.0001, 0.0002, 0.0002],
       device='cuda:2')
2022-12-23 10:02:13,059 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 3.270e+02 4.851e+02 5.650e+02 7.307e+02 2.258e+03, threshold=1.130e+03, percent-clipped=5.0
2022-12-23 10:02:20,763 WARNING [train.py:1060] (2/4) Exclude cut with ID 7255-291500-0008-134307-0_sp1.1 from training. Duration: 20.17275
2022-12-23 10:02:26,727 INFO [train.py:894] (2/4) Epoch 16, batch 3250, loss[loss=0.198, simple_loss=0.2804, pruned_loss=0.05777, over 18509.00 frames. ], tot_loss[loss=0.1944, simple_loss=0.2731, pruned_loss=0.05786, over 3714297.81 frames. ], batch size: 77, lr: 7.12e-03, grad_scale: 8.0
2022-12-23 10:02:26,767 WARNING [train.py:1060] (2/4) Exclude cut with ID 6978-92210-0019-146985-0_sp1.1 from training. Duration: 20.436375
2022-12-23 10:02:35,317 INFO [zipformer.py:660] (2/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=55848.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 10:02:41,111 INFO [zipformer.py:660] (2/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=55852.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 10:03:04,978 INFO [zipformer.py:660] (2/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=55867.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 10:03:22,724 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.8829, 1.7868, 2.1946, 1.2977, 2.1962, 2.1864, 1.6003, 2.4671],
       device='cuda:2'), covar=tensor([0.1112, 0.1740, 0.1146, 0.1767, 0.0659, 0.1114, 0.2034, 0.0535],
       device='cuda:2'), in_proj_covar=tensor([0.0197, 0.0208, 0.0207, 0.0195, 0.0176, 0.0218, 0.0213, 0.0187],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-23 10:03:43,294 INFO [train.py:894] (2/4) Epoch 16, batch 3300, loss[loss=0.1551, simple_loss=0.237, pruned_loss=0.03662, over 18556.00 frames. ], tot_loss[loss=0.1944, simple_loss=0.273, pruned_loss=0.05787, over 3713594.94 frames. ], batch size: 41, lr: 7.11e-03, grad_scale: 8.0
2022-12-23 10:03:48,679 WARNING [train.py:1060] (2/4) Exclude cut with ID 4234-40345-0022-142709-0_sp0.9 from training. Duration: 23.1055625
2022-12-23 10:03:52,945 WARNING [train.py:1060] (2/4) Exclude cut with ID 8291-282929-0007-12994-0_sp1.1 from training. Duration: 23.5
2022-12-23 10:04:00,830 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.9048, 1.9124, 1.4235, 2.0398, 2.0653, 1.7985, 2.6396, 2.0044],
       device='cuda:2'), covar=tensor([0.0919, 0.1616, 0.2805, 0.1693, 0.1808, 0.0915, 0.0932, 0.1178],
       device='cuda:2'), in_proj_covar=tensor([0.0176, 0.0202, 0.0246, 0.0286, 0.0231, 0.0186, 0.0208, 0.0201],
       device='cuda:2'), out_proj_covar=tensor([0.0001, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-23 10:04:03,446 WARNING [train.py:1060] (2/4) Exclude cut with ID 7255-291500-0009-134308-0_sp0.9 from training. Duration: 26.62775
2022-12-23 10:04:09,977 INFO [zipformer.py:660] (2/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=55909.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 10:04:16,140 WARNING [train.py:1060] (2/4) Exclude cut with ID 6951-79737-0018-132285-0 from training. Duration: 21.105
2022-12-23 10:04:19,056 INFO [zipformer.py:660] (2/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=55915.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 10:04:20,454 WARNING [train.py:1060] (2/4) Exclude cut with ID 4511-76322-0006-80011-0_sp0.9 from training. Duration: 24.411125
2022-12-23 10:04:34,194 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.9834, 1.9199, 2.0498, 1.9861, 1.5851, 4.5456, 2.2155, 2.5035],
       device='cuda:2'), covar=tensor([0.2944, 0.1888, 0.1755, 0.1846, 0.1312, 0.0131, 0.1484, 0.0840],
       device='cuda:2'), in_proj_covar=tensor([0.0136, 0.0117, 0.0127, 0.0121, 0.0104, 0.0099, 0.0094, 0.0090],
       device='cuda:2'), out_proj_covar=tensor([0.0005, 0.0005, 0.0005, 0.0005, 0.0004, 0.0003, 0.0004, 0.0004],
       device='cuda:2')
2022-12-23 10:04:47,336 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 2.602e+02 4.449e+02 5.484e+02 6.704e+02 1.621e+03, threshold=1.097e+03, percent-clipped=1.0
2022-12-23 10:04:47,377 WARNING [train.py:1060] (2/4) Exclude cut with ID 6758-72288-0033-108368-0_sp1.1 from training. Duration: 21.263625
2022-12-23 10:05:02,681 INFO [train.py:894] (2/4) Epoch 16, batch 3350, loss[loss=0.1771, simple_loss=0.2498, pruned_loss=0.05221, over 18656.00 frames. ], tot_loss[loss=0.1934, simple_loss=0.2721, pruned_loss=0.05731, over 3714062.13 frames. ], batch size: 41, lr: 7.11e-03, grad_scale: 8.0
2022-12-23 10:05:20,798 WARNING [train.py:1060] (2/4) Exclude cut with ID 4234-40345-0022-142709-0 from training. Duration: 20.795
2022-12-23 10:05:23,250 INFO [scaling.py:679] (2/4) Whitening: num_groups=8, num_channels=96, metric=1.41 vs. limit=2.0
2022-12-23 10:05:31,790 WARNING [train.py:1060] (2/4) Exclude cut with ID 7357-94126-0021-15852-0 from training. Duration: 24.76
2022-12-23 10:05:33,245 WARNING [train.py:1060] (2/4) Exclude cut with ID 3867-173237-0077-144769-0_sp0.9 from training. Duration: 22.25
2022-12-23 10:05:48,325 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.7853, 1.6572, 1.7264, 1.8654, 1.1344, 4.9401, 1.8840, 2.2524],
       device='cuda:2'), covar=tensor([0.3248, 0.2151, 0.1957, 0.2023, 0.1470, 0.0096, 0.1507, 0.0919],
       device='cuda:2'), in_proj_covar=tensor([0.0136, 0.0117, 0.0126, 0.0121, 0.0104, 0.0099, 0.0094, 0.0090],
       device='cuda:2'), out_proj_covar=tensor([0.0005, 0.0005, 0.0005, 0.0005, 0.0004, 0.0003, 0.0004, 0.0004],
       device='cuda:2')
2022-12-23 10:05:56,767 WARNING [train.py:1060] (2/4) Exclude cut with ID 7357-94126-0026-15857-0_sp1.1 from training. Duration: 20.5045625
2022-12-23 10:06:19,226 INFO [train.py:894] (2/4) Epoch 16, batch 3400, loss[loss=0.1925, simple_loss=0.2766, pruned_loss=0.05422, over 18679.00 frames. ], tot_loss[loss=0.1927, simple_loss=0.2713, pruned_loss=0.05704, over 3713113.76 frames. ], batch size: 50, lr: 7.11e-03, grad_scale: 8.0
2022-12-23 10:06:45,605 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([3.2188, 3.3106, 2.2462, 1.6746, 3.7682, 3.7425, 2.8289, 2.9569],
       device='cuda:2'), covar=tensor([0.0372, 0.0337, 0.0549, 0.0722, 0.0181, 0.0338, 0.0510, 0.0601],
       device='cuda:2'), in_proj_covar=tensor([0.0120, 0.0124, 0.0126, 0.0119, 0.0094, 0.0119, 0.0134, 0.0153],
       device='cuda:2'), out_proj_covar=tensor([0.0001, 0.0002, 0.0002, 0.0001, 0.0001, 0.0001, 0.0002, 0.0002],
       device='cuda:2')
2022-12-23 10:07:09,409 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([2.8124, 2.3301, 1.9565, 0.9711, 2.0030, 2.2673, 2.0302, 2.1745],
       device='cuda:2'), covar=tensor([0.0578, 0.0515, 0.1154, 0.1550, 0.1096, 0.1316, 0.1294, 0.0675],
       device='cuda:2'), in_proj_covar=tensor([0.0170, 0.0183, 0.0206, 0.0191, 0.0210, 0.0200, 0.0214, 0.0201],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-23 10:07:23,259 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 2.742e+02 4.730e+02 5.600e+02 6.470e+02 1.614e+03, threshold=1.120e+03, percent-clipped=3.0
2022-12-23 10:07:35,970 INFO [train.py:894] (2/4) Epoch 16, batch 3450, loss[loss=0.2423, simple_loss=0.3124, pruned_loss=0.08614, over 18699.00 frames. ], tot_loss[loss=0.1932, simple_loss=0.2716, pruned_loss=0.05736, over 3713248.87 frames. ], batch size: 62, lr: 7.11e-03, grad_scale: 8.0
2022-12-23 10:08:10,169 INFO [scaling.py:679] (2/4) Whitening: num_groups=1, num_channels=384, metric=4.55 vs. limit=5.0
2022-12-23 10:08:13,401 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([2.9728, 2.9338, 2.3788, 2.0138, 3.4963, 3.4853, 2.8492, 2.5809],
       device='cuda:2'), covar=tensor([0.0362, 0.0347, 0.0524, 0.0688, 0.0206, 0.0332, 0.0458, 0.0650],
       device='cuda:2'), in_proj_covar=tensor([0.0122, 0.0126, 0.0128, 0.0121, 0.0095, 0.0121, 0.0136, 0.0155],
       device='cuda:2'), out_proj_covar=tensor([0.0001, 0.0002, 0.0002, 0.0001, 0.0001, 0.0001, 0.0002, 0.0002],
       device='cuda:2')
2022-12-23 10:08:20,531 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([0.3140, 1.7134, 1.9797, 1.0819, 1.1160, 2.0322, 1.8730, 1.6523],
       device='cuda:2'), covar=tensor([0.0709, 0.0263, 0.0263, 0.0316, 0.0340, 0.0390, 0.0203, 0.0612],
       device='cuda:2'), in_proj_covar=tensor([0.0143, 0.0161, 0.0120, 0.0135, 0.0143, 0.0137, 0.0154, 0.0164],
       device='cuda:2'), out_proj_covar=tensor([1.1100e-04, 1.2586e-04, 9.2384e-05, 1.0235e-04, 1.0957e-04, 1.0780e-04,
        1.2166e-04, 1.2769e-04], device='cuda:2')
2022-12-23 10:08:48,636 INFO [train.py:894] (2/4) Epoch 16, batch 3500, loss[loss=0.2335, simple_loss=0.3013, pruned_loss=0.08286, over 18678.00 frames. ], tot_loss[loss=0.1953, simple_loss=0.2737, pruned_loss=0.05847, over 3714146.94 frames. ], batch size: 172, lr: 7.10e-03, grad_scale: 8.0
2022-12-23 10:09:09,256 WARNING [train.py:1060] (2/4) Exclude cut with ID 7859-102521-0017-7548-0_sp1.1 from training. Duration: 22.2954375
2022-12-23 10:09:18,371 INFO [train.py:894] (2/4) Epoch 17, batch 0, loss[loss=0.1928, simple_loss=0.2768, pruned_loss=0.05434, over 18719.00 frames. ], tot_loss[loss=0.1928, simple_loss=0.2768, pruned_loss=0.05434, over 18719.00 frames. ], batch size: 52, lr: 6.89e-03, grad_scale: 8.0
2022-12-23 10:09:18,371 INFO [train.py:919] (2/4) Computing validation loss
2022-12-23 10:09:30,655 INFO [train.py:928] (2/4) Epoch 17, validation: loss=0.1676, simple_loss=0.2661, pruned_loss=0.03449, over 944034.00 frames. 
2022-12-23 10:09:30,656 INFO [train.py:929] (2/4) Maximum memory allocated so far is 24680MB
2022-12-23 10:09:38,148 INFO [zipformer.py:660] (2/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=56104.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 10:10:18,862 WARNING [train.py:1060] (2/4) Exclude cut with ID 298-126791-0067-24026-0_sp0.9 from training. Duration: 21.438875
2022-12-23 10:10:23,559 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 3.084e+02 4.090e+02 5.146e+02 6.475e+02 1.382e+03, threshold=1.029e+03, percent-clipped=5.0
2022-12-23 10:10:23,622 WARNING [train.py:1060] (2/4) Exclude cut with ID 5652-39938-0025-23684-0_sp0.9 from training. Duration: 22.2055625
2022-12-23 10:10:39,948 INFO [zipformer.py:660] (2/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=56145.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 10:10:45,387 INFO [train.py:894] (2/4) Epoch 17, batch 50, loss[loss=0.1678, simple_loss=0.2592, pruned_loss=0.0382, over 18608.00 frames. ], tot_loss[loss=0.1893, simple_loss=0.2763, pruned_loss=0.05115, over 838589.39 frames. ], batch size: 51, lr: 6.88e-03, grad_scale: 8.0
2022-12-23 10:11:15,518 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([2.3943, 1.8177, 2.1173, 2.2039, 2.2812, 2.3468, 2.2697, 1.7450],
       device='cuda:2'), covar=tensor([0.2061, 0.3103, 0.2250, 0.2824, 0.1841, 0.0824, 0.3034, 0.1224],
       device='cuda:2'), in_proj_covar=tensor([0.0263, 0.0295, 0.0274, 0.0308, 0.0297, 0.0246, 0.0330, 0.0235],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-23 10:11:45,222 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([3.0215, 3.2553, 2.2876, 1.6840, 3.6087, 3.7168, 3.0031, 2.6063],
       device='cuda:2'), covar=tensor([0.0367, 0.0285, 0.0484, 0.0701, 0.0170, 0.0264, 0.0404, 0.0635],
       device='cuda:2'), in_proj_covar=tensor([0.0122, 0.0126, 0.0128, 0.0121, 0.0095, 0.0121, 0.0136, 0.0155],
       device='cuda:2'), out_proj_covar=tensor([0.0001, 0.0002, 0.0002, 0.0001, 0.0001, 0.0001, 0.0002, 0.0002],
       device='cuda:2')
2022-12-23 10:11:57,882 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([2.8107, 2.2850, 1.9205, 2.7275, 2.1191, 2.1155, 2.0487, 2.7950],
       device='cuda:2'), covar=tensor([0.1767, 0.2969, 0.1611, 0.2487, 0.3351, 0.0974, 0.3160, 0.0774],
       device='cuda:2'), in_proj_covar=tensor([0.0292, 0.0282, 0.0237, 0.0349, 0.0263, 0.0223, 0.0278, 0.0206],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0003, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-23 10:12:00,273 INFO [train.py:894] (2/4) Epoch 17, batch 100, loss[loss=0.1607, simple_loss=0.2425, pruned_loss=0.03942, over 18570.00 frames. ], tot_loss[loss=0.1858, simple_loss=0.2729, pruned_loss=0.04933, over 1476278.87 frames. ], batch size: 45, lr: 6.88e-03, grad_scale: 8.0
2022-12-23 10:12:08,291 INFO [zipformer.py:660] (2/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=56204.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 10:12:11,740 INFO [zipformer.py:660] (2/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=56206.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 10:12:53,868 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 2.099e+02 3.400e+02 4.151e+02 4.945e+02 1.658e+03, threshold=8.301e+02, percent-clipped=1.0
2022-12-23 10:13:16,884 INFO [train.py:894] (2/4) Epoch 17, batch 150, loss[loss=0.1851, simple_loss=0.2735, pruned_loss=0.04841, over 18556.00 frames. ], tot_loss[loss=0.1842, simple_loss=0.2715, pruned_loss=0.04842, over 1973208.58 frames. ], batch size: 57, lr: 6.88e-03, grad_scale: 8.0
2022-12-23 10:13:21,341 WARNING [train.py:1060] (2/4) Exclude cut with ID 7859-102521-0017-7548-0 from training. Duration: 24.525
2022-12-23 10:13:43,037 INFO [scaling.py:679] (2/4) Whitening: num_groups=1, num_channels=384, metric=2.71 vs. limit=5.0
2022-12-23 10:13:54,932 WARNING [train.py:1060] (2/4) Exclude cut with ID 3699-47246-0007-3408-0_sp0.9 from training. Duration: 20.26675
2022-12-23 10:14:08,252 WARNING [train.py:1060] (2/4) Exclude cut with ID 7859-102521-0017-7548-0_sp0.9 from training. Duration: 27.25
2022-12-23 10:14:32,300 INFO [train.py:894] (2/4) Epoch 17, batch 200, loss[loss=0.1862, simple_loss=0.2744, pruned_loss=0.04897, over 18673.00 frames. ], tot_loss[loss=0.1826, simple_loss=0.2694, pruned_loss=0.04793, over 2358675.74 frames. ], batch size: 69, lr: 6.87e-03, grad_scale: 8.0
2022-12-23 10:15:24,837 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 2.339e+02 3.583e+02 4.534e+02 5.413e+02 1.730e+03, threshold=9.068e+02, percent-clipped=8.0
2022-12-23 10:15:26,166 WARNING [train.py:1060] (2/4) Exclude cut with ID 6426-64292-0017-15984-0 from training. Duration: 21.68
2022-12-23 10:15:39,492 WARNING [train.py:1060] (2/4) Exclude cut with ID 4278-13270-0007-59342-0 from training. Duration: 21.6300625
2022-12-23 10:15:47,000 INFO [train.py:894] (2/4) Epoch 17, batch 250, loss[loss=0.1948, simple_loss=0.2892, pruned_loss=0.05017, over 18622.00 frames. ], tot_loss[loss=0.1802, simple_loss=0.2669, pruned_loss=0.0467, over 2658800.21 frames. ], batch size: 97, lr: 6.87e-03, grad_scale: 8.0
2022-12-23 10:16:01,825 WARNING [train.py:1060] (2/4) Exclude cut with ID 4278-13270-0007-59342-0_sp0.9 from training. Duration: 24.033375
2022-12-23 10:16:54,691 WARNING [train.py:1060] (2/4) Exclude cut with ID 4278-13270-0009-59344-0 from training. Duration: 22.905
2022-12-23 10:16:56,118 WARNING [train.py:1060] (2/4) Exclude cut with ID 5622-44585-0006-90525-0_sp1.1 from training. Duration: 23.4318125
2022-12-23 10:17:02,085 INFO [train.py:894] (2/4) Epoch 17, batch 300, loss[loss=0.15, simple_loss=0.2309, pruned_loss=0.03456, over 18610.00 frames. ], tot_loss[loss=0.1789, simple_loss=0.2655, pruned_loss=0.04617, over 2892749.19 frames. ], batch size: 41, lr: 6.87e-03, grad_scale: 8.0
2022-12-23 10:17:09,881 INFO [zipformer.py:660] (2/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=56404.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 10:17:53,853 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 2.464e+02 3.697e+02 4.364e+02 5.395e+02 9.387e+02, threshold=8.728e+02, percent-clipped=1.0
2022-12-23 10:18:16,089 INFO [train.py:894] (2/4) Epoch 17, batch 350, loss[loss=0.1638, simple_loss=0.2516, pruned_loss=0.03804, over 18677.00 frames. ], tot_loss[loss=0.1796, simple_loss=0.2666, pruned_loss=0.04625, over 3075277.20 frames. ], batch size: 50, lr: 6.87e-03, grad_scale: 16.0
2022-12-23 10:18:20,388 INFO [zipformer.py:660] (2/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=56452.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 10:18:56,160 WARNING [train.py:1060] (2/4) Exclude cut with ID 4278-13270-0009-59344-0_sp1.1 from training. Duration: 20.82275
2022-12-23 10:18:57,586 WARNING [train.py:1060] (2/4) Exclude cut with ID 4278-13270-0009-59344-0_sp0.9 from training. Duration: 25.45
2022-12-23 10:19:30,095 INFO [train.py:894] (2/4) Epoch 17, batch 400, loss[loss=0.1944, simple_loss=0.2799, pruned_loss=0.05443, over 18638.00 frames. ], tot_loss[loss=0.1812, simple_loss=0.2686, pruned_loss=0.04687, over 3217226.70 frames. ], batch size: 69, lr: 6.86e-03, grad_scale: 16.0
2022-12-23 10:19:32,019 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([2.4023, 2.4121, 2.1630, 1.4837, 2.8825, 2.6874, 2.3444, 1.9515],
       device='cuda:2'), covar=tensor([0.0372, 0.0329, 0.0428, 0.0769, 0.0217, 0.0311, 0.0447, 0.0714],
       device='cuda:2'), in_proj_covar=tensor([0.0122, 0.0125, 0.0127, 0.0120, 0.0095, 0.0120, 0.0136, 0.0155],
       device='cuda:2'), out_proj_covar=tensor([0.0001, 0.0002, 0.0002, 0.0001, 0.0001, 0.0001, 0.0002, 0.0002],
       device='cuda:2')
2022-12-23 10:19:33,347 INFO [zipformer.py:660] (2/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=56501.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 10:19:34,074 INFO [scaling.py:679] (2/4) Whitening: num_groups=1, num_channels=384, metric=3.03 vs. limit=5.0
2022-12-23 10:19:37,930 INFO [zipformer.py:660] (2/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=56504.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 10:19:54,109 WARNING [train.py:1060] (2/4) Exclude cut with ID 5622-44585-0006-90525-0 from training. Duration: 25.775
2022-12-23 10:20:07,551 INFO [zipformer.py:660] (2/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=56523.0, num_to_drop=1, layers_to_drop={0}
2022-12-23 10:20:15,563 WARNING [train.py:1060] (2/4) Exclude cut with ID 2929-85685-0071-62375-0_sp0.9 from training. Duration: 22.25
2022-12-23 10:20:23,539 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 2.488e+02 3.929e+02 4.700e+02 5.683e+02 9.730e+02, threshold=9.400e+02, percent-clipped=2.0
2022-12-23 10:20:39,815 INFO [scaling.py:679] (2/4) Whitening: num_groups=1, num_channels=384, metric=4.77 vs. limit=5.0
2022-12-23 10:20:43,011 WARNING [train.py:1060] (2/4) Exclude cut with ID 3972-170212-0014-23379-0 from training. Duration: 26.205
2022-12-23 10:20:45,874 INFO [train.py:894] (2/4) Epoch 17, batch 450, loss[loss=0.1658, simple_loss=0.2637, pruned_loss=0.03391, over 18724.00 frames. ], tot_loss[loss=0.1828, simple_loss=0.27, pruned_loss=0.04776, over 3327796.54 frames. ], batch size: 54, lr: 6.86e-03, grad_scale: 16.0
2022-12-23 10:20:50,343 INFO [zipformer.py:660] (2/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=56552.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 10:20:59,336 WARNING [train.py:1060] (2/4) Exclude cut with ID 5239-32139-0047-9341-0_sp0.9 from training. Duration: 30.1555625
2022-12-23 10:21:04,226 INFO [scaling.py:679] (2/4) Whitening: num_groups=1, num_channels=384, metric=3.67 vs. limit=5.0
2022-12-23 10:21:06,031 WARNING [train.py:1060] (2/4) Exclude cut with ID 1265-135635-0050-6781-0_sp0.9 from training. Duration: 21.8333125
2022-12-23 10:21:16,979 WARNING [train.py:1060] (2/4) Exclude cut with ID 1914-133440-0024-94914-0_sp1.1 from training. Duration: 20.6545625
2022-12-23 10:21:40,258 INFO [zipformer.py:660] (2/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=56584.0, num_to_drop=1, layers_to_drop={0}
2022-12-23 10:21:57,633 WARNING [train.py:1060] (2/4) Exclude cut with ID 7395-89880-0045-39920-0_sp0.9 from training. Duration: 20.52225
2022-12-23 10:21:58,550 INFO [scaling.py:679] (2/4) Whitening: num_groups=8, num_channels=96, metric=1.21 vs. limit=2.0
2022-12-23 10:22:02,113 INFO [train.py:894] (2/4) Epoch 17, batch 500, loss[loss=0.2231, simple_loss=0.2931, pruned_loss=0.07652, over 18572.00 frames. ], tot_loss[loss=0.1835, simple_loss=0.2701, pruned_loss=0.04841, over 3412716.01 frames. ], batch size: 177, lr: 6.86e-03, grad_scale: 16.0
2022-12-23 10:22:19,220 WARNING [train.py:1060] (2/4) Exclude cut with ID 3972-170212-0014-23379-0_sp0.9 from training. Duration: 29.1166875
2022-12-23 10:22:46,216 INFO [scaling.py:679] (2/4) Whitening: num_groups=8, num_channels=96, metric=1.23 vs. limit=2.0
2022-12-23 10:22:56,244 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 2.340e+02 3.538e+02 4.282e+02 5.017e+02 8.861e+02, threshold=8.564e+02, percent-clipped=0.0
2022-12-23 10:23:07,926 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([0.2214, 1.2647, 1.6146, 0.7615, 1.0241, 1.7145, 1.6243, 1.4524],
       device='cuda:2'), covar=tensor([0.0675, 0.0329, 0.0377, 0.0346, 0.0383, 0.0363, 0.0229, 0.0548],
       device='cuda:2'), in_proj_covar=tensor([0.0146, 0.0164, 0.0121, 0.0136, 0.0144, 0.0138, 0.0156, 0.0165],
       device='cuda:2'), out_proj_covar=tensor([1.1300e-04, 1.2774e-04, 9.2921e-05, 1.0332e-04, 1.1094e-04, 1.0846e-04,
        1.2310e-04, 1.2872e-04], device='cuda:2')
2022-12-23 10:23:19,230 INFO [train.py:894] (2/4) Epoch 17, batch 550, loss[loss=0.2075, simple_loss=0.3037, pruned_loss=0.05566, over 18591.00 frames. ], tot_loss[loss=0.1838, simple_loss=0.2702, pruned_loss=0.04872, over 3479498.73 frames. ], batch size: 57, lr: 6.85e-03, grad_scale: 16.0
2022-12-23 10:23:20,883 WARNING [train.py:1060] (2/4) Exclude cut with ID 543-133211-0007-59831-0_sp0.9 from training. Duration: 21.388875
2022-12-23 10:23:40,462 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.4970, 2.8393, 3.3703, 0.9941, 2.8127, 3.6172, 2.6227, 2.9730],
       device='cuda:2'), covar=tensor([0.0785, 0.0377, 0.0356, 0.0521, 0.0400, 0.0307, 0.0360, 0.0709],
       device='cuda:2'), in_proj_covar=tensor([0.0145, 0.0163, 0.0120, 0.0135, 0.0144, 0.0138, 0.0155, 0.0165],
       device='cuda:2'), out_proj_covar=tensor([1.1242e-04, 1.2716e-04, 9.2564e-05, 1.0283e-04, 1.1045e-04, 1.0801e-04,
        1.2245e-04, 1.2814e-04], device='cuda:2')
2022-12-23 10:23:56,095 WARNING [train.py:1060] (2/4) Exclude cut with ID 1914-133440-0024-94914-0 from training. Duration: 22.72
2022-12-23 10:23:57,495 WARNING [train.py:1060] (2/4) Exclude cut with ID 1914-133440-0031-94921-0_sp0.9 from training. Duration: 22.7444375
2022-12-23 10:24:15,081 INFO [scaling.py:679] (2/4) Whitening: num_groups=8, num_channels=96, metric=1.59 vs. limit=2.0
2022-12-23 10:24:34,985 INFO [train.py:894] (2/4) Epoch 17, batch 600, loss[loss=0.1823, simple_loss=0.2632, pruned_loss=0.05068, over 18530.00 frames. ], tot_loss[loss=0.1838, simple_loss=0.2699, pruned_loss=0.04879, over 3531132.98 frames. ], batch size: 47, lr: 6.85e-03, grad_scale: 16.0
2022-12-23 10:24:40,810 WARNING [train.py:1060] (2/4) Exclude cut with ID 4133-6541-0027-40495-0_sp1.1 from training. Duration: 0.9681875
2022-12-23 10:24:45,234 WARNING [train.py:1060] (2/4) Exclude cut with ID 6330-62851-0022-91297-0_sp0.9 from training. Duration: 22.3166875
2022-12-23 10:24:49,685 WARNING [train.py:1060] (2/4) Exclude cut with ID 543-133212-0015-59917-0_sp0.9 from training. Duration: 21.8166875
2022-12-23 10:25:12,805 INFO [scaling.py:679] (2/4) Whitening: num_groups=8, num_channels=96, metric=1.19 vs. limit=2.0
2022-12-23 10:25:29,261 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 2.559e+02 3.399e+02 4.174e+02 5.258e+02 9.875e+02, threshold=8.347e+02, percent-clipped=3.0
2022-12-23 10:25:51,055 INFO [train.py:894] (2/4) Epoch 17, batch 650, loss[loss=0.1606, simple_loss=0.2372, pruned_loss=0.04196, over 18479.00 frames. ], tot_loss[loss=0.1842, simple_loss=0.2705, pruned_loss=0.04893, over 3571051.94 frames. ], batch size: 43, lr: 6.85e-03, grad_scale: 16.0
2022-12-23 10:26:32,005 WARNING [train.py:1060] (2/4) Exclude cut with ID 4957-30119-0041-23990-0_sp0.9 from training. Duration: 20.22775
2022-12-23 10:27:07,829 INFO [train.py:894] (2/4) Epoch 17, batch 700, loss[loss=0.167, simple_loss=0.2565, pruned_loss=0.03875, over 18539.00 frames. ], tot_loss[loss=0.1839, simple_loss=0.2702, pruned_loss=0.04882, over 3602412.38 frames. ], batch size: 55, lr: 6.84e-03, grad_scale: 16.0
2022-12-23 10:27:11,973 INFO [zipformer.py:660] (2/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=56801.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 10:27:14,910 INFO [zipformer.py:660] (2/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=56803.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 10:27:16,593 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([2.1799, 1.5281, 1.8372, 1.8323, 2.2051, 2.1660, 2.0730, 1.6477],
       device='cuda:2'), covar=tensor([0.2066, 0.3022, 0.2365, 0.2626, 0.1723, 0.0824, 0.2742, 0.1198],
       device='cuda:2'), in_proj_covar=tensor([0.0262, 0.0295, 0.0274, 0.0308, 0.0297, 0.0245, 0.0329, 0.0234],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-23 10:27:19,307 WARNING [train.py:1060] (2/4) Exclude cut with ID 5239-32139-0047-9341-0_sp1.1 from training. Duration: 24.67275
2022-12-23 10:27:46,321 WARNING [train.py:1060] (2/4) Exclude cut with ID 3082-165428-0081-50734-0_sp0.9 from training. Duration: 21.8055625
2022-12-23 10:28:01,765 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 2.567e+02 3.551e+02 4.773e+02 5.686e+02 1.008e+03, threshold=9.547e+02, percent-clipped=4.0
2022-12-23 10:28:16,666 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([2.6681, 2.5176, 2.0030, 1.4248, 2.9623, 2.8714, 2.4289, 1.9272],
       device='cuda:2'), covar=tensor([0.0323, 0.0359, 0.0522, 0.0798, 0.0236, 0.0313, 0.0451, 0.0809],
       device='cuda:2'), in_proj_covar=tensor([0.0121, 0.0123, 0.0126, 0.0119, 0.0093, 0.0119, 0.0134, 0.0154],
       device='cuda:2'), out_proj_covar=tensor([0.0001, 0.0002, 0.0002, 0.0001, 0.0001, 0.0001, 0.0002, 0.0002],
       device='cuda:2')
2022-12-23 10:28:24,262 INFO [train.py:894] (2/4) Epoch 17, batch 750, loss[loss=0.1645, simple_loss=0.2474, pruned_loss=0.04078, over 18385.00 frames. ], tot_loss[loss=0.1834, simple_loss=0.2699, pruned_loss=0.04849, over 3627892.03 frames. ], batch size: 46, lr: 6.84e-03, grad_scale: 16.0
2022-12-23 10:28:24,282 WARNING [train.py:1060] (2/4) Exclude cut with ID 3340-169293-0054-76830-0_sp0.9 from training. Duration: 22.6666875
2022-12-23 10:28:24,994 INFO [zipformer.py:660] (2/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=56849.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 10:28:48,043 INFO [zipformer.py:660] (2/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=56864.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 10:29:10,380 INFO [zipformer.py:660] (2/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=56879.0, num_to_drop=1, layers_to_drop={2}
2022-12-23 10:29:28,229 WARNING [train.py:1060] (2/4) Exclude cut with ID 2411-132532-0017-82279-0_sp1.1 from training. Duration: 0.9681875
2022-12-23 10:29:40,594 INFO [train.py:894] (2/4) Epoch 17, batch 800, loss[loss=0.1692, simple_loss=0.2614, pruned_loss=0.03856, over 18447.00 frames. ], tot_loss[loss=0.1835, simple_loss=0.2703, pruned_loss=0.04838, over 3646942.93 frames. ], batch size: 50, lr: 6.84e-03, grad_scale: 16.0
2022-12-23 10:29:54,896 WARNING [train.py:1060] (2/4) Exclude cut with ID 6330-62850-0007-91323-0 from training. Duration: 22.485
2022-12-23 10:29:55,309 INFO [zipformer.py:660] (2/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=56908.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 10:30:33,211 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 2.478e+02 3.299e+02 4.044e+02 4.765e+02 1.058e+03, threshold=8.087e+02, percent-clipped=1.0
2022-12-23 10:30:34,490 WARNING [train.py:1060] (2/4) Exclude cut with ID 3972-170212-0014-23379-0_sp1.1 from training. Duration: 23.82275
2022-12-23 10:30:48,814 WARNING [train.py:1060] (2/4) Exclude cut with ID 4860-13185-0032-76709-0 from training. Duration: 20.77
2022-12-23 10:30:54,924 INFO [train.py:894] (2/4) Epoch 17, batch 850, loss[loss=0.1706, simple_loss=0.2569, pruned_loss=0.04213, over 18587.00 frames. ], tot_loss[loss=0.1831, simple_loss=0.2698, pruned_loss=0.04815, over 3661320.58 frames. ], batch size: 49, lr: 6.84e-03, grad_scale: 16.0
2022-12-23 10:30:58,321 WARNING [train.py:1060] (2/4) Exclude cut with ID 6426-64292-0017-15984-0_sp0.9 from training. Duration: 24.088875
2022-12-23 10:31:25,781 INFO [zipformer.py:660] (2/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=56969.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 10:31:26,932 WARNING [train.py:1060] (2/4) Exclude cut with ID 6330-62850-0007-91323-0_sp1.1 from training. Duration: 20.4409375
2022-12-23 10:31:29,737 INFO [scaling.py:679] (2/4) Whitening: num_groups=8, num_channels=96, metric=1.63 vs. limit=2.0
2022-12-23 10:32:01,322 INFO [zipformer.py:660] (2/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=56993.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 10:32:04,556 INFO [zipformer.py:660] (2/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=56995.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 10:32:10,248 INFO [train.py:894] (2/4) Epoch 17, batch 900, loss[loss=0.2069, simple_loss=0.2907, pruned_loss=0.06151, over 18614.00 frames. ], tot_loss[loss=0.1832, simple_loss=0.2698, pruned_loss=0.04832, over 3672631.30 frames. ], batch size: 53, lr: 6.83e-03, grad_scale: 16.0
2022-12-23 10:32:43,701 WARNING [train.py:1060] (2/4) Exclude cut with ID 2929-85685-0062-62366-0_sp0.9 from training. Duration: 22.511125
2022-12-23 10:32:43,716 WARNING [train.py:1060] (2/4) Exclude cut with ID 7395-89880-0031-39906-0 from training. Duration: 20.675
2022-12-23 10:33:01,588 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 2.430e+02 4.029e+02 4.883e+02 6.473e+02 1.483e+03, threshold=9.767e+02, percent-clipped=10.0
2022-12-23 10:33:24,988 INFO [train.py:894] (2/4) Epoch 17, batch 950, loss[loss=0.188, simple_loss=0.2753, pruned_loss=0.05034, over 18647.00 frames. ], tot_loss[loss=0.1832, simple_loss=0.2698, pruned_loss=0.04831, over 3681725.59 frames. ], batch size: 69, lr: 6.83e-03, grad_scale: 16.0
2022-12-23 10:33:33,573 INFO [zipformer.py:660] (2/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=57054.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 10:33:36,533 INFO [zipformer.py:660] (2/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=57056.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 10:33:51,291 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([2.0370, 2.0222, 1.4857, 2.2703, 2.2658, 1.9010, 2.8663, 2.0796],
       device='cuda:2'), covar=tensor([0.0873, 0.1660, 0.2819, 0.1772, 0.1669, 0.0916, 0.0919, 0.1245],
       device='cuda:2'), in_proj_covar=tensor([0.0175, 0.0204, 0.0245, 0.0285, 0.0231, 0.0185, 0.0207, 0.0200],
       device='cuda:2'), out_proj_covar=tensor([0.0001, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-23 10:34:02,092 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.8304, 1.5494, 1.6159, 2.0335, 1.7642, 3.5163, 1.4494, 1.5137],
       device='cuda:2'), covar=tensor([0.0788, 0.1800, 0.1128, 0.0895, 0.1407, 0.0217, 0.1371, 0.1564],
       device='cuda:2'), in_proj_covar=tensor([0.0072, 0.0081, 0.0073, 0.0074, 0.0090, 0.0074, 0.0084, 0.0077],
       device='cuda:2'), out_proj_covar=tensor([0.0003, 0.0004, 0.0004, 0.0004, 0.0004, 0.0003, 0.0004, 0.0004],
       device='cuda:2')
2022-12-23 10:34:18,831 INFO [scaling.py:679] (2/4) Whitening: num_groups=8, num_channels=96, metric=1.47 vs. limit=2.0
2022-12-23 10:34:20,866 WARNING [train.py:1060] (2/4) Exclude cut with ID 6330-62850-0007-91323-0_sp0.9 from training. Duration: 24.9833125
2022-12-23 10:34:41,009 INFO [train.py:894] (2/4) Epoch 17, batch 1000, loss[loss=0.2013, simple_loss=0.2918, pruned_loss=0.05536, over 18675.00 frames. ], tot_loss[loss=0.1825, simple_loss=0.2695, pruned_loss=0.04771, over 3688878.03 frames. ], batch size: 77, lr: 6.83e-03, grad_scale: 16.0
2022-12-23 10:34:53,930 WARNING [train.py:1060] (2/4) Exclude cut with ID 5239-32139-0047-9341-0 from training. Duration: 27.14
2022-12-23 10:35:09,940 WARNING [train.py:1060] (2/4) Exclude cut with ID 2929-85685-0044-62348-0 from training. Duration: 22.44
2022-12-23 10:35:33,883 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 2.085e+02 3.382e+02 4.120e+02 4.933e+02 1.139e+03, threshold=8.240e+02, percent-clipped=1.0
2022-12-23 10:35:52,149 INFO [scaling.py:679] (2/4) Whitening: num_groups=8, num_channels=96, metric=1.30 vs. limit=2.0
2022-12-23 10:35:56,744 INFO [train.py:894] (2/4) Epoch 17, batch 1050, loss[loss=0.209, simple_loss=0.2918, pruned_loss=0.06305, over 18644.00 frames. ], tot_loss[loss=0.1817, simple_loss=0.2688, pruned_loss=0.04735, over 3693759.29 frames. ], batch size: 97, lr: 6.82e-03, grad_scale: 16.0
2022-12-23 10:36:11,671 INFO [zipformer.py:660] (2/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=57159.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 10:36:29,362 WARNING [train.py:1060] (2/4) Exclude cut with ID 2929-85685-0060-62364-0_sp0.9 from training. Duration: 21.361125
2022-12-23 10:36:36,933 WARNING [train.py:1060] (2/4) Exclude cut with ID 2929-85685-0079-62383-0_sp1.1 from training. Duration: 27.0318125
2022-12-23 10:36:41,764 INFO [zipformer.py:660] (2/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=57179.0, num_to_drop=1, layers_to_drop={0}
2022-12-23 10:36:44,667 WARNING [train.py:1060] (2/4) Exclude cut with ID 5622-44585-0006-90525-0_sp0.9 from training. Duration: 28.638875
2022-12-23 10:36:52,251 INFO [zipformer.py:660] (2/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=57186.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 10:37:00,180 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.2610, 1.2907, 1.4949, 0.8640, 1.3223, 1.4108, 1.1238, 1.6475],
       device='cuda:2'), covar=tensor([0.0932, 0.1912, 0.1100, 0.1503, 0.0725, 0.0994, 0.2282, 0.0556],
       device='cuda:2'), in_proj_covar=tensor([0.0192, 0.0207, 0.0204, 0.0193, 0.0173, 0.0211, 0.0210, 0.0186],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-23 10:37:01,095 WARNING [train.py:1060] (2/4) Exclude cut with ID 3340-169293-0054-76830-0 from training. Duration: 20.4
2022-12-23 10:37:08,389 INFO [zipformer.py:660] (2/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=57197.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 10:37:10,969 INFO [train.py:894] (2/4) Epoch 17, batch 1100, loss[loss=0.1741, simple_loss=0.273, pruned_loss=0.0376, over 18479.00 frames. ], tot_loss[loss=0.1817, simple_loss=0.2687, pruned_loss=0.0473, over 3698203.81 frames. ], batch size: 54, lr: 6.82e-03, grad_scale: 16.0
2022-12-23 10:37:33,301 WARNING [train.py:1060] (2/4) Exclude cut with ID 2929-85685-0071-62375-0 from training. Duration: 20.025
2022-12-23 10:37:33,317 WARNING [train.py:1060] (2/4) Exclude cut with ID 2364-131735-0112-64612-0_sp0.9 from training. Duration: 20.488875
2022-12-23 10:37:39,981 WARNING [train.py:1060] (2/4) Exclude cut with ID 2929-85685-0079-62383-0 from training. Duration: 29.735
2022-12-23 10:37:54,211 INFO [zipformer.py:660] (2/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=57227.0, num_to_drop=1, layers_to_drop={1}
2022-12-23 10:38:03,731 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 2.132e+02 3.502e+02 4.489e+02 5.299e+02 1.026e+03, threshold=8.977e+02, percent-clipped=3.0
2022-12-23 10:38:19,577 INFO [zipformer.py:660] (2/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=57244.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 10:38:22,754 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.8872, 1.3156, 0.7262, 1.2753, 2.2698, 1.2840, 1.7745, 1.8093],
       device='cuda:2'), covar=tensor([0.1462, 0.1992, 0.2311, 0.1542, 0.1536, 0.1581, 0.1271, 0.1552],
       device='cuda:2'), in_proj_covar=tensor([0.0091, 0.0096, 0.0115, 0.0093, 0.0113, 0.0089, 0.0096, 0.0092],
       device='cuda:2'), out_proj_covar=tensor([0.0003, 0.0003, 0.0004, 0.0003, 0.0004, 0.0003, 0.0003, 0.0003],
       device='cuda:2')
2022-12-23 10:38:24,334 INFO [zipformer.py:660] (2/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=57247.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 10:38:26,749 INFO [train.py:894] (2/4) Epoch 17, batch 1150, loss[loss=0.183, simple_loss=0.2777, pruned_loss=0.04416, over 18513.00 frames. ], tot_loss[loss=0.1816, simple_loss=0.2685, pruned_loss=0.04731, over 3701318.75 frames. ], batch size: 64, lr: 6.82e-03, grad_scale: 8.0
2022-12-23 10:38:41,327 INFO [zipformer.py:660] (2/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=57258.0, num_to_drop=1, layers_to_drop={0}
2022-12-23 10:38:50,904 INFO [zipformer.py:660] (2/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=57264.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 10:39:02,659 WARNING [train.py:1060] (2/4) Exclude cut with ID 7276-92427-0014-12983-0_sp0.9 from training. Duration: 21.3055625
2022-12-23 10:39:02,702 WARNING [train.py:1060] (2/4) Exclude cut with ID 1025-75365-0008-79168-0_sp0.9 from training. Duration: 22.0666875
2022-12-23 10:39:30,371 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.6281, 1.4914, 1.3504, 1.5435, 1.9064, 1.7416, 1.7538, 1.1946],
       device='cuda:2'), covar=tensor([0.0264, 0.0210, 0.0478, 0.0178, 0.0177, 0.0321, 0.0257, 0.0286],
       device='cuda:2'), in_proj_covar=tensor([0.0089, 0.0120, 0.0146, 0.0122, 0.0114, 0.0114, 0.0094, 0.0123],
       device='cuda:2'), out_proj_covar=tensor([7.1781e-05, 9.6650e-05, 1.2282e-04, 9.8427e-05, 9.3305e-05, 8.8535e-05,
        7.4271e-05, 9.7945e-05], device='cuda:2')
2022-12-23 10:39:43,494 INFO [train.py:894] (2/4) Epoch 17, batch 1200, loss[loss=0.1681, simple_loss=0.2467, pruned_loss=0.04469, over 18364.00 frames. ], tot_loss[loss=0.1818, simple_loss=0.2688, pruned_loss=0.04737, over 3704029.98 frames. ], batch size: 46, lr: 6.81e-03, grad_scale: 8.0
2022-12-23 10:39:52,710 INFO [zipformer.py:660] (2/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=57305.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 10:40:06,484 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([3.2900, 3.1552, 2.0313, 1.6376, 3.7255, 3.8521, 2.9944, 2.6200],
       device='cuda:2'), covar=tensor([0.0361, 0.0349, 0.0607, 0.0770, 0.0177, 0.0286, 0.0461, 0.0678],
       device='cuda:2'), in_proj_covar=tensor([0.0122, 0.0124, 0.0128, 0.0119, 0.0095, 0.0121, 0.0135, 0.0155],
       device='cuda:2'), out_proj_covar=tensor([0.0001, 0.0002, 0.0002, 0.0001, 0.0001, 0.0001, 0.0002, 0.0002],
       device='cuda:2')
2022-12-23 10:40:37,781 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 2.106e+02 3.673e+02 4.461e+02 5.439e+02 1.096e+03, threshold=8.923e+02, percent-clipped=3.0
2022-12-23 10:40:52,555 WARNING [train.py:1060] (2/4) Exclude cut with ID 2929-85685-0062-62366-0 from training. Duration: 20.26
2022-12-23 10:40:58,950 INFO [train.py:894] (2/4) Epoch 17, batch 1250, loss[loss=0.2105, simple_loss=0.2985, pruned_loss=0.06122, over 18551.00 frames. ], tot_loss[loss=0.1823, simple_loss=0.2695, pruned_loss=0.04756, over 3706884.78 frames. ], batch size: 58, lr: 6.81e-03, grad_scale: 8.0
2022-12-23 10:40:59,126 INFO [zipformer.py:660] (2/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=57349.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 10:41:02,303 INFO [zipformer.py:660] (2/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=57351.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 10:41:06,573 WARNING [train.py:1060] (2/4) Exclude cut with ID 5239-32139-0030-9324-0_sp0.9 from training. Duration: 21.3444375
2022-12-23 10:41:19,527 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([2.6868, 2.4572, 1.8072, 1.2248, 3.0246, 2.8718, 2.5677, 2.0609],
       device='cuda:2'), covar=tensor([0.0318, 0.0354, 0.0546, 0.0796, 0.0215, 0.0305, 0.0383, 0.0740],
       device='cuda:2'), in_proj_covar=tensor([0.0123, 0.0124, 0.0128, 0.0120, 0.0096, 0.0121, 0.0136, 0.0156],
       device='cuda:2'), out_proj_covar=tensor([0.0001, 0.0002, 0.0002, 0.0001, 0.0001, 0.0001, 0.0002, 0.0002],
       device='cuda:2')
2022-12-23 10:42:02,283 WARNING [train.py:1060] (2/4) Exclude cut with ID 497-129325-0061-62254-0_sp1.1 from training. Duration: 0.97725
2022-12-23 10:42:11,909 INFO [zipformer.py:660] (2/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=57398.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 10:42:13,005 INFO [train.py:894] (2/4) Epoch 17, batch 1300, loss[loss=0.161, simple_loss=0.2374, pruned_loss=0.04232, over 18655.00 frames. ], tot_loss[loss=0.1812, simple_loss=0.2687, pruned_loss=0.04684, over 3708647.96 frames. ], batch size: 41, lr: 6.81e-03, grad_scale: 8.0
2022-12-23 10:42:45,614 WARNING [train.py:1060] (2/4) Exclude cut with ID 7395-89880-0031-39906-0_sp0.9 from training. Duration: 22.97225
2022-12-23 10:43:08,223 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 2.076e+02 3.308e+02 4.091e+02 4.945e+02 1.166e+03, threshold=8.182e+02, percent-clipped=4.0
2022-12-23 10:43:17,315 WARNING [train.py:1060] (2/4) Exclude cut with ID 7395-89880-0047-39922-0_sp0.9 from training. Duration: 21.97775
2022-12-23 10:43:28,397 WARNING [train.py:1060] (2/4) Exclude cut with ID 1112-1043-0006-89194-0_sp0.9 from training. Duration: 21.8333125
2022-12-23 10:43:29,888 INFO [train.py:894] (2/4) Epoch 17, batch 1350, loss[loss=0.193, simple_loss=0.2847, pruned_loss=0.05066, over 18532.00 frames. ], tot_loss[loss=0.1818, simple_loss=0.2693, pruned_loss=0.04713, over 3710066.27 frames. ], batch size: 55, lr: 6.81e-03, grad_scale: 8.0
2022-12-23 10:43:37,639 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([2.6056, 4.2002, 3.9831, 1.9889, 4.2616, 3.1956, 0.8121, 2.6385],
       device='cuda:2'), covar=tensor([0.1928, 0.0964, 0.1148, 0.3081, 0.0655, 0.0818, 0.4668, 0.1473],
       device='cuda:2'), in_proj_covar=tensor([0.0139, 0.0130, 0.0150, 0.0119, 0.0133, 0.0106, 0.0140, 0.0109],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0002, 0.0003, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-23 10:43:37,922 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.5017, 1.9404, 1.4794, 2.2029, 2.5825, 1.5575, 1.4197, 1.2620],
       device='cuda:2'), covar=tensor([0.2017, 0.1726, 0.1647, 0.1043, 0.1286, 0.1168, 0.2219, 0.1620],
       device='cuda:2'), in_proj_covar=tensor([0.0247, 0.0221, 0.0212, 0.0196, 0.0258, 0.0195, 0.0219, 0.0197],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-23 10:43:38,946 WARNING [train.py:1060] (2/4) Exclude cut with ID 1914-133440-0031-94921-0 from training. Duration: 20.47
2022-12-23 10:43:45,299 INFO [zipformer.py:660] (2/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=57459.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 10:43:45,475 INFO [zipformer.py:660] (2/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=57459.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 10:43:49,696 INFO [zipformer.py:660] (2/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=57462.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 10:43:55,382 INFO [scaling.py:679] (2/4) Whitening: num_groups=8, num_channels=96, metric=1.60 vs. limit=2.0
2022-12-23 10:44:39,369 INFO [zipformer.py:660] (2/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=57495.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 10:44:43,913 WARNING [train.py:1060] (2/4) Exclude cut with ID 7395-89880-0037-39912-0_sp0.9 from training. Duration: 20.67225
2022-12-23 10:44:45,324 INFO [train.py:894] (2/4) Epoch 17, batch 1400, loss[loss=0.1725, simple_loss=0.2677, pruned_loss=0.03868, over 18656.00 frames. ], tot_loss[loss=0.1824, simple_loss=0.27, pruned_loss=0.04736, over 3711213.23 frames. ], batch size: 78, lr: 6.80e-03, grad_scale: 8.0
2022-12-23 10:44:57,888 INFO [zipformer.py:660] (2/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=57507.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 10:45:05,451 WARNING [train.py:1060] (2/4) Exclude cut with ID 1914-133440-0024-94914-0_sp0.9 from training. Duration: 25.2444375
2022-12-23 10:45:22,316 INFO [zipformer.py:660] (2/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=57523.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 10:45:28,784 WARNING [train.py:1060] (2/4) Exclude cut with ID 3340-169293-0021-76797-0_sp0.9 from training. Duration: 21.1445
2022-12-23 10:45:40,194 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 2.185e+02 3.513e+02 4.155e+02 4.979e+02 8.906e+02, threshold=8.311e+02, percent-clipped=2.0
2022-12-23 10:45:51,106 INFO [zipformer.py:660] (2/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=57542.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 10:46:00,941 INFO [train.py:894] (2/4) Epoch 17, batch 1450, loss[loss=0.1809, simple_loss=0.2788, pruned_loss=0.04152, over 18706.00 frames. ], tot_loss[loss=0.1823, simple_loss=0.2702, pruned_loss=0.04722, over 3712684.50 frames. ], batch size: 52, lr: 6.80e-03, grad_scale: 8.0
2022-12-23 10:46:06,938 INFO [zipformer.py:660] (2/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=57553.0, num_to_drop=1, layers_to_drop={0}
2022-12-23 10:46:12,070 INFO [zipformer.py:660] (2/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=57556.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 10:46:23,886 INFO [zipformer.py:660] (2/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=57564.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 10:46:44,483 WARNING [train.py:1060] (2/4) Exclude cut with ID 2929-85685-0079-62383-0_sp0.9 from training. Duration: 33.038875
2022-12-23 10:47:15,626 INFO [train.py:894] (2/4) Epoch 17, batch 1500, loss[loss=0.1767, simple_loss=0.2671, pruned_loss=0.04314, over 18507.00 frames. ], tot_loss[loss=0.1812, simple_loss=0.269, pruned_loss=0.0467, over 3712313.46 frames. ], batch size: 52, lr: 6.80e-03, grad_scale: 8.0
2022-12-23 10:47:17,662 INFO [zipformer.py:660] (2/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=57600.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 10:47:22,896 WARNING [train.py:1060] (2/4) Exclude cut with ID 6426-64291-0000-16059-0_sp0.9 from training. Duration: 20.0944375
2022-12-23 10:47:36,326 WARNING [train.py:1060] (2/4) Exclude cut with ID 2929-85685-0044-62348-0_sp1.1 from training. Duration: 20.4
2022-12-23 10:47:36,530 INFO [zipformer.py:660] (2/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=57612.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 10:47:44,340 WARNING [train.py:1060] (2/4) Exclude cut with ID 6330-62851-0022-91297-0 from training. Duration: 20.085
2022-12-23 10:47:55,620 WARNING [train.py:1060] (2/4) Exclude cut with ID 4860-13185-0032-76709-0_sp0.9 from training. Duration: 23.07775
2022-12-23 10:48:12,657 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 2.389e+02 3.464e+02 4.346e+02 5.489e+02 1.288e+03, threshold=8.692e+02, percent-clipped=8.0
2022-12-23 10:48:34,271 INFO [train.py:894] (2/4) Epoch 17, batch 1550, loss[loss=0.2095, simple_loss=0.2988, pruned_loss=0.06007, over 18662.00 frames. ], tot_loss[loss=0.1807, simple_loss=0.2682, pruned_loss=0.04655, over 3712163.95 frames. ], batch size: 60, lr: 6.79e-03, grad_scale: 8.0
2022-12-23 10:48:34,538 INFO [zipformer.py:660] (2/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=57649.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 10:48:37,821 INFO [zipformer.py:660] (2/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=57651.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 10:48:42,086 WARNING [train.py:1060] (2/4) Exclude cut with ID 2929-85685-0044-62348-0_sp0.9 from training. Duration: 24.9333125
2022-12-23 10:49:04,669 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([2.0258, 1.9301, 1.4308, 2.1221, 2.1427, 1.8734, 2.6265, 2.0154],
       device='cuda:2'), covar=tensor([0.0820, 0.1654, 0.2786, 0.1719, 0.1658, 0.0865, 0.0928, 0.1153],
       device='cuda:2'), in_proj_covar=tensor([0.0174, 0.0204, 0.0244, 0.0284, 0.0230, 0.0186, 0.0207, 0.0200],
       device='cuda:2'), out_proj_covar=tensor([0.0001, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-23 10:49:06,001 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.4589, 1.8756, 1.4047, 2.2160, 2.3777, 1.4814, 1.4777, 1.2678],
       device='cuda:2'), covar=tensor([0.2089, 0.1820, 0.1799, 0.1039, 0.1301, 0.1278, 0.2248, 0.1712],
       device='cuda:2'), in_proj_covar=tensor([0.0245, 0.0220, 0.0212, 0.0195, 0.0257, 0.0194, 0.0218, 0.0196],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-23 10:49:28,285 WARNING [train.py:1060] (2/4) Exclude cut with ID 5118-111612-0016-124680-0_sp0.9 from training. Duration: 20.388875
2022-12-23 10:49:34,188 WARNING [train.py:1060] (2/4) Exclude cut with ID 432-122774-0017-62487-0_sp1.1 from training. Duration: 20.3590625
2022-12-23 10:49:43,446 INFO [scaling.py:679] (2/4) Whitening: num_groups=8, num_channels=192, metric=1.92 vs. limit=2.0
2022-12-23 10:49:47,395 INFO [zipformer.py:660] (2/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=57697.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 10:49:50,132 INFO [train.py:894] (2/4) Epoch 17, batch 1600, loss[loss=0.1829, simple_loss=0.2698, pruned_loss=0.04795, over 18441.00 frames. ], tot_loss[loss=0.1794, simple_loss=0.267, pruned_loss=0.04586, over 3711054.56 frames. ], batch size: 59, lr: 6.79e-03, grad_scale: 8.0
2022-12-23 10:49:50,313 INFO [zipformer.py:660] (2/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=57699.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 10:50:40,934 INFO [zipformer.py:660] (2/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=57733.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 10:50:41,042 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([3.1798, 2.3832, 1.8464, 3.0851, 2.1921, 2.2837, 2.3647, 3.3850],
       device='cuda:2'), covar=tensor([0.1820, 0.3225, 0.1808, 0.2891, 0.3917, 0.0994, 0.3235, 0.0670],
       device='cuda:2'), in_proj_covar=tensor([0.0290, 0.0284, 0.0240, 0.0347, 0.0263, 0.0223, 0.0280, 0.0207],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0003, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-23 10:50:43,432 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 2.172e+02 3.678e+02 4.212e+02 5.309e+02 1.200e+03, threshold=8.424e+02, percent-clipped=1.0
2022-12-23 10:50:45,119 WARNING [train.py:1060] (2/4) Exclude cut with ID 3557-8342-0013-54691-0_sp1.1 from training. Duration: 0.836375
2022-12-23 10:51:04,903 INFO [train.py:894] (2/4) Epoch 17, batch 1650, loss[loss=0.1978, simple_loss=0.2856, pruned_loss=0.055, over 18623.00 frames. ], tot_loss[loss=0.1805, simple_loss=0.2678, pruned_loss=0.04664, over 3712126.70 frames. ], batch size: 69, lr: 6.79e-03, grad_scale: 8.0
2022-12-23 10:51:12,984 INFO [zipformer.py:660] (2/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=57754.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 10:51:30,447 WARNING [train.py:1060] (2/4) Exclude cut with ID 8565-290391-0049-67394-0_sp0.9 from training. Duration: 21.3166875
2022-12-23 10:51:37,401 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([0.0623, 0.9015, 1.1444, 0.5695, 0.5887, 1.2139, 1.2386, 1.1770],
       device='cuda:2'), covar=tensor([0.0759, 0.0333, 0.0334, 0.0363, 0.0426, 0.0454, 0.0256, 0.0595],
       device='cuda:2'), in_proj_covar=tensor([0.0150, 0.0167, 0.0122, 0.0139, 0.0149, 0.0141, 0.0160, 0.0169],
       device='cuda:2'), out_proj_covar=tensor([1.1614e-04, 1.3073e-04, 9.3899e-05, 1.0530e-04, 1.1434e-04, 1.1067e-04,
        1.2547e-04, 1.3092e-04], device='cuda:2')
2022-12-23 10:51:45,949 INFO [scaling.py:679] (2/4) Whitening: num_groups=8, num_channels=96, metric=1.31 vs. limit=2.0
2022-12-23 10:51:59,963 WARNING [train.py:1060] (2/4) Exclude cut with ID 6533-399-0029-104863-0_sp0.9 from training. Duration: 22.1055625
2022-12-23 10:52:11,469 WARNING [train.py:1060] (2/4) Exclude cut with ID 7699-105389-0094-26379-0_sp1.1 from training. Duration: 21.77725
2022-12-23 10:52:13,215 INFO [zipformer.py:660] (2/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=57794.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 10:52:20,529 INFO [train.py:894] (2/4) Epoch 17, batch 1700, loss[loss=0.188, simple_loss=0.271, pruned_loss=0.05244, over 18571.00 frames. ], tot_loss[loss=0.1824, simple_loss=0.2686, pruned_loss=0.04812, over 3711937.10 frames. ], batch size: 49, lr: 6.79e-03, grad_scale: 8.0
2022-12-23 10:52:22,899 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([2.4618, 2.4366, 1.9121, 1.9251, 2.4080, 3.0425, 2.8747, 1.9423],
       device='cuda:2'), covar=tensor([0.0364, 0.0271, 0.0462, 0.0265, 0.0249, 0.0331, 0.0294, 0.0340],
       device='cuda:2'), in_proj_covar=tensor([0.0091, 0.0123, 0.0151, 0.0125, 0.0118, 0.0117, 0.0096, 0.0126],
       device='cuda:2'), out_proj_covar=tensor([7.3528e-05, 9.8455e-05, 1.2632e-04, 1.0093e-04, 9.6233e-05, 9.0991e-05,
        7.6207e-05, 1.0041e-04], device='cuda:2')
2022-12-23 10:52:32,573 WARNING [train.py:1060] (2/4) Exclude cut with ID 7255-291500-0005-134304-0_sp0.9 from training. Duration: 27.8166875
2022-12-23 10:52:49,802 INFO [zipformer.py:660] (2/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=57818.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 10:52:58,912 WARNING [train.py:1060] (2/4) Exclude cut with ID 7357-94126-0021-15852-0_sp1.1 from training. Duration: 22.5090625
2022-12-23 10:53:02,095 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([2.9109, 5.4630, 4.9850, 2.6762, 5.6371, 4.1716, 0.8420, 4.1322],
       device='cuda:2'), covar=tensor([0.1875, 0.0829, 0.1191, 0.2892, 0.0628, 0.0728, 0.4989, 0.0972],
       device='cuda:2'), in_proj_covar=tensor([0.0142, 0.0132, 0.0152, 0.0122, 0.0136, 0.0107, 0.0142, 0.0110],
       device='cuda:2'), out_proj_covar=tensor([0.0003, 0.0002, 0.0003, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-23 10:53:04,846 WARNING [train.py:1060] (2/4) Exclude cut with ID 7255-291500-0005-134304-0 from training. Duration: 25.035
2022-12-23 10:53:14,895 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 3.260e+02 4.546e+02 5.067e+02 6.355e+02 1.352e+03, threshold=1.013e+03, percent-clipped=7.0
2022-12-23 10:53:22,460 WARNING [train.py:1060] (2/4) Exclude cut with ID 774-127930-0014-10412-0_sp1.1 from training. Duration: 0.95
2022-12-23 10:53:25,479 INFO [zipformer.py:660] (2/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=57842.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 10:53:25,648 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.9553, 1.7087, 1.6850, 0.9173, 2.2831, 1.9881, 1.9098, 1.3418],
       device='cuda:2'), covar=tensor([0.0344, 0.0467, 0.0459, 0.0758, 0.0284, 0.0349, 0.0410, 0.0963],
       device='cuda:2'), in_proj_covar=tensor([0.0122, 0.0123, 0.0127, 0.0119, 0.0095, 0.0120, 0.0134, 0.0156],
       device='cuda:2'), out_proj_covar=tensor([0.0001, 0.0002, 0.0002, 0.0001, 0.0001, 0.0001, 0.0002, 0.0002],
       device='cuda:2')
2022-12-23 10:53:35,706 INFO [train.py:894] (2/4) Epoch 17, batch 1750, loss[loss=0.2129, simple_loss=0.3058, pruned_loss=0.05999, over 18571.00 frames. ], tot_loss[loss=0.1846, simple_loss=0.2698, pruned_loss=0.04974, over 3711951.53 frames. ], batch size: 56, lr: 6.78e-03, grad_scale: 8.0
2022-12-23 10:53:38,960 INFO [zipformer.py:660] (2/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=57851.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 10:53:40,230 WARNING [train.py:1060] (2/4) Exclude cut with ID 3033-130750-0096-55598-0_sp0.9 from training. Duration: 0.92225
2022-12-23 10:53:42,164 INFO [zipformer.py:660] (2/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=57853.0, num_to_drop=1, layers_to_drop={0}
2022-12-23 10:54:07,102 WARNING [train.py:1060] (2/4) Exclude cut with ID 4511-76322-0006-80011-0 from training. Duration: 21.97
2022-12-23 10:54:25,773 WARNING [train.py:1060] (2/4) Exclude cut with ID 7492-105653-0055-62765-0_sp0.9 from training. Duration: 21.97225
2022-12-23 10:54:25,807 WARNING [train.py:1060] (2/4) Exclude cut with ID 453-131332-0000-47844-0_sp0.9 from training. Duration: 25.3333125
2022-12-23 10:54:36,730 WARNING [train.py:1060] (2/4) Exclude cut with ID 5172-29468-0015-19128-0_sp0.9 from training. Duration: 21.5055625
2022-12-23 10:54:38,139 INFO [zipformer.py:660] (2/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=57890.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 10:54:45,855 WARNING [train.py:1060] (2/4) Exclude cut with ID 453-131332-0000-47844-0_sp1.1 from training. Duration: 20.72725
2022-12-23 10:54:52,431 INFO [train.py:894] (2/4) Epoch 17, batch 1800, loss[loss=0.1733, simple_loss=0.2528, pruned_loss=0.04687, over 18442.00 frames. ], tot_loss[loss=0.1875, simple_loss=0.2715, pruned_loss=0.05176, over 3711499.34 frames. ], batch size: 48, lr: 6.78e-03, grad_scale: 8.0
2022-12-23 10:54:54,299 INFO [zipformer.py:660] (2/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=57900.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 10:54:55,907 INFO [zipformer.py:660] (2/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=57901.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 10:55:18,173 WARNING [train.py:1060] (2/4) Exclude cut with ID 8631-249866-0030-130156-0_sp0.9 from training. Duration: 26.32775
2022-12-23 10:55:47,117 WARNING [train.py:1060] (2/4) Exclude cut with ID 3867-173237-0077-144769-0 from training. Duration: 20.025
2022-12-23 10:55:49,146 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 2.792e+02 4.677e+02 5.758e+02 7.423e+02 3.377e+03, threshold=1.152e+03, percent-clipped=6.0
2022-12-23 10:55:53,724 WARNING [train.py:1060] (2/4) Exclude cut with ID 6709-74022-0004-86860-0_sp1.1 from training. Duration: 0.9409375
2022-12-23 10:55:55,087 WARNING [train.py:1060] (2/4) Exclude cut with ID 4757-1811-0023-62229-0_sp0.9 from training. Duration: 21.37775
2022-12-23 10:56:07,934 INFO [zipformer.py:660] (2/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=57948.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 10:56:09,126 INFO [train.py:894] (2/4) Epoch 17, batch 1850, loss[loss=0.2083, simple_loss=0.2865, pruned_loss=0.06503, over 18375.00 frames. ], tot_loss[loss=0.1907, simple_loss=0.2731, pruned_loss=0.0541, over 3712363.73 frames. ], batch size: 51, lr: 6.78e-03, grad_scale: 8.0
2022-12-23 10:56:13,599 WARNING [train.py:1060] (2/4) Exclude cut with ID 1250-135782-0004-25974-0_sp0.9 from training. Duration: 21.17225
2022-12-23 10:56:15,418 WARNING [train.py:1060] (2/4) Exclude cut with ID 7357-94126-0021-15852-0_sp0.9 from training. Duration: 27.511125
2022-12-23 10:56:42,871 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([0.2887, 1.5847, 1.9141, 1.0767, 1.1012, 2.0424, 1.8830, 1.5522],
       device='cuda:2'), covar=tensor([0.0779, 0.0309, 0.0273, 0.0352, 0.0398, 0.0403, 0.0221, 0.0637],
       device='cuda:2'), in_proj_covar=tensor([0.0151, 0.0171, 0.0125, 0.0140, 0.0150, 0.0143, 0.0161, 0.0171],
       device='cuda:2'), out_proj_covar=tensor([1.1735e-04, 1.3336e-04, 9.5407e-05, 1.0566e-04, 1.1529e-04, 1.1228e-04,
        1.2683e-04, 1.3246e-04], device='cuda:2')
2022-12-23 10:56:48,272 WARNING [train.py:1060] (2/4) Exclude cut with ID 453-131332-0000-47844-0 from training. Duration: 22.8
2022-12-23 10:56:53,194 WARNING [train.py:1060] (2/4) Exclude cut with ID 4964-30587-0040-44509-0 from training. Duration: 22.585
2022-12-23 10:57:24,025 INFO [train.py:894] (2/4) Epoch 17, batch 1900, loss[loss=0.18, simple_loss=0.2661, pruned_loss=0.04694, over 18677.00 frames. ], tot_loss[loss=0.1927, simple_loss=0.2741, pruned_loss=0.05563, over 3712595.66 frames. ], batch size: 48, lr: 6.77e-03, grad_scale: 8.0
2022-12-23 10:57:24,066 WARNING [train.py:1060] (2/4) Exclude cut with ID 6978-92210-0001-146967-0_sp0.9 from training. Duration: 22.0166875
2022-12-23 10:57:42,056 WARNING [train.py:1060] (2/4) Exclude cut with ID 7255-291500-0003-134302-0_sp1.1 from training. Duration: 24.395375
2022-12-23 10:57:48,298 WARNING [train.py:1060] (2/4) Exclude cut with ID 3488-85273-0017-41203-0_sp0.9 from training. Duration: 27.47775
2022-12-23 10:57:52,414 WARNING [train.py:1060] (2/4) Exclude cut with ID 432-122774-0017-62487-0_sp0.9 from training. Duration: 24.8833125
2022-12-23 10:57:54,166 WARNING [train.py:1060] (2/4) Exclude cut with ID 6758-72288-0033-108368-0 from training. Duration: 23.39
2022-12-23 10:58:00,566 WARNING [train.py:1060] (2/4) Exclude cut with ID 8291-282929-0007-12994-0_sp0.9 from training. Duration: 28.72225
2022-12-23 10:58:09,590 INFO [scaling.py:679] (2/4) Whitening: num_groups=8, num_channels=96, metric=1.47 vs. limit=2.0
2022-12-23 10:58:11,273 WARNING [train.py:1060] (2/4) Exclude cut with ID 585-294811-0110-133686-0_sp0.9 from training. Duration: 20.8944375
2022-12-23 10:58:23,481 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 2.619e+02 4.765e+02 6.161e+02 7.873e+02 1.431e+03, threshold=1.232e+03, percent-clipped=2.0
2022-12-23 10:58:27,775 WARNING [train.py:1060] (2/4) Exclude cut with ID 5796-66357-0007-116447-0_sp0.9 from training. Duration: 23.8444375
2022-12-23 10:58:31,587 INFO [scaling.py:679] (2/4) Whitening: num_groups=8, num_channels=96, metric=1.13 vs. limit=2.0
2022-12-23 10:58:44,198 INFO [train.py:894] (2/4) Epoch 17, batch 1950, loss[loss=0.1991, simple_loss=0.2708, pruned_loss=0.06371, over 18445.00 frames. ], tot_loss[loss=0.194, simple_loss=0.2747, pruned_loss=0.05669, over 3713443.55 frames. ], batch size: 50, lr: 6.77e-03, grad_scale: 8.0
2022-12-23 10:58:52,394 INFO [zipformer.py:660] (2/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=58054.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 10:58:53,647 WARNING [train.py:1060] (2/4) Exclude cut with ID 8291-282929-0007-12994-0 from training. Duration: 25.85
2022-12-23 10:58:53,659 WARNING [train.py:1060] (2/4) Exclude cut with ID 8291-282929-0023-13010-0 from training. Duration: 21.39
2022-12-23 10:58:58,573 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([2.5536, 2.1667, 1.6352, 2.3963, 1.8785, 1.9739, 1.9735, 2.3736],
       device='cuda:2'), covar=tensor([0.2020, 0.2878, 0.1907, 0.2556, 0.3331, 0.1125, 0.2799, 0.0891],
       device='cuda:2'), in_proj_covar=tensor([0.0287, 0.0281, 0.0237, 0.0345, 0.0261, 0.0219, 0.0277, 0.0206],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0003, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-23 10:59:05,186 WARNING [train.py:1060] (2/4) Exclude cut with ID 7357-94126-0014-15845-0 from training. Duration: 27.92
2022-12-23 10:59:33,690 WARNING [train.py:1060] (2/4) Exclude cut with ID 8631-249866-0039-130165-0_sp0.9 from training. Duration: 20.661125
2022-12-23 10:59:45,964 INFO [zipformer.py:660] (2/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=58089.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 10:59:55,173 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.7252, 1.5135, 1.0545, 0.2717, 1.1776, 1.5747, 1.3769, 1.4867],
       device='cuda:2'), covar=tensor([0.0608, 0.0507, 0.1040, 0.1525, 0.1012, 0.1548, 0.1553, 0.0658],
       device='cuda:2'), in_proj_covar=tensor([0.0169, 0.0182, 0.0204, 0.0190, 0.0209, 0.0199, 0.0214, 0.0199],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-23 10:59:56,200 WARNING [train.py:1060] (2/4) Exclude cut with ID 7357-94126-0043-15874-0_sp0.9 from training. Duration: 20.07225
2022-12-23 11:00:01,377 INFO [train.py:894] (2/4) Epoch 17, batch 2000, loss[loss=0.2014, simple_loss=0.2902, pruned_loss=0.0563, over 18661.00 frames. ], tot_loss[loss=0.1943, simple_loss=0.2746, pruned_loss=0.05696, over 3713094.79 frames. ], batch size: 60, lr: 6.77e-03, grad_scale: 8.0
2022-12-23 11:00:06,322 WARNING [train.py:1060] (2/4) Exclude cut with ID 1085-156170-0017-128270-0 from training. Duration: 21.01
2022-12-23 11:00:06,435 INFO [zipformer.py:660] (2/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=58102.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 11:00:30,779 INFO [zipformer.py:660] (2/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=58118.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 11:00:57,068 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 3.188e+02 4.438e+02 5.445e+02 6.526e+02 1.482e+03, threshold=1.089e+03, percent-clipped=1.0
2022-12-23 11:01:04,372 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([2.2189, 2.7169, 2.7268, 1.1680, 2.8210, 2.0564, 0.5028, 1.8637],
       device='cuda:2'), covar=tensor([0.2292, 0.1696, 0.1778, 0.4181, 0.1515, 0.1340, 0.5088, 0.1863],
       device='cuda:2'), in_proj_covar=tensor([0.0143, 0.0134, 0.0154, 0.0123, 0.0138, 0.0109, 0.0144, 0.0111],
       device='cuda:2'), out_proj_covar=tensor([0.0003, 0.0002, 0.0003, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-23 11:01:05,952 INFO [zipformer.py:660] (2/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=58141.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 11:01:10,968 INFO [zipformer.py:660] (2/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=58144.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 11:01:13,808 WARNING [train.py:1060] (2/4) Exclude cut with ID 2195-150901-0045-59933-0 from training. Duration: 20.65
2022-12-23 11:01:18,091 INFO [train.py:894] (2/4) Epoch 17, batch 2050, loss[loss=0.2266, simple_loss=0.3088, pruned_loss=0.0722, over 18595.00 frames. ], tot_loss[loss=0.1945, simple_loss=0.2744, pruned_loss=0.05734, over 3714076.71 frames. ], batch size: 56, lr: 6.77e-03, grad_scale: 8.0
2022-12-23 11:01:21,044 WARNING [train.py:1060] (2/4) Exclude cut with ID 5796-66357-0007-116447-0 from training. Duration: 21.46
2022-12-23 11:01:21,246 INFO [zipformer.py:660] (2/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=58151.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 11:01:42,681 INFO [zipformer.py:660] (2/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=58166.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 11:02:08,995 WARNING [train.py:1060] (2/4) Exclude cut with ID 3557-8342-0013-54691-0 from training. Duration: 0.92
2022-12-23 11:02:13,625 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([2.2368, 2.0942, 1.5989, 1.7419, 2.1775, 2.7537, 2.5054, 1.8499],
       device='cuda:2'), covar=tensor([0.0323, 0.0291, 0.0476, 0.0293, 0.0267, 0.0328, 0.0322, 0.0355],
       device='cuda:2'), in_proj_covar=tensor([0.0092, 0.0126, 0.0152, 0.0125, 0.0118, 0.0118, 0.0097, 0.0127],
       device='cuda:2'), out_proj_covar=tensor([7.3986e-05, 1.0083e-04, 1.2705e-04, 1.0133e-04, 9.6528e-05, 9.1600e-05,
        7.6677e-05, 1.0154e-04], device='cuda:2')
2022-12-23 11:02:16,227 WARNING [train.py:1060] (2/4) Exclude cut with ID 8291-282929-0023-13010-0_sp0.9 from training. Duration: 23.7666875
2022-12-23 11:02:33,665 INFO [train.py:894] (2/4) Epoch 17, batch 2100, loss[loss=0.1726, simple_loss=0.2475, pruned_loss=0.04881, over 18426.00 frames. ], tot_loss[loss=0.194, simple_loss=0.2734, pruned_loss=0.05729, over 3712798.58 frames. ], batch size: 48, lr: 6.76e-03, grad_scale: 8.0
2022-12-23 11:02:33,795 INFO [zipformer.py:660] (2/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=58199.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 11:02:38,616 INFO [zipformer.py:660] (2/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=58202.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 11:02:43,266 INFO [zipformer.py:660] (2/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=58205.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 11:02:52,315 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([2.0372, 1.2230, 1.8046, 1.7332, 2.0926, 1.9979, 1.8296, 1.6091],
       device='cuda:2'), covar=tensor([0.1996, 0.3121, 0.2307, 0.2568, 0.1727, 0.0888, 0.2729, 0.1177],
       device='cuda:2'), in_proj_covar=tensor([0.0261, 0.0294, 0.0272, 0.0306, 0.0296, 0.0244, 0.0329, 0.0231],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-23 11:02:53,251 WARNING [train.py:1060] (2/4) Exclude cut with ID 8544-281189-0060-101339-0_sp0.9 from training. Duration: 20.861125
2022-12-23 11:03:02,873 WARNING [train.py:1060] (2/4) Exclude cut with ID 3488-65654-0031-41259-0_sp0.9 from training. Duration: 22.711125
2022-12-23 11:03:29,639 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 2.687e+02 4.700e+02 5.944e+02 7.223e+02 2.461e+03, threshold=1.189e+03, percent-clipped=2.0
2022-12-23 11:03:44,026 WARNING [train.py:1060] (2/4) Exclude cut with ID 6951-79737-0043-132310-0_sp1.1 from training. Duration: 22.986375
2022-12-23 11:03:51,319 INFO [train.py:894] (2/4) Epoch 17, batch 2150, loss[loss=0.2226, simple_loss=0.2936, pruned_loss=0.07576, over 18474.00 frames. ], tot_loss[loss=0.1964, simple_loss=0.2753, pruned_loss=0.05875, over 3713744.50 frames. ], batch size: 54, lr: 6.76e-03, grad_scale: 8.0
2022-12-23 11:04:00,558 WARNING [train.py:1060] (2/4) Exclude cut with ID 8040-260924-0003-80960-0_sp0.9 from training. Duration: 22.07225
2022-12-23 11:04:04,923 WARNING [train.py:1060] (2/4) Exclude cut with ID 7699-105389-0045-26330-0_sp0.9 from training. Duration: 20.3055625
2022-12-23 11:04:06,438 WARNING [train.py:1060] (2/4) Exclude cut with ID 6356-271890-0060-94317-0_sp0.9 from training. Duration: 20.72225
2022-12-23 11:04:25,644 WARNING [train.py:1060] (2/4) Exclude cut with ID 3488-85273-0017-41203-0_sp1.1 from training. Duration: 22.4818125
2022-12-23 11:04:44,186 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.8752, 1.4829, 0.7480, 1.3987, 2.1205, 1.3787, 1.6160, 1.8248],
       device='cuda:2'), covar=tensor([0.1566, 0.2051, 0.2519, 0.1541, 0.1792, 0.1777, 0.1489, 0.1637],
       device='cuda:2'), in_proj_covar=tensor([0.0092, 0.0097, 0.0116, 0.0094, 0.0115, 0.0091, 0.0097, 0.0093],
       device='cuda:2'), out_proj_covar=tensor([0.0003, 0.0003, 0.0004, 0.0003, 0.0004, 0.0003, 0.0003, 0.0003],
       device='cuda:2')
2022-12-23 11:04:51,395 WARNING [train.py:1060] (2/4) Exclude cut with ID 4964-30587-0040-44509-0_sp0.9 from training. Duration: 25.0944375
2022-12-23 11:04:54,696 WARNING [train.py:1060] (2/4) Exclude cut with ID 6533-399-0047-104881-0 from training. Duration: 21.515
2022-12-23 11:05:00,356 WARNING [train.py:1060] (2/4) Exclude cut with ID 7357-94126-0009-15840-0_sp0.9 from training. Duration: 27.02225
2022-12-23 11:05:06,268 WARNING [train.py:1060] (2/4) Exclude cut with ID 432-122774-0010-62480-0_sp0.9 from training. Duration: 22.22225
2022-12-23 11:05:07,552 INFO [train.py:894] (2/4) Epoch 17, batch 2200, loss[loss=0.2393, simple_loss=0.3066, pruned_loss=0.08601, over 18683.00 frames. ], tot_loss[loss=0.196, simple_loss=0.2749, pruned_loss=0.05852, over 3713519.40 frames. ], batch size: 60, lr: 6.76e-03, grad_scale: 8.0
2022-12-23 11:05:13,367 WARNING [train.py:1060] (2/4) Exclude cut with ID 4964-30587-0085-44554-0_sp0.9 from training. Duration: 20.85
2022-12-23 11:05:34,607 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([0.4435, 1.8675, 2.0546, 1.1873, 1.2939, 2.1697, 2.0344, 1.7167],
       device='cuda:2'), covar=tensor([0.0845, 0.0282, 0.0290, 0.0356, 0.0389, 0.0391, 0.0240, 0.0672],
       device='cuda:2'), in_proj_covar=tensor([0.0152, 0.0171, 0.0124, 0.0139, 0.0150, 0.0143, 0.0161, 0.0171],
       device='cuda:2'), out_proj_covar=tensor([1.1790e-04, 1.3332e-04, 9.4660e-05, 1.0513e-04, 1.1523e-04, 1.1171e-04,
        1.2659e-04, 1.3242e-04], device='cuda:2')
2022-12-23 11:05:47,385 WARNING [train.py:1060] (2/4) Exclude cut with ID 4295-39940-0007-92567-0 from training. Duration: 21.54
2022-12-23 11:05:53,228 WARNING [train.py:1060] (2/4) Exclude cut with ID 4964-30587-0040-44509-0_sp1.1 from training. Duration: 20.5318125
2022-12-23 11:06:03,154 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 2.624e+02 4.367e+02 5.459e+02 6.850e+02 1.946e+03, threshold=1.092e+03, percent-clipped=2.0
2022-12-23 11:06:03,251 WARNING [train.py:1060] (2/4) Exclude cut with ID 7255-291500-0012-134311-0_sp0.9 from training. Duration: 21.9333125
2022-12-23 11:06:23,581 INFO [train.py:894] (2/4) Epoch 17, batch 2250, loss[loss=0.2254, simple_loss=0.2987, pruned_loss=0.0761, over 18684.00 frames. ], tot_loss[loss=0.1966, simple_loss=0.2752, pruned_loss=0.05899, over 3713522.27 frames. ], batch size: 60, lr: 6.75e-03, grad_scale: 8.0
2022-12-23 11:06:36,167 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([3.2427, 2.8153, 2.8206, 3.2033, 2.8984, 2.8382, 3.3254, 1.0122],
       device='cuda:2'), covar=tensor([0.1060, 0.0858, 0.0950, 0.1052, 0.1757, 0.1479, 0.0877, 0.4951],
       device='cuda:2'), in_proj_covar=tensor([0.0333, 0.0220, 0.0231, 0.0258, 0.0314, 0.0263, 0.0281, 0.0277],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0001, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-23 11:06:51,936 WARNING [train.py:1060] (2/4) Exclude cut with ID 8631-249866-0025-130151-0_sp0.9 from training. Duration: 21.7944375
2022-12-23 11:07:03,587 WARNING [train.py:1060] (2/4) Exclude cut with ID 8291-282929-0002-12989-0_sp0.9 from training. Duration: 22.4666875
2022-12-23 11:07:11,558 INFO [scaling.py:679] (2/4) Whitening: num_groups=8, num_channels=96, metric=1.18 vs. limit=2.0
2022-12-23 11:07:12,199 WARNING [train.py:1060] (2/4) Exclude cut with ID 6121-9014-0076-24124-0 from training. Duration: 21.635
2022-12-23 11:07:16,405 WARNING [train.py:1060] (2/4) Exclude cut with ID 6121-9014-0076-24124-0_sp0.9 from training. Duration: 24.038875
2022-12-23 11:07:25,585 INFO [zipformer.py:660] (2/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=58389.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 11:07:40,064 INFO [train.py:894] (2/4) Epoch 17, batch 2300, loss[loss=0.1538, simple_loss=0.2269, pruned_loss=0.04034, over 18567.00 frames. ], tot_loss[loss=0.1954, simple_loss=0.2741, pruned_loss=0.05834, over 3713848.17 frames. ], batch size: 44, lr: 6.75e-03, grad_scale: 8.0
2022-12-23 11:08:02,226 WARNING [train.py:1060] (2/4) Exclude cut with ID 7255-291500-0009-134308-0_sp1.1 from training. Duration: 21.786375
2022-12-23 11:08:14,113 WARNING [train.py:1060] (2/4) Exclude cut with ID 8291-282929-0002-12989-0 from training. Duration: 20.22
2022-12-23 11:08:35,940 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 2.729e+02 4.639e+02 5.586e+02 7.120e+02 1.338e+03, threshold=1.117e+03, percent-clipped=3.0
2022-12-23 11:08:38,866 INFO [zipformer.py:660] (2/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=58437.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 11:08:52,628 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([2.2000, 1.6524, 0.9591, 1.6825, 2.5079, 1.7759, 2.0564, 2.2241],
       device='cuda:2'), covar=tensor([0.1433, 0.1963, 0.2281, 0.1456, 0.1585, 0.1608, 0.1303, 0.1539],
       device='cuda:2'), in_proj_covar=tensor([0.0091, 0.0096, 0.0115, 0.0093, 0.0114, 0.0090, 0.0096, 0.0092],
       device='cuda:2'), out_proj_covar=tensor([0.0003, 0.0003, 0.0004, 0.0003, 0.0004, 0.0003, 0.0003, 0.0003],
       device='cuda:2')
2022-12-23 11:08:56,501 INFO [train.py:894] (2/4) Epoch 17, batch 2350, loss[loss=0.1834, simple_loss=0.2696, pruned_loss=0.04859, over 18586.00 frames. ], tot_loss[loss=0.195, simple_loss=0.2741, pruned_loss=0.05795, over 3714159.19 frames. ], batch size: 56, lr: 6.75e-03, grad_scale: 8.0
2022-12-23 11:09:40,950 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([2.0273, 1.2585, 1.8009, 1.6923, 2.1299, 2.0077, 1.8178, 1.6552],
       device='cuda:2'), covar=tensor([0.2069, 0.3011, 0.2342, 0.2578, 0.1791, 0.0905, 0.2838, 0.1242],
       device='cuda:2'), in_proj_covar=tensor([0.0261, 0.0293, 0.0273, 0.0308, 0.0296, 0.0245, 0.0329, 0.0232],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-23 11:09:58,125 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([3.7373, 3.2413, 2.6604, 1.3966, 2.6524, 2.7639, 2.1827, 2.3400],
       device='cuda:2'), covar=tensor([0.0614, 0.0620, 0.1499, 0.1804, 0.1528, 0.1153, 0.1547, 0.1182],
       device='cuda:2'), in_proj_covar=tensor([0.0172, 0.0183, 0.0206, 0.0190, 0.0209, 0.0200, 0.0215, 0.0199],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-23 11:10:10,036 INFO [zipformer.py:660] (2/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=58497.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 11:10:12,651 INFO [train.py:894] (2/4) Epoch 17, batch 2400, loss[loss=0.1584, simple_loss=0.2453, pruned_loss=0.03577, over 18432.00 frames. ], tot_loss[loss=0.194, simple_loss=0.273, pruned_loss=0.05752, over 3713957.66 frames. ], batch size: 48, lr: 6.75e-03, grad_scale: 8.0
2022-12-23 11:10:14,899 WARNING [train.py:1060] (2/4) Exclude cut with ID 6951-79737-0043-132310-0 from training. Duration: 25.285
2022-12-23 11:10:15,048 INFO [zipformer.py:660] (2/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=58500.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 11:10:53,739 INFO [scaling.py:679] (2/4) Whitening: num_groups=8, num_channels=96, metric=1.37 vs. limit=2.0
2022-12-23 11:11:08,059 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 2.821e+02 4.687e+02 5.741e+02 6.736e+02 1.581e+03, threshold=1.148e+03, percent-clipped=3.0
2022-12-23 11:11:19,598 WARNING [train.py:1060] (2/4) Exclude cut with ID 811-130148-0001-63453-0_sp0.9 from training. Duration: 20.861125
2022-12-23 11:11:28,749 INFO [train.py:894] (2/4) Epoch 17, batch 2450, loss[loss=0.1741, simple_loss=0.256, pruned_loss=0.04609, over 18548.00 frames. ], tot_loss[loss=0.1939, simple_loss=0.273, pruned_loss=0.05736, over 3713872.84 frames. ], batch size: 49, lr: 6.74e-03, grad_scale: 8.0
2022-12-23 11:11:43,302 WARNING [train.py:1060] (2/4) Exclude cut with ID 6010-56788-0055-90261-0 from training. Duration: 20.88
2022-12-23 11:12:02,644 INFO [scaling.py:679] (2/4) Whitening: num_groups=8, num_channels=96, metric=1.47 vs. limit=2.0
2022-12-23 11:12:15,282 WARNING [train.py:1060] (2/4) Exclude cut with ID 7357-94126-0045-15876-0_sp0.9 from training. Duration: 23.4166875
2022-12-23 11:12:42,902 INFO [scaling.py:679] (2/4) Whitening: num_groups=8, num_channels=96, metric=1.22 vs. limit=2.0
2022-12-23 11:12:43,590 INFO [train.py:894] (2/4) Epoch 17, batch 2500, loss[loss=0.1887, simple_loss=0.2757, pruned_loss=0.05081, over 18543.00 frames. ], tot_loss[loss=0.1934, simple_loss=0.2727, pruned_loss=0.05701, over 3713609.44 frames. ], batch size: 55, lr: 6.74e-03, grad_scale: 8.0
2022-12-23 11:13:11,463 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([2.2861, 2.5810, 2.8158, 1.5395, 3.1555, 2.9703, 2.0155, 3.2979],
       device='cuda:2'), covar=tensor([0.1218, 0.1598, 0.1409, 0.2190, 0.0662, 0.1191, 0.2139, 0.0557],
       device='cuda:2'), in_proj_covar=tensor([0.0197, 0.0211, 0.0206, 0.0196, 0.0176, 0.0216, 0.0213, 0.0191],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-23 11:13:32,664 WARNING [train.py:1060] (2/4) Exclude cut with ID 7255-291500-0006-134305-0 from training. Duration: 21.24
2022-12-23 11:13:32,674 WARNING [train.py:1060] (2/4) Exclude cut with ID 6533-399-0047-104881-0_sp0.9 from training. Duration: 23.9055625
2022-12-23 11:13:37,239 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 2.626e+02 4.362e+02 5.589e+02 6.894e+02 1.162e+03, threshold=1.118e+03, percent-clipped=1.0
2022-12-23 11:13:38,053 INFO [scaling.py:679] (2/4) Whitening: num_groups=8, num_channels=96, metric=1.28 vs. limit=2.0
2022-12-23 11:13:59,180 INFO [train.py:894] (2/4) Epoch 17, batch 2550, loss[loss=0.1713, simple_loss=0.2538, pruned_loss=0.04435, over 18435.00 frames. ], tot_loss[loss=0.193, simple_loss=0.2725, pruned_loss=0.05679, over 3714031.44 frames. ], batch size: 48, lr: 6.74e-03, grad_scale: 8.0
2022-12-23 11:14:05,613 WARNING [train.py:1060] (2/4) Exclude cut with ID 6758-72288-0033-108368-0_sp0.9 from training. Duration: 25.988875
2022-12-23 11:14:12,020 WARNING [train.py:1060] (2/4) Exclude cut with ID 7255-291500-0001-134300-0_sp0.9 from training. Duration: 20.67225
2022-12-23 11:14:17,436 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.6056, 1.3892, 1.4106, 1.8992, 1.5825, 3.2591, 1.2731, 1.5715],
       device='cuda:2'), covar=tensor([0.0781, 0.1733, 0.1052, 0.0818, 0.1404, 0.0241, 0.1418, 0.1459],
       device='cuda:2'), in_proj_covar=tensor([0.0072, 0.0082, 0.0072, 0.0074, 0.0091, 0.0074, 0.0084, 0.0077],
       device='cuda:2'), out_proj_covar=tensor([0.0003, 0.0004, 0.0004, 0.0004, 0.0004, 0.0003, 0.0004, 0.0004],
       device='cuda:2')
2022-12-23 11:14:45,648 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([2.2127, 1.4502, 2.6997, 4.5762, 3.1328, 2.7594, 0.7847, 3.2984],
       device='cuda:2'), covar=tensor([0.1688, 0.1771, 0.1390, 0.0418, 0.0975, 0.1222, 0.2418, 0.0765],
       device='cuda:2'), in_proj_covar=tensor([0.0101, 0.0115, 0.0132, 0.0142, 0.0104, 0.0137, 0.0127, 0.0110],
       device='cuda:2'), out_proj_covar=tensor([0.0003, 0.0003, 0.0004, 0.0004, 0.0003, 0.0004, 0.0003, 0.0003],
       device='cuda:2')
2022-12-23 11:14:48,516 INFO [zipformer.py:660] (2/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=58681.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 11:14:59,654 WARNING [train.py:1060] (2/4) Exclude cut with ID 3488-85273-0038-41224-0 from training. Duration: 20.34
2022-12-23 11:15:14,333 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.8435, 0.7388, 1.5597, 1.5023, 1.7913, 1.8170, 1.5208, 1.4902],
       device='cuda:2'), covar=tensor([0.1954, 0.3020, 0.2505, 0.2425, 0.1980, 0.1010, 0.2685, 0.1326],
       device='cuda:2'), in_proj_covar=tensor([0.0262, 0.0294, 0.0275, 0.0310, 0.0298, 0.0245, 0.0330, 0.0233],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-23 11:15:17,178 INFO [train.py:894] (2/4) Epoch 17, batch 2600, loss[loss=0.184, simple_loss=0.2734, pruned_loss=0.0473, over 18476.00 frames. ], tot_loss[loss=0.1937, simple_loss=0.2731, pruned_loss=0.05714, over 3714034.28 frames. ], batch size: 54, lr: 6.73e-03, grad_scale: 8.0
2022-12-23 11:16:07,722 WARNING [train.py:1060] (2/4) Exclude cut with ID 7357-94126-0026-15857-0_sp0.9 from training. Duration: 25.061125
2022-12-23 11:16:12,108 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 2.664e+02 4.354e+02 5.588e+02 7.123e+02 1.427e+03, threshold=1.118e+03, percent-clipped=9.0
2022-12-23 11:16:12,522 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.4993, 1.4205, 1.4223, 1.8629, 1.7413, 3.3641, 1.3764, 1.5383],
       device='cuda:2'), covar=tensor([0.0885, 0.1758, 0.1107, 0.0912, 0.1372, 0.0253, 0.1462, 0.1517],
       device='cuda:2'), in_proj_covar=tensor([0.0072, 0.0082, 0.0072, 0.0073, 0.0090, 0.0074, 0.0084, 0.0077],
       device='cuda:2'), out_proj_covar=tensor([0.0003, 0.0004, 0.0004, 0.0004, 0.0004, 0.0003, 0.0004, 0.0004],
       device='cuda:2')
2022-12-23 11:16:18,682 WARNING [train.py:1060] (2/4) Exclude cut with ID 3033-130750-0096-55598-0 from training. Duration: 0.83
2022-12-23 11:16:23,793 INFO [zipformer.py:660] (2/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=58742.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 11:16:34,624 INFO [train.py:894] (2/4) Epoch 17, batch 2650, loss[loss=0.1844, simple_loss=0.2597, pruned_loss=0.05454, over 18706.00 frames. ], tot_loss[loss=0.1942, simple_loss=0.2737, pruned_loss=0.0573, over 3714614.81 frames. ], batch size: 50, lr: 6.73e-03, grad_scale: 8.0
2022-12-23 11:16:39,928 INFO [scaling.py:679] (2/4) Whitening: num_groups=8, num_channels=96, metric=1.40 vs. limit=2.0
2022-12-23 11:16:45,034 WARNING [train.py:1060] (2/4) Exclude cut with ID 3488-85273-0017-41203-0 from training. Duration: 24.73
2022-12-23 11:16:57,908 WARNING [train.py:1060] (2/4) Exclude cut with ID 7255-291500-0009-134308-0 from training. Duration: 23.965
2022-12-23 11:17:06,715 WARNING [train.py:1060] (2/4) Exclude cut with ID 6978-92210-0030-146996-0_sp0.9 from training. Duration: 22.088875
2022-12-23 11:17:13,269 INFO [zipformer.py:660] (2/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=58775.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 11:17:22,687 WARNING [train.py:1060] (2/4) Exclude cut with ID 7255-291500-0006-134305-0_sp0.9 from training. Duration: 23.6
2022-12-23 11:17:47,864 INFO [zipformer.py:660] (2/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=58797.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 11:17:50,517 INFO [train.py:894] (2/4) Epoch 17, batch 2700, loss[loss=0.1944, simple_loss=0.2741, pruned_loss=0.05735, over 18695.00 frames. ], tot_loss[loss=0.1939, simple_loss=0.2735, pruned_loss=0.05711, over 3714910.52 frames. ], batch size: 50, lr: 6.73e-03, grad_scale: 8.0
2022-12-23 11:17:52,637 INFO [zipformer.py:660] (2/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=58800.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 11:18:45,635 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 2.763e+02 4.526e+02 5.821e+02 7.032e+02 2.176e+03, threshold=1.164e+03, percent-clipped=2.0
2022-12-23 11:18:48,103 INFO [zipformer.py:660] (2/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=58836.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 11:19:01,378 INFO [zipformer.py:660] (2/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=58845.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 11:19:05,886 INFO [zipformer.py:660] (2/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=58848.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 11:19:07,368 INFO [train.py:894] (2/4) Epoch 17, batch 2750, loss[loss=0.2033, simple_loss=0.2852, pruned_loss=0.06075, over 18671.00 frames. ], tot_loss[loss=0.1938, simple_loss=0.2731, pruned_loss=0.05728, over 3715425.32 frames. ], batch size: 99, lr: 6.73e-03, grad_scale: 8.0
2022-12-23 11:19:08,948 WARNING [train.py:1060] (2/4) Exclude cut with ID 8291-282929-0024-13011-0 from training. Duration: 23.795
2022-12-23 11:19:24,368 WARNING [train.py:1060] (2/4) Exclude cut with ID 8631-249866-0030-130156-0_sp1.1 from training. Duration: 21.5409375
2022-12-23 11:19:27,242 WARNING [train.py:1060] (2/4) Exclude cut with ID 6978-92210-0019-146985-0_sp0.9 from training. Duration: 24.97775
2022-12-23 11:19:38,130 WARNING [train.py:1060] (2/4) Exclude cut with ID 1085-156170-0017-128270-0_sp0.9 from training. Duration: 23.3444375
2022-12-23 11:20:06,116 WARNING [train.py:1060] (2/4) Exclude cut with ID 6010-56788-0055-90261-0_sp0.9 from training. Duration: 23.2
2022-12-23 11:20:13,647 WARNING [train.py:1060] (2/4) Exclude cut with ID 5653-46179-0060-117930-0_sp0.9 from training. Duration: 21.17225
2022-12-23 11:20:22,314 INFO [train.py:894] (2/4) Epoch 17, batch 2800, loss[loss=0.2097, simple_loss=0.2913, pruned_loss=0.06407, over 18564.00 frames. ], tot_loss[loss=0.1927, simple_loss=0.272, pruned_loss=0.05664, over 3714903.70 frames. ], batch size: 78, lr: 6.72e-03, grad_scale: 8.0
2022-12-23 11:20:30,132 WARNING [train.py:1060] (2/4) Exclude cut with ID 7255-291500-0008-134307-0_sp0.9 from training. Duration: 24.6555625
2022-12-23 11:20:33,429 INFO [zipformer.py:660] (2/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=58906.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 11:20:36,171 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([2.2934, 2.7655, 2.7567, 1.5846, 3.0553, 3.2841, 2.0255, 3.1923],
       device='cuda:2'), covar=tensor([0.1077, 0.1500, 0.1356, 0.2044, 0.0681, 0.0971, 0.2093, 0.0569],
       device='cuda:2'), in_proj_covar=tensor([0.0197, 0.0210, 0.0204, 0.0195, 0.0174, 0.0215, 0.0214, 0.0189],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-23 11:21:17,307 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 2.392e+02 4.668e+02 5.683e+02 7.379e+02 2.321e+03, threshold=1.137e+03, percent-clipped=4.0
2022-12-23 11:21:17,843 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.5680, 1.3289, 1.1706, 1.3063, 1.6771, 1.6379, 1.7258, 1.1321],
       device='cuda:2'), covar=tensor([0.0274, 0.0296, 0.0511, 0.0234, 0.0207, 0.0371, 0.0224, 0.0345],
       device='cuda:2'), in_proj_covar=tensor([0.0091, 0.0123, 0.0149, 0.0123, 0.0115, 0.0116, 0.0095, 0.0125],
       device='cuda:2'), out_proj_covar=tensor([7.3326e-05, 9.8928e-05, 1.2497e-04, 9.9373e-05, 9.3833e-05, 9.0711e-05,
        7.5388e-05, 9.9248e-05], device='cuda:2')
2022-12-23 11:21:26,413 WARNING [train.py:1060] (2/4) Exclude cut with ID 3488-65654-0031-41259-0 from training. Duration: 20.44
2022-12-23 11:21:38,058 INFO [train.py:894] (2/4) Epoch 17, batch 2850, loss[loss=0.2021, simple_loss=0.2797, pruned_loss=0.06228, over 18599.00 frames. ], tot_loss[loss=0.1932, simple_loss=0.2728, pruned_loss=0.05682, over 3715826.30 frames. ], batch size: 57, lr: 6.72e-03, grad_scale: 8.0
2022-12-23 11:21:39,642 WARNING [train.py:1060] (2/4) Exclude cut with ID 6951-79737-0018-132285-0_sp0.9 from training. Duration: 23.45
2022-12-23 11:22:07,223 INFO [zipformer.py:660] (2/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=58967.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 11:22:11,623 WARNING [train.py:1060] (2/4) Exclude cut with ID 6945-60535-0076-12784-0_sp0.9 from training. Duration: 20.52225
2022-12-23 11:22:20,675 WARNING [train.py:1060] (2/4) Exclude cut with ID 7255-291500-0008-134307-0 from training. Duration: 22.19
2022-12-23 11:22:21,014 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.5551, 1.4289, 1.4612, 1.8009, 1.5740, 3.4473, 1.3875, 1.5496],
       device='cuda:2'), covar=tensor([0.0800, 0.1717, 0.1027, 0.0879, 0.1429, 0.0222, 0.1339, 0.1427],
       device='cuda:2'), in_proj_covar=tensor([0.0071, 0.0081, 0.0071, 0.0073, 0.0089, 0.0074, 0.0083, 0.0076],
       device='cuda:2'), out_proj_covar=tensor([0.0003, 0.0004, 0.0004, 0.0003, 0.0004, 0.0003, 0.0004, 0.0004],
       device='cuda:2')
2022-12-23 11:22:31,500 WARNING [train.py:1060] (2/4) Exclude cut with ID 7357-94126-0014-15845-0_sp1.1 from training. Duration: 25.3818125
2022-12-23 11:22:36,676 INFO [scaling.py:679] (2/4) Whitening: num_groups=8, num_channels=96, metric=1.38 vs. limit=2.0
2022-12-23 11:22:37,425 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([0.2776, 1.6701, 1.9869, 1.0595, 1.1362, 2.0732, 1.7915, 1.6034],
       device='cuda:2'), covar=tensor([0.0796, 0.0311, 0.0285, 0.0353, 0.0364, 0.0424, 0.0235, 0.0733],
       device='cuda:2'), in_proj_covar=tensor([0.0150, 0.0167, 0.0123, 0.0137, 0.0147, 0.0140, 0.0158, 0.0169],
       device='cuda:2'), out_proj_covar=tensor([1.1593e-04, 1.3018e-04, 9.4013e-05, 1.0332e-04, 1.1242e-04, 1.0940e-04,
        1.2386e-04, 1.3062e-04], device='cuda:2')
2022-12-23 11:22:40,857 INFO [scaling.py:679] (2/4) Whitening: num_groups=8, num_channels=96, metric=1.18 vs. limit=2.0
2022-12-23 11:22:47,244 WARNING [train.py:1060] (2/4) Exclude cut with ID 6951-79737-0043-132310-0_sp0.9 from training. Duration: 28.0944375
2022-12-23 11:22:52,916 WARNING [train.py:1060] (2/4) Exclude cut with ID 2195-150901-0045-59933-0_sp0.9 from training. Duration: 22.9444375
2022-12-23 11:22:54,347 INFO [train.py:894] (2/4) Epoch 17, batch 2900, loss[loss=0.2062, simple_loss=0.2792, pruned_loss=0.06657, over 18569.00 frames. ], tot_loss[loss=0.1928, simple_loss=0.2724, pruned_loss=0.05655, over 3714472.46 frames. ], batch size: 49, lr: 6.72e-03, grad_scale: 8.0
2022-12-23 11:23:02,086 WARNING [train.py:1060] (2/4) Exclude cut with ID 8291-282929-0024-13011-0_sp1.1 from training. Duration: 21.6318125
2022-12-23 11:23:20,763 WARNING [train.py:1060] (2/4) Exclude cut with ID 8631-249866-0030-130156-0 from training. Duration: 23.695
2022-12-23 11:23:24,517 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([2.0486, 1.2981, 1.6827, 1.7565, 2.0859, 1.9891, 1.8460, 1.5326],
       device='cuda:2'), covar=tensor([0.2038, 0.3027, 0.2454, 0.2418, 0.1797, 0.0903, 0.2715, 0.1256],
       device='cuda:2'), in_proj_covar=tensor([0.0264, 0.0296, 0.0276, 0.0311, 0.0299, 0.0248, 0.0333, 0.0235],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-23 11:23:48,051 WARNING [train.py:1060] (2/4) Exclude cut with ID 7699-105389-0094-26379-0 from training. Duration: 23.955
2022-12-23 11:23:49,461 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 2.824e+02 4.704e+02 5.257e+02 6.358e+02 1.532e+03, threshold=1.051e+03, percent-clipped=1.0
2022-12-23 11:23:52,507 INFO [zipformer.py:660] (2/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=59037.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 11:24:09,841 INFO [train.py:894] (2/4) Epoch 17, batch 2950, loss[loss=0.2085, simple_loss=0.2937, pruned_loss=0.06166, over 18716.00 frames. ], tot_loss[loss=0.1933, simple_loss=0.2731, pruned_loss=0.05676, over 3714285.39 frames. ], batch size: 54, lr: 6.71e-03, grad_scale: 8.0
2022-12-23 11:24:18,091 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([2.2739, 2.0499, 1.5176, 0.7070, 1.5334, 1.9836, 1.6015, 1.9806],
       device='cuda:2'), covar=tensor([0.0589, 0.0433, 0.0945, 0.1330, 0.0939, 0.1270, 0.1432, 0.0545],
       device='cuda:2'), in_proj_covar=tensor([0.0170, 0.0181, 0.0203, 0.0189, 0.0208, 0.0198, 0.0212, 0.0199],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-23 11:24:20,918 WARNING [train.py:1060] (2/4) Exclude cut with ID 8291-282929-0024-13011-0_sp0.9 from training. Duration: 26.438875
2022-12-23 11:25:06,033 WARNING [train.py:1060] (2/4) Exclude cut with ID 7699-105389-0021-26306-0_sp0.9 from training. Duration: 21.2444375
2022-12-23 11:25:06,061 WARNING [train.py:1060] (2/4) Exclude cut with ID 7357-94126-0014-15845-0_sp0.9 from training. Duration: 31.02225
2022-12-23 11:25:14,841 WARNING [train.py:1060] (2/4) Exclude cut with ID 432-122774-0017-62487-0 from training. Duration: 22.395
2022-12-23 11:25:27,144 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.8606, 1.2732, 0.7199, 1.3419, 2.1035, 1.2287, 1.5648, 1.7326],
       device='cuda:2'), covar=tensor([0.1571, 0.2095, 0.2284, 0.1575, 0.1745, 0.1741, 0.1397, 0.1626],
       device='cuda:2'), in_proj_covar=tensor([0.0093, 0.0097, 0.0117, 0.0095, 0.0116, 0.0092, 0.0098, 0.0094],
       device='cuda:2'), out_proj_covar=tensor([0.0003, 0.0003, 0.0004, 0.0003, 0.0004, 0.0003, 0.0003, 0.0003],
       device='cuda:2')
2022-12-23 11:25:28,221 INFO [train.py:894] (2/4) Epoch 17, batch 3000, loss[loss=0.2132, simple_loss=0.2861, pruned_loss=0.07021, over 18591.00 frames. ], tot_loss[loss=0.1923, simple_loss=0.2722, pruned_loss=0.0562, over 3715320.18 frames. ], batch size: 69, lr: 6.71e-03, grad_scale: 8.0
2022-12-23 11:25:28,222 INFO [train.py:919] (2/4) Computing validation loss
2022-12-23 11:25:39,070 INFO [train.py:928] (2/4) Epoch 17, validation: loss=0.1653, simple_loss=0.2636, pruned_loss=0.03352, over 944034.00 frames. 
2022-12-23 11:25:39,072 INFO [train.py:929] (2/4) Maximum memory allocated so far is 24680MB
2022-12-23 11:25:40,595 WARNING [train.py:1060] (2/4) Exclude cut with ID 7357-94126-0045-15876-0 from training. Duration: 21.075
2022-12-23 11:25:47,734 WARNING [train.py:1060] (2/4) Exclude cut with ID 6482-98857-0025-147532-0_sp0.9 from training. Duration: 20.0055625
2022-12-23 11:25:47,746 WARNING [train.py:1060] (2/4) Exclude cut with ID 6951-79737-0037-132304-0_sp0.9 from training. Duration: 22.05
2022-12-23 11:25:47,753 WARNING [train.py:1060] (2/4) Exclude cut with ID 7255-291500-0003-134302-0 from training. Duration: 26.8349375
2022-12-23 11:25:50,620 WARNING [train.py:1060] (2/4) Exclude cut with ID 7357-94126-0009-15840-0_sp1.1 from training. Duration: 22.1090625
2022-12-23 11:25:58,343 WARNING [train.py:1060] (2/4) Exclude cut with ID 7699-105389-0094-26379-0_sp0.9 from training. Duration: 26.6166875
2022-12-23 11:26:14,767 WARNING [train.py:1060] (2/4) Exclude cut with ID 2046-178027-0000-53705-0_sp0.9 from training. Duration: 20.3055625
2022-12-23 11:26:16,403 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.9089, 1.8846, 2.1431, 1.2823, 2.1144, 2.3266, 1.5327, 2.4994],
       device='cuda:2'), covar=tensor([0.1007, 0.1654, 0.1136, 0.1699, 0.0723, 0.1030, 0.2202, 0.0487],
       device='cuda:2'), in_proj_covar=tensor([0.0194, 0.0207, 0.0202, 0.0193, 0.0173, 0.0214, 0.0210, 0.0186],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-23 11:26:28,215 INFO [zipformer.py:660] (2/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=59131.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 11:26:34,378 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 3.258e+02 4.780e+02 5.724e+02 6.904e+02 1.208e+03, threshold=1.145e+03, percent-clipped=1.0
2022-12-23 11:26:40,089 WARNING [train.py:1060] (2/4) Exclude cut with ID 7205-50138-0008-5373-0_sp0.9 from training. Duration: 20.7
2022-12-23 11:26:56,016 INFO [train.py:894] (2/4) Epoch 17, batch 3050, loss[loss=0.1842, simple_loss=0.2633, pruned_loss=0.05259, over 18686.00 frames. ], tot_loss[loss=0.1932, simple_loss=0.2726, pruned_loss=0.05694, over 3715004.52 frames. ], batch size: 46, lr: 6.71e-03, grad_scale: 8.0
2022-12-23 11:27:15,732 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([2.5289, 1.8609, 0.8700, 1.6820, 2.3858, 1.7811, 2.1912, 2.1407],
       device='cuda:2'), covar=tensor([0.1449, 0.1932, 0.2338, 0.1534, 0.1835, 0.1611, 0.1369, 0.1645],
       device='cuda:2'), in_proj_covar=tensor([0.0093, 0.0097, 0.0117, 0.0095, 0.0116, 0.0092, 0.0098, 0.0093],
       device='cuda:2'), out_proj_covar=tensor([0.0003, 0.0003, 0.0004, 0.0003, 0.0004, 0.0003, 0.0003, 0.0003],
       device='cuda:2')
2022-12-23 11:27:18,528 INFO [zipformer.py:660] (2/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=59164.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 11:27:20,882 INFO [scaling.py:679] (2/4) Whitening: num_groups=1, num_channels=384, metric=4.55 vs. limit=5.0
2022-12-23 11:27:22,952 WARNING [train.py:1060] (2/4) Exclude cut with ID 6978-92210-0019-146985-0 from training. Duration: 22.48
2022-12-23 11:27:30,985 INFO [scaling.py:679] (2/4) Whitening: num_groups=8, num_channels=192, metric=1.68 vs. limit=2.0
2022-12-23 11:27:38,515 WARNING [train.py:1060] (2/4) Exclude cut with ID 7255-291500-0003-134302-0_sp0.9 from training. Duration: 29.816625
2022-12-23 11:27:58,605 WARNING [train.py:1060] (2/4) Exclude cut with ID 7255-291500-0005-134304-0_sp1.1 from training. Duration: 22.7590625
2022-12-23 11:28:03,406 WARNING [train.py:1060] (2/4) Exclude cut with ID 7357-94126-0026-15857-0 from training. Duration: 22.555
2022-12-23 11:28:12,954 INFO [train.py:894] (2/4) Epoch 17, batch 3100, loss[loss=0.1705, simple_loss=0.249, pruned_loss=0.04599, over 18538.00 frames. ], tot_loss[loss=0.194, simple_loss=0.2734, pruned_loss=0.0573, over 3714559.46 frames. ], batch size: 47, lr: 6.71e-03, grad_scale: 8.0
2022-12-23 11:28:26,522 WARNING [train.py:1060] (2/4) Exclude cut with ID 1250-135782-0005-25975-0_sp0.9 from training. Duration: 21.688875
2022-12-23 11:28:35,713 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([2.3173, 2.5652, 2.8694, 1.6297, 2.7114, 2.9457, 1.9781, 3.1386],
       device='cuda:2'), covar=tensor([0.1165, 0.1620, 0.1279, 0.1952, 0.0771, 0.1212, 0.2191, 0.0503],
       device='cuda:2'), in_proj_covar=tensor([0.0197, 0.0210, 0.0205, 0.0195, 0.0175, 0.0216, 0.0213, 0.0188],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-23 11:28:52,846 INFO [zipformer.py:660] (2/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=59225.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 11:28:57,784 WARNING [train.py:1060] (2/4) Exclude cut with ID 3488-85273-0038-41224-0_sp0.9 from training. Duration: 22.6
2022-12-23 11:29:08,315 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 3.058e+02 4.490e+02 5.519e+02 6.910e+02 1.551e+03, threshold=1.104e+03, percent-clipped=2.0
2022-12-23 11:29:27,477 INFO [scaling.py:679] (2/4) Whitening: num_groups=8, num_channels=96, metric=1.57 vs. limit=2.0
2022-12-23 11:29:30,670 INFO [train.py:894] (2/4) Epoch 17, batch 3150, loss[loss=0.1911, simple_loss=0.2777, pruned_loss=0.05223, over 18671.00 frames. ], tot_loss[loss=0.1935, simple_loss=0.2725, pruned_loss=0.05723, over 3714065.45 frames. ], batch size: 60, lr: 6.70e-03, grad_scale: 16.0
2022-12-23 11:29:35,427 WARNING [train.py:1060] (2/4) Exclude cut with ID 7357-94126-0009-15840-0 from training. Duration: 24.32
2022-12-23 11:29:50,230 INFO [zipformer.py:660] (2/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=59262.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 11:29:56,332 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([2.1264, 1.2241, 2.4880, 4.3862, 3.2186, 2.8002, 0.4934, 3.0069],
       device='cuda:2'), covar=tensor([0.1883, 0.1985, 0.1552, 0.0500, 0.0990, 0.1254, 0.2761, 0.0969],
       device='cuda:2'), in_proj_covar=tensor([0.0101, 0.0115, 0.0131, 0.0143, 0.0105, 0.0137, 0.0127, 0.0110],
       device='cuda:2'), out_proj_covar=tensor([0.0003, 0.0003, 0.0004, 0.0004, 0.0003, 0.0004, 0.0003, 0.0003],
       device='cuda:2')
2022-12-23 11:30:35,056 WARNING [train.py:1060] (2/4) Exclude cut with ID 8291-276745-0093-13116-0_sp0.9 from training. Duration: 21.061125
2022-12-23 11:30:47,415 INFO [train.py:894] (2/4) Epoch 17, batch 3200, loss[loss=0.1889, simple_loss=0.2599, pruned_loss=0.05891, over 18424.00 frames. ], tot_loss[loss=0.1923, simple_loss=0.2716, pruned_loss=0.05648, over 3713961.28 frames. ], batch size: 48, lr: 6.70e-03, grad_scale: 16.0
2022-12-23 11:30:47,471 WARNING [train.py:1060] (2/4) Exclude cut with ID 7357-94126-0024-15855-0_sp0.9 from training. Duration: 20.32225
2022-12-23 11:30:59,485 WARNING [train.py:1060] (2/4) Exclude cut with ID 3033-130750-0096-55598-0_sp1.1 from training. Duration: 0.7545625
2022-12-23 11:31:16,158 WARNING [train.py:1060] (2/4) Exclude cut with ID 4295-39940-0007-92567-0_sp0.9 from training. Duration: 23.9333125
2022-12-23 11:31:42,114 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 2.749e+02 4.792e+02 5.650e+02 6.896e+02 1.712e+03, threshold=1.130e+03, percent-clipped=1.0
2022-12-23 11:31:45,487 INFO [zipformer.py:660] (2/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=59337.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 11:31:49,527 WARNING [train.py:1060] (2/4) Exclude cut with ID 7255-291500-0008-134307-0_sp1.1 from training. Duration: 20.17275
2022-12-23 11:31:53,676 WARNING [train.py:1060] (2/4) Exclude cut with ID 6978-92210-0019-146985-0_sp1.1 from training. Duration: 20.436375
2022-12-23 11:32:02,415 INFO [train.py:894] (2/4) Epoch 17, batch 3250, loss[loss=0.2062, simple_loss=0.2871, pruned_loss=0.06265, over 18723.00 frames. ], tot_loss[loss=0.1928, simple_loss=0.2721, pruned_loss=0.05675, over 3713459.71 frames. ], batch size: 60, lr: 6.70e-03, grad_scale: 16.0
2022-12-23 11:32:58,985 INFO [zipformer.py:660] (2/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=59385.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 11:33:02,146 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.7968, 2.3402, 1.5811, 2.6610, 2.9535, 1.6579, 1.9908, 1.4601],
       device='cuda:2'), covar=tensor([0.1942, 0.1645, 0.1590, 0.0926, 0.1443, 0.1139, 0.1877, 0.1554],
       device='cuda:2'), in_proj_covar=tensor([0.0244, 0.0222, 0.0209, 0.0196, 0.0258, 0.0193, 0.0220, 0.0196],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-23 11:33:17,664 WARNING [train.py:1060] (2/4) Exclude cut with ID 4234-40345-0022-142709-0_sp0.9 from training. Duration: 23.1055625
2022-12-23 11:33:19,052 INFO [train.py:894] (2/4) Epoch 17, batch 3300, loss[loss=0.1543, simple_loss=0.2276, pruned_loss=0.04052, over 18372.00 frames. ], tot_loss[loss=0.1922, simple_loss=0.2718, pruned_loss=0.05635, over 3713850.10 frames. ], batch size: 42, lr: 6.69e-03, grad_scale: 16.0
2022-12-23 11:33:20,692 WARNING [train.py:1060] (2/4) Exclude cut with ID 8291-282929-0007-12994-0_sp1.1 from training. Duration: 23.5
2022-12-23 11:33:31,074 WARNING [train.py:1060] (2/4) Exclude cut with ID 7255-291500-0009-134308-0_sp0.9 from training. Duration: 26.62775
2022-12-23 11:33:43,489 WARNING [train.py:1060] (2/4) Exclude cut with ID 6951-79737-0018-132285-0 from training. Duration: 21.105
2022-12-23 11:33:48,310 WARNING [train.py:1060] (2/4) Exclude cut with ID 4511-76322-0006-80011-0_sp0.9 from training. Duration: 24.411125
2022-12-23 11:34:09,105 INFO [zipformer.py:660] (2/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=59431.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 11:34:10,791 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([3.0477, 2.4658, 1.7660, 2.8893, 2.1832, 2.2639, 2.4211, 3.2111],
       device='cuda:2'), covar=tensor([0.1938, 0.3224, 0.1954, 0.2944, 0.3730, 0.1066, 0.2990, 0.0801],
       device='cuda:2'), in_proj_covar=tensor([0.0290, 0.0284, 0.0239, 0.0347, 0.0263, 0.0223, 0.0280, 0.0208],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0003, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-23 11:34:14,482 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 2.836e+02 4.538e+02 5.968e+02 7.307e+02 1.624e+03, threshold=1.194e+03, percent-clipped=6.0
2022-12-23 11:34:17,503 WARNING [train.py:1060] (2/4) Exclude cut with ID 6758-72288-0033-108368-0_sp1.1 from training. Duration: 21.263625
2022-12-23 11:34:35,003 INFO [train.py:894] (2/4) Epoch 17, batch 3350, loss[loss=0.1647, simple_loss=0.2406, pruned_loss=0.04441, over 18423.00 frames. ], tot_loss[loss=0.1916, simple_loss=0.2712, pruned_loss=0.05598, over 3714081.38 frames. ], batch size: 42, lr: 6.69e-03, grad_scale: 16.0
2022-12-23 11:34:41,484 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.8952, 1.8334, 1.6229, 1.6470, 2.0156, 2.1379, 2.1517, 1.4962],
       device='cuda:2'), covar=tensor([0.0275, 0.0232, 0.0417, 0.0211, 0.0183, 0.0315, 0.0220, 0.0294],
       device='cuda:2'), in_proj_covar=tensor([0.0091, 0.0123, 0.0149, 0.0123, 0.0115, 0.0116, 0.0095, 0.0124],
       device='cuda:2'), out_proj_covar=tensor([7.3262e-05, 9.8850e-05, 1.2404e-04, 9.9170e-05, 9.3570e-05, 8.9902e-05,
        7.5192e-05, 9.8816e-05], device='cuda:2')
2022-12-23 11:34:51,693 WARNING [train.py:1060] (2/4) Exclude cut with ID 4234-40345-0022-142709-0 from training. Duration: 20.795
2022-12-23 11:34:56,459 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.5585, 1.8098, 1.4435, 2.2358, 2.4796, 1.6319, 1.4356, 1.3481],
       device='cuda:2'), covar=tensor([0.1967, 0.1863, 0.1677, 0.0988, 0.1156, 0.1151, 0.2202, 0.1551],
       device='cuda:2'), in_proj_covar=tensor([0.0245, 0.0223, 0.0211, 0.0196, 0.0259, 0.0194, 0.0222, 0.0197],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-23 11:35:01,977 WARNING [train.py:1060] (2/4) Exclude cut with ID 7357-94126-0021-15852-0 from training. Duration: 24.76
2022-12-23 11:35:01,991 WARNING [train.py:1060] (2/4) Exclude cut with ID 3867-173237-0077-144769-0_sp0.9 from training. Duration: 22.25
2022-12-23 11:35:09,126 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([2.6290, 3.2964, 3.2597, 1.4008, 3.4454, 2.5596, 0.8108, 2.3437],
       device='cuda:2'), covar=tensor([0.1905, 0.1491, 0.1615, 0.3765, 0.1050, 0.1036, 0.4828, 0.1614],
       device='cuda:2'), in_proj_covar=tensor([0.0144, 0.0139, 0.0159, 0.0124, 0.0140, 0.0111, 0.0145, 0.0113],
       device='cuda:2'), out_proj_covar=tensor([0.0003, 0.0002, 0.0003, 0.0002, 0.0002, 0.0002, 0.0003, 0.0002],
       device='cuda:2')
2022-12-23 11:35:12,549 INFO [scaling.py:679] (2/4) Whitening: num_groups=8, num_channels=192, metric=1.64 vs. limit=2.0
2022-12-23 11:35:19,033 INFO [scaling.py:679] (2/4) Whitening: num_groups=8, num_channels=192, metric=1.86 vs. limit=2.0
2022-12-23 11:35:21,129 INFO [zipformer.py:660] (2/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=59479.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 11:35:28,612 WARNING [train.py:1060] (2/4) Exclude cut with ID 7357-94126-0026-15857-0_sp1.1 from training. Duration: 20.5045625
2022-12-23 11:35:50,236 INFO [train.py:894] (2/4) Epoch 17, batch 3400, loss[loss=0.2504, simple_loss=0.3148, pruned_loss=0.09302, over 18569.00 frames. ], tot_loss[loss=0.1916, simple_loss=0.2712, pruned_loss=0.05596, over 3713793.10 frames. ], batch size: 170, lr: 6.69e-03, grad_scale: 16.0
2022-12-23 11:35:54,976 INFO [zipformer.py:660] (2/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=59502.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 11:35:58,029 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.6822, 1.6192, 1.2230, 1.6531, 1.7189, 1.5005, 2.2190, 1.7916],
       device='cuda:2'), covar=tensor([0.1056, 0.1583, 0.2960, 0.1678, 0.2032, 0.1147, 0.1060, 0.1360],
       device='cuda:2'), in_proj_covar=tensor([0.0177, 0.0204, 0.0247, 0.0288, 0.0236, 0.0189, 0.0212, 0.0201],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-23 11:36:21,717 INFO [zipformer.py:660] (2/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=59520.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 11:36:43,135 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 3.219e+02 4.503e+02 5.475e+02 7.433e+02 1.825e+03, threshold=1.095e+03, percent-clipped=2.0
2022-12-23 11:37:02,779 INFO [train.py:894] (2/4) Epoch 17, batch 3450, loss[loss=0.2093, simple_loss=0.2928, pruned_loss=0.0629, over 18689.00 frames. ], tot_loss[loss=0.1916, simple_loss=0.2711, pruned_loss=0.05603, over 3714157.31 frames. ], batch size: 79, lr: 6.69e-03, grad_scale: 16.0
2022-12-23 11:37:23,179 INFO [zipformer.py:660] (2/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=59562.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 11:37:24,651 INFO [zipformer.py:660] (2/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=59563.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 11:38:17,578 INFO [train.py:894] (2/4) Epoch 17, batch 3500, loss[loss=0.2177, simple_loss=0.296, pruned_loss=0.06974, over 18656.00 frames. ], tot_loss[loss=0.1913, simple_loss=0.2709, pruned_loss=0.05583, over 3715654.01 frames. ], batch size: 176, lr: 6.68e-03, grad_scale: 16.0
2022-12-23 11:38:38,351 WARNING [train.py:1060] (2/4) Exclude cut with ID 7859-102521-0017-7548-0_sp1.1 from training. Duration: 22.2954375
2022-12-23 11:38:46,854 INFO [train.py:894] (2/4) Epoch 18, batch 0, loss[loss=0.1931, simple_loss=0.2694, pruned_loss=0.05836, over 18607.00 frames. ], tot_loss[loss=0.1931, simple_loss=0.2694, pruned_loss=0.05836, over 18607.00 frames. ], batch size: 45, lr: 6.49e-03, grad_scale: 16.0
2022-12-23 11:38:46,854 INFO [train.py:919] (2/4) Computing validation loss
2022-12-23 11:38:50,103 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([2.1215, 2.1194, 2.5841, 1.6225, 2.3670, 2.4708, 1.8486, 2.5540],
       device='cuda:2'), covar=tensor([0.1165, 0.1630, 0.1538, 0.2122, 0.0631, 0.1144, 0.2137, 0.0570],
       device='cuda:2'), in_proj_covar=tensor([0.0197, 0.0210, 0.0205, 0.0195, 0.0174, 0.0216, 0.0213, 0.0191],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-23 11:38:57,548 INFO [train.py:928] (2/4) Epoch 18, validation: loss=0.1653, simple_loss=0.2643, pruned_loss=0.03315, over 944034.00 frames. 
2022-12-23 11:38:57,549 INFO [train.py:929] (2/4) Maximum memory allocated so far is 24680MB
2022-12-23 11:39:04,896 INFO [zipformer.py:660] (2/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=59610.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 11:39:43,084 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 2.453e+02 4.121e+02 4.989e+02 6.015e+02 1.398e+03, threshold=9.978e+02, percent-clipped=2.0
2022-12-23 11:39:48,891 WARNING [train.py:1060] (2/4) Exclude cut with ID 298-126791-0067-24026-0_sp0.9 from training. Duration: 21.438875
2022-12-23 11:39:53,349 WARNING [train.py:1060] (2/4) Exclude cut with ID 5652-39938-0025-23684-0_sp0.9 from training. Duration: 22.2055625
2022-12-23 11:39:54,427 INFO [scaling.py:679] (2/4) Whitening: num_groups=1, num_channels=384, metric=4.80 vs. limit=5.0
2022-12-23 11:40:12,950 INFO [train.py:894] (2/4) Epoch 18, batch 50, loss[loss=0.1732, simple_loss=0.2643, pruned_loss=0.04104, over 18724.00 frames. ], tot_loss[loss=0.1832, simple_loss=0.2704, pruned_loss=0.04796, over 838083.13 frames. ], batch size: 54, lr: 6.49e-03, grad_scale: 16.0
2022-12-23 11:40:57,197 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([4.1802, 3.6474, 3.6094, 4.1212, 3.7792, 3.6071, 4.3516, 1.2635],
       device='cuda:2'), covar=tensor([0.0802, 0.0716, 0.0732, 0.0825, 0.1607, 0.1374, 0.0611, 0.5435],
       device='cuda:2'), in_proj_covar=tensor([0.0335, 0.0220, 0.0233, 0.0258, 0.0316, 0.0263, 0.0281, 0.0276],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0001, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-23 11:41:01,653 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([4.5498, 3.9403, 3.9159, 4.5158, 4.1020, 3.9889, 4.7356, 1.3557],
       device='cuda:2'), covar=tensor([0.0757, 0.0659, 0.0695, 0.0803, 0.1504, 0.1219, 0.0533, 0.5580],
       device='cuda:2'), in_proj_covar=tensor([0.0335, 0.0220, 0.0232, 0.0258, 0.0316, 0.0263, 0.0281, 0.0276],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0001, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-23 11:41:29,462 INFO [train.py:894] (2/4) Epoch 18, batch 100, loss[loss=0.1685, simple_loss=0.257, pruned_loss=0.04004, over 18686.00 frames. ], tot_loss[loss=0.1825, simple_loss=0.2695, pruned_loss=0.04773, over 1476648.38 frames. ], batch size: 48, lr: 6.49e-03, grad_scale: 16.0
2022-12-23 11:42:02,146 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.5905, 1.4961, 1.6004, 2.0824, 1.9081, 3.7421, 1.5069, 1.6033],
       device='cuda:2'), covar=tensor([0.0870, 0.1797, 0.1087, 0.0879, 0.1324, 0.0192, 0.1380, 0.1513],
       device='cuda:2'), in_proj_covar=tensor([0.0072, 0.0081, 0.0072, 0.0073, 0.0089, 0.0073, 0.0083, 0.0076],
       device='cuda:2'), out_proj_covar=tensor([0.0003, 0.0004, 0.0004, 0.0004, 0.0004, 0.0003, 0.0004, 0.0004],
       device='cuda:2')
2022-12-23 11:42:14,866 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 2.193e+02 3.350e+02 4.029e+02 4.665e+02 1.011e+03, threshold=8.057e+02, percent-clipped=1.0
2022-12-23 11:42:44,687 INFO [train.py:894] (2/4) Epoch 18, batch 150, loss[loss=0.172, simple_loss=0.255, pruned_loss=0.04456, over 18401.00 frames. ], tot_loss[loss=0.1822, simple_loss=0.2691, pruned_loss=0.04771, over 1971066.44 frames. ], batch size: 46, lr: 6.48e-03, grad_scale: 16.0
2022-12-23 11:42:46,603 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([2.4403, 1.7904, 2.1075, 2.0735, 2.2746, 2.2881, 2.0927, 1.7006],
       device='cuda:2'), covar=tensor([0.2111, 0.3380, 0.2458, 0.3019, 0.1869, 0.0949, 0.3315, 0.1287],
       device='cuda:2'), in_proj_covar=tensor([0.0263, 0.0295, 0.0275, 0.0311, 0.0299, 0.0247, 0.0331, 0.0234],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-23 11:42:55,515 WARNING [train.py:1060] (2/4) Exclude cut with ID 7859-102521-0017-7548-0 from training. Duration: 24.525
2022-12-23 11:43:05,620 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([2.2706, 1.5410, 1.9481, 1.8269, 2.1531, 2.1972, 1.9882, 1.7650],
       device='cuda:2'), covar=tensor([0.2189, 0.3169, 0.2385, 0.2827, 0.1903, 0.0905, 0.3065, 0.1225],
       device='cuda:2'), in_proj_covar=tensor([0.0263, 0.0295, 0.0274, 0.0311, 0.0299, 0.0247, 0.0331, 0.0235],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-23 11:43:30,691 WARNING [train.py:1060] (2/4) Exclude cut with ID 3699-47246-0007-3408-0_sp0.9 from training. Duration: 20.26675
2022-12-23 11:43:37,154 INFO [scaling.py:679] (2/4) Whitening: num_groups=8, num_channels=192, metric=1.93 vs. limit=2.0
2022-12-23 11:43:43,503 WARNING [train.py:1060] (2/4) Exclude cut with ID 7859-102521-0017-7548-0_sp0.9 from training. Duration: 27.25
2022-12-23 11:44:01,074 INFO [train.py:894] (2/4) Epoch 18, batch 200, loss[loss=0.1569, simple_loss=0.2465, pruned_loss=0.03365, over 18712.00 frames. ], tot_loss[loss=0.1815, simple_loss=0.2682, pruned_loss=0.04738, over 2357270.52 frames. ], batch size: 50, lr: 6.48e-03, grad_scale: 16.0
2022-12-23 11:44:07,674 INFO [scaling.py:679] (2/4) Whitening: num_groups=8, num_channels=192, metric=1.70 vs. limit=2.0
2022-12-23 11:44:25,634 INFO [zipformer.py:660] (2/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=59820.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 11:44:47,321 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 2.080e+02 3.472e+02 4.148e+02 5.017e+02 1.205e+03, threshold=8.296e+02, percent-clipped=2.0
2022-12-23 11:44:57,337 WARNING [train.py:1060] (2/4) Exclude cut with ID 6426-64292-0017-15984-0 from training. Duration: 21.68
2022-12-23 11:45:06,652 INFO [zipformer.py:660] (2/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=59848.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 11:45:07,908 WARNING [train.py:1060] (2/4) Exclude cut with ID 4278-13270-0007-59342-0 from training. Duration: 21.6300625
2022-12-23 11:45:16,503 INFO [train.py:894] (2/4) Epoch 18, batch 250, loss[loss=0.1899, simple_loss=0.2745, pruned_loss=0.05264, over 18667.00 frames. ], tot_loss[loss=0.1801, simple_loss=0.2673, pruned_loss=0.04651, over 2658199.63 frames. ], batch size: 60, lr: 6.48e-03, grad_scale: 16.0
2022-12-23 11:45:21,532 INFO [zipformer.py:660] (2/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=59858.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 11:45:32,266 WARNING [train.py:1060] (2/4) Exclude cut with ID 4278-13270-0007-59342-0_sp0.9 from training. Duration: 24.033375
2022-12-23 11:45:36,650 INFO [zipformer.py:660] (2/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=59868.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 11:46:01,640 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([4.5667, 3.9214, 3.9971, 4.5068, 4.1988, 4.1085, 4.7458, 1.5940],
       device='cuda:2'), covar=tensor([0.0715, 0.0731, 0.0622, 0.0722, 0.1325, 0.1086, 0.0500, 0.4889],
       device='cuda:2'), in_proj_covar=tensor([0.0333, 0.0217, 0.0229, 0.0254, 0.0310, 0.0258, 0.0278, 0.0271],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0001, 0.0001, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-23 11:46:07,740 INFO [zipformer.py:660] (2/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=59889.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 11:46:27,788 WARNING [train.py:1060] (2/4) Exclude cut with ID 4278-13270-0009-59344-0 from training. Duration: 22.905
2022-12-23 11:46:29,273 WARNING [train.py:1060] (2/4) Exclude cut with ID 5622-44585-0006-90525-0_sp1.1 from training. Duration: 23.4318125
2022-12-23 11:46:30,710 INFO [train.py:894] (2/4) Epoch 18, batch 300, loss[loss=0.1657, simple_loss=0.2502, pruned_loss=0.04057, over 18671.00 frames. ], tot_loss[loss=0.1796, simple_loss=0.2664, pruned_loss=0.04635, over 2892668.77 frames. ], batch size: 46, lr: 6.48e-03, grad_scale: 16.0
2022-12-23 11:46:38,740 INFO [zipformer.py:660] (2/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=59909.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 11:46:58,069 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([2.1514, 2.0655, 2.1332, 1.1521, 2.2160, 2.3651, 1.7000, 2.7004],
       device='cuda:2'), covar=tensor([0.1189, 0.1967, 0.1518, 0.2314, 0.0827, 0.1421, 0.2251, 0.0566],
       device='cuda:2'), in_proj_covar=tensor([0.0194, 0.0207, 0.0203, 0.0193, 0.0172, 0.0215, 0.0209, 0.0188],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-23 11:47:05,345 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([3.6771, 3.0300, 2.8192, 1.3030, 2.6568, 2.3361, 2.0782, 2.7039],
       device='cuda:2'), covar=tensor([0.0517, 0.0609, 0.1278, 0.1767, 0.1513, 0.1363, 0.1485, 0.0927],
       device='cuda:2'), in_proj_covar=tensor([0.0165, 0.0179, 0.0200, 0.0187, 0.0203, 0.0195, 0.0207, 0.0194],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-23 11:47:16,510 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 1.844e+02 3.499e+02 4.281e+02 5.083e+02 1.083e+03, threshold=8.562e+02, percent-clipped=4.0
2022-12-23 11:47:33,661 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([2.2152, 1.7013, 1.9522, 1.8636, 2.1788, 2.1985, 2.0401, 1.8363],
       device='cuda:2'), covar=tensor([0.1956, 0.2907, 0.2166, 0.2652, 0.1783, 0.0837, 0.3007, 0.1125],
       device='cuda:2'), in_proj_covar=tensor([0.0260, 0.0293, 0.0272, 0.0308, 0.0298, 0.0245, 0.0330, 0.0232],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-23 11:47:39,580 INFO [zipformer.py:660] (2/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=59950.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 11:47:47,293 INFO [train.py:894] (2/4) Epoch 18, batch 350, loss[loss=0.1787, simple_loss=0.2674, pruned_loss=0.04496, over 18559.00 frames. ], tot_loss[loss=0.1795, simple_loss=0.2664, pruned_loss=0.04632, over 3074425.53 frames. ], batch size: 49, lr: 6.47e-03, grad_scale: 16.0
2022-12-23 11:48:28,465 WARNING [train.py:1060] (2/4) Exclude cut with ID 4278-13270-0009-59344-0_sp1.1 from training. Duration: 20.82275
2022-12-23 11:48:29,897 WARNING [train.py:1060] (2/4) Exclude cut with ID 4278-13270-0009-59344-0_sp0.9 from training. Duration: 25.45
2022-12-23 11:48:47,994 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.5256, 1.7001, 1.4320, 1.9916, 2.4688, 1.5135, 1.5221, 1.2399],
       device='cuda:2'), covar=tensor([0.2166, 0.2164, 0.2015, 0.1270, 0.1388, 0.1482, 0.2390, 0.1881],
       device='cuda:2'), in_proj_covar=tensor([0.0244, 0.0221, 0.0211, 0.0194, 0.0257, 0.0193, 0.0222, 0.0196],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-23 11:49:06,026 INFO [train.py:894] (2/4) Epoch 18, batch 400, loss[loss=0.1742, simple_loss=0.2676, pruned_loss=0.04034, over 18578.00 frames. ], tot_loss[loss=0.1797, simple_loss=0.2668, pruned_loss=0.04634, over 3216983.67 frames. ], batch size: 51, lr: 6.47e-03, grad_scale: 16.0
2022-12-23 11:49:33,077 WARNING [train.py:1060] (2/4) Exclude cut with ID 5622-44585-0006-90525-0 from training. Duration: 25.775
2022-12-23 11:49:50,713 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 2.438e+02 3.518e+02 4.134e+02 5.271e+02 1.666e+03, threshold=8.268e+02, percent-clipped=1.0
2022-12-23 11:49:55,870 WARNING [train.py:1060] (2/4) Exclude cut with ID 2929-85685-0071-62375-0_sp0.9 from training. Duration: 22.25
2022-12-23 11:49:59,092 INFO [zipformer.py:660] (2/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=60040.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 11:50:17,492 INFO [zipformer.py:660] (2/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=60052.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 11:50:21,181 INFO [train.py:894] (2/4) Epoch 18, batch 450, loss[loss=0.2023, simple_loss=0.2855, pruned_loss=0.0596, over 18715.00 frames. ], tot_loss[loss=0.1802, simple_loss=0.267, pruned_loss=0.04666, over 3325979.35 frames. ], batch size: 50, lr: 6.47e-03, grad_scale: 16.0
2022-12-23 11:50:21,299 WARNING [train.py:1060] (2/4) Exclude cut with ID 3972-170212-0014-23379-0 from training. Duration: 26.205
2022-12-23 11:50:37,094 INFO [zipformer.py:660] (2/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=60065.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 11:50:38,084 WARNING [train.py:1060] (2/4) Exclude cut with ID 5239-32139-0047-9341-0_sp0.9 from training. Duration: 30.1555625
2022-12-23 11:50:43,035 INFO [scaling.py:679] (2/4) Whitening: num_groups=8, num_channels=96, metric=1.62 vs. limit=2.0
2022-12-23 11:50:43,780 WARNING [train.py:1060] (2/4) Exclude cut with ID 1265-135635-0050-6781-0_sp0.9 from training. Duration: 21.8333125
2022-12-23 11:50:52,491 WARNING [train.py:1060] (2/4) Exclude cut with ID 1914-133440-0024-94914-0_sp1.1 from training. Duration: 20.6545625
2022-12-23 11:50:55,655 INFO [zipformer.py:660] (2/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=60078.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 11:51:30,783 INFO [zipformer.py:660] (2/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=60101.0, num_to_drop=1, layers_to_drop={3}
2022-12-23 11:51:31,892 WARNING [train.py:1060] (2/4) Exclude cut with ID 7395-89880-0045-39920-0_sp0.9 from training. Duration: 20.52225
2022-12-23 11:51:37,530 INFO [train.py:894] (2/4) Epoch 18, batch 500, loss[loss=0.1982, simple_loss=0.287, pruned_loss=0.05475, over 18641.00 frames. ], tot_loss[loss=0.1808, simple_loss=0.2677, pruned_loss=0.04693, over 3412003.64 frames. ], batch size: 175, lr: 6.46e-03, grad_scale: 16.0
2022-12-23 11:51:49,201 INFO [zipformer.py:660] (2/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=60113.0, num_to_drop=1, layers_to_drop={3}
2022-12-23 11:51:54,231 WARNING [train.py:1060] (2/4) Exclude cut with ID 3972-170212-0014-23379-0_sp0.9 from training. Duration: 29.1166875
2022-12-23 11:52:07,641 INFO [zipformer.py:660] (2/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=60126.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 11:52:20,897 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 2.520e+02 3.872e+02 4.827e+02 5.955e+02 1.289e+03, threshold=9.654e+02, percent-clipped=8.0
2022-12-23 11:52:27,207 INFO [zipformer.py:660] (2/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=60139.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 11:52:50,638 INFO [train.py:894] (2/4) Epoch 18, batch 550, loss[loss=0.2065, simple_loss=0.298, pruned_loss=0.05749, over 18485.00 frames. ], tot_loss[loss=0.1812, simple_loss=0.2682, pruned_loss=0.04711, over 3478500.30 frames. ], batch size: 54, lr: 6.46e-03, grad_scale: 16.0
2022-12-23 11:52:53,618 WARNING [train.py:1060] (2/4) Exclude cut with ID 543-133211-0007-59831-0_sp0.9 from training. Duration: 21.388875
2022-12-23 11:52:55,459 INFO [zipformer.py:660] (2/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=60158.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 11:53:29,332 WARNING [train.py:1060] (2/4) Exclude cut with ID 1914-133440-0024-94914-0 from training. Duration: 22.72
2022-12-23 11:53:30,792 WARNING [train.py:1060] (2/4) Exclude cut with ID 1914-133440-0031-94921-0_sp0.9 from training. Duration: 22.7444375
2022-12-23 11:53:55,200 INFO [zipformer.py:660] (2/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=60198.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 11:54:04,710 INFO [zipformer.py:660] (2/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=60204.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 11:54:06,014 INFO [train.py:894] (2/4) Epoch 18, batch 600, loss[loss=0.187, simple_loss=0.2769, pruned_loss=0.04849, over 18705.00 frames. ], tot_loss[loss=0.1808, simple_loss=0.2678, pruned_loss=0.04695, over 3531027.58 frames. ], batch size: 52, lr: 6.46e-03, grad_scale: 16.0
2022-12-23 11:54:07,619 INFO [zipformer.py:660] (2/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=60206.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 11:54:13,478 WARNING [train.py:1060] (2/4) Exclude cut with ID 4133-6541-0027-40495-0_sp1.1 from training. Duration: 0.9681875
2022-12-23 11:54:16,371 WARNING [train.py:1060] (2/4) Exclude cut with ID 6330-62851-0022-91297-0_sp0.9 from training. Duration: 22.3166875
2022-12-23 11:54:22,230 WARNING [train.py:1060] (2/4) Exclude cut with ID 543-133212-0015-59917-0_sp0.9 from training. Duration: 21.8166875
2022-12-23 11:54:50,621 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 2.513e+02 3.538e+02 4.331e+02 5.491e+02 1.076e+03, threshold=8.662e+02, percent-clipped=2.0
2022-12-23 11:55:05,749 INFO [zipformer.py:660] (2/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=60245.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 11:55:21,581 INFO [train.py:894] (2/4) Epoch 18, batch 650, loss[loss=0.2124, simple_loss=0.2973, pruned_loss=0.06376, over 18556.00 frames. ], tot_loss[loss=0.181, simple_loss=0.2679, pruned_loss=0.04703, over 3571929.19 frames. ], batch size: 98, lr: 6.46e-03, grad_scale: 16.0
2022-12-23 11:55:27,689 INFO [zipformer.py:660] (2/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=60259.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 11:55:49,715 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.6108, 1.9652, 1.5206, 2.2335, 2.6747, 1.5901, 1.6761, 1.2594],
       device='cuda:2'), covar=tensor([0.1784, 0.1569, 0.1484, 0.0941, 0.1133, 0.1083, 0.1903, 0.1454],
       device='cuda:2'), in_proj_covar=tensor([0.0242, 0.0219, 0.0210, 0.0193, 0.0255, 0.0193, 0.0219, 0.0194],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-23 11:55:59,573 INFO [scaling.py:679] (2/4) Whitening: num_groups=8, num_channels=192, metric=1.92 vs. limit=2.0
2022-12-23 11:56:04,637 WARNING [train.py:1060] (2/4) Exclude cut with ID 4957-30119-0041-23990-0_sp0.9 from training. Duration: 20.22775
2022-12-23 11:56:36,960 INFO [train.py:894] (2/4) Epoch 18, batch 700, loss[loss=0.1926, simple_loss=0.2764, pruned_loss=0.05438, over 18642.00 frames. ], tot_loss[loss=0.1816, simple_loss=0.2686, pruned_loss=0.0473, over 3603351.39 frames. ], batch size: 53, lr: 6.45e-03, grad_scale: 16.0
2022-12-23 11:56:46,954 WARNING [train.py:1060] (2/4) Exclude cut with ID 5239-32139-0047-9341-0_sp1.1 from training. Duration: 24.67275
2022-12-23 11:56:47,236 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.5381, 1.0945, 0.6470, 1.0969, 2.1386, 0.5805, 1.2147, 1.3183],
       device='cuda:2'), covar=tensor([0.1770, 0.2361, 0.2110, 0.1734, 0.1556, 0.1809, 0.1689, 0.2016],
       device='cuda:2'), in_proj_covar=tensor([0.0093, 0.0097, 0.0117, 0.0094, 0.0115, 0.0091, 0.0098, 0.0094],
       device='cuda:2'), out_proj_covar=tensor([0.0003, 0.0003, 0.0004, 0.0003, 0.0004, 0.0003, 0.0003, 0.0003],
       device='cuda:2')
2022-12-23 11:57:00,948 INFO [scaling.py:679] (2/4) Whitening: num_groups=1, num_channels=384, metric=4.43 vs. limit=5.0
2022-12-23 11:57:15,070 WARNING [train.py:1060] (2/4) Exclude cut with ID 3082-165428-0081-50734-0_sp0.9 from training. Duration: 21.8055625
2022-12-23 11:57:20,764 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 1.961e+02 3.464e+02 4.226e+02 5.168e+02 1.038e+03, threshold=8.451e+02, percent-clipped=1.0
2022-12-23 11:57:22,613 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.5044, 1.2317, 1.6910, 2.6654, 1.9067, 2.2654, 0.8349, 1.8067],
       device='cuda:2'), covar=tensor([0.1781, 0.1665, 0.1391, 0.0634, 0.1123, 0.1150, 0.2135, 0.1207],
       device='cuda:2'), in_proj_covar=tensor([0.0099, 0.0113, 0.0130, 0.0141, 0.0104, 0.0135, 0.0127, 0.0108],
       device='cuda:2'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0004, 0.0003, 0.0004, 0.0003, 0.0003],
       device='cuda:2')
2022-12-23 11:57:51,731 INFO [train.py:894] (2/4) Epoch 18, batch 750, loss[loss=0.2019, simple_loss=0.2897, pruned_loss=0.05705, over 18659.00 frames. ], tot_loss[loss=0.1821, simple_loss=0.269, pruned_loss=0.0476, over 3628050.15 frames. ], batch size: 62, lr: 6.45e-03, grad_scale: 16.0
2022-12-23 11:57:51,735 WARNING [train.py:1060] (2/4) Exclude cut with ID 3340-169293-0054-76830-0_sp0.9 from training. Duration: 22.6666875
2022-12-23 11:58:22,957 INFO [scaling.py:679] (2/4) Whitening: num_groups=8, num_channels=96, metric=1.14 vs. limit=2.0
2022-12-23 11:58:39,415 INFO [scaling.py:679] (2/4) Whitening: num_groups=8, num_channels=96, metric=1.24 vs. limit=2.0
2022-12-23 11:58:52,765 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([3.1444, 5.5796, 4.9092, 2.5395, 5.5589, 4.1411, 0.7538, 3.8889],
       device='cuda:2'), covar=tensor([0.1763, 0.0781, 0.1276, 0.3198, 0.0554, 0.0811, 0.5076, 0.1208],
       device='cuda:2'), in_proj_covar=tensor([0.0143, 0.0136, 0.0155, 0.0122, 0.0138, 0.0110, 0.0141, 0.0112],
       device='cuda:2'), out_proj_covar=tensor([0.0003, 0.0002, 0.0003, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-23 11:58:52,772 INFO [zipformer.py:660] (2/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=60396.0, num_to_drop=1, layers_to_drop={0}
2022-12-23 11:58:56,799 WARNING [train.py:1060] (2/4) Exclude cut with ID 2411-132532-0017-82279-0_sp1.1 from training. Duration: 0.9681875
2022-12-23 11:59:05,837 INFO [train.py:894] (2/4) Epoch 18, batch 800, loss[loss=0.196, simple_loss=0.2816, pruned_loss=0.05525, over 18695.00 frames. ], tot_loss[loss=0.1817, simple_loss=0.2688, pruned_loss=0.04727, over 3647174.24 frames. ], batch size: 62, lr: 6.45e-03, grad_scale: 16.0
2022-12-23 11:59:10,494 INFO [zipformer.py:660] (2/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=60408.0, num_to_drop=1, layers_to_drop={0}
2022-12-23 11:59:22,508 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([2.4729, 2.6699, 1.6809, 2.9779, 2.7553, 2.4778, 3.6688, 2.5314],
       device='cuda:2'), covar=tensor([0.0794, 0.1640, 0.2713, 0.1783, 0.1553, 0.0772, 0.0799, 0.1122],
       device='cuda:2'), in_proj_covar=tensor([0.0175, 0.0202, 0.0244, 0.0283, 0.0231, 0.0187, 0.0206, 0.0199],
       device='cuda:2'), out_proj_covar=tensor([0.0001, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-23 11:59:23,573 WARNING [train.py:1060] (2/4) Exclude cut with ID 6330-62850-0007-91323-0 from training. Duration: 22.485
2022-12-23 11:59:29,447 INFO [zipformer.py:660] (2/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=60421.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 11:59:48,840 INFO [zipformer.py:660] (2/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=60434.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 11:59:50,046 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 2.393e+02 3.573e+02 4.387e+02 5.388e+02 1.358e+03, threshold=8.775e+02, percent-clipped=3.0
2022-12-23 12:00:02,376 WARNING [train.py:1060] (2/4) Exclude cut with ID 3972-170212-0014-23379-0_sp1.1 from training. Duration: 23.82275
2022-12-23 12:00:14,283 WARNING [train.py:1060] (2/4) Exclude cut with ID 4860-13185-0032-76709-0 from training. Duration: 20.77
2022-12-23 12:00:16,160 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([3.4744, 2.8777, 2.5117, 1.1740, 2.6480, 2.1402, 1.7439, 2.3216],
       device='cuda:2'), covar=tensor([0.0701, 0.0735, 0.1861, 0.2170, 0.1697, 0.1861, 0.2156, 0.1254],
       device='cuda:2'), in_proj_covar=tensor([0.0170, 0.0182, 0.0205, 0.0190, 0.0208, 0.0201, 0.0212, 0.0198],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-23 12:00:20,317 INFO [train.py:894] (2/4) Epoch 18, batch 850, loss[loss=0.1517, simple_loss=0.2291, pruned_loss=0.03711, over 18394.00 frames. ], tot_loss[loss=0.1814, simple_loss=0.2686, pruned_loss=0.04709, over 3661920.36 frames. ], batch size: 42, lr: 6.45e-03, grad_scale: 16.0
2022-12-23 12:00:20,321 WARNING [train.py:1060] (2/4) Exclude cut with ID 6426-64292-0017-15984-0_sp0.9 from training. Duration: 24.088875
2022-12-23 12:00:43,037 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([2.5613, 2.8971, 3.0397, 1.5580, 3.2617, 3.3802, 2.1799, 3.4439],
       device='cuda:2'), covar=tensor([0.1190, 0.1564, 0.1387, 0.2292, 0.0681, 0.1009, 0.2074, 0.0523],
       device='cuda:2'), in_proj_covar=tensor([0.0196, 0.0208, 0.0205, 0.0195, 0.0173, 0.0214, 0.0210, 0.0191],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-23 12:00:49,784 WARNING [train.py:1060] (2/4) Exclude cut with ID 6330-62850-0007-91323-0_sp1.1 from training. Duration: 20.4409375
2022-12-23 12:01:34,003 INFO [zipformer.py:660] (2/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=60504.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 12:01:35,268 INFO [train.py:894] (2/4) Epoch 18, batch 900, loss[loss=0.1799, simple_loss=0.2688, pruned_loss=0.04551, over 18575.00 frames. ], tot_loss[loss=0.1812, simple_loss=0.2683, pruned_loss=0.04701, over 3672995.59 frames. ], batch size: 177, lr: 6.44e-03, grad_scale: 16.0
2022-12-23 12:01:43,118 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([2.4554, 3.9191, 3.8328, 1.7881, 4.0316, 2.9038, 0.7797, 2.7175],
       device='cuda:2'), covar=tensor([0.2157, 0.1153, 0.1481, 0.3455, 0.0791, 0.0976, 0.5001, 0.1439],
       device='cuda:2'), in_proj_covar=tensor([0.0144, 0.0137, 0.0155, 0.0123, 0.0138, 0.0110, 0.0143, 0.0113],
       device='cuda:2'), out_proj_covar=tensor([0.0003, 0.0002, 0.0003, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-23 12:02:07,120 WARNING [train.py:1060] (2/4) Exclude cut with ID 2929-85685-0062-62366-0_sp0.9 from training. Duration: 22.511125
2022-12-23 12:02:08,556 WARNING [train.py:1060] (2/4) Exclude cut with ID 7395-89880-0031-39906-0 from training. Duration: 20.675
2022-12-23 12:02:20,152 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 2.232e+02 3.736e+02 4.464e+02 5.276e+02 1.113e+03, threshold=8.928e+02, percent-clipped=1.0
2022-12-23 12:02:36,114 INFO [zipformer.py:660] (2/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=60545.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 12:02:38,295 INFO [scaling.py:679] (2/4) Whitening: num_groups=1, num_channels=384, metric=4.02 vs. limit=5.0
2022-12-23 12:02:45,830 INFO [zipformer.py:660] (2/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=60552.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 12:02:48,659 INFO [zipformer.py:660] (2/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=60554.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 12:02:49,903 INFO [train.py:894] (2/4) Epoch 18, batch 950, loss[loss=0.1986, simple_loss=0.2863, pruned_loss=0.05546, over 18574.00 frames. ], tot_loss[loss=0.182, simple_loss=0.2694, pruned_loss=0.04732, over 3682070.77 frames. ], batch size: 51, lr: 6.44e-03, grad_scale: 16.0
2022-12-23 12:03:47,400 WARNING [train.py:1060] (2/4) Exclude cut with ID 6330-62850-0007-91323-0_sp0.9 from training. Duration: 24.9833125
2022-12-23 12:03:47,535 INFO [zipformer.py:660] (2/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=60593.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 12:04:05,106 INFO [train.py:894] (2/4) Epoch 18, batch 1000, loss[loss=0.1668, simple_loss=0.2564, pruned_loss=0.03863, over 18465.00 frames. ], tot_loss[loss=0.1812, simple_loss=0.2684, pruned_loss=0.04694, over 3688478.96 frames. ], batch size: 50, lr: 6.44e-03, grad_scale: 16.0
2022-12-23 12:04:16,371 WARNING [train.py:1060] (2/4) Exclude cut with ID 5239-32139-0047-9341-0 from training. Duration: 27.14
2022-12-23 12:04:32,402 WARNING [train.py:1060] (2/4) Exclude cut with ID 2929-85685-0044-62348-0 from training. Duration: 22.44
2022-12-23 12:04:48,804 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 2.280e+02 3.696e+02 4.340e+02 5.163e+02 1.095e+03, threshold=8.681e+02, percent-clipped=1.0
2022-12-23 12:05:19,539 INFO [train.py:894] (2/4) Epoch 18, batch 1050, loss[loss=0.1747, simple_loss=0.2684, pruned_loss=0.04056, over 18630.00 frames. ], tot_loss[loss=0.1814, simple_loss=0.2686, pruned_loss=0.04707, over 3694399.60 frames. ], batch size: 53, lr: 6.44e-03, grad_scale: 16.0
2022-12-23 12:05:22,972 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.4640, 1.6373, 1.7287, 1.0871, 1.7227, 1.8085, 1.3331, 2.0482],
       device='cuda:2'), covar=tensor([0.1124, 0.1869, 0.1241, 0.1743, 0.0716, 0.1118, 0.2419, 0.0557],
       device='cuda:2'), in_proj_covar=tensor([0.0198, 0.0209, 0.0205, 0.0196, 0.0173, 0.0214, 0.0212, 0.0190],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-23 12:05:30,732 INFO [scaling.py:679] (2/4) Whitening: num_groups=8, num_channels=192, metric=1.94 vs. limit=2.0
2022-12-23 12:05:45,204 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([2.9247, 2.3901, 1.8391, 2.9507, 2.1587, 2.2974, 2.4411, 3.3344],
       device='cuda:2'), covar=tensor([0.2118, 0.3120, 0.1814, 0.2756, 0.3738, 0.1048, 0.2921, 0.0777],
       device='cuda:2'), in_proj_covar=tensor([0.0290, 0.0283, 0.0239, 0.0343, 0.0265, 0.0224, 0.0281, 0.0207],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0003, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-23 12:05:47,869 WARNING [train.py:1060] (2/4) Exclude cut with ID 2929-85685-0060-62364-0_sp0.9 from training. Duration: 21.361125
2022-12-23 12:05:52,243 WARNING [train.py:1060] (2/4) Exclude cut with ID 2929-85685-0079-62383-0_sp1.1 from training. Duration: 27.0318125
2022-12-23 12:06:04,049 WARNING [train.py:1060] (2/4) Exclude cut with ID 5622-44585-0006-90525-0_sp0.9 from training. Duration: 28.638875
2022-12-23 12:06:19,150 WARNING [train.py:1060] (2/4) Exclude cut with ID 3340-169293-0054-76830-0 from training. Duration: 20.4
2022-12-23 12:06:22,338 INFO [zipformer.py:660] (2/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=60696.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 12:06:34,839 INFO [train.py:894] (2/4) Epoch 18, batch 1100, loss[loss=0.1672, simple_loss=0.2556, pruned_loss=0.03944, over 18395.00 frames. ], tot_loss[loss=0.1808, simple_loss=0.2683, pruned_loss=0.04671, over 3699155.69 frames. ], batch size: 46, lr: 6.43e-03, grad_scale: 16.0
2022-12-23 12:06:39,655 INFO [zipformer.py:660] (2/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=60708.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 12:06:51,818 WARNING [train.py:1060] (2/4) Exclude cut with ID 2929-85685-0071-62375-0 from training. Duration: 20.025
2022-12-23 12:06:51,837 WARNING [train.py:1060] (2/4) Exclude cut with ID 2364-131735-0112-64612-0_sp0.9 from training. Duration: 20.488875
2022-12-23 12:06:55,261 INFO [zipformer.py:660] (2/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=60718.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 12:06:57,644 WARNING [train.py:1060] (2/4) Exclude cut with ID 2929-85685-0079-62383-0 from training. Duration: 29.735
2022-12-23 12:06:59,231 INFO [zipformer.py:660] (2/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=60721.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 12:07:00,817 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([2.6937, 2.5648, 1.8114, 3.1892, 2.8775, 2.5279, 3.9412, 2.6055],
       device='cuda:2'), covar=tensor([0.0752, 0.1806, 0.2707, 0.1685, 0.1580, 0.0808, 0.0770, 0.1085],
       device='cuda:2'), in_proj_covar=tensor([0.0177, 0.0206, 0.0248, 0.0287, 0.0235, 0.0189, 0.0208, 0.0202],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-23 12:07:11,080 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.6820, 1.2417, 0.8440, 1.2649, 2.0531, 1.0147, 1.5056, 1.6332],
       device='cuda:2'), covar=tensor([0.1615, 0.1917, 0.2106, 0.1501, 0.1710, 0.1721, 0.1415, 0.1592],
       device='cuda:2'), in_proj_covar=tensor([0.0093, 0.0097, 0.0116, 0.0094, 0.0114, 0.0091, 0.0098, 0.0093],
       device='cuda:2'), out_proj_covar=tensor([0.0003, 0.0003, 0.0004, 0.0003, 0.0004, 0.0003, 0.0003, 0.0003],
       device='cuda:2')
2022-12-23 12:07:19,410 INFO [zipformer.py:660] (2/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=60734.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 12:07:20,615 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 2.164e+02 3.194e+02 3.933e+02 5.010e+02 1.354e+03, threshold=7.866e+02, percent-clipped=2.0
2022-12-23 12:07:32,459 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([2.4730, 1.6596, 0.6122, 1.8056, 2.4155, 1.6665, 2.2121, 2.2171],
       device='cuda:2'), covar=tensor([0.1387, 0.1765, 0.2407, 0.1415, 0.1560, 0.1632, 0.1298, 0.1544],
       device='cuda:2'), in_proj_covar=tensor([0.0093, 0.0097, 0.0116, 0.0095, 0.0114, 0.0091, 0.0098, 0.0093],
       device='cuda:2'), out_proj_covar=tensor([0.0003, 0.0003, 0.0004, 0.0003, 0.0004, 0.0003, 0.0003, 0.0003],
       device='cuda:2')
2022-12-23 12:07:33,610 INFO [zipformer.py:660] (2/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=60744.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 12:07:49,971 INFO [train.py:894] (2/4) Epoch 18, batch 1150, loss[loss=0.1836, simple_loss=0.267, pruned_loss=0.05012, over 18427.00 frames. ], tot_loss[loss=0.1806, simple_loss=0.2677, pruned_loss=0.04674, over 3701238.07 frames. ], batch size: 48, lr: 6.43e-03, grad_scale: 16.0
2022-12-23 12:07:51,619 INFO [zipformer.py:660] (2/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=60756.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 12:08:10,452 INFO [zipformer.py:660] (2/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=60769.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 12:08:20,302 WARNING [train.py:1060] (2/4) Exclude cut with ID 7276-92427-0014-12983-0_sp0.9 from training. Duration: 21.3055625
2022-12-23 12:08:21,824 WARNING [train.py:1060] (2/4) Exclude cut with ID 1025-75365-0008-79168-0_sp0.9 from training. Duration: 22.0666875
2022-12-23 12:08:25,803 INFO [zipformer.py:660] (2/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=60779.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 12:08:29,425 INFO [zipformer.py:660] (2/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=60782.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 12:08:43,826 INFO [scaling.py:679] (2/4) Whitening: num_groups=8, num_channels=192, metric=1.91 vs. limit=2.0
2022-12-23 12:09:03,248 INFO [train.py:894] (2/4) Epoch 18, batch 1200, loss[loss=0.1641, simple_loss=0.2443, pruned_loss=0.042, over 18411.00 frames. ], tot_loss[loss=0.1806, simple_loss=0.2679, pruned_loss=0.04666, over 3703343.15 frames. ], batch size: 42, lr: 6.43e-03, grad_scale: 16.0
2022-12-23 12:09:48,496 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 2.177e+02 3.323e+02 4.383e+02 5.760e+02 1.405e+03, threshold=8.765e+02, percent-clipped=5.0
2022-12-23 12:09:58,879 INFO [scaling.py:679] (2/4) Whitening: num_groups=8, num_channels=192, metric=1.71 vs. limit=2.0
2022-12-23 12:10:12,637 WARNING [train.py:1060] (2/4) Exclude cut with ID 2929-85685-0062-62366-0 from training. Duration: 20.26
2022-12-23 12:10:15,861 INFO [zipformer.py:660] (2/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=60854.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 12:10:17,016 INFO [train.py:894] (2/4) Epoch 18, batch 1250, loss[loss=0.1655, simple_loss=0.2532, pruned_loss=0.03888, over 18565.00 frames. ], tot_loss[loss=0.1807, simple_loss=0.2684, pruned_loss=0.04653, over 3704681.36 frames. ], batch size: 49, lr: 6.43e-03, grad_scale: 16.0
2022-12-23 12:10:27,054 WARNING [train.py:1060] (2/4) Exclude cut with ID 5239-32139-0030-9324-0_sp0.9 from training. Duration: 21.3444375
2022-12-23 12:10:46,935 INFO [scaling.py:679] (2/4) Whitening: num_groups=1, num_channels=384, metric=2.69 vs. limit=5.0
2022-12-23 12:11:24,444 WARNING [train.py:1060] (2/4) Exclude cut with ID 497-129325-0061-62254-0_sp1.1 from training. Duration: 0.97725
2022-12-23 12:11:27,564 INFO [zipformer.py:660] (2/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=60902.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 12:11:32,078 INFO [train.py:894] (2/4) Epoch 18, batch 1300, loss[loss=0.1646, simple_loss=0.2495, pruned_loss=0.03982, over 18561.00 frames. ], tot_loss[loss=0.1796, simple_loss=0.2672, pruned_loss=0.04597, over 3706038.79 frames. ], batch size: 49, lr: 6.42e-03, grad_scale: 16.0
2022-12-23 12:12:06,997 WARNING [train.py:1060] (2/4) Exclude cut with ID 7395-89880-0031-39906-0_sp0.9 from training. Duration: 22.97225
2022-12-23 12:12:16,807 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 2.396e+02 3.264e+02 4.038e+02 4.760e+02 9.694e+02, threshold=8.075e+02, percent-clipped=2.0
2022-12-23 12:12:17,255 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([3.6685, 3.3393, 2.9934, 1.5452, 2.6863, 2.6225, 2.1181, 2.8306],
       device='cuda:2'), covar=tensor([0.0569, 0.0498, 0.1073, 0.1544, 0.1410, 0.1223, 0.1438, 0.0861],
       device='cuda:2'), in_proj_covar=tensor([0.0171, 0.0183, 0.0205, 0.0189, 0.0210, 0.0199, 0.0213, 0.0199],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-23 12:12:25,010 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([6.0694, 5.1528, 5.3540, 6.0156, 5.5669, 5.3230, 6.0963, 1.8861],
       device='cuda:2'), covar=tensor([0.0474, 0.0489, 0.0444, 0.0568, 0.1103, 0.1038, 0.0338, 0.4880],
       device='cuda:2'), in_proj_covar=tensor([0.0331, 0.0215, 0.0229, 0.0251, 0.0310, 0.0260, 0.0277, 0.0272],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0001, 0.0001, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-23 12:12:36,640 WARNING [train.py:1060] (2/4) Exclude cut with ID 7395-89880-0047-39922-0_sp0.9 from training. Duration: 21.97775
2022-12-23 12:12:46,589 INFO [train.py:894] (2/4) Epoch 18, batch 1350, loss[loss=0.1469, simple_loss=0.2293, pruned_loss=0.03224, over 18485.00 frames. ], tot_loss[loss=0.1796, simple_loss=0.2673, pruned_loss=0.0459, over 3707604.76 frames. ], batch size: 43, lr: 6.42e-03, grad_scale: 16.0
2022-12-23 12:12:49,592 WARNING [train.py:1060] (2/4) Exclude cut with ID 1112-1043-0006-89194-0_sp0.9 from training. Duration: 21.8333125
2022-12-23 12:13:00,727 WARNING [train.py:1060] (2/4) Exclude cut with ID 1914-133440-0031-94921-0 from training. Duration: 20.47
2022-12-23 12:13:27,495 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([2.3216, 2.0265, 1.4871, 0.5860, 1.4613, 2.0401, 1.6437, 1.8447],
       device='cuda:2'), covar=tensor([0.0575, 0.0454, 0.1022, 0.1472, 0.1101, 0.1333, 0.1495, 0.0615],
       device='cuda:2'), in_proj_covar=tensor([0.0172, 0.0184, 0.0206, 0.0190, 0.0211, 0.0200, 0.0213, 0.0201],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-23 12:13:30,508 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([3.7328, 3.2290, 2.9588, 1.3570, 2.7302, 2.7427, 2.3679, 2.7623],
       device='cuda:2'), covar=tensor([0.0553, 0.0585, 0.1218, 0.1816, 0.1567, 0.1249, 0.1362, 0.0961],
       device='cuda:2'), in_proj_covar=tensor([0.0172, 0.0184, 0.0206, 0.0190, 0.0211, 0.0200, 0.0213, 0.0201],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-23 12:14:01,336 INFO [train.py:894] (2/4) Epoch 18, batch 1400, loss[loss=0.1852, simple_loss=0.2756, pruned_loss=0.04737, over 18552.00 frames. ], tot_loss[loss=0.1805, simple_loss=0.2685, pruned_loss=0.04624, over 3709346.98 frames. ], batch size: 58, lr: 6.42e-03, grad_scale: 16.0
2022-12-23 12:14:08,860 WARNING [train.py:1060] (2/4) Exclude cut with ID 7395-89880-0037-39912-0_sp0.9 from training. Duration: 20.67225
2022-12-23 12:14:27,286 WARNING [train.py:1060] (2/4) Exclude cut with ID 1914-133440-0024-94914-0_sp0.9 from training. Duration: 25.2444375
2022-12-23 12:14:48,188 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 2.343e+02 3.490e+02 4.029e+02 5.313e+02 7.188e+02, threshold=8.059e+02, percent-clipped=0.0
2022-12-23 12:14:51,218 WARNING [train.py:1060] (2/4) Exclude cut with ID 3340-169293-0021-76797-0_sp0.9 from training. Duration: 21.1445
2022-12-23 12:14:54,840 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([3.2326, 2.4250, 1.8798, 2.8999, 2.2714, 2.3706, 2.5083, 3.4565],
       device='cuda:2'), covar=tensor([0.1842, 0.3033, 0.1815, 0.2782, 0.3590, 0.0944, 0.3094, 0.0713],
       device='cuda:2'), in_proj_covar=tensor([0.0289, 0.0281, 0.0238, 0.0342, 0.0264, 0.0222, 0.0281, 0.0206],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0003, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-23 12:15:13,936 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([2.6011, 2.2324, 2.1807, 1.5354, 2.8169, 2.6902, 2.3005, 1.8780],
       device='cuda:2'), covar=tensor([0.0316, 0.0399, 0.0402, 0.0669, 0.0237, 0.0299, 0.0434, 0.0758],
       device='cuda:2'), in_proj_covar=tensor([0.0123, 0.0124, 0.0127, 0.0120, 0.0096, 0.0121, 0.0134, 0.0155],
       device='cuda:2'), out_proj_covar=tensor([0.0001, 0.0002, 0.0002, 0.0001, 0.0001, 0.0001, 0.0002, 0.0002],
       device='cuda:2')
2022-12-23 12:15:17,583 INFO [train.py:894] (2/4) Epoch 18, batch 1450, loss[loss=0.1899, simple_loss=0.281, pruned_loss=0.0494, over 18578.00 frames. ], tot_loss[loss=0.1802, simple_loss=0.2682, pruned_loss=0.04611, over 3710180.19 frames. ], batch size: 77, lr: 6.41e-03, grad_scale: 16.0
2022-12-23 12:15:27,214 INFO [zipformer.py:660] (2/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=61061.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 12:15:45,920 INFO [scaling.py:679] (2/4) Whitening: num_groups=1, num_channels=384, metric=3.65 vs. limit=5.0
2022-12-23 12:15:46,560 INFO [zipformer.py:660] (2/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=61074.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 12:16:05,179 WARNING [train.py:1060] (2/4) Exclude cut with ID 2929-85685-0079-62383-0_sp0.9 from training. Duration: 33.038875
2022-12-23 12:16:32,532 INFO [train.py:894] (2/4) Epoch 18, batch 1500, loss[loss=0.1869, simple_loss=0.2716, pruned_loss=0.05112, over 18706.00 frames. ], tot_loss[loss=0.1797, simple_loss=0.2676, pruned_loss=0.04593, over 3711557.75 frames. ], batch size: 50, lr: 6.41e-03, grad_scale: 16.0
2022-12-23 12:16:42,751 WARNING [train.py:1060] (2/4) Exclude cut with ID 6426-64291-0000-16059-0_sp0.9 from training. Duration: 20.0944375
2022-12-23 12:16:55,077 WARNING [train.py:1060] (2/4) Exclude cut with ID 2929-85685-0044-62348-0_sp1.1 from training. Duration: 20.4
2022-12-23 12:16:58,873 INFO [zipformer.py:660] (2/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=61122.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 12:17:00,553 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([2.1948, 1.5837, 2.6139, 4.6026, 3.4753, 2.9914, 1.0760, 3.2538],
       device='cuda:2'), covar=tensor([0.1689, 0.1709, 0.1398, 0.0391, 0.0797, 0.1006, 0.2217, 0.0773],
       device='cuda:2'), in_proj_covar=tensor([0.0102, 0.0116, 0.0133, 0.0145, 0.0106, 0.0137, 0.0129, 0.0111],
       device='cuda:2'), out_proj_covar=tensor([0.0003, 0.0003, 0.0004, 0.0004, 0.0003, 0.0004, 0.0003, 0.0003],
       device='cuda:2')
2022-12-23 12:17:03,314 WARNING [train.py:1060] (2/4) Exclude cut with ID 6330-62851-0022-91297-0 from training. Duration: 20.085
2022-12-23 12:17:14,085 WARNING [train.py:1060] (2/4) Exclude cut with ID 4860-13185-0032-76709-0_sp0.9 from training. Duration: 23.07775
2022-12-23 12:17:18,337 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 2.189e+02 3.774e+02 4.544e+02 5.492e+02 1.131e+03, threshold=9.087e+02, percent-clipped=3.0
2022-12-23 12:17:48,323 INFO [train.py:894] (2/4) Epoch 18, batch 1550, loss[loss=0.1872, simple_loss=0.2605, pruned_loss=0.05696, over 18616.00 frames. ], tot_loss[loss=0.1801, simple_loss=0.2678, pruned_loss=0.04623, over 3711611.28 frames. ], batch size: 45, lr: 6.41e-03, grad_scale: 16.0
2022-12-23 12:17:59,725 WARNING [train.py:1060] (2/4) Exclude cut with ID 2929-85685-0044-62348-0_sp0.9 from training. Duration: 24.9333125
2022-12-23 12:18:11,672 INFO [scaling.py:679] (2/4) Whitening: num_groups=8, num_channels=192, metric=1.95 vs. limit=2.0
2022-12-23 12:18:46,682 WARNING [train.py:1060] (2/4) Exclude cut with ID 5118-111612-0016-124680-0_sp0.9 from training. Duration: 20.388875
2022-12-23 12:18:52,549 WARNING [train.py:1060] (2/4) Exclude cut with ID 432-122774-0017-62487-0_sp1.1 from training. Duration: 20.3590625
2022-12-23 12:19:04,961 INFO [train.py:894] (2/4) Epoch 18, batch 1600, loss[loss=0.152, simple_loss=0.2336, pruned_loss=0.03522, over 18529.00 frames. ], tot_loss[loss=0.1797, simple_loss=0.2672, pruned_loss=0.04607, over 3711435.62 frames. ], batch size: 41, lr: 6.41e-03, grad_scale: 16.0
2022-12-23 12:19:48,907 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 2.328e+02 3.675e+02 4.504e+02 5.800e+02 1.343e+03, threshold=9.009e+02, percent-clipped=8.0
2022-12-23 12:20:02,099 WARNING [train.py:1060] (2/4) Exclude cut with ID 3557-8342-0013-54691-0_sp1.1 from training. Duration: 0.836375
2022-12-23 12:20:18,433 INFO [train.py:894] (2/4) Epoch 18, batch 1650, loss[loss=0.2087, simple_loss=0.289, pruned_loss=0.06425, over 18526.00 frames. ], tot_loss[loss=0.1811, simple_loss=0.2681, pruned_loss=0.04704, over 3712563.99 frames. ], batch size: 58, lr: 6.40e-03, grad_scale: 32.0
2022-12-23 12:20:45,149 WARNING [train.py:1060] (2/4) Exclude cut with ID 8565-290391-0049-67394-0_sp0.9 from training. Duration: 21.3166875
2022-12-23 12:21:14,410 WARNING [train.py:1060] (2/4) Exclude cut with ID 6533-399-0029-104863-0_sp0.9 from training. Duration: 22.1055625
2022-12-23 12:21:26,655 WARNING [train.py:1060] (2/4) Exclude cut with ID 7699-105389-0094-26379-0_sp1.1 from training. Duration: 21.77725
2022-12-23 12:21:34,229 INFO [train.py:894] (2/4) Epoch 18, batch 1700, loss[loss=0.1979, simple_loss=0.2833, pruned_loss=0.05622, over 18554.00 frames. ], tot_loss[loss=0.1822, simple_loss=0.2684, pruned_loss=0.04796, over 3712314.70 frames. ], batch size: 55, lr: 6.40e-03, grad_scale: 16.0
2022-12-23 12:21:44,829 WARNING [train.py:1060] (2/4) Exclude cut with ID 7255-291500-0005-134304-0_sp0.9 from training. Duration: 27.8166875
2022-12-23 12:22:11,122 WARNING [train.py:1060] (2/4) Exclude cut with ID 7357-94126-0021-15852-0_sp1.1 from training. Duration: 22.5090625
2022-12-23 12:22:18,179 WARNING [train.py:1060] (2/4) Exclude cut with ID 7255-291500-0005-134304-0 from training. Duration: 25.035
2022-12-23 12:22:21,104 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 2.660e+02 4.296e+02 5.030e+02 5.943e+02 1.306e+03, threshold=1.006e+03, percent-clipped=4.0
2022-12-23 12:22:35,357 WARNING [train.py:1060] (2/4) Exclude cut with ID 774-127930-0014-10412-0_sp1.1 from training. Duration: 0.95
2022-12-23 12:22:48,764 INFO [train.py:894] (2/4) Epoch 18, batch 1750, loss[loss=0.1974, simple_loss=0.2788, pruned_loss=0.05804, over 18578.00 frames. ], tot_loss[loss=0.1846, simple_loss=0.2697, pruned_loss=0.04975, over 3712799.12 frames. ], batch size: 49, lr: 6.40e-03, grad_scale: 16.0
2022-12-23 12:22:54,878 WARNING [train.py:1060] (2/4) Exclude cut with ID 3033-130750-0096-55598-0_sp0.9 from training. Duration: 0.92225
2022-12-23 12:23:17,350 INFO [zipformer.py:660] (2/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=61374.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 12:23:21,314 WARNING [train.py:1060] (2/4) Exclude cut with ID 4511-76322-0006-80011-0 from training. Duration: 21.97
2022-12-23 12:23:40,819 WARNING [train.py:1060] (2/4) Exclude cut with ID 7492-105653-0055-62765-0_sp0.9 from training. Duration: 21.97225
2022-12-23 12:23:42,287 WARNING [train.py:1060] (2/4) Exclude cut with ID 453-131332-0000-47844-0_sp0.9 from training. Duration: 25.3333125
2022-12-23 12:23:52,019 WARNING [train.py:1060] (2/4) Exclude cut with ID 5172-29468-0015-19128-0_sp0.9 from training. Duration: 21.5055625
2022-12-23 12:24:01,703 WARNING [train.py:1060] (2/4) Exclude cut with ID 453-131332-0000-47844-0_sp1.1 from training. Duration: 20.72725
2022-12-23 12:24:04,530 INFO [train.py:894] (2/4) Epoch 18, batch 1800, loss[loss=0.18, simple_loss=0.2649, pruned_loss=0.04759, over 18580.00 frames. ], tot_loss[loss=0.1873, simple_loss=0.2709, pruned_loss=0.05185, over 3713925.23 frames. ], batch size: 57, lr: 6.40e-03, grad_scale: 16.0
2022-12-23 12:24:23,206 INFO [zipformer.py:660] (2/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=61417.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 12:24:30,563 INFO [zipformer.py:660] (2/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=61422.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 12:24:36,674 WARNING [train.py:1060] (2/4) Exclude cut with ID 8631-249866-0030-130156-0_sp0.9 from training. Duration: 26.32775
2022-12-23 12:24:37,389 INFO [scaling.py:679] (2/4) Whitening: num_groups=1, num_channels=384, metric=2.98 vs. limit=5.0
2022-12-23 12:24:50,827 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 2.906e+02 4.710e+02 5.748e+02 7.409e+02 1.844e+03, threshold=1.150e+03, percent-clipped=6.0
2022-12-23 12:24:57,603 INFO [scaling.py:679] (2/4) Whitening: num_groups=8, num_channels=96, metric=1.61 vs. limit=2.0
2022-12-23 12:25:06,392 WARNING [train.py:1060] (2/4) Exclude cut with ID 3867-173237-0077-144769-0 from training. Duration: 20.025
2022-12-23 12:25:12,132 WARNING [train.py:1060] (2/4) Exclude cut with ID 6709-74022-0004-86860-0_sp1.1 from training. Duration: 0.9409375
2022-12-23 12:25:12,143 WARNING [train.py:1060] (2/4) Exclude cut with ID 4757-1811-0023-62229-0_sp0.9 from training. Duration: 21.37775
2022-12-23 12:25:18,727 INFO [train.py:894] (2/4) Epoch 18, batch 1850, loss[loss=0.1895, simple_loss=0.2763, pruned_loss=0.05134, over 18400.00 frames. ], tot_loss[loss=0.1894, simple_loss=0.2718, pruned_loss=0.05353, over 3713765.39 frames. ], batch size: 53, lr: 6.39e-03, grad_scale: 16.0
2022-12-23 12:25:31,306 WARNING [train.py:1060] (2/4) Exclude cut with ID 1250-135782-0004-25974-0_sp0.9 from training. Duration: 21.17225
2022-12-23 12:25:31,313 WARNING [train.py:1060] (2/4) Exclude cut with ID 7357-94126-0021-15852-0_sp0.9 from training. Duration: 27.511125
2022-12-23 12:26:04,817 WARNING [train.py:1060] (2/4) Exclude cut with ID 453-131332-0000-47844-0 from training. Duration: 22.8
2022-12-23 12:26:09,169 WARNING [train.py:1060] (2/4) Exclude cut with ID 4964-30587-0040-44509-0 from training. Duration: 22.585
2022-12-23 12:26:33,711 INFO [train.py:894] (2/4) Epoch 18, batch 1900, loss[loss=0.2132, simple_loss=0.2902, pruned_loss=0.06804, over 18638.00 frames. ], tot_loss[loss=0.1906, simple_loss=0.2721, pruned_loss=0.05448, over 3714114.55 frames. ], batch size: 173, lr: 6.39e-03, grad_scale: 16.0
2022-12-23 12:26:42,041 WARNING [train.py:1060] (2/4) Exclude cut with ID 6978-92210-0001-146967-0_sp0.9 from training. Duration: 22.0166875
2022-12-23 12:26:52,723 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([3.8615, 3.3775, 3.3557, 3.7974, 3.5167, 3.4831, 4.0038, 1.2822],
       device='cuda:2'), covar=tensor([0.0754, 0.0739, 0.0706, 0.0771, 0.1445, 0.1168, 0.0674, 0.4501],
       device='cuda:2'), in_proj_covar=tensor([0.0336, 0.0219, 0.0232, 0.0258, 0.0315, 0.0263, 0.0279, 0.0274],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0001, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-23 12:26:57,048 WARNING [train.py:1060] (2/4) Exclude cut with ID 7255-291500-0003-134302-0_sp1.1 from training. Duration: 24.395375
2022-12-23 12:27:03,649 WARNING [train.py:1060] (2/4) Exclude cut with ID 3488-85273-0017-41203-0_sp0.9 from training. Duration: 27.47775
2022-12-23 12:27:09,426 WARNING [train.py:1060] (2/4) Exclude cut with ID 432-122774-0017-62487-0_sp0.9 from training. Duration: 24.8833125
2022-12-23 12:27:12,311 WARNING [train.py:1060] (2/4) Exclude cut with ID 6758-72288-0033-108368-0 from training. Duration: 23.39
2022-12-23 12:27:18,061 WARNING [train.py:1060] (2/4) Exclude cut with ID 8291-282929-0007-12994-0_sp0.9 from training. Duration: 28.72225
2022-12-23 12:27:20,884 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 3.500e+02 4.898e+02 5.639e+02 7.232e+02 1.821e+03, threshold=1.128e+03, percent-clipped=2.0
2022-12-23 12:27:27,578 WARNING [train.py:1060] (2/4) Exclude cut with ID 585-294811-0110-133686-0_sp0.9 from training. Duration: 20.8944375
2022-12-23 12:27:42,094 WARNING [train.py:1060] (2/4) Exclude cut with ID 5796-66357-0007-116447-0_sp0.9 from training. Duration: 23.8444375
2022-12-23 12:27:44,224 INFO [scaling.py:679] (2/4) Whitening: num_groups=8, num_channels=96, metric=1.27 vs. limit=2.0
2022-12-23 12:27:49,723 INFO [train.py:894] (2/4) Epoch 18, batch 1950, loss[loss=0.2075, simple_loss=0.2847, pruned_loss=0.06521, over 18569.00 frames. ], tot_loss[loss=0.1909, simple_loss=0.2723, pruned_loss=0.05473, over 3714296.40 frames. ], batch size: 51, lr: 6.39e-03, grad_scale: 16.0
2022-12-23 12:28:06,341 WARNING [train.py:1060] (2/4) Exclude cut with ID 8291-282929-0007-12994-0 from training. Duration: 25.85
2022-12-23 12:28:06,347 WARNING [train.py:1060] (2/4) Exclude cut with ID 8291-282929-0023-13010-0 from training. Duration: 21.39
2022-12-23 12:28:17,416 WARNING [train.py:1060] (2/4) Exclude cut with ID 7357-94126-0014-15845-0 from training. Duration: 27.92
2022-12-23 12:28:30,834 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([2.6909, 2.3054, 1.7400, 2.4958, 1.9876, 2.0893, 2.1060, 2.6426],
       device='cuda:2'), covar=tensor([0.1787, 0.2705, 0.1681, 0.2489, 0.3198, 0.0970, 0.2671, 0.0781],
       device='cuda:2'), in_proj_covar=tensor([0.0289, 0.0282, 0.0238, 0.0345, 0.0265, 0.0223, 0.0281, 0.0207],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0003, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-23 12:28:45,132 WARNING [train.py:1060] (2/4) Exclude cut with ID 8631-249866-0039-130165-0_sp0.9 from training. Duration: 20.661125
2022-12-23 12:29:05,149 INFO [train.py:894] (2/4) Epoch 18, batch 2000, loss[loss=0.1619, simple_loss=0.2396, pruned_loss=0.0421, over 18550.00 frames. ], tot_loss[loss=0.1917, simple_loss=0.2727, pruned_loss=0.05539, over 3714143.09 frames. ], batch size: 44, lr: 6.39e-03, grad_scale: 16.0
2022-12-23 12:29:08,309 WARNING [train.py:1060] (2/4) Exclude cut with ID 7357-94126-0043-15874-0_sp0.9 from training. Duration: 20.07225
2022-12-23 12:29:15,865 WARNING [train.py:1060] (2/4) Exclude cut with ID 1085-156170-0017-128270-0 from training. Duration: 21.01
2022-12-23 12:29:52,147 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 3.258e+02 4.574e+02 5.379e+02 6.112e+02 1.044e+03, threshold=1.076e+03, percent-clipped=0.0
2022-12-23 12:30:21,755 INFO [train.py:894] (2/4) Epoch 18, batch 2050, loss[loss=0.1706, simple_loss=0.2567, pruned_loss=0.04226, over 18724.00 frames. ], tot_loss[loss=0.1914, simple_loss=0.2722, pruned_loss=0.05524, over 3713805.37 frames. ], batch size: 54, lr: 6.38e-03, grad_scale: 16.0
2022-12-23 12:30:21,764 WARNING [train.py:1060] (2/4) Exclude cut with ID 2195-150901-0045-59933-0 from training. Duration: 20.65
2022-12-23 12:30:29,792 WARNING [train.py:1060] (2/4) Exclude cut with ID 5796-66357-0007-116447-0 from training. Duration: 21.46
2022-12-23 12:30:33,138 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([2.5718, 2.4342, 2.0558, 1.7485, 2.6103, 3.1114, 2.9613, 2.1494],
       device='cuda:2'), covar=tensor([0.0294, 0.0282, 0.0396, 0.0267, 0.0213, 0.0234, 0.0217, 0.0274],
       device='cuda:2'), in_proj_covar=tensor([0.0093, 0.0126, 0.0152, 0.0124, 0.0115, 0.0119, 0.0096, 0.0126],
       device='cuda:2'), out_proj_covar=tensor([7.4914e-05, 1.0069e-04, 1.2664e-04, 9.9523e-05, 9.3892e-05, 9.2096e-05,
        7.6017e-05, 1.0001e-04], device='cuda:2')
2022-12-23 12:30:46,420 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([2.7315, 2.2247, 2.2367, 2.3045, 2.5468, 2.5244, 2.4565, 1.9738],
       device='cuda:2'), covar=tensor([0.1871, 0.3008, 0.2403, 0.2688, 0.1683, 0.0816, 0.3029, 0.1159],
       device='cuda:2'), in_proj_covar=tensor([0.0262, 0.0294, 0.0273, 0.0309, 0.0297, 0.0247, 0.0334, 0.0234],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-23 12:31:16,686 WARNING [train.py:1060] (2/4) Exclude cut with ID 3557-8342-0013-54691-0 from training. Duration: 0.92
2022-12-23 12:31:22,822 WARNING [train.py:1060] (2/4) Exclude cut with ID 8291-282929-0023-13010-0_sp0.9 from training. Duration: 23.7666875
2022-12-23 12:31:36,350 INFO [train.py:894] (2/4) Epoch 18, batch 2100, loss[loss=0.2009, simple_loss=0.2888, pruned_loss=0.05646, over 18468.00 frames. ], tot_loss[loss=0.1922, simple_loss=0.2726, pruned_loss=0.05589, over 3713371.74 frames. ], batch size: 64, lr: 6.38e-03, grad_scale: 16.0
2022-12-23 12:31:40,641 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([2.4748, 2.1937, 1.7625, 0.7849, 1.7022, 1.9077, 1.7449, 1.9643],
       device='cuda:2'), covar=tensor([0.0594, 0.0515, 0.1109, 0.1590, 0.1182, 0.1509, 0.1551, 0.0752],
       device='cuda:2'), in_proj_covar=tensor([0.0173, 0.0186, 0.0208, 0.0192, 0.0214, 0.0203, 0.0217, 0.0203],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-23 12:31:55,738 INFO [zipformer.py:660] (2/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=61717.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 12:31:58,303 WARNING [train.py:1060] (2/4) Exclude cut with ID 8544-281189-0060-101339-0_sp0.9 from training. Duration: 20.861125
2022-12-23 12:32:08,003 WARNING [train.py:1060] (2/4) Exclude cut with ID 3488-65654-0031-41259-0_sp0.9 from training. Duration: 22.711125
2022-12-23 12:32:18,440 INFO [zipformer.py:660] (2/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=61732.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 12:32:24,050 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 3.133e+02 4.990e+02 6.097e+02 8.007e+02 2.031e+03, threshold=1.219e+03, percent-clipped=4.0
2022-12-23 12:32:36,033 INFO [zipformer.py:660] (2/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=61743.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 12:32:53,501 INFO [train.py:894] (2/4) Epoch 18, batch 2150, loss[loss=0.1978, simple_loss=0.2864, pruned_loss=0.05459, over 18720.00 frames. ], tot_loss[loss=0.1928, simple_loss=0.2731, pruned_loss=0.05627, over 3713680.15 frames. ], batch size: 54, lr: 6.38e-03, grad_scale: 16.0
2022-12-23 12:32:53,540 WARNING [train.py:1060] (2/4) Exclude cut with ID 6951-79737-0043-132310-0_sp1.1 from training. Duration: 22.986375
2022-12-23 12:33:08,538 WARNING [train.py:1060] (2/4) Exclude cut with ID 8040-260924-0003-80960-0_sp0.9 from training. Duration: 22.07225
2022-12-23 12:33:08,646 INFO [zipformer.py:660] (2/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=61765.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 12:33:13,418 WARNING [train.py:1060] (2/4) Exclude cut with ID 7699-105389-0045-26330-0_sp0.9 from training. Duration: 20.3055625
2022-12-23 12:33:16,197 WARNING [train.py:1060] (2/4) Exclude cut with ID 6356-271890-0060-94317-0_sp0.9 from training. Duration: 20.72225
2022-12-23 12:33:31,382 WARNING [train.py:1060] (2/4) Exclude cut with ID 3488-85273-0017-41203-0_sp1.1 from training. Duration: 22.4818125
2022-12-23 12:33:34,851 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.5509, 1.5301, 1.6164, 1.4620, 1.0593, 3.6186, 1.4906, 1.9861],
       device='cuda:2'), covar=tensor([0.3237, 0.2013, 0.1977, 0.2164, 0.1533, 0.0181, 0.1646, 0.0884],
       device='cuda:2'), in_proj_covar=tensor([0.0132, 0.0116, 0.0125, 0.0120, 0.0102, 0.0097, 0.0092, 0.0089],
       device='cuda:2'), out_proj_covar=tensor([0.0005, 0.0005, 0.0005, 0.0005, 0.0004, 0.0003, 0.0004, 0.0004],
       device='cuda:2')
2022-12-23 12:33:52,220 INFO [zipformer.py:660] (2/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=61793.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 12:33:57,861 WARNING [train.py:1060] (2/4) Exclude cut with ID 4964-30587-0040-44509-0_sp0.9 from training. Duration: 25.0944375
2022-12-23 12:34:02,324 WARNING [train.py:1060] (2/4) Exclude cut with ID 6533-399-0047-104881-0 from training. Duration: 21.515
2022-12-23 12:34:05,235 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([4.2023, 3.6571, 3.5873, 4.1455, 3.8484, 3.7011, 4.3403, 1.2671],
       device='cuda:2'), covar=tensor([0.0684, 0.0644, 0.0664, 0.0723, 0.1331, 0.1110, 0.0565, 0.4885],
       device='cuda:2'), in_proj_covar=tensor([0.0337, 0.0219, 0.0233, 0.0258, 0.0314, 0.0263, 0.0278, 0.0275],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0001, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-23 12:34:09,301 WARNING [train.py:1060] (2/4) Exclude cut with ID 7357-94126-0009-15840-0_sp0.9 from training. Duration: 27.02225
2022-12-23 12:34:09,745 INFO [zipformer.py:660] (2/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=61804.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 12:34:10,770 INFO [train.py:894] (2/4) Epoch 18, batch 2200, loss[loss=0.2118, simple_loss=0.2865, pruned_loss=0.06853, over 18594.00 frames. ], tot_loss[loss=0.1936, simple_loss=0.2737, pruned_loss=0.05675, over 3714552.01 frames. ], batch size: 51, lr: 6.38e-03, grad_scale: 16.0
2022-12-23 12:34:11,246 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([0.0471, 0.9619, 1.1553, 0.5373, 0.5319, 1.1959, 1.2120, 1.1485],
       device='cuda:2'), covar=tensor([0.0737, 0.0331, 0.0363, 0.0380, 0.0465, 0.0504, 0.0285, 0.0553],
       device='cuda:2'), in_proj_covar=tensor([0.0149, 0.0168, 0.0125, 0.0138, 0.0148, 0.0141, 0.0159, 0.0169],
       device='cuda:2'), out_proj_covar=tensor([1.1454e-04, 1.3063e-04, 9.5675e-05, 1.0381e-04, 1.1237e-04, 1.0944e-04,
        1.2426e-04, 1.3033e-04], device='cuda:2')
2022-12-23 12:34:15,198 WARNING [train.py:1060] (2/4) Exclude cut with ID 432-122774-0010-62480-0_sp0.9 from training. Duration: 22.22225
2022-12-23 12:34:21,185 WARNING [train.py:1060] (2/4) Exclude cut with ID 4964-30587-0085-44554-0_sp0.9 from training. Duration: 20.85
2022-12-23 12:34:25,920 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([3.0108, 5.4317, 4.8737, 2.6585, 5.3976, 3.9924, 0.8975, 3.6166],
       device='cuda:2'), covar=tensor([0.1863, 0.0968, 0.1174, 0.3025, 0.0672, 0.0885, 0.5323, 0.1308],
       device='cuda:2'), in_proj_covar=tensor([0.0145, 0.0140, 0.0155, 0.0124, 0.0141, 0.0112, 0.0145, 0.0113],
       device='cuda:2'), out_proj_covar=tensor([0.0003, 0.0002, 0.0003, 0.0002, 0.0002, 0.0002, 0.0003, 0.0002],
       device='cuda:2')
2022-12-23 12:34:53,506 WARNING [train.py:1060] (2/4) Exclude cut with ID 4295-39940-0007-92567-0 from training. Duration: 21.54
2022-12-23 12:34:57,838 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 2.687e+02 4.531e+02 5.497e+02 7.224e+02 2.180e+03, threshold=1.099e+03, percent-clipped=3.0
2022-12-23 12:34:57,910 WARNING [train.py:1060] (2/4) Exclude cut with ID 4964-30587-0040-44509-0_sp1.1 from training. Duration: 20.5318125
2022-12-23 12:35:08,036 WARNING [train.py:1060] (2/4) Exclude cut with ID 7255-291500-0012-134311-0_sp0.9 from training. Duration: 21.9333125
2022-12-23 12:35:26,412 INFO [train.py:894] (2/4) Epoch 18, batch 2250, loss[loss=0.2072, simple_loss=0.2865, pruned_loss=0.06395, over 18465.00 frames. ], tot_loss[loss=0.1936, simple_loss=0.2738, pruned_loss=0.05674, over 3714012.07 frames. ], batch size: 64, lr: 6.37e-03, grad_scale: 16.0
2022-12-23 12:35:55,300 WARNING [train.py:1060] (2/4) Exclude cut with ID 8631-249866-0025-130151-0_sp0.9 from training. Duration: 21.7944375
2022-12-23 12:36:08,063 WARNING [train.py:1060] (2/4) Exclude cut with ID 8291-282929-0002-12989-0_sp0.9 from training. Duration: 22.4666875
2022-12-23 12:36:13,800 WARNING [train.py:1060] (2/4) Exclude cut with ID 6121-9014-0076-24124-0 from training. Duration: 21.635
2022-12-23 12:36:20,877 WARNING [train.py:1060] (2/4) Exclude cut with ID 6121-9014-0076-24124-0_sp0.9 from training. Duration: 24.038875
2022-12-23 12:36:40,755 INFO [train.py:894] (2/4) Epoch 18, batch 2300, loss[loss=0.1615, simple_loss=0.238, pruned_loss=0.04253, over 18550.00 frames. ], tot_loss[loss=0.1928, simple_loss=0.2729, pruned_loss=0.05634, over 3712845.14 frames. ], batch size: 44, lr: 6.37e-03, grad_scale: 16.0
2022-12-23 12:36:55,239 INFO [zipformer.py:660] (2/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=61914.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 12:37:02,392 WARNING [train.py:1060] (2/4) Exclude cut with ID 7255-291500-0009-134308-0_sp1.1 from training. Duration: 21.786375
2022-12-23 12:37:14,229 WARNING [train.py:1060] (2/4) Exclude cut with ID 8291-282929-0002-12989-0 from training. Duration: 20.22
2022-12-23 12:37:26,933 INFO [scaling.py:679] (2/4) Whitening: num_groups=8, num_channels=192, metric=1.57 vs. limit=2.0
2022-12-23 12:37:28,974 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 2.260e+02 4.610e+02 5.551e+02 6.861e+02 1.067e+03, threshold=1.110e+03, percent-clipped=0.0
2022-12-23 12:37:58,690 INFO [train.py:894] (2/4) Epoch 18, batch 2350, loss[loss=0.1849, simple_loss=0.2535, pruned_loss=0.05817, over 18384.00 frames. ], tot_loss[loss=0.1927, simple_loss=0.2728, pruned_loss=0.05625, over 3712627.24 frames. ], batch size: 46, lr: 6.37e-03, grad_scale: 16.0
2022-12-23 12:38:30,614 INFO [zipformer.py:660] (2/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=61975.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 12:39:15,660 WARNING [train.py:1060] (2/4) Exclude cut with ID 6951-79737-0043-132310-0 from training. Duration: 25.285
2022-12-23 12:39:18,286 INFO [train.py:894] (2/4) Epoch 18, batch 2400, loss[loss=0.2339, simple_loss=0.3118, pruned_loss=0.07801, over 18611.00 frames. ], tot_loss[loss=0.1932, simple_loss=0.2729, pruned_loss=0.05674, over 3712422.46 frames. ], batch size: 69, lr: 6.37e-03, grad_scale: 16.0
2022-12-23 12:40:05,027 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 2.629e+02 4.370e+02 5.333e+02 6.634e+02 1.277e+03, threshold=1.067e+03, percent-clipped=2.0
2022-12-23 12:40:16,413 INFO [scaling.py:679] (2/4) Whitening: num_groups=8, num_channels=96, metric=1.22 vs. limit=2.0
2022-12-23 12:40:20,617 WARNING [train.py:1060] (2/4) Exclude cut with ID 811-130148-0001-63453-0_sp0.9 from training. Duration: 20.861125
2022-12-23 12:40:24,586 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([2.4371, 2.1676, 1.7188, 0.7639, 1.7286, 1.9280, 1.7164, 1.9156],
       device='cuda:2'), covar=tensor([0.0604, 0.0504, 0.1127, 0.1580, 0.1132, 0.1419, 0.1526, 0.0715],
       device='cuda:2'), in_proj_covar=tensor([0.0171, 0.0183, 0.0207, 0.0191, 0.0211, 0.0200, 0.0214, 0.0202],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-23 12:40:34,667 INFO [train.py:894] (2/4) Epoch 18, batch 2450, loss[loss=0.1495, simple_loss=0.23, pruned_loss=0.03452, over 18698.00 frames. ], tot_loss[loss=0.1924, simple_loss=0.2725, pruned_loss=0.05613, over 3712256.52 frames. ], batch size: 46, lr: 6.36e-03, grad_scale: 16.0
2022-12-23 12:40:45,009 WARNING [train.py:1060] (2/4) Exclude cut with ID 6010-56788-0055-90261-0 from training. Duration: 20.88
2022-12-23 12:41:11,115 INFO [scaling.py:679] (2/4) Whitening: num_groups=8, num_channels=192, metric=1.80 vs. limit=2.0
2022-12-23 12:41:15,901 WARNING [train.py:1060] (2/4) Exclude cut with ID 7357-94126-0045-15876-0_sp0.9 from training. Duration: 23.4166875
2022-12-23 12:41:24,157 INFO [zipformer.py:660] (2/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=62088.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 12:41:41,124 INFO [zipformer.py:660] (2/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=62099.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 12:41:50,015 INFO [train.py:894] (2/4) Epoch 18, batch 2500, loss[loss=0.1881, simple_loss=0.2737, pruned_loss=0.05128, over 18658.00 frames. ], tot_loss[loss=0.1928, simple_loss=0.273, pruned_loss=0.05631, over 3713206.40 frames. ], batch size: 69, lr: 6.36e-03, grad_scale: 16.0
2022-12-23 12:42:15,777 INFO [zipformer.py:660] (2/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=62122.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 12:42:33,516 WARNING [train.py:1060] (2/4) Exclude cut with ID 7255-291500-0006-134305-0 from training. Duration: 21.24
2022-12-23 12:42:34,885 WARNING [train.py:1060] (2/4) Exclude cut with ID 6533-399-0047-104881-0_sp0.9 from training. Duration: 23.9055625
2022-12-23 12:42:36,305 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 3.067e+02 4.462e+02 5.167e+02 6.894e+02 1.650e+03, threshold=1.033e+03, percent-clipped=6.0
2022-12-23 12:43:04,727 INFO [scaling.py:679] (2/4) Whitening: num_groups=8, num_channels=96, metric=1.23 vs. limit=2.0
2022-12-23 12:43:05,400 INFO [train.py:894] (2/4) Epoch 18, batch 2550, loss[loss=0.2153, simple_loss=0.2964, pruned_loss=0.0671, over 18663.00 frames. ], tot_loss[loss=0.1937, simple_loss=0.2737, pruned_loss=0.05683, over 3712781.49 frames. ], batch size: 60, lr: 6.36e-03, grad_scale: 16.0
2022-12-23 12:43:09,696 WARNING [train.py:1060] (2/4) Exclude cut with ID 6758-72288-0033-108368-0_sp0.9 from training. Duration: 25.988875
2022-12-23 12:43:18,032 WARNING [train.py:1060] (2/4) Exclude cut with ID 7255-291500-0001-134300-0_sp0.9 from training. Duration: 20.67225
2022-12-23 12:43:36,360 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([2.6565, 2.5964, 1.8026, 2.9744, 2.9376, 2.4088, 3.8852, 2.7158],
       device='cuda:2'), covar=tensor([0.0789, 0.1637, 0.2663, 0.1901, 0.1529, 0.0837, 0.0749, 0.1086],
       device='cuda:2'), in_proj_covar=tensor([0.0176, 0.0205, 0.0246, 0.0288, 0.0236, 0.0189, 0.0209, 0.0201],
       device='cuda:2'), out_proj_covar=tensor([0.0001, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-23 12:43:48,363 INFO [zipformer.py:660] (2/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=62183.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 12:43:56,695 INFO [scaling.py:679] (2/4) Whitening: num_groups=8, num_channels=96, metric=1.53 vs. limit=2.0
2022-12-23 12:44:06,137 WARNING [train.py:1060] (2/4) Exclude cut with ID 3488-85273-0038-41224-0 from training. Duration: 20.34
2022-12-23 12:44:21,578 INFO [train.py:894] (2/4) Epoch 18, batch 2600, loss[loss=0.1685, simple_loss=0.2401, pruned_loss=0.04843, over 18438.00 frames. ], tot_loss[loss=0.1922, simple_loss=0.2722, pruned_loss=0.05607, over 3712201.17 frames. ], batch size: 42, lr: 6.36e-03, grad_scale: 16.0
2022-12-23 12:44:24,834 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.4552, 1.2138, 1.7631, 3.3445, 2.4012, 2.5672, 0.6968, 2.2129],
       device='cuda:2'), covar=tensor([0.2037, 0.1888, 0.1668, 0.0642, 0.1062, 0.1186, 0.2438, 0.1193],
       device='cuda:2'), in_proj_covar=tensor([0.0103, 0.0117, 0.0134, 0.0147, 0.0107, 0.0140, 0.0130, 0.0112],
       device='cuda:2'), out_proj_covar=tensor([0.0003, 0.0003, 0.0004, 0.0004, 0.0003, 0.0004, 0.0003, 0.0003],
       device='cuda:2')
2022-12-23 12:44:53,219 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.7042, 1.7644, 1.8039, 1.6851, 1.2184, 3.8125, 1.6710, 2.1508],
       device='cuda:2'), covar=tensor([0.2962, 0.1881, 0.1775, 0.1957, 0.1420, 0.0162, 0.1503, 0.0813],
       device='cuda:2'), in_proj_covar=tensor([0.0132, 0.0116, 0.0125, 0.0120, 0.0102, 0.0097, 0.0092, 0.0089],
       device='cuda:2'), out_proj_covar=tensor([0.0005, 0.0005, 0.0005, 0.0005, 0.0004, 0.0003, 0.0004, 0.0004],
       device='cuda:2')
2022-12-23 12:45:07,937 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 3.177e+02 4.427e+02 5.358e+02 6.353e+02 1.453e+03, threshold=1.072e+03, percent-clipped=5.0
2022-12-23 12:45:15,288 WARNING [train.py:1060] (2/4) Exclude cut with ID 7357-94126-0026-15857-0_sp0.9 from training. Duration: 25.061125
2022-12-23 12:45:27,804 WARNING [train.py:1060] (2/4) Exclude cut with ID 3033-130750-0096-55598-0 from training. Duration: 0.83
2022-12-23 12:45:37,358 INFO [train.py:894] (2/4) Epoch 18, batch 2650, loss[loss=0.1817, simple_loss=0.2711, pruned_loss=0.04618, over 18591.00 frames. ], tot_loss[loss=0.1919, simple_loss=0.272, pruned_loss=0.05588, over 3713432.58 frames. ], batch size: 56, lr: 6.35e-03, grad_scale: 16.0
2022-12-23 12:45:48,441 INFO [scaling.py:679] (2/4) Whitening: num_groups=8, num_channels=96, metric=1.32 vs. limit=2.0
2022-12-23 12:45:52,001 WARNING [train.py:1060] (2/4) Exclude cut with ID 3488-85273-0017-41203-0 from training. Duration: 24.73
2022-12-23 12:46:00,184 INFO [zipformer.py:660] (2/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=62270.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 12:46:04,587 INFO [zipformer.py:660] (2/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=62273.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 12:46:07,758 WARNING [train.py:1060] (2/4) Exclude cut with ID 7255-291500-0009-134308-0 from training. Duration: 23.965
2022-12-23 12:46:15,103 WARNING [train.py:1060] (2/4) Exclude cut with ID 6978-92210-0030-146996-0_sp0.9 from training. Duration: 22.088875
2022-12-23 12:46:31,021 WARNING [train.py:1060] (2/4) Exclude cut with ID 7255-291500-0006-134305-0_sp0.9 from training. Duration: 23.6
2022-12-23 12:46:52,865 INFO [train.py:894] (2/4) Epoch 18, batch 2700, loss[loss=0.1991, simple_loss=0.287, pruned_loss=0.05562, over 18468.00 frames. ], tot_loss[loss=0.1916, simple_loss=0.2717, pruned_loss=0.05569, over 3713418.45 frames. ], batch size: 54, lr: 6.35e-03, grad_scale: 16.0
2022-12-23 12:47:36,962 INFO [zipformer.py:660] (2/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=62334.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 12:47:39,999 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 2.427e+02 4.380e+02 5.434e+02 6.774e+02 1.412e+03, threshold=1.087e+03, percent-clipped=1.0
2022-12-23 12:47:40,939 INFO [scaling.py:679] (2/4) Whitening: num_groups=8, num_channels=192, metric=1.93 vs. limit=2.0
2022-12-23 12:47:54,257 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([2.3707, 1.4901, 0.6629, 1.8683, 2.4449, 1.7990, 2.0102, 1.9918],
       device='cuda:2'), covar=tensor([0.1540, 0.2209, 0.2513, 0.1425, 0.1757, 0.1603, 0.1468, 0.1723],
       device='cuda:2'), in_proj_covar=tensor([0.0094, 0.0097, 0.0117, 0.0095, 0.0116, 0.0091, 0.0099, 0.0094],
       device='cuda:2'), out_proj_covar=tensor([0.0003, 0.0003, 0.0004, 0.0003, 0.0004, 0.0003, 0.0003, 0.0003],
       device='cuda:2')
2022-12-23 12:47:55,834 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([2.1729, 2.1812, 1.5516, 2.3996, 2.3066, 1.8768, 2.9070, 2.2132],
       device='cuda:2'), covar=tensor([0.0872, 0.1608, 0.2732, 0.1785, 0.1692, 0.0963, 0.0926, 0.1181],
       device='cuda:2'), in_proj_covar=tensor([0.0176, 0.0204, 0.0245, 0.0287, 0.0236, 0.0188, 0.0207, 0.0201],
       device='cuda:2'), out_proj_covar=tensor([0.0001, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-23 12:48:08,465 INFO [train.py:894] (2/4) Epoch 18, batch 2750, loss[loss=0.1709, simple_loss=0.2626, pruned_loss=0.03965, over 18707.00 frames. ], tot_loss[loss=0.1915, simple_loss=0.2716, pruned_loss=0.05573, over 3714016.19 frames. ], batch size: 54, lr: 6.35e-03, grad_scale: 16.0
2022-12-23 12:48:09,846 WARNING [train.py:1060] (2/4) Exclude cut with ID 8291-282929-0024-13011-0 from training. Duration: 23.795
2022-12-23 12:48:26,704 WARNING [train.py:1060] (2/4) Exclude cut with ID 8631-249866-0030-130156-0_sp1.1 from training. Duration: 21.5409375
2022-12-23 12:48:28,756 WARNING [train.py:1060] (2/4) Exclude cut with ID 6978-92210-0019-146985-0_sp0.9 from training. Duration: 24.97775
2022-12-23 12:48:37,946 WARNING [train.py:1060] (2/4) Exclude cut with ID 1085-156170-0017-128270-0_sp0.9 from training. Duration: 23.3444375
2022-12-23 12:49:00,718 INFO [zipformer.py:660] (2/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=62388.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 12:49:05,377 INFO [zipformer.py:660] (2/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=62391.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 12:49:06,686 WARNING [train.py:1060] (2/4) Exclude cut with ID 6010-56788-0055-90261-0_sp0.9 from training. Duration: 23.2
2022-12-23 12:49:12,406 WARNING [train.py:1060] (2/4) Exclude cut with ID 5653-46179-0060-117930-0_sp0.9 from training. Duration: 21.17225
2022-12-23 12:49:17,546 INFO [zipformer.py:660] (2/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=62399.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 12:49:25,893 INFO [train.py:894] (2/4) Epoch 18, batch 2800, loss[loss=0.2197, simple_loss=0.3027, pruned_loss=0.06839, over 18504.00 frames. ], tot_loss[loss=0.1915, simple_loss=0.2714, pruned_loss=0.05584, over 3714482.39 frames. ], batch size: 52, lr: 6.35e-03, grad_scale: 16.0
2022-12-23 12:49:33,327 WARNING [train.py:1060] (2/4) Exclude cut with ID 7255-291500-0008-134307-0_sp0.9 from training. Duration: 24.6555625
2022-12-23 12:50:03,970 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([2.5868, 2.2946, 1.6390, 2.5336, 1.9968, 2.0655, 2.0696, 2.6100],
       device='cuda:2'), covar=tensor([0.2050, 0.2916, 0.1879, 0.2621, 0.3554, 0.1010, 0.2909, 0.0874],
       device='cuda:2'), in_proj_covar=tensor([0.0293, 0.0285, 0.0242, 0.0350, 0.0267, 0.0225, 0.0284, 0.0210],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0003, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-23 12:50:14,069 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 2.980e+02 4.461e+02 5.405e+02 6.441e+02 1.681e+03, threshold=1.081e+03, percent-clipped=2.0
2022-12-23 12:50:14,262 INFO [zipformer.py:660] (2/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=62436.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 12:50:24,483 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.7153, 1.6628, 1.7769, 1.7320, 1.2516, 3.7825, 1.7254, 2.1540],
       device='cuda:2'), covar=tensor([0.3302, 0.1959, 0.1874, 0.2043, 0.1435, 0.0178, 0.1582, 0.0822],
       device='cuda:2'), in_proj_covar=tensor([0.0134, 0.0117, 0.0126, 0.0121, 0.0103, 0.0098, 0.0093, 0.0089],
       device='cuda:2'), out_proj_covar=tensor([0.0005, 0.0005, 0.0005, 0.0005, 0.0004, 0.0003, 0.0004, 0.0004],
       device='cuda:2')
2022-12-23 12:50:25,669 WARNING [train.py:1060] (2/4) Exclude cut with ID 3488-65654-0031-41259-0 from training. Duration: 20.44
2022-12-23 12:50:30,098 INFO [zipformer.py:660] (2/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=62447.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 12:50:37,705 INFO [zipformer.py:660] (2/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=62452.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 12:50:42,216 INFO [train.py:894] (2/4) Epoch 18, batch 2850, loss[loss=0.1811, simple_loss=0.2659, pruned_loss=0.04814, over 18723.00 frames. ], tot_loss[loss=0.1907, simple_loss=0.271, pruned_loss=0.05519, over 3714588.75 frames. ], batch size: 69, lr: 6.34e-03, grad_scale: 16.0
2022-12-23 12:50:42,242 WARNING [train.py:1060] (2/4) Exclude cut with ID 6951-79737-0018-132285-0_sp0.9 from training. Duration: 23.45
2022-12-23 12:51:11,495 WARNING [train.py:1060] (2/4) Exclude cut with ID 6945-60535-0076-12784-0_sp0.9 from training. Duration: 20.52225
2022-12-23 12:51:17,465 INFO [zipformer.py:660] (2/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=62478.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 12:51:20,444 WARNING [train.py:1060] (2/4) Exclude cut with ID 7255-291500-0008-134307-0 from training. Duration: 22.19
2022-12-23 12:51:26,559 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([2.1439, 1.9745, 1.5493, 1.6019, 1.9956, 2.3204, 2.2150, 1.9837],
       device='cuda:2'), covar=tensor([0.0255, 0.0266, 0.0452, 0.0274, 0.0266, 0.0287, 0.0253, 0.0289],
       device='cuda:2'), in_proj_covar=tensor([0.0093, 0.0125, 0.0150, 0.0123, 0.0115, 0.0118, 0.0095, 0.0125],
       device='cuda:2'), out_proj_covar=tensor([7.4567e-05, 1.0021e-04, 1.2524e-04, 9.8437e-05, 9.4035e-05, 9.1232e-05,
        7.5232e-05, 9.9249e-05], device='cuda:2')
2022-12-23 12:51:29,044 WARNING [train.py:1060] (2/4) Exclude cut with ID 7357-94126-0014-15845-0_sp1.1 from training. Duration: 25.3818125
2022-12-23 12:51:29,380 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([2.2962, 2.1372, 1.8649, 1.1411, 2.5395, 2.3751, 2.1938, 1.5757],
       device='cuda:2'), covar=tensor([0.0349, 0.0385, 0.0446, 0.0699, 0.0255, 0.0320, 0.0371, 0.0906],
       device='cuda:2'), in_proj_covar=tensor([0.0124, 0.0125, 0.0127, 0.0120, 0.0097, 0.0121, 0.0134, 0.0155],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0001, 0.0001, 0.0001, 0.0002, 0.0002],
       device='cuda:2')
2022-12-23 12:51:45,483 WARNING [train.py:1060] (2/4) Exclude cut with ID 6951-79737-0043-132310-0_sp0.9 from training. Duration: 28.0944375
2022-12-23 12:51:51,422 WARNING [train.py:1060] (2/4) Exclude cut with ID 2195-150901-0045-59933-0_sp0.9 from training. Duration: 22.9444375
2022-12-23 12:51:57,906 INFO [train.py:894] (2/4) Epoch 18, batch 2900, loss[loss=0.2181, simple_loss=0.2921, pruned_loss=0.07203, over 18667.00 frames. ], tot_loss[loss=0.1913, simple_loss=0.2712, pruned_loss=0.05575, over 3713973.06 frames. ], batch size: 69, lr: 6.34e-03, grad_scale: 16.0
2022-12-23 12:52:00,828 WARNING [train.py:1060] (2/4) Exclude cut with ID 8291-282929-0024-13011-0_sp1.1 from training. Duration: 21.6318125
2022-12-23 12:52:18,911 WARNING [train.py:1060] (2/4) Exclude cut with ID 8631-249866-0030-130156-0 from training. Duration: 23.695
2022-12-23 12:52:44,255 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 2.769e+02 4.491e+02 5.311e+02 7.328e+02 1.636e+03, threshold=1.062e+03, percent-clipped=8.0
2022-12-23 12:52:45,915 WARNING [train.py:1060] (2/4) Exclude cut with ID 7699-105389-0094-26379-0 from training. Duration: 23.955
2022-12-23 12:52:58,216 INFO [zipformer.py:660] (2/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=62545.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 12:53:13,512 INFO [train.py:894] (2/4) Epoch 18, batch 2950, loss[loss=0.2003, simple_loss=0.2776, pruned_loss=0.0615, over 18523.00 frames. ], tot_loss[loss=0.1911, simple_loss=0.2706, pruned_loss=0.05583, over 3713767.04 frames. ], batch size: 58, lr: 6.34e-03, grad_scale: 16.0
2022-12-23 12:53:19,226 WARNING [train.py:1060] (2/4) Exclude cut with ID 8291-282929-0024-13011-0_sp0.9 from training. Duration: 26.438875
2022-12-23 12:53:37,037 INFO [zipformer.py:660] (2/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=62570.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 12:53:58,837 INFO [zipformer.py:660] (2/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=62585.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 12:54:04,344 WARNING [train.py:1060] (2/4) Exclude cut with ID 7699-105389-0021-26306-0_sp0.9 from training. Duration: 21.2444375
2022-12-23 12:54:04,366 WARNING [train.py:1060] (2/4) Exclude cut with ID 7357-94126-0014-15845-0_sp0.9 from training. Duration: 31.02225
2022-12-23 12:54:14,569 WARNING [train.py:1060] (2/4) Exclude cut with ID 432-122774-0017-62487-0 from training. Duration: 22.395
2022-12-23 12:54:29,660 INFO [train.py:894] (2/4) Epoch 18, batch 3000, loss[loss=0.175, simple_loss=0.2532, pruned_loss=0.04837, over 18674.00 frames. ], tot_loss[loss=0.1906, simple_loss=0.2702, pruned_loss=0.05553, over 3714097.14 frames. ], batch size: 46, lr: 6.34e-03, grad_scale: 16.0
2022-12-23 12:54:29,660 INFO [train.py:919] (2/4) Computing validation loss
2022-12-23 12:54:40,507 INFO [train.py:928] (2/4) Epoch 18, validation: loss=0.164, simple_loss=0.2628, pruned_loss=0.0326, over 944034.00 frames. 
2022-12-23 12:54:40,508 INFO [train.py:929] (2/4) Maximum memory allocated so far is 24680MB
2022-12-23 12:54:42,669 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.6067, 1.4523, 1.3348, 1.5122, 1.7658, 1.6634, 1.7432, 1.1939],
       device='cuda:2'), covar=tensor([0.0303, 0.0233, 0.0469, 0.0195, 0.0206, 0.0356, 0.0244, 0.0286],
       device='cuda:2'), in_proj_covar=tensor([0.0093, 0.0125, 0.0150, 0.0123, 0.0115, 0.0118, 0.0096, 0.0125],
       device='cuda:2'), out_proj_covar=tensor([7.4292e-05, 9.9911e-05, 1.2494e-04, 9.8545e-05, 9.3927e-05, 9.1377e-05,
        7.5410e-05, 9.9519e-05], device='cuda:2')
2022-12-23 12:54:42,680 INFO [zipformer.py:660] (2/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=62606.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 12:54:43,706 WARNING [train.py:1060] (2/4) Exclude cut with ID 7357-94126-0045-15876-0 from training. Duration: 21.075
2022-12-23 12:54:48,078 WARNING [train.py:1060] (2/4) Exclude cut with ID 6482-98857-0025-147532-0_sp0.9 from training. Duration: 20.0055625
2022-12-23 12:54:48,089 WARNING [train.py:1060] (2/4) Exclude cut with ID 6951-79737-0037-132304-0_sp0.9 from training. Duration: 22.05
2022-12-23 12:54:48,100 WARNING [train.py:1060] (2/4) Exclude cut with ID 7255-291500-0003-134302-0 from training. Duration: 26.8349375
2022-12-23 12:54:51,423 WARNING [train.py:1060] (2/4) Exclude cut with ID 7357-94126-0009-15840-0_sp1.1 from training. Duration: 22.1090625
2022-12-23 12:54:59,128 WARNING [train.py:1060] (2/4) Exclude cut with ID 7699-105389-0094-26379-0_sp0.9 from training. Duration: 26.6166875
2022-12-23 12:55:00,658 INFO [zipformer.py:660] (2/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=62618.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 12:55:15,886 WARNING [train.py:1060] (2/4) Exclude cut with ID 2046-178027-0000-53705-0_sp0.9 from training. Duration: 20.3055625
2022-12-23 12:55:18,384 INFO [zipformer.py:660] (2/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=62629.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 12:55:28,400 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 3.034e+02 4.348e+02 5.116e+02 6.564e+02 1.235e+03, threshold=1.023e+03, percent-clipped=2.0
2022-12-23 12:55:40,192 WARNING [train.py:1060] (2/4) Exclude cut with ID 7205-50138-0008-5373-0_sp0.9 from training. Duration: 20.7
2022-12-23 12:55:44,099 INFO [zipformer.py:660] (2/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=62646.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 12:55:57,533 INFO [train.py:894] (2/4) Epoch 18, batch 3050, loss[loss=0.2155, simple_loss=0.2914, pruned_loss=0.06976, over 18671.00 frames. ], tot_loss[loss=0.1896, simple_loss=0.2694, pruned_loss=0.05491, over 3714078.37 frames. ], batch size: 69, lr: 6.33e-03, grad_scale: 16.0
2022-12-23 12:56:14,089 INFO [zipformer.py:660] (2/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=62666.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 12:56:22,836 WARNING [train.py:1060] (2/4) Exclude cut with ID 6978-92210-0019-146985-0 from training. Duration: 22.48
2022-12-23 12:56:39,817 WARNING [train.py:1060] (2/4) Exclude cut with ID 7255-291500-0003-134302-0_sp0.9 from training. Duration: 29.816625
2022-12-23 12:56:58,135 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([2.1374, 1.4755, 1.7759, 1.7784, 2.1337, 2.0994, 1.9270, 1.6698],
       device='cuda:2'), covar=tensor([0.2090, 0.3182, 0.2533, 0.2778, 0.1905, 0.0944, 0.3055, 0.1267],
       device='cuda:2'), in_proj_covar=tensor([0.0268, 0.0300, 0.0279, 0.0313, 0.0305, 0.0252, 0.0340, 0.0238],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-23 12:56:59,050 WARNING [train.py:1060] (2/4) Exclude cut with ID 7255-291500-0005-134304-0_sp1.1 from training. Duration: 22.7590625
2022-12-23 12:57:05,075 WARNING [train.py:1060] (2/4) Exclude cut with ID 7357-94126-0026-15857-0 from training. Duration: 22.555
2022-12-23 12:57:09,771 INFO [scaling.py:679] (2/4) Whitening: num_groups=8, num_channels=192, metric=1.89 vs. limit=2.0
2022-12-23 12:57:13,176 INFO [train.py:894] (2/4) Epoch 18, batch 3100, loss[loss=0.1862, simple_loss=0.2737, pruned_loss=0.04935, over 18722.00 frames. ], tot_loss[loss=0.1905, simple_loss=0.2704, pruned_loss=0.05532, over 3715318.61 frames. ], batch size: 50, lr: 6.33e-03, grad_scale: 8.0
2022-12-23 12:57:26,222 WARNING [train.py:1060] (2/4) Exclude cut with ID 1250-135782-0005-25975-0_sp0.9 from training. Duration: 21.688875
2022-12-23 12:57:46,917 INFO [zipformer.py:660] (2/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=62727.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 12:58:00,959 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 2.728e+02 4.372e+02 5.368e+02 7.060e+02 1.480e+03, threshold=1.074e+03, percent-clipped=5.0
2022-12-23 12:58:04,372 WARNING [train.py:1060] (2/4) Exclude cut with ID 3488-85273-0038-41224-0_sp0.9 from training. Duration: 22.6
2022-12-23 12:58:16,139 INFO [zipformer.py:660] (2/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=62747.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 12:58:28,827 INFO [train.py:894] (2/4) Epoch 18, batch 3150, loss[loss=0.1676, simple_loss=0.2454, pruned_loss=0.04493, over 18530.00 frames. ], tot_loss[loss=0.1899, simple_loss=0.2697, pruned_loss=0.05502, over 3714355.09 frames. ], batch size: 47, lr: 6.33e-03, grad_scale: 8.0
2022-12-23 12:58:38,976 WARNING [train.py:1060] (2/4) Exclude cut with ID 7357-94126-0009-15840-0 from training. Duration: 24.32
2022-12-23 12:58:56,073 INFO [zipformer.py:660] (2/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=62773.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 12:59:03,039 INFO [zipformer.py:660] (2/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=62778.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 12:59:06,381 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([0.1295, 0.9841, 1.1335, 0.5368, 0.5873, 1.1563, 1.1260, 1.1233],
       device='cuda:2'), covar=tensor([0.0688, 0.0312, 0.0315, 0.0353, 0.0398, 0.0460, 0.0239, 0.0537],
       device='cuda:2'), in_proj_covar=tensor([0.0148, 0.0167, 0.0124, 0.0136, 0.0146, 0.0140, 0.0158, 0.0167],
       device='cuda:2'), out_proj_covar=tensor([1.1429e-04, 1.2947e-04, 9.4841e-05, 1.0183e-04, 1.1049e-04, 1.0842e-04,
        1.2307e-04, 1.2856e-04], device='cuda:2')
2022-12-23 12:59:34,699 WARNING [train.py:1060] (2/4) Exclude cut with ID 8291-276745-0093-13116-0_sp0.9 from training. Duration: 21.061125
2022-12-23 12:59:44,255 INFO [train.py:894] (2/4) Epoch 18, batch 3200, loss[loss=0.1875, simple_loss=0.276, pruned_loss=0.04945, over 18576.00 frames. ], tot_loss[loss=0.1896, simple_loss=0.2695, pruned_loss=0.05486, over 3714410.29 frames. ], batch size: 56, lr: 6.33e-03, grad_scale: 8.0
2022-12-23 12:59:48,828 WARNING [train.py:1060] (2/4) Exclude cut with ID 7357-94126-0024-15855-0_sp0.9 from training. Duration: 20.32225
2022-12-23 13:00:02,512 WARNING [train.py:1060] (2/4) Exclude cut with ID 3033-130750-0096-55598-0_sp1.1 from training. Duration: 0.7545625
2022-12-23 13:00:16,786 WARNING [train.py:1060] (2/4) Exclude cut with ID 4295-39940-0007-92567-0_sp0.9 from training. Duration: 23.9333125
2022-12-23 13:00:16,886 INFO [zipformer.py:660] (2/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=62826.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 13:00:29,767 INFO [zipformer.py:660] (2/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=62834.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 13:00:33,649 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 2.693e+02 4.436e+02 5.531e+02 6.977e+02 1.575e+03, threshold=1.106e+03, percent-clipped=1.0
2022-12-23 13:00:49,243 WARNING [train.py:1060] (2/4) Exclude cut with ID 7255-291500-0008-134307-0_sp1.1 from training. Duration: 20.17275
2022-12-23 13:00:54,967 WARNING [train.py:1060] (2/4) Exclude cut with ID 6978-92210-0019-146985-0_sp1.1 from training. Duration: 20.436375
2022-12-23 13:01:00,868 INFO [train.py:894] (2/4) Epoch 18, batch 3250, loss[loss=0.2193, simple_loss=0.2962, pruned_loss=0.07124, over 18476.00 frames. ], tot_loss[loss=0.1906, simple_loss=0.2704, pruned_loss=0.05537, over 3715269.02 frames. ], batch size: 54, lr: 6.32e-03, grad_scale: 8.0
2022-12-23 13:01:01,887 INFO [scaling.py:679] (2/4) Whitening: num_groups=8, num_channels=96, metric=1.33 vs. limit=2.0
2022-12-23 13:01:19,677 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([2.1716, 1.5180, 1.8920, 1.8778, 2.1425, 2.1353, 2.0283, 1.7315],
       device='cuda:2'), covar=tensor([0.2081, 0.3165, 0.2325, 0.2659, 0.1898, 0.0920, 0.2829, 0.1263],
       device='cuda:2'), in_proj_covar=tensor([0.0268, 0.0299, 0.0277, 0.0313, 0.0304, 0.0250, 0.0339, 0.0238],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-23 13:02:10,870 INFO [zipformer.py:660] (2/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=62901.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 13:02:16,425 INFO [train.py:894] (2/4) Epoch 18, batch 3300, loss[loss=0.2187, simple_loss=0.3052, pruned_loss=0.06607, over 18684.00 frames. ], tot_loss[loss=0.1906, simple_loss=0.2708, pruned_loss=0.05525, over 3715696.43 frames. ], batch size: 69, lr: 6.32e-03, grad_scale: 8.0
2022-12-23 13:02:17,782 WARNING [train.py:1060] (2/4) Exclude cut with ID 4234-40345-0022-142709-0_sp0.9 from training. Duration: 23.1055625
2022-12-23 13:02:19,793 WARNING [train.py:1060] (2/4) Exclude cut with ID 8291-282929-0007-12994-0_sp1.1 from training. Duration: 23.5
2022-12-23 13:02:32,424 WARNING [train.py:1060] (2/4) Exclude cut with ID 7255-291500-0009-134308-0_sp0.9 from training. Duration: 26.62775
2022-12-23 13:02:41,697 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.3448, 1.0742, 1.4027, 2.1852, 1.6584, 2.1998, 0.7331, 1.6606],
       device='cuda:2'), covar=tensor([0.1770, 0.1695, 0.1280, 0.0804, 0.1129, 0.0992, 0.1876, 0.1187],
       device='cuda:2'), in_proj_covar=tensor([0.0101, 0.0115, 0.0133, 0.0146, 0.0106, 0.0139, 0.0130, 0.0111],
       device='cuda:2'), out_proj_covar=tensor([0.0003, 0.0003, 0.0004, 0.0004, 0.0003, 0.0004, 0.0003, 0.0003],
       device='cuda:2')
2022-12-23 13:02:45,638 WARNING [train.py:1060] (2/4) Exclude cut with ID 6951-79737-0018-132285-0 from training. Duration: 21.105
2022-12-23 13:02:48,682 WARNING [train.py:1060] (2/4) Exclude cut with ID 4511-76322-0006-80011-0_sp0.9 from training. Duration: 24.411125
2022-12-23 13:02:53,568 INFO [zipformer.py:660] (2/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=62929.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 13:03:04,359 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.6584, 1.0699, 0.9793, 1.3844, 2.1055, 1.3645, 1.2623, 1.5837],
       device='cuda:2'), covar=tensor([0.1885, 0.2642, 0.2257, 0.1656, 0.1864, 0.1741, 0.1821, 0.1947],
       device='cuda:2'), in_proj_covar=tensor([0.0093, 0.0097, 0.0115, 0.0094, 0.0116, 0.0090, 0.0098, 0.0093],
       device='cuda:2'), out_proj_covar=tensor([0.0003, 0.0003, 0.0004, 0.0003, 0.0004, 0.0003, 0.0003, 0.0003],
       device='cuda:2')
2022-12-23 13:03:05,528 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 2.669e+02 4.438e+02 5.623e+02 6.897e+02 1.767e+03, threshold=1.125e+03, percent-clipped=3.0
2022-12-23 13:03:11,436 INFO [zipformer.py:660] (2/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=62941.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 13:03:15,311 WARNING [train.py:1060] (2/4) Exclude cut with ID 6758-72288-0033-108368-0_sp1.1 from training. Duration: 21.263625
2022-12-23 13:03:31,687 INFO [train.py:894] (2/4) Epoch 18, batch 3350, loss[loss=0.1647, simple_loss=0.2452, pruned_loss=0.04211, over 18424.00 frames. ], tot_loss[loss=0.1907, simple_loss=0.2709, pruned_loss=0.05527, over 3716086.49 frames. ], batch size: 48, lr: 6.32e-03, grad_scale: 8.0
2022-12-23 13:03:48,516 WARNING [train.py:1060] (2/4) Exclude cut with ID 4234-40345-0022-142709-0 from training. Duration: 20.795
2022-12-23 13:03:59,213 WARNING [train.py:1060] (2/4) Exclude cut with ID 7357-94126-0021-15852-0 from training. Duration: 24.76
2022-12-23 13:03:59,230 WARNING [train.py:1060] (2/4) Exclude cut with ID 3867-173237-0077-144769-0_sp0.9 from training. Duration: 22.25
2022-12-23 13:04:05,348 INFO [zipformer.py:660] (2/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=62977.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 13:04:23,855 WARNING [train.py:1060] (2/4) Exclude cut with ID 7357-94126-0026-15857-0_sp1.1 from training. Duration: 20.5045625
2022-12-23 13:04:48,413 INFO [train.py:894] (2/4) Epoch 18, batch 3400, loss[loss=0.1784, simple_loss=0.2604, pruned_loss=0.04819, over 18584.00 frames. ], tot_loss[loss=0.1899, simple_loss=0.27, pruned_loss=0.0549, over 3715452.60 frames. ], batch size: 78, lr: 6.32e-03, grad_scale: 8.0
2022-12-23 13:05:08,297 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.8167, 1.6195, 2.0385, 1.7673, 1.3034, 5.0214, 2.1100, 2.3457],
       device='cuda:2'), covar=tensor([0.3295, 0.2119, 0.1769, 0.2061, 0.1425, 0.0099, 0.1484, 0.0886],
       device='cuda:2'), in_proj_covar=tensor([0.0135, 0.0117, 0.0127, 0.0121, 0.0103, 0.0098, 0.0093, 0.0090],
       device='cuda:2'), out_proj_covar=tensor([0.0005, 0.0005, 0.0005, 0.0005, 0.0004, 0.0003, 0.0004, 0.0004],
       device='cuda:2')
2022-12-23 13:05:13,720 INFO [zipformer.py:660] (2/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=63022.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 13:05:36,154 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 2.859e+02 4.260e+02 5.426e+02 6.876e+02 1.239e+03, threshold=1.085e+03, percent-clipped=1.0
2022-12-23 13:05:50,728 INFO [zipformer.py:660] (2/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=63047.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 13:05:57,199 INFO [scaling.py:679] (2/4) Whitening: num_groups=8, num_channels=192, metric=1.63 vs. limit=2.0
2022-12-23 13:06:01,233 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([2.5934, 2.4064, 1.9754, 1.2912, 3.0151, 2.8051, 2.4008, 1.9246],
       device='cuda:2'), covar=tensor([0.0369, 0.0400, 0.0547, 0.0773, 0.0225, 0.0327, 0.0466, 0.0809],
       device='cuda:2'), in_proj_covar=tensor([0.0126, 0.0127, 0.0130, 0.0122, 0.0098, 0.0123, 0.0137, 0.0158],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0001, 0.0001, 0.0002, 0.0002],
       device='cuda:2')
2022-12-23 13:06:02,086 INFO [train.py:894] (2/4) Epoch 18, batch 3450, loss[loss=0.2159, simple_loss=0.3024, pruned_loss=0.06474, over 18713.00 frames. ], tot_loss[loss=0.1901, simple_loss=0.2701, pruned_loss=0.05508, over 3714311.22 frames. ], batch size: 52, lr: 6.31e-03, grad_scale: 8.0
2022-12-23 13:06:58,861 INFO [zipformer.py:660] (2/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=63095.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 13:07:14,062 INFO [train.py:894] (2/4) Epoch 18, batch 3500, loss[loss=0.1887, simple_loss=0.2706, pruned_loss=0.05341, over 18571.00 frames. ], tot_loss[loss=0.19, simple_loss=0.2701, pruned_loss=0.05494, over 3714728.84 frames. ], batch size: 77, lr: 6.31e-03, grad_scale: 8.0
2022-12-23 13:07:34,992 WARNING [train.py:1060] (2/4) Exclude cut with ID 7859-102521-0017-7548-0_sp1.1 from training. Duration: 22.2954375
2022-12-23 13:07:45,951 INFO [train.py:894] (2/4) Epoch 19, batch 0, loss[loss=0.1986, simple_loss=0.2763, pruned_loss=0.06047, over 18668.00 frames. ], tot_loss[loss=0.1986, simple_loss=0.2763, pruned_loss=0.06047, over 18668.00 frames. ], batch size: 48, lr: 6.14e-03, grad_scale: 8.0
2022-12-23 13:07:45,951 INFO [train.py:919] (2/4) Computing validation loss
2022-12-23 13:07:56,915 INFO [train.py:928] (2/4) Epoch 19, validation: loss=0.1656, simple_loss=0.2641, pruned_loss=0.03356, over 944034.00 frames. 
2022-12-23 13:07:56,916 INFO [train.py:929] (2/4) Maximum memory allocated so far is 24680MB
2022-12-23 13:08:24,352 INFO [zipformer.py:660] (2/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=63129.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 13:08:35,788 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 2.218e+02 4.228e+02 5.499e+02 7.176e+02 2.026e+03, threshold=1.100e+03, percent-clipped=7.0
2022-12-23 13:08:53,090 WARNING [train.py:1060] (2/4) Exclude cut with ID 298-126791-0067-24026-0_sp0.9 from training. Duration: 21.438875
2022-12-23 13:08:57,626 WARNING [train.py:1060] (2/4) Exclude cut with ID 5652-39938-0025-23684-0_sp0.9 from training. Duration: 22.2055625
2022-12-23 13:09:12,658 INFO [train.py:894] (2/4) Epoch 19, batch 50, loss[loss=0.1617, simple_loss=0.242, pruned_loss=0.04068, over 18475.00 frames. ], tot_loss[loss=0.1822, simple_loss=0.2684, pruned_loss=0.04805, over 838448.79 frames. ], batch size: 43, lr: 6.14e-03, grad_scale: 8.0
2022-12-23 13:10:15,222 INFO [zipformer.py:660] (2/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=63201.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 13:10:29,630 INFO [train.py:894] (2/4) Epoch 19, batch 100, loss[loss=0.1703, simple_loss=0.2693, pruned_loss=0.03562, over 18600.00 frames. ], tot_loss[loss=0.1784, simple_loss=0.2656, pruned_loss=0.0456, over 1476394.81 frames. ], batch size: 56, lr: 6.13e-03, grad_scale: 8.0
2022-12-23 13:10:45,470 INFO [zipformer.py:660] (2/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=63221.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 13:11:07,781 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 2.052e+02 3.397e+02 4.166e+02 5.229e+02 9.928e+02, threshold=8.333e+02, percent-clipped=0.0
2022-12-23 13:11:14,340 INFO [zipformer.py:660] (2/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=63241.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 13:11:26,420 INFO [zipformer.py:660] (2/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=63249.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 13:11:43,018 INFO [train.py:894] (2/4) Epoch 19, batch 150, loss[loss=0.1583, simple_loss=0.2484, pruned_loss=0.03412, over 18701.00 frames. ], tot_loss[loss=0.1772, simple_loss=0.2646, pruned_loss=0.04494, over 1973119.23 frames. ], batch size: 50, lr: 6.13e-03, grad_scale: 8.0
2022-12-23 13:12:01,479 WARNING [train.py:1060] (2/4) Exclude cut with ID 7859-102521-0017-7548-0 from training. Duration: 24.525
2022-12-23 13:12:14,844 INFO [zipformer.py:660] (2/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=63282.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 13:12:25,514 INFO [zipformer.py:660] (2/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=63289.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 13:12:32,680 WARNING [train.py:1060] (2/4) Exclude cut with ID 3699-47246-0007-3408-0_sp0.9 from training. Duration: 20.26675
2022-12-23 13:12:46,697 WARNING [train.py:1060] (2/4) Exclude cut with ID 7859-102521-0017-7548-0_sp0.9 from training. Duration: 27.25
2022-12-23 13:12:59,142 INFO [train.py:894] (2/4) Epoch 19, batch 200, loss[loss=0.179, simple_loss=0.2742, pruned_loss=0.04189, over 18584.00 frames. ], tot_loss[loss=0.1775, simple_loss=0.2647, pruned_loss=0.04514, over 2358733.50 frames. ], batch size: 56, lr: 6.13e-03, grad_scale: 8.0
2022-12-23 13:13:15,715 INFO [zipformer.py:660] (2/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=63322.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 13:13:29,288 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.7464, 1.9179, 2.0192, 1.0590, 1.9910, 2.0926, 1.5346, 2.3510],
       device='cuda:2'), covar=tensor([0.1122, 0.1730, 0.1161, 0.1948, 0.0782, 0.1071, 0.2216, 0.0537],
       device='cuda:2'), in_proj_covar=tensor([0.0197, 0.0210, 0.0204, 0.0194, 0.0176, 0.0215, 0.0212, 0.0187],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-23 13:13:38,108 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 2.297e+02 3.501e+02 4.294e+02 5.225e+02 1.195e+03, threshold=8.588e+02, percent-clipped=3.0
2022-12-23 13:14:02,321 WARNING [train.py:1060] (2/4) Exclude cut with ID 6426-64292-0017-15984-0 from training. Duration: 21.68
2022-12-23 13:14:13,268 WARNING [train.py:1060] (2/4) Exclude cut with ID 4278-13270-0007-59342-0 from training. Duration: 21.6300625
2022-12-23 13:14:14,635 INFO [train.py:894] (2/4) Epoch 19, batch 250, loss[loss=0.1708, simple_loss=0.2615, pruned_loss=0.04003, over 18521.00 frames. ], tot_loss[loss=0.1768, simple_loss=0.2642, pruned_loss=0.04468, over 2659391.13 frames. ], batch size: 55, lr: 6.13e-03, grad_scale: 8.0
2022-12-23 13:14:28,122 INFO [zipformer.py:660] (2/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=63370.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 13:14:31,266 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([4.5420, 3.9301, 3.9329, 4.4858, 4.1703, 4.0455, 4.6908, 1.4434],
       device='cuda:2'), covar=tensor([0.0731, 0.0660, 0.0619, 0.0749, 0.1324, 0.1143, 0.0574, 0.4937],
       device='cuda:2'), in_proj_covar=tensor([0.0342, 0.0222, 0.0236, 0.0263, 0.0321, 0.0265, 0.0286, 0.0278],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0001, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-23 13:14:35,405 WARNING [train.py:1060] (2/4) Exclude cut with ID 4278-13270-0007-59342-0_sp0.9 from training. Duration: 24.033375
2022-12-23 13:15:30,909 INFO [train.py:894] (2/4) Epoch 19, batch 300, loss[loss=0.185, simple_loss=0.2789, pruned_loss=0.04558, over 18530.00 frames. ], tot_loss[loss=0.1766, simple_loss=0.2642, pruned_loss=0.04446, over 2893250.83 frames. ], batch size: 55, lr: 6.12e-03, grad_scale: 8.0
2022-12-23 13:15:30,968 WARNING [train.py:1060] (2/4) Exclude cut with ID 4278-13270-0009-59344-0 from training. Duration: 22.905
2022-12-23 13:15:32,322 WARNING [train.py:1060] (2/4) Exclude cut with ID 5622-44585-0006-90525-0_sp1.1 from training. Duration: 23.4318125
2022-12-23 13:15:37,734 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([4.3915, 3.7611, 3.7382, 4.2919, 3.9612, 3.8517, 4.5285, 1.3061],
       device='cuda:2'), covar=tensor([0.0718, 0.0707, 0.0734, 0.0840, 0.1416, 0.1203, 0.0586, 0.5192],
       device='cuda:2'), in_proj_covar=tensor([0.0339, 0.0221, 0.0235, 0.0262, 0.0319, 0.0264, 0.0284, 0.0276],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0001, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-23 13:15:42,757 INFO [scaling.py:679] (2/4) Whitening: num_groups=8, num_channels=96, metric=1.31 vs. limit=2.0
2022-12-23 13:15:44,738 INFO [zipformer.py:660] (2/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=63420.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 13:15:47,491 INFO [zipformer.py:660] (2/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=63422.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 13:15:58,079 INFO [zipformer.py:660] (2/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=63429.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 13:16:10,932 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 2.316e+02 3.457e+02 4.109e+02 5.023e+02 1.116e+03, threshold=8.218e+02, percent-clipped=3.0
2022-12-23 13:16:46,628 INFO [train.py:894] (2/4) Epoch 19, batch 350, loss[loss=0.2012, simple_loss=0.2885, pruned_loss=0.05693, over 18680.00 frames. ], tot_loss[loss=0.1772, simple_loss=0.2651, pruned_loss=0.04468, over 3076102.62 frames. ], batch size: 60, lr: 6.12e-03, grad_scale: 8.0
2022-12-23 13:17:09,882 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.5353, 1.6455, 1.9029, 1.0143, 1.7756, 1.9044, 1.4154, 2.1825],
       device='cuda:2'), covar=tensor([0.1116, 0.1917, 0.1042, 0.1663, 0.0756, 0.1002, 0.2231, 0.0506],
       device='cuda:2'), in_proj_covar=tensor([0.0197, 0.0211, 0.0204, 0.0193, 0.0176, 0.0215, 0.0213, 0.0188],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-23 13:17:11,092 INFO [zipformer.py:660] (2/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=63477.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 13:17:18,002 INFO [zipformer.py:660] (2/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=63481.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 13:17:21,423 INFO [zipformer.py:660] (2/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=63483.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 13:17:31,478 WARNING [train.py:1060] (2/4) Exclude cut with ID 4278-13270-0009-59344-0_sp1.1 from training. Duration: 20.82275
2022-12-23 13:17:33,120 WARNING [train.py:1060] (2/4) Exclude cut with ID 4278-13270-0009-59344-0_sp0.9 from training. Duration: 25.45
2022-12-23 13:18:02,714 INFO [train.py:894] (2/4) Epoch 19, batch 400, loss[loss=0.1641, simple_loss=0.2487, pruned_loss=0.0398, over 18676.00 frames. ], tot_loss[loss=0.1778, simple_loss=0.2652, pruned_loss=0.0452, over 3216875.05 frames. ], batch size: 46, lr: 6.12e-03, grad_scale: 8.0
2022-12-23 13:18:28,582 WARNING [train.py:1060] (2/4) Exclude cut with ID 5622-44585-0006-90525-0 from training. Duration: 25.775
2022-12-23 13:18:42,196 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 2.168e+02 3.327e+02 4.266e+02 5.162e+02 9.583e+02, threshold=8.532e+02, percent-clipped=3.0
2022-12-23 13:18:50,565 WARNING [train.py:1060] (2/4) Exclude cut with ID 2929-85685-0071-62375-0_sp0.9 from training. Duration: 22.25
2022-12-23 13:19:16,829 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([2.7296, 2.2326, 1.7113, 2.6829, 2.0404, 2.1353, 2.0961, 2.5836],
       device='cuda:2'), covar=tensor([0.1851, 0.3250, 0.1906, 0.2373, 0.3537, 0.1046, 0.3134, 0.0888],
       device='cuda:2'), in_proj_covar=tensor([0.0291, 0.0286, 0.0240, 0.0345, 0.0267, 0.0224, 0.0285, 0.0207],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0003, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-23 13:19:17,712 INFO [train.py:894] (2/4) Epoch 19, batch 450, loss[loss=0.1566, simple_loss=0.246, pruned_loss=0.03353, over 18534.00 frames. ], tot_loss[loss=0.1792, simple_loss=0.2669, pruned_loss=0.04581, over 3326710.78 frames. ], batch size: 47, lr: 6.12e-03, grad_scale: 8.0
2022-12-23 13:19:17,789 WARNING [train.py:1060] (2/4) Exclude cut with ID 3972-170212-0014-23379-0 from training. Duration: 26.205
2022-12-23 13:19:29,461 INFO [zipformer.py:660] (2/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=63569.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 13:19:33,656 WARNING [train.py:1060] (2/4) Exclude cut with ID 5239-32139-0047-9341-0_sp0.9 from training. Duration: 30.1555625
2022-12-23 13:19:40,194 WARNING [train.py:1060] (2/4) Exclude cut with ID 1265-135635-0050-6781-0_sp0.9 from training. Duration: 21.8333125
2022-12-23 13:19:41,646 INFO [zipformer.py:660] (2/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=63577.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 13:19:48,992 WARNING [train.py:1060] (2/4) Exclude cut with ID 1914-133440-0024-94914-0_sp1.1 from training. Duration: 20.6545625
2022-12-23 13:19:55,489 INFO [zipformer.py:660] (2/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=63586.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 13:20:30,936 WARNING [train.py:1060] (2/4) Exclude cut with ID 7395-89880-0045-39920-0_sp0.9 from training. Duration: 20.52225
2022-12-23 13:20:32,293 INFO [train.py:894] (2/4) Epoch 19, batch 500, loss[loss=0.1619, simple_loss=0.252, pruned_loss=0.03587, over 18449.00 frames. ], tot_loss[loss=0.1792, simple_loss=0.2667, pruned_loss=0.04582, over 3410869.40 frames. ], batch size: 50, lr: 6.11e-03, grad_scale: 8.0
2022-12-23 13:20:52,424 WARNING [train.py:1060] (2/4) Exclude cut with ID 3972-170212-0014-23379-0_sp0.9 from training. Duration: 29.1166875
2022-12-23 13:21:01,474 INFO [zipformer.py:660] (2/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=63630.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 13:21:07,677 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.8453, 1.3584, 1.3309, 1.9328, 1.6579, 3.2290, 1.4198, 1.5341],
       device='cuda:2'), covar=tensor([0.0753, 0.1837, 0.1128, 0.0864, 0.1398, 0.0219, 0.1335, 0.1602],
       device='cuda:2'), in_proj_covar=tensor([0.0073, 0.0082, 0.0073, 0.0074, 0.0090, 0.0075, 0.0085, 0.0077],
       device='cuda:2'), out_proj_covar=tensor([0.0003, 0.0004, 0.0004, 0.0004, 0.0004, 0.0003, 0.0004, 0.0004],
       device='cuda:2')
2022-12-23 13:21:11,693 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 2.157e+02 3.552e+02 4.116e+02 4.841e+02 8.719e+02, threshold=8.233e+02, percent-clipped=1.0
2022-12-23 13:21:27,166 INFO [zipformer.py:660] (2/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=63647.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 13:21:48,211 INFO [train.py:894] (2/4) Epoch 19, batch 550, loss[loss=0.186, simple_loss=0.2789, pruned_loss=0.04659, over 18566.00 frames. ], tot_loss[loss=0.179, simple_loss=0.2663, pruned_loss=0.04579, over 3478015.03 frames. ], batch size: 55, lr: 6.11e-03, grad_scale: 8.0
2022-12-23 13:21:51,224 WARNING [train.py:1060] (2/4) Exclude cut with ID 543-133211-0007-59831-0_sp0.9 from training. Duration: 21.388875
2022-12-23 13:22:28,600 WARNING [train.py:1060] (2/4) Exclude cut with ID 1914-133440-0024-94914-0 from training. Duration: 22.72
2022-12-23 13:22:30,050 WARNING [train.py:1060] (2/4) Exclude cut with ID 1914-133440-0031-94921-0_sp0.9 from training. Duration: 22.7444375
2022-12-23 13:23:04,733 INFO [train.py:894] (2/4) Epoch 19, batch 600, loss[loss=0.1655, simple_loss=0.2504, pruned_loss=0.04029, over 18408.00 frames. ], tot_loss[loss=0.1794, simple_loss=0.267, pruned_loss=0.04589, over 3530403.74 frames. ], batch size: 46, lr: 6.11e-03, grad_scale: 8.0
2022-12-23 13:23:14,954 WARNING [train.py:1060] (2/4) Exclude cut with ID 4133-6541-0027-40495-0_sp1.1 from training. Duration: 0.9681875
2022-12-23 13:23:18,802 WARNING [train.py:1060] (2/4) Exclude cut with ID 6330-62851-0022-91297-0_sp0.9 from training. Duration: 22.3166875
2022-12-23 13:23:25,119 WARNING [train.py:1060] (2/4) Exclude cut with ID 543-133212-0015-59917-0_sp0.9 from training. Duration: 21.8166875
2022-12-23 13:23:46,946 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 1.867e+02 3.213e+02 3.819e+02 4.458e+02 9.536e+02, threshold=7.638e+02, percent-clipped=0.0
2022-12-23 13:24:22,566 INFO [train.py:894] (2/4) Epoch 19, batch 650, loss[loss=0.1914, simple_loss=0.2794, pruned_loss=0.05174, over 18648.00 frames. ], tot_loss[loss=0.1798, simple_loss=0.2675, pruned_loss=0.04605, over 3571193.74 frames. ], batch size: 62, lr: 6.11e-03, grad_scale: 8.0
2022-12-23 13:24:45,295 INFO [zipformer.py:660] (2/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=63776.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 13:24:48,846 INFO [zipformer.py:660] (2/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=63778.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 13:25:10,811 WARNING [train.py:1060] (2/4) Exclude cut with ID 4957-30119-0041-23990-0_sp0.9 from training. Duration: 20.22775
2022-12-23 13:25:29,989 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([3.0328, 5.4166, 4.8554, 2.6179, 5.5860, 4.1508, 0.5532, 3.4937],
       device='cuda:2'), covar=tensor([0.1785, 0.0710, 0.1149, 0.2863, 0.0505, 0.0683, 0.5162, 0.1206],
       device='cuda:2'), in_proj_covar=tensor([0.0141, 0.0135, 0.0152, 0.0121, 0.0138, 0.0110, 0.0141, 0.0111],
       device='cuda:2'), out_proj_covar=tensor([0.0003, 0.0002, 0.0003, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-23 13:25:35,772 INFO [zipformer.py:660] (2/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=63810.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 13:25:36,897 INFO [train.py:894] (2/4) Epoch 19, batch 700, loss[loss=0.1626, simple_loss=0.2504, pruned_loss=0.03738, over 18377.00 frames. ], tot_loss[loss=0.1792, simple_loss=0.267, pruned_loss=0.04575, over 3601997.84 frames. ], batch size: 46, lr: 6.11e-03, grad_scale: 8.0
2022-12-23 13:25:53,651 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([4.7380, 4.0539, 4.0228, 4.6025, 4.3082, 4.1185, 4.8294, 1.5094],
       device='cuda:2'), covar=tensor([0.0606, 0.0648, 0.0617, 0.0662, 0.1282, 0.1177, 0.0490, 0.4891],
       device='cuda:2'), in_proj_covar=tensor([0.0334, 0.0219, 0.0232, 0.0257, 0.0316, 0.0261, 0.0282, 0.0274],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0001, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-23 13:25:56,232 WARNING [train.py:1060] (2/4) Exclude cut with ID 5239-32139-0047-9341-0_sp1.1 from training. Duration: 24.67275
2022-12-23 13:26:15,080 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 2.104e+02 3.773e+02 4.476e+02 5.280e+02 9.303e+02, threshold=8.951e+02, percent-clipped=10.0
2022-12-23 13:26:20,194 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([5.8422, 4.9021, 5.0425, 5.8012, 5.3660, 5.1304, 5.8759, 1.7412],
       device='cuda:2'), covar=tensor([0.0558, 0.0654, 0.0528, 0.0550, 0.1244, 0.1050, 0.0389, 0.4953],
       device='cuda:2'), in_proj_covar=tensor([0.0332, 0.0218, 0.0231, 0.0255, 0.0314, 0.0260, 0.0281, 0.0272],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0001, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-23 13:26:26,121 WARNING [train.py:1060] (2/4) Exclude cut with ID 3082-165428-0081-50734-0_sp0.9 from training. Duration: 21.8055625
2022-12-23 13:26:27,773 INFO [zipformer.py:660] (2/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=63845.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 13:26:51,613 INFO [train.py:894] (2/4) Epoch 19, batch 750, loss[loss=0.1595, simple_loss=0.2458, pruned_loss=0.03659, over 18383.00 frames. ], tot_loss[loss=0.1799, simple_loss=0.2674, pruned_loss=0.04621, over 3626981.38 frames. ], batch size: 46, lr: 6.10e-03, grad_scale: 8.0
2022-12-23 13:27:00,986 WARNING [train.py:1060] (2/4) Exclude cut with ID 3340-169293-0054-76830-0_sp0.9 from training. Duration: 22.6666875
2022-12-23 13:27:07,484 INFO [zipformer.py:660] (2/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=63871.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 13:27:16,832 INFO [zipformer.py:660] (2/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=63877.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 13:27:34,815 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.5906, 1.4508, 1.5983, 1.5189, 1.0823, 2.9539, 1.3861, 1.8740],
       device='cuda:2'), covar=tensor([0.3154, 0.2116, 0.2009, 0.2100, 0.1491, 0.0231, 0.1588, 0.0814],
       device='cuda:2'), in_proj_covar=tensor([0.0133, 0.0117, 0.0125, 0.0121, 0.0102, 0.0096, 0.0092, 0.0088],
       device='cuda:2'), out_proj_covar=tensor([0.0005, 0.0005, 0.0005, 0.0005, 0.0004, 0.0003, 0.0004, 0.0004],
       device='cuda:2')
2022-12-23 13:28:00,177 INFO [zipformer.py:660] (2/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=63906.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 13:28:04,169 WARNING [train.py:1060] (2/4) Exclude cut with ID 2411-132532-0017-82279-0_sp1.1 from training. Duration: 0.9681875
2022-12-23 13:28:06,917 INFO [train.py:894] (2/4) Epoch 19, batch 800, loss[loss=0.182, simple_loss=0.2721, pruned_loss=0.04594, over 18627.00 frames. ], tot_loss[loss=0.1795, simple_loss=0.2669, pruned_loss=0.04603, over 3646691.09 frames. ], batch size: 53, lr: 6.10e-03, grad_scale: 8.0
2022-12-23 13:28:15,474 INFO [zipformer.py:660] (2/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=63916.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 13:28:29,379 INFO [zipformer.py:660] (2/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=63925.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 13:28:29,446 INFO [zipformer.py:660] (2/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=63925.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 13:28:32,218 WARNING [train.py:1060] (2/4) Exclude cut with ID 6330-62850-0007-91323-0 from training. Duration: 22.485
2022-12-23 13:28:47,233 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 2.190e+02 3.396e+02 4.192e+02 4.908e+02 9.047e+02, threshold=8.383e+02, percent-clipped=1.0
2022-12-23 13:28:47,749 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([2.0069, 1.8531, 1.5450, 1.6499, 1.7276, 1.8195, 1.7182, 1.8234],
       device='cuda:2'), covar=tensor([0.2175, 0.3165, 0.1964, 0.2478, 0.3391, 0.1128, 0.2675, 0.1028],
       device='cuda:2'), in_proj_covar=tensor([0.0293, 0.0288, 0.0243, 0.0348, 0.0269, 0.0225, 0.0287, 0.0210],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0003, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-23 13:28:54,640 INFO [zipformer.py:660] (2/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=63942.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 13:29:08,718 WARNING [train.py:1060] (2/4) Exclude cut with ID 3972-170212-0014-23379-0_sp1.1 from training. Duration: 23.82275
2022-12-23 13:29:19,120 WARNING [train.py:1060] (2/4) Exclude cut with ID 4860-13185-0032-76709-0 from training. Duration: 20.77
2022-12-23 13:29:23,352 INFO [train.py:894] (2/4) Epoch 19, batch 850, loss[loss=0.2191, simple_loss=0.2974, pruned_loss=0.07043, over 18619.00 frames. ], tot_loss[loss=0.1805, simple_loss=0.2681, pruned_loss=0.04651, over 3661274.28 frames. ], batch size: 173, lr: 6.10e-03, grad_scale: 8.0
2022-12-23 13:29:26,963 WARNING [train.py:1060] (2/4) Exclude cut with ID 6426-64292-0017-15984-0_sp0.9 from training. Duration: 24.088875
2022-12-23 13:29:48,306 INFO [zipformer.py:660] (2/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=63977.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 13:29:53,763 WARNING [train.py:1060] (2/4) Exclude cut with ID 6330-62850-0007-91323-0_sp1.1 from training. Duration: 20.4409375
2022-12-23 13:30:43,306 INFO [train.py:894] (2/4) Epoch 19, batch 900, loss[loss=0.1933, simple_loss=0.2872, pruned_loss=0.04971, over 18634.00 frames. ], tot_loss[loss=0.1799, simple_loss=0.2677, pruned_loss=0.04609, over 3673720.06 frames. ], batch size: 60, lr: 6.10e-03, grad_scale: 8.0
2022-12-23 13:31:01,890 INFO [zipformer.py:660] (2/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=64023.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 13:31:12,271 WARNING [train.py:1060] (2/4) Exclude cut with ID 2929-85685-0062-62366-0_sp0.9 from training. Duration: 22.511125
2022-12-23 13:31:13,683 WARNING [train.py:1060] (2/4) Exclude cut with ID 7395-89880-0031-39906-0 from training. Duration: 20.675
2022-12-23 13:31:22,423 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 2.188e+02 3.638e+02 4.436e+02 5.508e+02 1.633e+03, threshold=8.872e+02, percent-clipped=5.0
2022-12-23 13:31:58,434 INFO [train.py:894] (2/4) Epoch 19, batch 950, loss[loss=0.1601, simple_loss=0.247, pruned_loss=0.03655, over 18593.00 frames. ], tot_loss[loss=0.1809, simple_loss=0.2686, pruned_loss=0.04656, over 3682474.35 frames. ], batch size: 45, lr: 6.09e-03, grad_scale: 8.0
2022-12-23 13:31:59,751 INFO [scaling.py:679] (2/4) Whitening: num_groups=8, num_channels=96, metric=1.45 vs. limit=2.0
2022-12-23 13:32:02,755 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([2.6816, 2.2350, 1.6060, 2.4588, 1.9658, 2.1421, 2.0999, 2.6791],
       device='cuda:2'), covar=tensor([0.1938, 0.3222, 0.1900, 0.3008, 0.3657, 0.1033, 0.2977, 0.0865],
       device='cuda:2'), in_proj_covar=tensor([0.0291, 0.0286, 0.0241, 0.0345, 0.0267, 0.0223, 0.0284, 0.0208],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0003, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-23 13:32:21,452 INFO [zipformer.py:660] (2/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=64076.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 13:32:24,299 INFO [zipformer.py:660] (2/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=64078.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 13:32:32,961 INFO [zipformer.py:660] (2/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=64084.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 13:32:51,807 WARNING [train.py:1060] (2/4) Exclude cut with ID 6330-62850-0007-91323-0_sp0.9 from training. Duration: 24.9833125
2022-12-23 13:33:12,930 INFO [train.py:894] (2/4) Epoch 19, batch 1000, loss[loss=0.2027, simple_loss=0.29, pruned_loss=0.05768, over 18715.00 frames. ], tot_loss[loss=0.1804, simple_loss=0.2681, pruned_loss=0.04635, over 3689862.65 frames. ], batch size: 52, lr: 6.09e-03, grad_scale: 8.0
2022-12-23 13:33:23,953 WARNING [train.py:1060] (2/4) Exclude cut with ID 5239-32139-0047-9341-0 from training. Duration: 27.14
2022-12-23 13:33:33,143 INFO [zipformer.py:660] (2/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=64124.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 13:33:36,269 INFO [zipformer.py:660] (2/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=64126.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 13:33:37,769 WARNING [train.py:1060] (2/4) Exclude cut with ID 2929-85685-0044-62348-0 from training. Duration: 22.44
2022-12-23 13:33:52,922 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 2.707e+02 3.878e+02 4.599e+02 5.845e+02 2.549e+03, threshold=9.198e+02, percent-clipped=7.0
2022-12-23 13:34:29,461 INFO [train.py:894] (2/4) Epoch 19, batch 1050, loss[loss=0.1885, simple_loss=0.2763, pruned_loss=0.05039, over 18664.00 frames. ], tot_loss[loss=0.18, simple_loss=0.268, pruned_loss=0.046, over 3694434.77 frames. ], batch size: 175, lr: 6.09e-03, grad_scale: 8.0
2022-12-23 13:34:37,098 INFO [zipformer.py:660] (2/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=64166.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 13:34:55,627 WARNING [train.py:1060] (2/4) Exclude cut with ID 2929-85685-0060-62364-0_sp0.9 from training. Duration: 21.361125
2022-12-23 13:35:01,886 WARNING [train.py:1060] (2/4) Exclude cut with ID 2929-85685-0079-62383-0_sp1.1 from training. Duration: 27.0318125
2022-12-23 13:35:12,199 WARNING [train.py:1060] (2/4) Exclude cut with ID 5622-44585-0006-90525-0_sp0.9 from training. Duration: 28.638875
2022-12-23 13:35:27,742 WARNING [train.py:1060] (2/4) Exclude cut with ID 3340-169293-0054-76830-0 from training. Duration: 20.4
2022-12-23 13:35:29,576 INFO [zipformer.py:660] (2/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=64201.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 13:35:44,103 INFO [train.py:894] (2/4) Epoch 19, batch 1100, loss[loss=0.19, simple_loss=0.2821, pruned_loss=0.04891, over 18600.00 frames. ], tot_loss[loss=0.1798, simple_loss=0.2681, pruned_loss=0.04579, over 3699082.75 frames. ], batch size: 56, lr: 6.09e-03, grad_scale: 8.0
2022-12-23 13:36:01,178 WARNING [train.py:1060] (2/4) Exclude cut with ID 2929-85685-0071-62375-0 from training. Duration: 20.025
2022-12-23 13:36:01,191 WARNING [train.py:1060] (2/4) Exclude cut with ID 2364-131735-0112-64612-0_sp0.9 from training. Duration: 20.488875
2022-12-23 13:36:05,826 WARNING [train.py:1060] (2/4) Exclude cut with ID 2929-85685-0079-62383-0 from training. Duration: 29.735
2022-12-23 13:36:06,020 INFO [zipformer.py:660] (2/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=64225.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 13:36:17,433 INFO [zipformer.py:660] (2/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=64232.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 13:36:24,141 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 2.216e+02 3.480e+02 4.012e+02 4.856e+02 7.809e+02, threshold=8.025e+02, percent-clipped=0.0
2022-12-23 13:36:31,549 INFO [zipformer.py:660] (2/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=64242.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 13:36:37,529 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([2.2391, 2.7825, 2.8252, 1.4913, 3.0509, 3.0122, 1.8557, 3.3149],
       device='cuda:2'), covar=tensor([0.1347, 0.1534, 0.1420, 0.2260, 0.0713, 0.1281, 0.2218, 0.0542],
       device='cuda:2'), in_proj_covar=tensor([0.0195, 0.0208, 0.0204, 0.0193, 0.0174, 0.0213, 0.0211, 0.0187],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-23 13:36:59,351 INFO [train.py:894] (2/4) Epoch 19, batch 1150, loss[loss=0.1526, simple_loss=0.2384, pruned_loss=0.03346, over 18711.00 frames. ], tot_loss[loss=0.1801, simple_loss=0.2683, pruned_loss=0.04594, over 3702775.86 frames. ], batch size: 46, lr: 6.08e-03, grad_scale: 8.0
2022-12-23 13:37:13,945 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([2.4563, 2.2489, 1.6134, 0.5974, 1.6062, 2.0029, 1.6203, 1.8520],
       device='cuda:2'), covar=tensor([0.0604, 0.0548, 0.1221, 0.1734, 0.1319, 0.1449, 0.1751, 0.0767],
       device='cuda:2'), in_proj_covar=tensor([0.0170, 0.0180, 0.0200, 0.0187, 0.0206, 0.0195, 0.0210, 0.0196],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-23 13:37:16,521 INFO [zipformer.py:660] (2/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=64272.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 13:37:17,959 INFO [zipformer.py:660] (2/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=64273.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 13:37:26,006 WARNING [train.py:1060] (2/4) Exclude cut with ID 7276-92427-0014-12983-0_sp0.9 from training. Duration: 21.3055625
2022-12-23 13:37:27,633 WARNING [train.py:1060] (2/4) Exclude cut with ID 1025-75365-0008-79168-0_sp0.9 from training. Duration: 22.0666875
2022-12-23 13:37:43,687 INFO [zipformer.py:660] (2/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=64290.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 13:37:48,632 INFO [zipformer.py:660] (2/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=64293.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 13:38:15,990 INFO [train.py:894] (2/4) Epoch 19, batch 1200, loss[loss=0.2105, simple_loss=0.2958, pruned_loss=0.06258, over 18654.00 frames. ], tot_loss[loss=0.1799, simple_loss=0.268, pruned_loss=0.04587, over 3705000.79 frames. ], batch size: 181, lr: 6.08e-03, grad_scale: 8.0
2022-12-23 13:38:54,836 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 2.279e+02 3.709e+02 4.310e+02 5.274e+02 1.425e+03, threshold=8.620e+02, percent-clipped=2.0
2022-12-23 13:39:07,231 INFO [zipformer.py:660] (2/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=64345.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 13:39:17,347 WARNING [train.py:1060] (2/4) Exclude cut with ID 2929-85685-0062-62366-0 from training. Duration: 20.26
2022-12-23 13:39:29,938 WARNING [train.py:1060] (2/4) Exclude cut with ID 5239-32139-0030-9324-0_sp0.9 from training. Duration: 21.3444375
2022-12-23 13:39:31,049 INFO [train.py:894] (2/4) Epoch 19, batch 1250, loss[loss=0.184, simple_loss=0.2592, pruned_loss=0.05444, over 18593.00 frames. ], tot_loss[loss=0.179, simple_loss=0.2672, pruned_loss=0.04538, over 3706790.17 frames. ], batch size: 45, lr: 6.08e-03, grad_scale: 8.0
2022-12-23 13:39:32,900 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.8687, 1.7222, 1.9401, 1.8020, 1.4898, 5.0040, 2.0010, 2.4685],
       device='cuda:2'), covar=tensor([0.3118, 0.2023, 0.1786, 0.1972, 0.1241, 0.0090, 0.1471, 0.0779],
       device='cuda:2'), in_proj_covar=tensor([0.0132, 0.0116, 0.0125, 0.0120, 0.0101, 0.0096, 0.0091, 0.0088],
       device='cuda:2'), out_proj_covar=tensor([0.0005, 0.0005, 0.0005, 0.0005, 0.0004, 0.0003, 0.0004, 0.0004],
       device='cuda:2')
2022-12-23 13:39:52,880 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.6150, 1.3578, 1.4112, 2.0881, 1.7558, 3.4676, 1.3821, 1.5235],
       device='cuda:2'), covar=tensor([0.0795, 0.1760, 0.1032, 0.0804, 0.1333, 0.0209, 0.1311, 0.1560],
       device='cuda:2'), in_proj_covar=tensor([0.0072, 0.0081, 0.0072, 0.0074, 0.0089, 0.0074, 0.0084, 0.0076],
       device='cuda:2'), out_proj_covar=tensor([0.0003, 0.0004, 0.0004, 0.0004, 0.0004, 0.0003, 0.0004, 0.0004],
       device='cuda:2')
2022-12-23 13:39:58,486 INFO [zipformer.py:660] (2/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=64379.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 13:39:59,188 INFO [scaling.py:679] (2/4) Whitening: num_groups=8, num_channels=96, metric=1.22 vs. limit=2.0
2022-12-23 13:40:26,936 WARNING [train.py:1060] (2/4) Exclude cut with ID 497-129325-0061-62254-0_sp1.1 from training. Duration: 0.97725
2022-12-23 13:40:27,326 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([2.0822, 2.5071, 1.9892, 3.0905, 3.0573, 2.0238, 2.1472, 1.6269],
       device='cuda:2'), covar=tensor([0.1708, 0.1496, 0.1393, 0.0803, 0.1293, 0.1035, 0.1729, 0.1425],
       device='cuda:2'), in_proj_covar=tensor([0.0241, 0.0220, 0.0212, 0.0196, 0.0258, 0.0193, 0.0219, 0.0194],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-23 13:40:39,208 INFO [zipformer.py:660] (2/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=64406.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 13:40:47,314 INFO [train.py:894] (2/4) Epoch 19, batch 1300, loss[loss=0.1708, simple_loss=0.2618, pruned_loss=0.03995, over 18725.00 frames. ], tot_loss[loss=0.1794, simple_loss=0.2676, pruned_loss=0.04563, over 3708218.76 frames. ], batch size: 52, lr: 6.08e-03, grad_scale: 8.0
2022-12-23 13:41:09,968 WARNING [train.py:1060] (2/4) Exclude cut with ID 7395-89880-0031-39906-0_sp0.9 from training. Duration: 22.97225
2022-12-23 13:41:24,359 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 2.334e+02 3.220e+02 4.296e+02 5.683e+02 9.700e+02, threshold=8.592e+02, percent-clipped=5.0
2022-12-23 13:41:40,674 WARNING [train.py:1060] (2/4) Exclude cut with ID 7395-89880-0047-39922-0_sp0.9 from training. Duration: 21.97775
2022-12-23 13:41:55,146 WARNING [train.py:1060] (2/4) Exclude cut with ID 1112-1043-0006-89194-0_sp0.9 from training. Duration: 21.8333125
2022-12-23 13:42:00,555 INFO [train.py:894] (2/4) Epoch 19, batch 1350, loss[loss=0.1633, simple_loss=0.26, pruned_loss=0.03327, over 18399.00 frames. ], tot_loss[loss=0.1783, simple_loss=0.2664, pruned_loss=0.04507, over 3709469.55 frames. ], batch size: 53, lr: 6.07e-03, grad_scale: 8.0
2022-12-23 13:42:06,409 WARNING [train.py:1060] (2/4) Exclude cut with ID 1914-133440-0031-94921-0 from training. Duration: 20.47
2022-12-23 13:42:06,722 INFO [zipformer.py:660] (2/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=64465.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 13:42:07,980 INFO [zipformer.py:660] (2/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=64466.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 13:42:22,479 INFO [scaling.py:679] (2/4) Whitening: num_groups=8, num_channels=96, metric=1.65 vs. limit=2.0
2022-12-23 13:42:36,555 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.8121, 1.7071, 1.3446, 1.5995, 1.8000, 1.6446, 2.1061, 1.8608],
       device='cuda:2'), covar=tensor([0.0904, 0.1678, 0.2826, 0.1773, 0.2000, 0.0969, 0.1079, 0.1326],
       device='cuda:2'), in_proj_covar=tensor([0.0176, 0.0205, 0.0245, 0.0286, 0.0234, 0.0189, 0.0206, 0.0201],
       device='cuda:2'), out_proj_covar=tensor([0.0001, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-23 13:42:59,949 INFO [zipformer.py:660] (2/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=64501.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 13:43:11,929 WARNING [train.py:1060] (2/4) Exclude cut with ID 7395-89880-0037-39912-0_sp0.9 from training. Duration: 20.67225
2022-12-23 13:43:15,488 INFO [train.py:894] (2/4) Epoch 19, batch 1400, loss[loss=0.1977, simple_loss=0.2879, pruned_loss=0.05375, over 18459.00 frames. ], tot_loss[loss=0.1785, simple_loss=0.2666, pruned_loss=0.04521, over 3710302.70 frames. ], batch size: 64, lr: 6.07e-03, grad_scale: 8.0
2022-12-23 13:43:19,650 INFO [zipformer.py:660] (2/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=64514.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 13:43:34,286 WARNING [train.py:1060] (2/4) Exclude cut with ID 1914-133440-0024-94914-0_sp0.9 from training. Duration: 25.2444375
2022-12-23 13:43:40,631 INFO [zipformer.py:660] (2/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=64526.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 13:43:55,840 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 1.643e+02 3.417e+02 3.883e+02 4.818e+02 7.997e+02, threshold=7.766e+02, percent-clipped=0.0
2022-12-23 13:43:57,352 WARNING [train.py:1060] (2/4) Exclude cut with ID 3340-169293-0021-76797-0_sp0.9 from training. Duration: 21.1445
2022-12-23 13:44:13,675 INFO [zipformer.py:660] (2/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=64549.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 13:44:31,859 INFO [train.py:894] (2/4) Epoch 19, batch 1450, loss[loss=0.1797, simple_loss=0.2765, pruned_loss=0.04141, over 18513.00 frames. ], tot_loss[loss=0.1782, simple_loss=0.2661, pruned_loss=0.04515, over 3709852.35 frames. ], batch size: 78, lr: 6.07e-03, grad_scale: 8.0
2022-12-23 13:44:48,593 INFO [zipformer.py:660] (2/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=64572.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 13:45:11,644 WARNING [train.py:1060] (2/4) Exclude cut with ID 2929-85685-0079-62383-0_sp0.9 from training. Duration: 33.038875
2022-12-23 13:45:11,812 INFO [zipformer.py:660] (2/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=64588.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 13:45:13,381 INFO [zipformer.py:660] (2/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=64589.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 13:45:23,761 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([3.6990, 2.9956, 2.1010, 1.9576, 3.7711, 3.8922, 3.2105, 2.7928],
       device='cuda:2'), covar=tensor([0.0276, 0.0392, 0.0592, 0.0701, 0.0185, 0.0284, 0.0415, 0.0650],
       device='cuda:2'), in_proj_covar=tensor([0.0123, 0.0126, 0.0130, 0.0121, 0.0097, 0.0121, 0.0135, 0.0157],
       device='cuda:2'), out_proj_covar=tensor([0.0001, 0.0002, 0.0002, 0.0001, 0.0001, 0.0001, 0.0002, 0.0002],
       device='cuda:2')
2022-12-23 13:45:46,436 INFO [train.py:894] (2/4) Epoch 19, batch 1500, loss[loss=0.1672, simple_loss=0.2613, pruned_loss=0.03652, over 18636.00 frames. ], tot_loss[loss=0.1769, simple_loss=0.265, pruned_loss=0.04442, over 3711155.81 frames. ], batch size: 53, lr: 6.07e-03, grad_scale: 8.0
2022-12-23 13:45:48,023 WARNING [train.py:1060] (2/4) Exclude cut with ID 6426-64291-0000-16059-0_sp0.9 from training. Duration: 20.0944375
2022-12-23 13:46:00,456 INFO [zipformer.py:660] (2/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=64620.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 13:46:03,164 WARNING [train.py:1060] (2/4) Exclude cut with ID 2929-85685-0044-62348-0_sp1.1 from training. Duration: 20.4
2022-12-23 13:46:12,157 WARNING [train.py:1060] (2/4) Exclude cut with ID 6330-62851-0022-91297-0 from training. Duration: 20.085
2022-12-23 13:46:22,187 WARNING [train.py:1060] (2/4) Exclude cut with ID 4860-13185-0032-76709-0_sp0.9 from training. Duration: 23.07775
2022-12-23 13:46:24,995 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 2.225e+02 3.474e+02 4.325e+02 5.384e+02 8.400e+02, threshold=8.650e+02, percent-clipped=2.0
2022-12-23 13:46:42,675 INFO [zipformer.py:660] (2/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=64648.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 13:46:45,626 INFO [zipformer.py:660] (2/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=64650.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 13:47:02,830 INFO [train.py:894] (2/4) Epoch 19, batch 1550, loss[loss=0.2173, simple_loss=0.2974, pruned_loss=0.06861, over 18612.00 frames. ], tot_loss[loss=0.1777, simple_loss=0.2662, pruned_loss=0.04461, over 3712683.51 frames. ], batch size: 99, lr: 6.07e-03, grad_scale: 8.0
2022-12-23 13:47:07,202 WARNING [train.py:1060] (2/4) Exclude cut with ID 2929-85685-0044-62348-0_sp0.9 from training. Duration: 24.9333125
2022-12-23 13:47:29,687 INFO [zipformer.py:660] (2/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=64679.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 13:47:51,487 WARNING [train.py:1060] (2/4) Exclude cut with ID 5118-111612-0016-124680-0_sp0.9 from training. Duration: 20.388875
2022-12-23 13:47:59,757 WARNING [train.py:1060] (2/4) Exclude cut with ID 432-122774-0017-62487-0_sp1.1 from training. Duration: 20.3590625
2022-12-23 13:48:02,588 INFO [zipformer.py:660] (2/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=64701.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 13:48:05,787 INFO [zipformer.py:660] (2/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=64703.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 13:48:15,369 INFO [zipformer.py:660] (2/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=64709.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 13:48:17,777 INFO [train.py:894] (2/4) Epoch 19, batch 1600, loss[loss=0.1746, simple_loss=0.2639, pruned_loss=0.04264, over 18581.00 frames. ], tot_loss[loss=0.1777, simple_loss=0.2661, pruned_loss=0.04464, over 3713462.96 frames. ], batch size: 51, lr: 6.06e-03, grad_scale: 16.0
2022-12-23 13:48:41,811 INFO [zipformer.py:660] (2/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=64727.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 13:48:57,058 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 2.475e+02 3.648e+02 4.518e+02 5.484e+02 9.669e+02, threshold=9.035e+02, percent-clipped=5.0
2022-12-23 13:49:07,554 WARNING [train.py:1060] (2/4) Exclude cut with ID 3557-8342-0013-54691-0_sp1.1 from training. Duration: 0.836375
2022-12-23 13:49:09,514 INFO [zipformer.py:660] (2/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=64745.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 13:49:19,824 INFO [zipformer.py:660] (2/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=64752.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 13:49:29,313 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.5904, 1.5420, 1.7192, 0.9977, 1.7085, 1.7497, 1.3306, 2.0617],
       device='cuda:2'), covar=tensor([0.0963, 0.1683, 0.1238, 0.1622, 0.0699, 0.1013, 0.2148, 0.0525],
       device='cuda:2'), in_proj_covar=tensor([0.0195, 0.0209, 0.0204, 0.0191, 0.0174, 0.0212, 0.0210, 0.0186],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-23 13:49:30,787 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([2.2365, 2.1975, 1.6812, 2.6114, 2.4014, 2.0508, 3.1083, 2.2169],
       device='cuda:2'), covar=tensor([0.0783, 0.1568, 0.2524, 0.1584, 0.1648, 0.0885, 0.0821, 0.1111],
       device='cuda:2'), in_proj_covar=tensor([0.0176, 0.0205, 0.0246, 0.0285, 0.0234, 0.0190, 0.0204, 0.0201],
       device='cuda:2'), out_proj_covar=tensor([0.0001, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-23 13:49:33,374 INFO [train.py:894] (2/4) Epoch 19, batch 1650, loss[loss=0.1825, simple_loss=0.2566, pruned_loss=0.0542, over 18538.00 frames. ], tot_loss[loss=0.1787, simple_loss=0.2663, pruned_loss=0.04552, over 3713784.08 frames. ], batch size: 44, lr: 6.06e-03, grad_scale: 16.0
2022-12-23 13:49:35,020 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([4.7346, 4.1171, 3.9994, 4.6535, 4.3294, 4.1550, 4.8630, 1.4759],
       device='cuda:2'), covar=tensor([0.0735, 0.0653, 0.0630, 0.0738, 0.1425, 0.1173, 0.0555, 0.5156],
       device='cuda:2'), in_proj_covar=tensor([0.0337, 0.0220, 0.0232, 0.0257, 0.0319, 0.0264, 0.0283, 0.0276],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0001, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-23 13:49:38,538 INFO [zipformer.py:660] (2/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=64764.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 13:49:50,868 WARNING [train.py:1060] (2/4) Exclude cut with ID 8565-290391-0049-67394-0_sp0.9 from training. Duration: 21.3166875
2022-12-23 13:50:12,629 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([2.2021, 1.5411, 1.9568, 1.8957, 2.2566, 2.1632, 2.0235, 1.8178],
       device='cuda:2'), covar=tensor([0.2021, 0.3192, 0.2253, 0.2589, 0.1785, 0.0921, 0.2886, 0.1190],
       device='cuda:2'), in_proj_covar=tensor([0.0266, 0.0297, 0.0276, 0.0313, 0.0303, 0.0249, 0.0338, 0.0237],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-23 13:50:21,058 WARNING [train.py:1060] (2/4) Exclude cut with ID 6533-399-0029-104863-0_sp0.9 from training. Duration: 22.1055625
2022-12-23 13:50:31,377 WARNING [train.py:1060] (2/4) Exclude cut with ID 7699-105389-0094-26379-0_sp1.1 from training. Duration: 21.77725
2022-12-23 13:50:42,106 INFO [zipformer.py:660] (2/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=64806.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 13:50:49,675 INFO [train.py:894] (2/4) Epoch 19, batch 1700, loss[loss=0.1721, simple_loss=0.2441, pruned_loss=0.05005, over 18695.00 frames. ], tot_loss[loss=0.1812, simple_loss=0.2676, pruned_loss=0.0474, over 3714663.69 frames. ], batch size: 41, lr: 6.06e-03, grad_scale: 16.0
2022-12-23 13:50:52,520 WARNING [train.py:1060] (2/4) Exclude cut with ID 7255-291500-0005-134304-0_sp0.9 from training. Duration: 27.8166875
2022-12-23 13:50:52,907 INFO [zipformer.py:660] (2/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=64813.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 13:51:04,225 INFO [zipformer.py:660] (2/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=64821.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 13:51:18,622 WARNING [train.py:1060] (2/4) Exclude cut with ID 7357-94126-0021-15852-0_sp1.1 from training. Duration: 22.5090625
2022-12-23 13:51:26,319 WARNING [train.py:1060] (2/4) Exclude cut with ID 7255-291500-0005-134304-0 from training. Duration: 25.035
2022-12-23 13:51:29,133 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 2.637e+02 3.847e+02 4.863e+02 6.028e+02 1.438e+03, threshold=9.726e+02, percent-clipped=3.0
2022-12-23 13:51:44,040 WARNING [train.py:1060] (2/4) Exclude cut with ID 774-127930-0014-10412-0_sp1.1 from training. Duration: 0.95
2022-12-23 13:52:03,754 INFO [train.py:894] (2/4) Epoch 19, batch 1750, loss[loss=0.1939, simple_loss=0.2785, pruned_loss=0.05464, over 18720.00 frames. ], tot_loss[loss=0.1827, simple_loss=0.2683, pruned_loss=0.04858, over 3714256.99 frames. ], batch size: 52, lr: 6.06e-03, grad_scale: 8.0
2022-12-23 13:52:03,773 WARNING [train.py:1060] (2/4) Exclude cut with ID 3033-130750-0096-55598-0_sp0.9 from training. Duration: 0.92225
2022-12-23 13:52:29,229 WARNING [train.py:1060] (2/4) Exclude cut with ID 4511-76322-0006-80011-0 from training. Duration: 21.97
2022-12-23 13:52:44,687 INFO [zipformer.py:660] (2/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=64888.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 13:52:47,270 WARNING [train.py:1060] (2/4) Exclude cut with ID 7492-105653-0055-62765-0_sp0.9 from training. Duration: 21.97225
2022-12-23 13:52:48,735 WARNING [train.py:1060] (2/4) Exclude cut with ID 453-131332-0000-47844-0_sp0.9 from training. Duration: 25.3333125
2022-12-23 13:52:58,768 WARNING [train.py:1060] (2/4) Exclude cut with ID 5172-29468-0015-19128-0_sp0.9 from training. Duration: 21.5055625
2022-12-23 13:53:08,131 WARNING [train.py:1060] (2/4) Exclude cut with ID 453-131332-0000-47844-0_sp1.1 from training. Duration: 20.72725
2022-12-23 13:53:20,447 INFO [train.py:894] (2/4) Epoch 19, batch 1800, loss[loss=0.1934, simple_loss=0.2742, pruned_loss=0.05629, over 18446.00 frames. ], tot_loss[loss=0.1853, simple_loss=0.2698, pruned_loss=0.05038, over 3714483.13 frames. ], batch size: 54, lr: 6.05e-03, grad_scale: 8.0
2022-12-23 13:53:41,098 WARNING [train.py:1060] (2/4) Exclude cut with ID 8631-249866-0030-130156-0_sp0.9 from training. Duration: 26.32775
2022-12-23 13:53:57,609 INFO [zipformer.py:660] (2/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=64936.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 13:54:00,265 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 2.796e+02 4.805e+02 5.504e+02 6.542e+02 1.511e+03, threshold=1.101e+03, percent-clipped=5.0
2022-12-23 13:54:11,131 INFO [zipformer.py:660] (2/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=64945.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 13:54:15,909 WARNING [train.py:1060] (2/4) Exclude cut with ID 3867-173237-0077-144769-0 from training. Duration: 20.025
2022-12-23 13:54:20,107 WARNING [train.py:1060] (2/4) Exclude cut with ID 6709-74022-0004-86860-0_sp1.1 from training. Duration: 0.9409375
2022-12-23 13:54:20,118 WARNING [train.py:1060] (2/4) Exclude cut with ID 4757-1811-0023-62229-0_sp0.9 from training. Duration: 21.37775
2022-12-23 13:54:35,442 INFO [train.py:894] (2/4) Epoch 19, batch 1850, loss[loss=0.2227, simple_loss=0.2916, pruned_loss=0.07684, over 18590.00 frames. ], tot_loss[loss=0.1856, simple_loss=0.2689, pruned_loss=0.05121, over 3713796.92 frames. ], batch size: 178, lr: 6.05e-03, grad_scale: 8.0
2022-12-23 13:54:41,256 WARNING [train.py:1060] (2/4) Exclude cut with ID 1250-135782-0004-25974-0_sp0.9 from training. Duration: 21.17225
2022-12-23 13:54:42,603 WARNING [train.py:1060] (2/4) Exclude cut with ID 7357-94126-0021-15852-0_sp0.9 from training. Duration: 27.511125
2022-12-23 13:54:57,636 INFO [zipformer.py:660] (2/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=64976.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 13:55:14,979 WARNING [train.py:1060] (2/4) Exclude cut with ID 453-131332-0000-47844-0 from training. Duration: 22.8
2022-12-23 13:55:19,229 WARNING [train.py:1060] (2/4) Exclude cut with ID 4964-30587-0040-44509-0 from training. Duration: 22.585
2022-12-23 13:55:35,646 INFO [zipformer.py:660] (2/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=65001.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 13:55:40,038 INFO [zipformer.py:660] (2/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=65004.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 13:55:50,581 INFO [train.py:894] (2/4) Epoch 19, batch 1900, loss[loss=0.2019, simple_loss=0.2877, pruned_loss=0.058, over 18480.00 frames. ], tot_loss[loss=0.1868, simple_loss=0.2693, pruned_loss=0.05209, over 3714623.65 frames. ], batch size: 54, lr: 6.05e-03, grad_scale: 8.0
2022-12-23 13:55:50,639 WARNING [train.py:1060] (2/4) Exclude cut with ID 6978-92210-0001-146967-0_sp0.9 from training. Duration: 22.0166875
2022-12-23 13:56:05,610 WARNING [train.py:1060] (2/4) Exclude cut with ID 7255-291500-0003-134302-0_sp1.1 from training. Duration: 24.395375
2022-12-23 13:56:11,691 WARNING [train.py:1060] (2/4) Exclude cut with ID 3488-85273-0017-41203-0_sp0.9 from training. Duration: 27.47775
2022-12-23 13:56:16,107 WARNING [train.py:1060] (2/4) Exclude cut with ID 432-122774-0017-62487-0_sp0.9 from training. Duration: 24.8833125
2022-12-23 13:56:19,161 WARNING [train.py:1060] (2/4) Exclude cut with ID 6758-72288-0033-108368-0 from training. Duration: 23.39
2022-12-23 13:56:25,099 WARNING [train.py:1060] (2/4) Exclude cut with ID 8291-282929-0007-12994-0_sp0.9 from training. Duration: 28.72225
2022-12-23 13:56:29,886 INFO [zipformer.py:660] (2/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=65037.0, num_to_drop=1, layers_to_drop={3}
2022-12-23 13:56:30,837 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 2.978e+02 4.683e+02 6.050e+02 7.570e+02 2.134e+03, threshold=1.210e+03, percent-clipped=7.0
2022-12-23 13:56:35,816 WARNING [train.py:1060] (2/4) Exclude cut with ID 585-294811-0110-133686-0_sp0.9 from training. Duration: 20.8944375
2022-12-23 13:56:49,159 WARNING [train.py:1060] (2/4) Exclude cut with ID 5796-66357-0007-116447-0_sp0.9 from training. Duration: 23.8444375
2022-12-23 13:56:49,251 INFO [zipformer.py:660] (2/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=65049.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 13:56:58,550 INFO [zipformer.py:660] (2/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=65055.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 13:57:04,427 INFO [zipformer.py:660] (2/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=65059.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 13:57:07,079 INFO [train.py:894] (2/4) Epoch 19, batch 1950, loss[loss=0.2028, simple_loss=0.2819, pruned_loss=0.0619, over 18457.00 frames. ], tot_loss[loss=0.1879, simple_loss=0.2697, pruned_loss=0.05301, over 3714762.85 frames. ], batch size: 50, lr: 6.05e-03, grad_scale: 8.0
2022-12-23 13:57:12,667 WARNING [train.py:1060] (2/4) Exclude cut with ID 8291-282929-0007-12994-0 from training. Duration: 25.85
2022-12-23 13:57:12,674 WARNING [train.py:1060] (2/4) Exclude cut with ID 8291-282929-0023-13010-0 from training. Duration: 21.39
2022-12-23 13:57:16,181 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([3.4357, 2.8948, 2.3516, 1.8729, 3.7412, 3.8730, 3.1770, 2.4284],
       device='cuda:2'), covar=tensor([0.0323, 0.0407, 0.0509, 0.0707, 0.0190, 0.0272, 0.0390, 0.0756],
       device='cuda:2'), in_proj_covar=tensor([0.0122, 0.0124, 0.0128, 0.0120, 0.0096, 0.0121, 0.0134, 0.0155],
       device='cuda:2'), out_proj_covar=tensor([0.0001, 0.0002, 0.0002, 0.0001, 0.0001, 0.0001, 0.0002, 0.0002],
       device='cuda:2')
2022-12-23 13:57:23,882 WARNING [train.py:1060] (2/4) Exclude cut with ID 7357-94126-0014-15845-0 from training. Duration: 27.92
2022-12-23 13:57:42,357 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.3530, 1.1544, 1.4944, 2.2360, 1.6378, 2.3494, 0.7394, 1.6969],
       device='cuda:2'), covar=tensor([0.1772, 0.1590, 0.1231, 0.0775, 0.1126, 0.0826, 0.1923, 0.1174],
       device='cuda:2'), in_proj_covar=tensor([0.0103, 0.0117, 0.0135, 0.0148, 0.0107, 0.0141, 0.0131, 0.0113],
       device='cuda:2'), out_proj_covar=tensor([0.0003, 0.0003, 0.0004, 0.0004, 0.0003, 0.0004, 0.0003, 0.0003],
       device='cuda:2')
2022-12-23 13:57:53,279 WARNING [train.py:1060] (2/4) Exclude cut with ID 8631-249866-0039-130165-0_sp0.9 from training. Duration: 20.661125
2022-12-23 13:58:08,844 INFO [zipformer.py:660] (2/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=65101.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 13:58:18,589 WARNING [train.py:1060] (2/4) Exclude cut with ID 7357-94126-0043-15874-0_sp0.9 from training. Duration: 20.07225
2022-12-23 13:58:18,811 INFO [zipformer.py:660] (2/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=65108.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 13:58:23,137 INFO [train.py:894] (2/4) Epoch 19, batch 2000, loss[loss=0.1966, simple_loss=0.2701, pruned_loss=0.06156, over 18424.00 frames. ], tot_loss[loss=0.1882, simple_loss=0.2696, pruned_loss=0.05343, over 3714794.29 frames. ], batch size: 48, lr: 6.04e-03, grad_scale: 8.0
2022-12-23 13:58:24,952 WARNING [train.py:1060] (2/4) Exclude cut with ID 1085-156170-0017-128270-0 from training. Duration: 21.01
2022-12-23 13:58:31,404 INFO [zipformer.py:660] (2/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=65116.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 13:58:38,841 INFO [zipformer.py:660] (2/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=65121.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 13:59:05,242 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 2.657e+02 4.505e+02 5.231e+02 6.101e+02 1.148e+03, threshold=1.046e+03, percent-clipped=0.0
2022-12-23 13:59:35,668 WARNING [train.py:1060] (2/4) Exclude cut with ID 2195-150901-0045-59933-0 from training. Duration: 20.65
2022-12-23 13:59:38,663 INFO [train.py:894] (2/4) Epoch 19, batch 2050, loss[loss=0.2107, simple_loss=0.2899, pruned_loss=0.06572, over 18492.00 frames. ], tot_loss[loss=0.1895, simple_loss=0.2703, pruned_loss=0.05441, over 3715002.27 frames. ], batch size: 78, lr: 6.04e-03, grad_scale: 8.0
2022-12-23 13:59:41,627 WARNING [train.py:1060] (2/4) Exclude cut with ID 5796-66357-0007-116447-0 from training. Duration: 21.46
2022-12-23 13:59:51,015 INFO [zipformer.py:660] (2/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=65169.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 14:00:06,211 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([2.2236, 2.0177, 1.8372, 1.2217, 2.5427, 2.4324, 2.0459, 1.6669],
       device='cuda:2'), covar=tensor([0.0407, 0.0449, 0.0494, 0.0731, 0.0294, 0.0349, 0.0442, 0.0829],
       device='cuda:2'), in_proj_covar=tensor([0.0122, 0.0124, 0.0128, 0.0120, 0.0096, 0.0121, 0.0133, 0.0155],
       device='cuda:2'), out_proj_covar=tensor([0.0001, 0.0002, 0.0002, 0.0001, 0.0001, 0.0001, 0.0002, 0.0002],
       device='cuda:2')
2022-12-23 14:00:28,426 WARNING [train.py:1060] (2/4) Exclude cut with ID 3557-8342-0013-54691-0 from training. Duration: 0.92
2022-12-23 14:00:35,595 WARNING [train.py:1060] (2/4) Exclude cut with ID 8291-282929-0023-13010-0_sp0.9 from training. Duration: 23.7666875
2022-12-23 14:00:49,571 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.7036, 1.4619, 1.5265, 2.1081, 1.7069, 3.3889, 1.4012, 1.5286],
       device='cuda:2'), covar=tensor([0.0841, 0.1804, 0.1112, 0.0823, 0.1378, 0.0268, 0.1405, 0.1579],
       device='cuda:2'), in_proj_covar=tensor([0.0073, 0.0082, 0.0073, 0.0074, 0.0090, 0.0075, 0.0085, 0.0077],
       device='cuda:2'), out_proj_covar=tensor([0.0003, 0.0004, 0.0004, 0.0004, 0.0004, 0.0003, 0.0004, 0.0004],
       device='cuda:2')
2022-12-23 14:00:54,014 INFO [train.py:894] (2/4) Epoch 19, batch 2100, loss[loss=0.1583, simple_loss=0.2391, pruned_loss=0.03877, over 18670.00 frames. ], tot_loss[loss=0.1907, simple_loss=0.2708, pruned_loss=0.05528, over 3714726.03 frames. ], batch size: 48, lr: 6.04e-03, grad_scale: 8.0
2022-12-23 14:01:07,523 WARNING [train.py:1060] (2/4) Exclude cut with ID 8544-281189-0060-101339-0_sp0.9 from training. Duration: 20.861125
2022-12-23 14:01:18,783 WARNING [train.py:1060] (2/4) Exclude cut with ID 3488-65654-0031-41259-0_sp0.9 from training. Duration: 22.711125
2022-12-23 14:01:36,005 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 2.467e+02 4.618e+02 5.775e+02 7.305e+02 1.300e+03, threshold=1.155e+03, percent-clipped=2.0
2022-12-23 14:01:46,568 INFO [zipformer.py:660] (2/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=65245.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 14:01:59,677 WARNING [train.py:1060] (2/4) Exclude cut with ID 6951-79737-0043-132310-0_sp1.1 from training. Duration: 22.986375
2022-12-23 14:02:10,569 INFO [train.py:894] (2/4) Epoch 19, batch 2150, loss[loss=0.2034, simple_loss=0.2918, pruned_loss=0.05752, over 18458.00 frames. ], tot_loss[loss=0.1904, simple_loss=0.2707, pruned_loss=0.05505, over 3714871.07 frames. ], batch size: 54, lr: 6.04e-03, grad_scale: 8.0
2022-12-23 14:02:15,169 WARNING [train.py:1060] (2/4) Exclude cut with ID 8040-260924-0003-80960-0_sp0.9 from training. Duration: 22.07225
2022-12-23 14:02:19,379 WARNING [train.py:1060] (2/4) Exclude cut with ID 7699-105389-0045-26330-0_sp0.9 from training. Duration: 20.3055625
2022-12-23 14:02:23,255 WARNING [train.py:1060] (2/4) Exclude cut with ID 6356-271890-0060-94317-0_sp0.9 from training. Duration: 20.72225
2022-12-23 14:02:40,924 WARNING [train.py:1060] (2/4) Exclude cut with ID 3488-85273-0017-41203-0_sp1.1 from training. Duration: 22.4818125
2022-12-23 14:02:45,268 INFO [scaling.py:679] (2/4) Whitening: num_groups=8, num_channels=192, metric=1.85 vs. limit=2.0
2022-12-23 14:02:47,960 INFO [zipformer.py:660] (2/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=65285.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 14:02:59,885 INFO [zipformer.py:660] (2/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=65293.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 14:03:07,058 WARNING [train.py:1060] (2/4) Exclude cut with ID 4964-30587-0040-44509-0_sp0.9 from training. Duration: 25.0944375
2022-12-23 14:03:11,863 WARNING [train.py:1060] (2/4) Exclude cut with ID 6533-399-0047-104881-0 from training. Duration: 21.515
2022-12-23 14:03:16,594 INFO [zipformer.py:660] (2/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=65304.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 14:03:17,741 WARNING [train.py:1060] (2/4) Exclude cut with ID 7357-94126-0009-15840-0_sp0.9 from training. Duration: 27.02225
2022-12-23 14:03:24,811 WARNING [train.py:1060] (2/4) Exclude cut with ID 432-122774-0010-62480-0_sp0.9 from training. Duration: 22.22225
2022-12-23 14:03:27,775 INFO [train.py:894] (2/4) Epoch 19, batch 2200, loss[loss=0.1999, simple_loss=0.2818, pruned_loss=0.05897, over 18551.00 frames. ], tot_loss[loss=0.1898, simple_loss=0.2698, pruned_loss=0.05489, over 3714117.07 frames. ], batch size: 57, lr: 6.04e-03, grad_scale: 8.0
2022-12-23 14:03:30,657 WARNING [train.py:1060] (2/4) Exclude cut with ID 4964-30587-0085-44554-0_sp0.9 from training. Duration: 20.85
2022-12-23 14:04:00,716 INFO [zipformer.py:660] (2/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=65332.0, num_to_drop=1, layers_to_drop={1}
2022-12-23 14:04:04,805 WARNING [train.py:1060] (2/4) Exclude cut with ID 4295-39940-0007-92567-0 from training. Duration: 21.54
2022-12-23 14:04:09,155 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 2.710e+02 4.175e+02 5.130e+02 6.233e+02 1.122e+03, threshold=1.026e+03, percent-clipped=0.0
2022-12-23 14:04:09,207 WARNING [train.py:1060] (2/4) Exclude cut with ID 4964-30587-0040-44509-0_sp1.1 from training. Duration: 20.5318125
2022-12-23 14:04:17,682 WARNING [train.py:1060] (2/4) Exclude cut with ID 7255-291500-0012-134311-0_sp0.9 from training. Duration: 21.9333125
2022-12-23 14:04:21,768 INFO [zipformer.py:660] (2/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=65346.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 14:04:30,292 INFO [zipformer.py:660] (2/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=65352.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 14:04:40,320 INFO [zipformer.py:660] (2/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=65359.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 14:04:43,094 INFO [train.py:894] (2/4) Epoch 19, batch 2250, loss[loss=0.1722, simple_loss=0.2466, pruned_loss=0.04892, over 18455.00 frames. ], tot_loss[loss=0.1904, simple_loss=0.2705, pruned_loss=0.0552, over 3714057.33 frames. ], batch size: 42, lr: 6.03e-03, grad_scale: 8.0
2022-12-23 14:05:09,291 WARNING [train.py:1060] (2/4) Exclude cut with ID 8631-249866-0025-130151-0_sp0.9 from training. Duration: 21.7944375
2022-12-23 14:05:20,891 WARNING [train.py:1060] (2/4) Exclude cut with ID 8291-282929-0002-12989-0_sp0.9 from training. Duration: 22.4666875
2022-12-23 14:05:24,748 INFO [scaling.py:679] (2/4) Whitening: num_groups=8, num_channels=96, metric=1.16 vs. limit=2.0
2022-12-23 14:05:29,167 WARNING [train.py:1060] (2/4) Exclude cut with ID 6121-9014-0076-24124-0 from training. Duration: 21.635
2022-12-23 14:05:35,092 WARNING [train.py:1060] (2/4) Exclude cut with ID 6121-9014-0076-24124-0_sp0.9 from training. Duration: 24.038875
2022-12-23 14:05:44,118 INFO [zipformer.py:660] (2/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=65401.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 14:05:52,555 INFO [zipformer.py:660] (2/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=65407.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 14:05:54,030 INFO [zipformer.py:660] (2/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=65408.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 14:05:58,204 INFO [train.py:894] (2/4) Epoch 19, batch 2300, loss[loss=0.1747, simple_loss=0.2632, pruned_loss=0.04308, over 18575.00 frames. ], tot_loss[loss=0.1897, simple_loss=0.2702, pruned_loss=0.05464, over 3713754.92 frames. ], batch size: 56, lr: 6.03e-03, grad_scale: 8.0
2022-12-23 14:05:58,396 INFO [zipformer.py:660] (2/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=65411.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 14:06:20,080 WARNING [train.py:1060] (2/4) Exclude cut with ID 7255-291500-0009-134308-0_sp1.1 from training. Duration: 21.786375
2022-12-23 14:06:30,499 WARNING [train.py:1060] (2/4) Exclude cut with ID 8291-282929-0002-12989-0 from training. Duration: 20.22
2022-12-23 14:06:39,237 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 2.902e+02 4.557e+02 5.632e+02 7.135e+02 1.827e+03, threshold=1.126e+03, percent-clipped=4.0
2022-12-23 14:06:56,380 INFO [zipformer.py:660] (2/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=65449.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 14:07:06,642 INFO [zipformer.py:660] (2/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=65456.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 14:07:14,131 INFO [train.py:894] (2/4) Epoch 19, batch 2350, loss[loss=0.1995, simple_loss=0.2857, pruned_loss=0.05672, over 18516.00 frames. ], tot_loss[loss=0.1893, simple_loss=0.2699, pruned_loss=0.05437, over 3713768.70 frames. ], batch size: 55, lr: 6.03e-03, grad_scale: 8.0
2022-12-23 14:08:30,813 INFO [train.py:894] (2/4) Epoch 19, batch 2400, loss[loss=0.213, simple_loss=0.2842, pruned_loss=0.07092, over 18671.00 frames. ], tot_loss[loss=0.1903, simple_loss=0.2705, pruned_loss=0.05499, over 3714183.10 frames. ], batch size: 62, lr: 6.03e-03, grad_scale: 8.0
2022-12-23 14:08:30,827 WARNING [train.py:1060] (2/4) Exclude cut with ID 6951-79737-0043-132310-0 from training. Duration: 25.285
2022-12-23 14:09:02,189 INFO [zipformer.py:660] (2/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=65532.0, num_to_drop=1, layers_to_drop={1}
2022-12-23 14:09:10,951 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 3.058e+02 4.637e+02 5.730e+02 7.137e+02 1.631e+03, threshold=1.146e+03, percent-clipped=2.0
2022-12-23 14:09:35,639 WARNING [train.py:1060] (2/4) Exclude cut with ID 811-130148-0001-63453-0_sp0.9 from training. Duration: 20.861125
2022-12-23 14:09:46,575 INFO [train.py:894] (2/4) Epoch 19, batch 2450, loss[loss=0.1797, simple_loss=0.2641, pruned_loss=0.04762, over 18520.00 frames. ], tot_loss[loss=0.1904, simple_loss=0.2708, pruned_loss=0.05502, over 3714360.54 frames. ], batch size: 52, lr: 6.02e-03, grad_scale: 8.0
2022-12-23 14:09:48,278 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([2.5065, 3.5708, 3.5778, 1.4525, 3.7072, 2.7655, 0.5234, 2.4734],
       device='cuda:2'), covar=tensor([0.1997, 0.1217, 0.1393, 0.3557, 0.0960, 0.0989, 0.5109, 0.1441],
       device='cuda:2'), in_proj_covar=tensor([0.0145, 0.0140, 0.0157, 0.0124, 0.0142, 0.0114, 0.0145, 0.0113],
       device='cuda:2'), out_proj_covar=tensor([0.0003, 0.0002, 0.0003, 0.0002, 0.0002, 0.0002, 0.0003, 0.0002],
       device='cuda:2')
2022-12-23 14:09:57,692 WARNING [train.py:1060] (2/4) Exclude cut with ID 6010-56788-0055-90261-0 from training. Duration: 20.88
2022-12-23 14:10:31,134 WARNING [train.py:1060] (2/4) Exclude cut with ID 7357-94126-0045-15876-0_sp0.9 from training. Duration: 23.4166875
2022-12-23 14:10:35,885 INFO [zipformer.py:660] (2/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=65593.0, num_to_drop=1, layers_to_drop={2}
2022-12-23 14:10:47,560 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.5831, 1.2530, 1.8293, 3.1012, 2.2137, 2.5355, 0.8185, 2.1348],
       device='cuda:2'), covar=tensor([0.1943, 0.1757, 0.1583, 0.0678, 0.1158, 0.1155, 0.2366, 0.1206],
       device='cuda:2'), in_proj_covar=tensor([0.0102, 0.0115, 0.0134, 0.0147, 0.0106, 0.0140, 0.0130, 0.0112],
       device='cuda:2'), out_proj_covar=tensor([0.0003, 0.0003, 0.0004, 0.0004, 0.0003, 0.0004, 0.0003, 0.0003],
       device='cuda:2')
2022-12-23 14:11:04,066 INFO [train.py:894] (2/4) Epoch 19, batch 2500, loss[loss=0.1881, simple_loss=0.2733, pruned_loss=0.05143, over 18517.00 frames. ], tot_loss[loss=0.1903, simple_loss=0.2708, pruned_loss=0.05493, over 3714201.42 frames. ], batch size: 52, lr: 6.02e-03, grad_scale: 8.0
2022-12-23 14:11:33,329 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([2.4471, 2.1466, 1.8057, 0.8747, 1.8083, 1.9969, 1.7021, 1.9664],
       device='cuda:2'), covar=tensor([0.0609, 0.0500, 0.1111, 0.1524, 0.1091, 0.1353, 0.1483, 0.0736],
       device='cuda:2'), in_proj_covar=tensor([0.0174, 0.0184, 0.0206, 0.0192, 0.0210, 0.0201, 0.0214, 0.0202],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-23 14:11:36,339 INFO [zipformer.py:660] (2/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=65632.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 14:11:44,824 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 2.293e+02 4.318e+02 5.164e+02 6.581e+02 1.348e+03, threshold=1.033e+03, percent-clipped=4.0
2022-12-23 14:11:44,910 WARNING [train.py:1060] (2/4) Exclude cut with ID 7255-291500-0006-134305-0 from training. Duration: 21.24
2022-12-23 14:11:46,282 WARNING [train.py:1060] (2/4) Exclude cut with ID 6533-399-0047-104881-0_sp0.9 from training. Duration: 23.9055625
2022-12-23 14:11:49,381 INFO [zipformer.py:660] (2/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=65641.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 14:12:19,898 INFO [train.py:894] (2/4) Epoch 19, batch 2550, loss[loss=0.1819, simple_loss=0.2511, pruned_loss=0.05632, over 18535.00 frames. ], tot_loss[loss=0.1897, simple_loss=0.2699, pruned_loss=0.05479, over 3713778.54 frames. ], batch size: 41, lr: 6.02e-03, grad_scale: 8.0
2022-12-23 14:12:19,968 WARNING [train.py:1060] (2/4) Exclude cut with ID 6758-72288-0033-108368-0_sp0.9 from training. Duration: 25.988875
2022-12-23 14:12:27,843 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([4.4130, 3.8391, 3.7854, 4.2451, 3.9557, 3.9031, 4.5542, 1.4378],
       device='cuda:2'), covar=tensor([0.0723, 0.0680, 0.0718, 0.0881, 0.1481, 0.1193, 0.0563, 0.5058],
       device='cuda:2'), in_proj_covar=tensor([0.0347, 0.0227, 0.0239, 0.0269, 0.0328, 0.0275, 0.0291, 0.0284],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0001, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-23 14:12:28,998 WARNING [train.py:1060] (2/4) Exclude cut with ID 7255-291500-0001-134300-0_sp0.9 from training. Duration: 20.67225
2022-12-23 14:12:48,991 INFO [zipformer.py:660] (2/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=65680.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 14:12:52,364 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.7916, 1.7663, 1.4962, 1.7248, 2.0092, 1.9656, 1.9826, 1.5170],
       device='cuda:2'), covar=tensor([0.0315, 0.0281, 0.0428, 0.0221, 0.0183, 0.0353, 0.0273, 0.0303],
       device='cuda:2'), in_proj_covar=tensor([0.0093, 0.0127, 0.0152, 0.0124, 0.0117, 0.0119, 0.0097, 0.0127],
       device='cuda:2'), out_proj_covar=tensor([7.4728e-05, 1.0133e-04, 1.2603e-04, 9.9602e-05, 9.5027e-05, 9.1634e-05,
        7.6200e-05, 9.9990e-05], device='cuda:2')
2022-12-23 14:12:55,188 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.4806, 1.7281, 1.3124, 2.0593, 2.1492, 1.5344, 1.2419, 1.2332],
       device='cuda:2'), covar=tensor([0.1902, 0.1765, 0.1704, 0.0995, 0.1289, 0.1115, 0.2177, 0.1515],
       device='cuda:2'), in_proj_covar=tensor([0.0243, 0.0222, 0.0210, 0.0196, 0.0259, 0.0192, 0.0218, 0.0193],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-23 14:13:07,748 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.3294, 1.5713, 1.3122, 1.9109, 1.6979, 1.5111, 1.0253, 1.2134],
       device='cuda:2'), covar=tensor([0.1985, 0.1739, 0.1619, 0.1039, 0.1164, 0.1156, 0.2147, 0.1583],
       device='cuda:2'), in_proj_covar=tensor([0.0243, 0.0222, 0.0210, 0.0196, 0.0259, 0.0192, 0.0218, 0.0193],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-23 14:13:17,286 WARNING [train.py:1060] (2/4) Exclude cut with ID 3488-85273-0038-41224-0 from training. Duration: 20.34
2022-12-23 14:13:35,174 INFO [train.py:894] (2/4) Epoch 19, batch 2600, loss[loss=0.1743, simple_loss=0.2587, pruned_loss=0.04494, over 18570.00 frames. ], tot_loss[loss=0.1895, simple_loss=0.2698, pruned_loss=0.05459, over 3714351.65 frames. ], batch size: 49, lr: 6.02e-03, grad_scale: 8.0
2022-12-23 14:13:35,436 INFO [zipformer.py:660] (2/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=65711.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 14:14:16,261 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 2.847e+02 4.482e+02 5.428e+02 6.828e+02 2.111e+03, threshold=1.086e+03, percent-clipped=5.0
2022-12-23 14:14:29,313 INFO [scaling.py:679] (2/4) Whitening: num_groups=8, num_channels=96, metric=1.54 vs. limit=2.0
2022-12-23 14:14:29,802 WARNING [train.py:1060] (2/4) Exclude cut with ID 7357-94126-0026-15857-0_sp0.9 from training. Duration: 25.061125
2022-12-23 14:14:42,554 WARNING [train.py:1060] (2/4) Exclude cut with ID 3033-130750-0096-55598-0 from training. Duration: 0.83
2022-12-23 14:14:48,294 INFO [zipformer.py:660] (2/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=65759.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 14:14:50,975 INFO [train.py:894] (2/4) Epoch 19, batch 2650, loss[loss=0.1581, simple_loss=0.2337, pruned_loss=0.04123, over 18411.00 frames. ], tot_loss[loss=0.1891, simple_loss=0.2692, pruned_loss=0.05454, over 3714240.47 frames. ], batch size: 42, lr: 6.01e-03, grad_scale: 8.0
2022-12-23 14:14:59,278 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([2.3650, 2.1465, 1.6893, 2.0249, 1.8536, 2.0121, 1.8967, 2.0974],
       device='cuda:2'), covar=tensor([0.1937, 0.2848, 0.1885, 0.2442, 0.3176, 0.1079, 0.2734, 0.0978],
       device='cuda:2'), in_proj_covar=tensor([0.0293, 0.0290, 0.0244, 0.0350, 0.0270, 0.0227, 0.0287, 0.0211],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0003, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-23 14:15:06,303 WARNING [train.py:1060] (2/4) Exclude cut with ID 3488-85273-0017-41203-0 from training. Duration: 24.73
2022-12-23 14:15:17,892 WARNING [train.py:1060] (2/4) Exclude cut with ID 7255-291500-0009-134308-0 from training. Duration: 23.965
2022-12-23 14:15:25,843 WARNING [train.py:1060] (2/4) Exclude cut with ID 6978-92210-0030-146996-0_sp0.9 from training. Duration: 22.088875
2022-12-23 14:15:43,002 WARNING [train.py:1060] (2/4) Exclude cut with ID 7255-291500-0006-134305-0_sp0.9 from training. Duration: 23.6
2022-12-23 14:16:01,278 INFO [scaling.py:679] (2/4) Whitening: num_groups=8, num_channels=96, metric=1.44 vs. limit=2.0
2022-12-23 14:16:06,598 INFO [train.py:894] (2/4) Epoch 19, batch 2700, loss[loss=0.2197, simple_loss=0.3086, pruned_loss=0.06539, over 18470.00 frames. ], tot_loss[loss=0.1896, simple_loss=0.2699, pruned_loss=0.05469, over 3714488.25 frames. ], batch size: 64, lr: 6.01e-03, grad_scale: 8.0
2022-12-23 14:16:19,893 INFO [scaling.py:679] (2/4) Whitening: num_groups=8, num_channels=96, metric=1.49 vs. limit=2.0
2022-12-23 14:16:28,685 INFO [scaling.py:679] (2/4) Whitening: num_groups=1, num_channels=384, metric=2.55 vs. limit=5.0
2022-12-23 14:16:43,917 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.4904, 1.3947, 1.0578, 1.6958, 1.6610, 3.0036, 1.3878, 1.6224],
       device='cuda:2'), covar=tensor([0.0816, 0.1780, 0.1117, 0.0883, 0.1373, 0.0258, 0.1354, 0.1441],
       device='cuda:2'), in_proj_covar=tensor([0.0073, 0.0083, 0.0073, 0.0075, 0.0090, 0.0076, 0.0086, 0.0078],
       device='cuda:2'), out_proj_covar=tensor([0.0004, 0.0004, 0.0004, 0.0004, 0.0004, 0.0003, 0.0004, 0.0004],
       device='cuda:2')
2022-12-23 14:16:47,933 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 2.953e+02 4.141e+02 5.298e+02 6.563e+02 1.106e+03, threshold=1.060e+03, percent-clipped=1.0
2022-12-23 14:16:48,369 INFO [zipformer.py:660] (2/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=65838.0, num_to_drop=1, layers_to_drop={0}
2022-12-23 14:16:58,322 INFO [scaling.py:679] (2/4) Whitening: num_groups=8, num_channels=96, metric=1.13 vs. limit=2.0
2022-12-23 14:17:23,086 INFO [train.py:894] (2/4) Epoch 19, batch 2750, loss[loss=0.1471, simple_loss=0.2302, pruned_loss=0.03197, over 18582.00 frames. ], tot_loss[loss=0.1893, simple_loss=0.2698, pruned_loss=0.05441, over 3715274.84 frames. ], batch size: 41, lr: 6.01e-03, grad_scale: 8.0
2022-12-23 14:17:24,531 WARNING [train.py:1060] (2/4) Exclude cut with ID 8291-282929-0024-13011-0 from training. Duration: 23.795
2022-12-23 14:17:39,857 WARNING [train.py:1060] (2/4) Exclude cut with ID 8631-249866-0030-130156-0_sp1.1 from training. Duration: 21.5409375
2022-12-23 14:17:42,727 WARNING [train.py:1060] (2/4) Exclude cut with ID 6978-92210-0019-146985-0_sp0.9 from training. Duration: 24.97775
2022-12-23 14:17:53,627 WARNING [train.py:1060] (2/4) Exclude cut with ID 1085-156170-0017-128270-0_sp0.9 from training. Duration: 23.3444375
2022-12-23 14:18:04,862 INFO [zipformer.py:660] (2/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=65888.0, num_to_drop=1, layers_to_drop={2}
2022-12-23 14:18:20,173 WARNING [train.py:1060] (2/4) Exclude cut with ID 6010-56788-0055-90261-0_sp0.9 from training. Duration: 23.2
2022-12-23 14:18:21,994 INFO [zipformer.py:660] (2/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=65899.0, num_to_drop=1, layers_to_drop={0}
2022-12-23 14:18:27,468 WARNING [train.py:1060] (2/4) Exclude cut with ID 5653-46179-0060-117930-0_sp0.9 from training. Duration: 21.17225
2022-12-23 14:18:39,156 INFO [train.py:894] (2/4) Epoch 19, batch 2800, loss[loss=0.1971, simple_loss=0.28, pruned_loss=0.05713, over 18590.00 frames. ], tot_loss[loss=0.1898, simple_loss=0.27, pruned_loss=0.05477, over 3716121.56 frames. ], batch size: 51, lr: 6.01e-03, grad_scale: 8.0
2022-12-23 14:18:48,162 WARNING [train.py:1060] (2/4) Exclude cut with ID 7255-291500-0008-134307-0_sp0.9 from training. Duration: 24.6555625
2022-12-23 14:19:20,967 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 2.932e+02 4.793e+02 6.010e+02 7.523e+02 1.271e+03, threshold=1.202e+03, percent-clipped=4.0
2022-12-23 14:19:25,621 INFO [zipformer.py:660] (2/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=65941.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 14:19:41,726 INFO [scaling.py:679] (2/4) Whitening: num_groups=8, num_channels=192, metric=1.82 vs. limit=2.0
2022-12-23 14:19:43,696 WARNING [train.py:1060] (2/4) Exclude cut with ID 3488-65654-0031-41259-0 from training. Duration: 20.44
2022-12-23 14:19:56,140 INFO [train.py:894] (2/4) Epoch 19, batch 2850, loss[loss=0.1889, simple_loss=0.2735, pruned_loss=0.0522, over 18646.00 frames. ], tot_loss[loss=0.1897, simple_loss=0.2701, pruned_loss=0.05466, over 3715608.22 frames. ], batch size: 60, lr: 6.01e-03, grad_scale: 8.0
2022-12-23 14:19:57,822 WARNING [train.py:1060] (2/4) Exclude cut with ID 6951-79737-0018-132285-0_sp0.9 from training. Duration: 23.45
2022-12-23 14:19:59,611 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.1278, 1.2029, 0.8452, 1.3393, 1.3908, 2.4356, 1.1587, 1.3793],
       device='cuda:2'), covar=tensor([0.0960, 0.1929, 0.1138, 0.0993, 0.1588, 0.0363, 0.1587, 0.1614],
       device='cuda:2'), in_proj_covar=tensor([0.0073, 0.0083, 0.0073, 0.0075, 0.0090, 0.0076, 0.0085, 0.0077],
       device='cuda:2'), out_proj_covar=tensor([0.0003, 0.0004, 0.0004, 0.0004, 0.0004, 0.0003, 0.0004, 0.0004],
       device='cuda:2')
2022-12-23 14:20:27,894 WARNING [train.py:1060] (2/4) Exclude cut with ID 6945-60535-0076-12784-0_sp0.9 from training. Duration: 20.52225
2022-12-23 14:20:35,751 WARNING [train.py:1060] (2/4) Exclude cut with ID 7255-291500-0008-134307-0 from training. Duration: 22.19
2022-12-23 14:20:37,182 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([3.8796, 3.3692, 3.3232, 3.7679, 3.5327, 3.3650, 3.9735, 1.1670],
       device='cuda:2'), covar=tensor([0.0857, 0.0804, 0.0759, 0.0861, 0.1601, 0.1358, 0.0754, 0.5322],
       device='cuda:2'), in_proj_covar=tensor([0.0342, 0.0226, 0.0235, 0.0264, 0.0321, 0.0269, 0.0288, 0.0281],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0001, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-23 14:20:39,878 INFO [zipformer.py:660] (2/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=65989.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 14:20:43,466 INFO [scaling.py:679] (2/4) Whitening: num_groups=8, num_channels=96, metric=1.43 vs. limit=2.0
2022-12-23 14:20:45,949 WARNING [train.py:1060] (2/4) Exclude cut with ID 7357-94126-0014-15845-0_sp1.1 from training. Duration: 25.3818125
2022-12-23 14:21:03,258 WARNING [train.py:1060] (2/4) Exclude cut with ID 6951-79737-0043-132310-0_sp0.9 from training. Duration: 28.0944375
2022-12-23 14:21:09,511 WARNING [train.py:1060] (2/4) Exclude cut with ID 2195-150901-0045-59933-0_sp0.9 from training. Duration: 22.9444375
2022-12-23 14:21:15,603 INFO [train.py:894] (2/4) Epoch 19, batch 2900, loss[loss=0.1768, simple_loss=0.2583, pruned_loss=0.04768, over 18704.00 frames. ], tot_loss[loss=0.1885, simple_loss=0.269, pruned_loss=0.05398, over 3715098.31 frames. ], batch size: 50, lr: 6.00e-03, grad_scale: 8.0
2022-12-23 14:21:19,013 WARNING [train.py:1060] (2/4) Exclude cut with ID 8291-282929-0024-13011-0_sp1.1 from training. Duration: 21.6318125
2022-12-23 14:21:31,622 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.4041, 1.7579, 1.3088, 2.0178, 2.1727, 1.5008, 1.1405, 1.2811],
       device='cuda:2'), covar=tensor([0.1960, 0.1697, 0.1707, 0.1010, 0.1248, 0.1087, 0.2366, 0.1518],
       device='cuda:2'), in_proj_covar=tensor([0.0245, 0.0222, 0.0212, 0.0197, 0.0259, 0.0193, 0.0220, 0.0194],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-23 14:21:36,742 WARNING [train.py:1060] (2/4) Exclude cut with ID 8631-249866-0030-130156-0 from training. Duration: 23.695
2022-12-23 14:21:47,723 INFO [scaling.py:679] (2/4) Whitening: num_groups=8, num_channels=192, metric=2.00 vs. limit=2.0
2022-12-23 14:21:55,733 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 2.668e+02 4.307e+02 5.420e+02 6.494e+02 1.228e+03, threshold=1.084e+03, percent-clipped=1.0
2022-12-23 14:22:01,911 INFO [scaling.py:679] (2/4) Whitening: num_groups=1, num_channels=384, metric=4.70 vs. limit=5.0
2022-12-23 14:22:02,344 WARNING [train.py:1060] (2/4) Exclude cut with ID 7699-105389-0094-26379-0 from training. Duration: 23.955
2022-12-23 14:22:04,130 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.6462, 1.6065, 1.6088, 1.6806, 1.2965, 3.9521, 1.6514, 2.3519],
       device='cuda:2'), covar=tensor([0.4435, 0.2689, 0.2585, 0.2842, 0.1595, 0.0241, 0.1710, 0.0850],
       device='cuda:2'), in_proj_covar=tensor([0.0133, 0.0117, 0.0127, 0.0121, 0.0103, 0.0097, 0.0092, 0.0089],
       device='cuda:2'), out_proj_covar=tensor([0.0005, 0.0005, 0.0005, 0.0005, 0.0004, 0.0003, 0.0004, 0.0004],
       device='cuda:2')
2022-12-23 14:22:09,105 INFO [scaling.py:679] (2/4) Whitening: num_groups=8, num_channels=96, metric=1.41 vs. limit=2.0
2022-12-23 14:22:30,196 INFO [train.py:894] (2/4) Epoch 19, batch 2950, loss[loss=0.1758, simple_loss=0.2568, pruned_loss=0.04744, over 18540.00 frames. ], tot_loss[loss=0.1884, simple_loss=0.2692, pruned_loss=0.05382, over 3713752.14 frames. ], batch size: 47, lr: 6.00e-03, grad_scale: 8.0
2022-12-23 14:22:36,401 WARNING [train.py:1060] (2/4) Exclude cut with ID 8291-282929-0024-13011-0_sp0.9 from training. Duration: 26.438875
2022-12-23 14:23:21,799 WARNING [train.py:1060] (2/4) Exclude cut with ID 7699-105389-0021-26306-0_sp0.9 from training. Duration: 21.2444375
2022-12-23 14:23:21,823 WARNING [train.py:1060] (2/4) Exclude cut with ID 7357-94126-0014-15845-0_sp0.9 from training. Duration: 31.02225
2022-12-23 14:23:31,803 WARNING [train.py:1060] (2/4) Exclude cut with ID 432-122774-0017-62487-0 from training. Duration: 22.395
2022-12-23 14:23:44,320 INFO [train.py:894] (2/4) Epoch 19, batch 3000, loss[loss=0.1807, simple_loss=0.2624, pruned_loss=0.04946, over 18617.00 frames. ], tot_loss[loss=0.1894, simple_loss=0.27, pruned_loss=0.05441, over 3713664.74 frames. ], batch size: 53, lr: 6.00e-03, grad_scale: 8.0
2022-12-23 14:23:44,320 INFO [train.py:919] (2/4) Computing validation loss
2022-12-23 14:23:48,653 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([2.6838, 2.1875, 1.7869, 2.5403, 1.9460, 2.1559, 2.0278, 2.5917],
       device='cuda:2'), covar=tensor([0.2004, 0.3534, 0.2017, 0.2742, 0.3929, 0.1075, 0.3318, 0.0912],
       device='cuda:2'), in_proj_covar=tensor([0.0291, 0.0287, 0.0245, 0.0348, 0.0269, 0.0226, 0.0285, 0.0210],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0003, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-23 14:23:55,322 INFO [train.py:928] (2/4) Epoch 19, validation: loss=0.1652, simple_loss=0.263, pruned_loss=0.03365, over 944034.00 frames. 
2022-12-23 14:23:55,323 INFO [train.py:929] (2/4) Maximum memory allocated so far is 24680MB
2022-12-23 14:24:00,893 WARNING [train.py:1060] (2/4) Exclude cut with ID 7357-94126-0045-15876-0 from training. Duration: 21.075
2022-12-23 14:24:05,215 WARNING [train.py:1060] (2/4) Exclude cut with ID 6482-98857-0025-147532-0_sp0.9 from training. Duration: 20.0055625
2022-12-23 14:24:06,481 WARNING [train.py:1060] (2/4) Exclude cut with ID 6951-79737-0037-132304-0_sp0.9 from training. Duration: 22.05
2022-12-23 14:24:06,491 WARNING [train.py:1060] (2/4) Exclude cut with ID 7255-291500-0003-134302-0 from training. Duration: 26.8349375
2022-12-23 14:24:09,896 WARNING [train.py:1060] (2/4) Exclude cut with ID 7357-94126-0009-15840-0_sp1.1 from training. Duration: 22.1090625
2022-12-23 14:24:17,288 WARNING [train.py:1060] (2/4) Exclude cut with ID 7699-105389-0094-26379-0_sp0.9 from training. Duration: 26.6166875
2022-12-23 14:24:33,094 WARNING [train.py:1060] (2/4) Exclude cut with ID 2046-178027-0000-53705-0_sp0.9 from training. Duration: 20.3055625
2022-12-23 14:24:34,460 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 2.524e+02 4.579e+02 5.252e+02 6.316e+02 1.044e+03, threshold=1.050e+03, percent-clipped=0.0
2022-12-23 14:24:45,291 INFO [zipformer.py:660] (2/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=66144.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 14:24:57,016 WARNING [train.py:1060] (2/4) Exclude cut with ID 7205-50138-0008-5373-0_sp0.9 from training. Duration: 20.7
2022-12-23 14:25:09,453 INFO [zipformer.py:660] (2/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=66160.0, num_to_drop=1, layers_to_drop={1}
2022-12-23 14:25:10,768 INFO [train.py:894] (2/4) Epoch 19, batch 3050, loss[loss=0.1885, simple_loss=0.2648, pruned_loss=0.05612, over 18541.00 frames. ], tot_loss[loss=0.1894, simple_loss=0.2701, pruned_loss=0.05437, over 3713513.08 frames. ], batch size: 47, lr: 6.00e-03, grad_scale: 8.0
2022-12-23 14:25:38,966 WARNING [train.py:1060] (2/4) Exclude cut with ID 6978-92210-0019-146985-0 from training. Duration: 22.48
2022-12-23 14:25:51,606 INFO [zipformer.py:660] (2/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=66188.0, num_to_drop=1, layers_to_drop={1}
2022-12-23 14:25:54,769 WARNING [train.py:1060] (2/4) Exclude cut with ID 7255-291500-0003-134302-0_sp0.9 from training. Duration: 29.816625
2022-12-23 14:26:00,739 INFO [zipformer.py:660] (2/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=66194.0, num_to_drop=1, layers_to_drop={2}
2022-12-23 14:26:12,376 WARNING [train.py:1060] (2/4) Exclude cut with ID 7255-291500-0005-134304-0_sp1.1 from training. Duration: 22.7590625
2022-12-23 14:26:18,798 WARNING [train.py:1060] (2/4) Exclude cut with ID 7357-94126-0026-15857-0 from training. Duration: 22.555
2022-12-23 14:26:19,096 INFO [zipformer.py:660] (2/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=66205.0, num_to_drop=1, layers_to_drop={2}
2022-12-23 14:26:27,926 INFO [train.py:894] (2/4) Epoch 19, batch 3100, loss[loss=0.1963, simple_loss=0.276, pruned_loss=0.05836, over 18659.00 frames. ], tot_loss[loss=0.1891, simple_loss=0.2696, pruned_loss=0.05424, over 3713769.46 frames. ], batch size: 48, lr: 5.99e-03, grad_scale: 8.0
2022-12-23 14:26:38,744 WARNING [train.py:1060] (2/4) Exclude cut with ID 1250-135782-0005-25975-0_sp0.9 from training. Duration: 21.688875
2022-12-23 14:26:43,379 INFO [zipformer.py:660] (2/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=66221.0, num_to_drop=1, layers_to_drop={0}
2022-12-23 14:27:01,265 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([3.0247, 5.2833, 4.7052, 2.4558, 5.2677, 4.2151, 0.7337, 3.3749],
       device='cuda:2'), covar=tensor([0.2091, 0.0913, 0.1330, 0.3222, 0.0782, 0.0744, 0.5825, 0.1393],
       device='cuda:2'), in_proj_covar=tensor([0.0149, 0.0144, 0.0159, 0.0125, 0.0146, 0.0115, 0.0147, 0.0116],
       device='cuda:2'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0002, 0.0003, 0.0002, 0.0003, 0.0002],
       device='cuda:2')
2022-12-23 14:27:05,964 INFO [zipformer.py:660] (2/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=66236.0, num_to_drop=1, layers_to_drop={1}
2022-12-23 14:27:08,463 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 2.484e+02 4.466e+02 5.278e+02 6.458e+02 1.421e+03, threshold=1.056e+03, percent-clipped=3.0
2022-12-23 14:27:15,234 WARNING [train.py:1060] (2/4) Exclude cut with ID 3488-85273-0038-41224-0_sp0.9 from training. Duration: 22.6
2022-12-23 14:27:38,846 INFO [zipformer.py:660] (2/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=66258.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 14:27:43,446 INFO [train.py:894] (2/4) Epoch 19, batch 3150, loss[loss=0.2051, simple_loss=0.2673, pruned_loss=0.07149, over 18563.00 frames. ], tot_loss[loss=0.189, simple_loss=0.2693, pruned_loss=0.05438, over 3713741.68 frames. ], batch size: 49, lr: 5.99e-03, grad_scale: 8.0
2022-12-23 14:27:54,073 WARNING [train.py:1060] (2/4) Exclude cut with ID 7357-94126-0009-15840-0 from training. Duration: 24.32
2022-12-23 14:28:50,574 WARNING [train.py:1060] (2/4) Exclude cut with ID 8291-276745-0093-13116-0_sp0.9 from training. Duration: 21.061125
2022-12-23 14:28:59,664 INFO [train.py:894] (2/4) Epoch 19, batch 3200, loss[loss=0.2451, simple_loss=0.3073, pruned_loss=0.09143, over 18597.00 frames. ], tot_loss[loss=0.1882, simple_loss=0.2682, pruned_loss=0.05405, over 3713633.91 frames. ], batch size: 175, lr: 5.99e-03, grad_scale: 8.0
2022-12-23 14:29:02,890 WARNING [train.py:1060] (2/4) Exclude cut with ID 7357-94126-0024-15855-0_sp0.9 from training. Duration: 20.32225
2022-12-23 14:29:12,274 INFO [zipformer.py:660] (2/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=66319.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 14:29:16,424 WARNING [train.py:1060] (2/4) Exclude cut with ID 3033-130750-0096-55598-0_sp1.1 from training. Duration: 0.7545625
2022-12-23 14:29:32,264 WARNING [train.py:1060] (2/4) Exclude cut with ID 4295-39940-0007-92567-0_sp0.9 from training. Duration: 23.9333125
2022-12-23 14:29:40,642 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 3.023e+02 4.549e+02 5.620e+02 6.931e+02 1.616e+03, threshold=1.124e+03, percent-clipped=4.0
2022-12-23 14:29:44,980 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([2.6256, 3.6346, 3.5130, 1.4291, 3.7463, 2.7971, 0.5760, 2.4489],
       device='cuda:2'), covar=tensor([0.1896, 0.1226, 0.1400, 0.3422, 0.0932, 0.0961, 0.4899, 0.1461],
       device='cuda:2'), in_proj_covar=tensor([0.0147, 0.0142, 0.0157, 0.0123, 0.0145, 0.0114, 0.0145, 0.0114],
       device='cuda:2'), out_proj_covar=tensor([0.0003, 0.0002, 0.0003, 0.0002, 0.0003, 0.0002, 0.0003, 0.0002],
       device='cuda:2')
2022-12-23 14:30:04,037 WARNING [train.py:1060] (2/4) Exclude cut with ID 7255-291500-0008-134307-0_sp1.1 from training. Duration: 20.17275
2022-12-23 14:30:04,289 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([4.6027, 3.9521, 3.9788, 4.4872, 4.1789, 4.1042, 4.7231, 1.3291],
       device='cuda:2'), covar=tensor([0.0711, 0.0707, 0.0618, 0.0779, 0.1422, 0.1129, 0.0543, 0.5170],
       device='cuda:2'), in_proj_covar=tensor([0.0347, 0.0228, 0.0238, 0.0268, 0.0327, 0.0271, 0.0293, 0.0282],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0001, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-23 14:30:11,106 WARNING [train.py:1060] (2/4) Exclude cut with ID 6978-92210-0019-146985-0_sp1.1 from training. Duration: 20.436375
2022-12-23 14:30:13,764 INFO [train.py:894] (2/4) Epoch 19, batch 3250, loss[loss=0.191, simple_loss=0.2811, pruned_loss=0.05047, over 18688.00 frames. ], tot_loss[loss=0.189, simple_loss=0.2691, pruned_loss=0.05446, over 3712872.87 frames. ], batch size: 99, lr: 5.99e-03, grad_scale: 8.0
2022-12-23 14:31:28,781 WARNING [train.py:1060] (2/4) Exclude cut with ID 4234-40345-0022-142709-0_sp0.9 from training. Duration: 23.1055625
2022-12-23 14:31:30,471 INFO [train.py:894] (2/4) Epoch 19, batch 3300, loss[loss=0.1664, simple_loss=0.2541, pruned_loss=0.03934, over 18468.00 frames. ], tot_loss[loss=0.1888, simple_loss=0.2689, pruned_loss=0.05433, over 3713111.84 frames. ], batch size: 68, lr: 5.99e-03, grad_scale: 8.0
2022-12-23 14:31:30,526 WARNING [train.py:1060] (2/4) Exclude cut with ID 8291-282929-0007-12994-0_sp1.1 from training. Duration: 23.5
2022-12-23 14:31:42,048 WARNING [train.py:1060] (2/4) Exclude cut with ID 7255-291500-0009-134308-0_sp0.9 from training. Duration: 26.62775
2022-12-23 14:31:46,825 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([3.0889, 3.1498, 1.9863, 1.6865, 3.5188, 3.5632, 3.0175, 2.5278],
       device='cuda:2'), covar=tensor([0.0387, 0.0346, 0.0591, 0.0733, 0.0231, 0.0316, 0.0430, 0.0681],
       device='cuda:2'), in_proj_covar=tensor([0.0126, 0.0128, 0.0132, 0.0124, 0.0099, 0.0124, 0.0137, 0.0159],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0001, 0.0001, 0.0002, 0.0002],
       device='cuda:2')
2022-12-23 14:31:56,731 WARNING [train.py:1060] (2/4) Exclude cut with ID 6951-79737-0018-132285-0 from training. Duration: 21.105
2022-12-23 14:32:01,007 WARNING [train.py:1060] (2/4) Exclude cut with ID 4511-76322-0006-80011-0_sp0.9 from training. Duration: 24.411125
2022-12-23 14:32:10,907 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 2.913e+02 4.378e+02 5.345e+02 6.831e+02 1.350e+03, threshold=1.069e+03, percent-clipped=1.0
2022-12-23 14:32:24,092 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.8962, 1.9114, 1.4646, 2.1014, 2.0913, 1.7680, 2.6125, 1.9779],
       device='cuda:2'), covar=tensor([0.0984, 0.1635, 0.2850, 0.1732, 0.1850, 0.0992, 0.0937, 0.1325],
       device='cuda:2'), in_proj_covar=tensor([0.0180, 0.0208, 0.0252, 0.0293, 0.0240, 0.0193, 0.0211, 0.0205],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-23 14:32:28,270 WARNING [train.py:1060] (2/4) Exclude cut with ID 6758-72288-0033-108368-0_sp1.1 from training. Duration: 21.263625
2022-12-23 14:32:45,771 INFO [train.py:894] (2/4) Epoch 19, batch 3350, loss[loss=0.2098, simple_loss=0.2791, pruned_loss=0.07026, over 18503.00 frames. ], tot_loss[loss=0.1887, simple_loss=0.2687, pruned_loss=0.0544, over 3713375.41 frames. ], batch size: 52, lr: 5.98e-03, grad_scale: 8.0
2022-12-23 14:32:58,602 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([2.2219, 2.2609, 1.5770, 2.6372, 2.4659, 2.0131, 3.0572, 2.1631],
       device='cuda:2'), covar=tensor([0.0856, 0.1713, 0.2799, 0.1765, 0.1620, 0.0919, 0.0851, 0.1287],
       device='cuda:2'), in_proj_covar=tensor([0.0179, 0.0207, 0.0251, 0.0292, 0.0239, 0.0192, 0.0210, 0.0204],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-23 14:32:59,720 WARNING [train.py:1060] (2/4) Exclude cut with ID 4234-40345-0022-142709-0 from training. Duration: 20.795
2022-12-23 14:33:10,331 WARNING [train.py:1060] (2/4) Exclude cut with ID 7357-94126-0021-15852-0 from training. Duration: 24.76
2022-12-23 14:33:10,345 WARNING [train.py:1060] (2/4) Exclude cut with ID 3867-173237-0077-144769-0_sp0.9 from training. Duration: 22.25
2022-12-23 14:33:34,426 WARNING [train.py:1060] (2/4) Exclude cut with ID 7357-94126-0026-15857-0_sp1.1 from training. Duration: 20.5045625
2022-12-23 14:33:37,555 INFO [zipformer.py:660] (2/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=66494.0, num_to_drop=1, layers_to_drop={2}
2022-12-23 14:33:46,493 INFO [zipformer.py:660] (2/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=66500.0, num_to_drop=1, layers_to_drop={3}
2022-12-23 14:34:03,484 INFO [train.py:894] (2/4) Epoch 19, batch 3400, loss[loss=0.1932, simple_loss=0.2754, pruned_loss=0.05552, over 18645.00 frames. ], tot_loss[loss=0.1882, simple_loss=0.2683, pruned_loss=0.05402, over 3713210.88 frames. ], batch size: 98, lr: 5.98e-03, grad_scale: 8.0
2022-12-23 14:34:10,366 INFO [zipformer.py:660] (2/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=66516.0, num_to_drop=1, layers_to_drop={3}
2022-12-23 14:34:17,782 INFO [zipformer.py:660] (2/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=66521.0, num_to_drop=1, layers_to_drop={0}
2022-12-23 14:34:33,118 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.6499, 1.9965, 1.6587, 2.3821, 2.5672, 1.6410, 1.6785, 1.4298],
       device='cuda:2'), covar=tensor([0.1794, 0.1661, 0.1480, 0.0899, 0.1260, 0.1083, 0.1939, 0.1428],
       device='cuda:2'), in_proj_covar=tensor([0.0243, 0.0221, 0.0212, 0.0196, 0.0258, 0.0192, 0.0220, 0.0194],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-23 14:34:40,814 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 2.665e+02 4.804e+02 5.705e+02 7.377e+02 1.281e+03, threshold=1.141e+03, percent-clipped=2.0
2022-12-23 14:34:46,516 INFO [zipformer.py:660] (2/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=66542.0, num_to_drop=1, layers_to_drop={1}
2022-12-23 14:35:13,228 INFO [train.py:894] (2/4) Epoch 19, batch 3450, loss[loss=0.2059, simple_loss=0.2889, pruned_loss=0.06144, over 18508.00 frames. ], tot_loss[loss=0.1885, simple_loss=0.2687, pruned_loss=0.0541, over 3713218.81 frames. ], batch size: 52, lr: 5.98e-03, grad_scale: 8.0
2022-12-23 14:35:14,298 INFO [scaling.py:679] (2/4) Whitening: num_groups=8, num_channels=96, metric=1.37 vs. limit=2.0
2022-12-23 14:35:23,545 INFO [scaling.py:679] (2/4) Whitening: num_groups=8, num_channels=192, metric=1.84 vs. limit=2.0
2022-12-23 14:35:32,536 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([2.0145, 2.0021, 1.5687, 2.0264, 2.1660, 1.8757, 2.6711, 2.0695],
       device='cuda:2'), covar=tensor([0.0888, 0.1614, 0.2724, 0.1721, 0.1854, 0.0929, 0.0965, 0.1268],
       device='cuda:2'), in_proj_covar=tensor([0.0178, 0.0208, 0.0251, 0.0292, 0.0239, 0.0192, 0.0209, 0.0204],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-23 14:35:42,620 INFO [zipformer.py:660] (2/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=66582.0, num_to_drop=1, layers_to_drop={2}
2022-12-23 14:36:26,024 INFO [train.py:894] (2/4) Epoch 19, batch 3500, loss[loss=0.1776, simple_loss=0.2706, pruned_loss=0.04233, over 18559.00 frames. ], tot_loss[loss=0.1886, simple_loss=0.2693, pruned_loss=0.05396, over 3714687.30 frames. ], batch size: 98, lr: 5.98e-03, grad_scale: 8.0
2022-12-23 14:36:31,101 INFO [zipformer.py:660] (2/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=66614.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 14:36:46,513 WARNING [train.py:1060] (2/4) Exclude cut with ID 7859-102521-0017-7548-0_sp1.1 from training. Duration: 22.2954375
2022-12-23 14:36:56,201 INFO [scaling.py:679] (2/4) Whitening: num_groups=8, num_channels=96, metric=1.20 vs. limit=2.0
2022-12-23 14:36:56,742 INFO [scaling.py:679] (2/4) Whitening: num_groups=8, num_channels=96, metric=1.60 vs. limit=2.0
2022-12-23 14:36:56,851 INFO [train.py:894] (2/4) Epoch 20, batch 0, loss[loss=0.1724, simple_loss=0.2552, pruned_loss=0.04484, over 18564.00 frames. ], tot_loss[loss=0.1724, simple_loss=0.2552, pruned_loss=0.04484, over 18564.00 frames. ], batch size: 45, lr: 5.82e-03, grad_scale: 8.0
2022-12-23 14:36:56,851 INFO [train.py:919] (2/4) Computing validation loss
2022-12-23 14:37:07,740 INFO [train.py:928] (2/4) Epoch 20, validation: loss=0.1637, simple_loss=0.2619, pruned_loss=0.03275, over 944034.00 frames. 
2022-12-23 14:37:07,740 INFO [train.py:929] (2/4) Maximum memory allocated so far is 24680MB
2022-12-23 14:37:31,895 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([2.5004, 2.1179, 0.6133, 1.6360, 2.6074, 1.7685, 2.4423, 2.4858],
       device='cuda:2'), covar=tensor([0.1446, 0.1806, 0.2614, 0.1561, 0.1536, 0.1586, 0.1259, 0.1638],
       device='cuda:2'), in_proj_covar=tensor([0.0093, 0.0096, 0.0116, 0.0095, 0.0114, 0.0090, 0.0097, 0.0093],
       device='cuda:2'), out_proj_covar=tensor([0.0003, 0.0003, 0.0004, 0.0003, 0.0004, 0.0003, 0.0003, 0.0003],
       device='cuda:2')
2022-12-23 14:37:39,090 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 2.159e+02 4.005e+02 5.121e+02 6.959e+02 1.364e+03, threshold=1.024e+03, percent-clipped=1.0
2022-12-23 14:37:52,783 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([2.7930, 2.8366, 2.0813, 1.8471, 3.2525, 3.2976, 3.0545, 2.3276],
       device='cuda:2'), covar=tensor([0.0400, 0.0353, 0.0587, 0.0707, 0.0245, 0.0321, 0.0373, 0.0709],
       device='cuda:2'), in_proj_covar=tensor([0.0124, 0.0126, 0.0130, 0.0121, 0.0098, 0.0122, 0.0135, 0.0157],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0001, 0.0001, 0.0001, 0.0002, 0.0002],
       device='cuda:2')
2022-12-23 14:37:58,000 WARNING [train.py:1060] (2/4) Exclude cut with ID 298-126791-0067-24026-0_sp0.9 from training. Duration: 21.438875
2022-12-23 14:38:03,262 WARNING [train.py:1060] (2/4) Exclude cut with ID 5652-39938-0025-23684-0_sp0.9 from training. Duration: 22.2055625
2022-12-23 14:38:23,332 INFO [train.py:894] (2/4) Epoch 20, batch 50, loss[loss=0.173, simple_loss=0.2544, pruned_loss=0.04577, over 18691.00 frames. ], tot_loss[loss=0.1803, simple_loss=0.2677, pruned_loss=0.04646, over 838980.23 frames. ], batch size: 50, lr: 5.82e-03, grad_scale: 8.0
2022-12-23 14:38:42,148 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([2.1137, 2.1355, 1.4984, 2.5489, 2.3637, 1.9566, 2.9171, 2.0998],
       device='cuda:2'), covar=tensor([0.0834, 0.1741, 0.2844, 0.1706, 0.1723, 0.0928, 0.0860, 0.1255],
       device='cuda:2'), in_proj_covar=tensor([0.0178, 0.0208, 0.0251, 0.0292, 0.0238, 0.0191, 0.0209, 0.0204],
       device='cuda:2'), out_proj_covar=tensor([0.0001, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-23 14:39:03,260 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([0.2798, 1.6004, 1.9158, 0.9037, 1.2037, 2.0366, 1.9122, 1.6325],
       device='cuda:2'), covar=tensor([0.0775, 0.0312, 0.0324, 0.0404, 0.0390, 0.0436, 0.0231, 0.0706],
       device='cuda:2'), in_proj_covar=tensor([0.0150, 0.0170, 0.0126, 0.0140, 0.0148, 0.0142, 0.0162, 0.0171],
       device='cuda:2'), out_proj_covar=tensor([1.1545e-04, 1.3118e-04, 9.5413e-05, 1.0494e-04, 1.1183e-04, 1.0945e-04,
        1.2573e-04, 1.3078e-04], device='cuda:2')
2022-12-23 14:39:37,538 INFO [train.py:894] (2/4) Epoch 20, batch 100, loss[loss=0.1614, simple_loss=0.2467, pruned_loss=0.03808, over 18518.00 frames. ], tot_loss[loss=0.1755, simple_loss=0.2642, pruned_loss=0.0434, over 1475940.41 frames. ], batch size: 44, lr: 5.82e-03, grad_scale: 8.0
2022-12-23 14:39:59,305 INFO [zipformer.py:660] (2/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=66731.0, num_to_drop=1, layers_to_drop={0}
2022-12-23 14:40:09,559 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 2.446e+02 3.334e+02 3.825e+02 4.671e+02 9.107e+02, threshold=7.651e+02, percent-clipped=0.0
2022-12-23 14:40:53,548 INFO [train.py:894] (2/4) Epoch 20, batch 150, loss[loss=0.1729, simple_loss=0.2644, pruned_loss=0.04067, over 18507.00 frames. ], tot_loss[loss=0.1755, simple_loss=0.2642, pruned_loss=0.04338, over 1971855.13 frames. ], batch size: 52, lr: 5.82e-03, grad_scale: 8.0
2022-12-23 14:41:02,693 WARNING [train.py:1060] (2/4) Exclude cut with ID 7859-102521-0017-7548-0 from training. Duration: 24.525
2022-12-23 14:41:30,337 INFO [zipformer.py:660] (2/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=66792.0, num_to_drop=1, layers_to_drop={3}
2022-12-23 14:41:37,616 WARNING [train.py:1060] (2/4) Exclude cut with ID 3699-47246-0007-3408-0_sp0.9 from training. Duration: 20.26675
2022-12-23 14:41:42,898 INFO [zipformer.py:660] (2/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=66800.0, num_to_drop=1, layers_to_drop={0}
2022-12-23 14:41:52,164 WARNING [train.py:1060] (2/4) Exclude cut with ID 7859-102521-0017-7548-0_sp0.9 from training. Duration: 27.25
2022-12-23 14:42:07,226 INFO [zipformer.py:660] (2/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=66816.0, num_to_drop=1, layers_to_drop={0}
2022-12-23 14:42:08,306 INFO [train.py:894] (2/4) Epoch 20, batch 200, loss[loss=0.1895, simple_loss=0.2808, pruned_loss=0.04908, over 18684.00 frames. ], tot_loss[loss=0.1747, simple_loss=0.2631, pruned_loss=0.04316, over 2358518.29 frames. ], batch size: 69, lr: 5.81e-03, grad_scale: 8.0
2022-12-23 14:42:39,485 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 2.187e+02 3.272e+02 3.815e+02 5.495e+02 8.905e+02, threshold=7.631e+02, percent-clipped=2.0
2022-12-23 14:42:50,379 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.4840, 1.4126, 1.4157, 1.5024, 1.3391, 3.6875, 1.6321, 2.1547],
       device='cuda:2'), covar=tensor([0.4456, 0.2961, 0.2777, 0.2815, 0.1557, 0.0223, 0.1648, 0.0950],
       device='cuda:2'), in_proj_covar=tensor([0.0133, 0.0117, 0.0125, 0.0120, 0.0102, 0.0097, 0.0091, 0.0089],
       device='cuda:2'), out_proj_covar=tensor([0.0005, 0.0005, 0.0005, 0.0005, 0.0004, 0.0003, 0.0004, 0.0004],
       device='cuda:2')
2022-12-23 14:42:54,452 INFO [zipformer.py:660] (2/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=66848.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 14:43:07,107 WARNING [train.py:1060] (2/4) Exclude cut with ID 6426-64292-0017-15984-0 from training. Duration: 21.68
2022-12-23 14:43:18,206 WARNING [train.py:1060] (2/4) Exclude cut with ID 4278-13270-0007-59342-0 from training. Duration: 21.6300625
2022-12-23 14:43:19,965 INFO [zipformer.py:660] (2/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=66864.0, num_to_drop=1, layers_to_drop={0}
2022-12-23 14:43:24,561 INFO [train.py:894] (2/4) Epoch 20, batch 250, loss[loss=0.1604, simple_loss=0.2491, pruned_loss=0.03586, over 18381.00 frames. ], tot_loss[loss=0.174, simple_loss=0.2628, pruned_loss=0.04261, over 2658221.23 frames. ], batch size: 46, lr: 5.81e-03, grad_scale: 16.0
2022-12-23 14:43:39,184 INFO [zipformer.py:660] (2/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=66877.0, num_to_drop=1, layers_to_drop={3}
2022-12-23 14:43:41,623 WARNING [train.py:1060] (2/4) Exclude cut with ID 4278-13270-0007-59342-0_sp0.9 from training. Duration: 24.033375
2022-12-23 14:44:34,323 INFO [zipformer.py:660] (2/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=66914.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 14:44:38,952 INFO [train.py:894] (2/4) Epoch 20, batch 300, loss[loss=0.167, simple_loss=0.2578, pruned_loss=0.03806, over 18590.00 frames. ], tot_loss[loss=0.1741, simple_loss=0.2623, pruned_loss=0.04296, over 2893002.71 frames. ], batch size: 51, lr: 5.81e-03, grad_scale: 16.0
2022-12-23 14:44:43,217 WARNING [train.py:1060] (2/4) Exclude cut with ID 4278-13270-0009-59344-0 from training. Duration: 22.905
2022-12-23 14:44:43,258 WARNING [train.py:1060] (2/4) Exclude cut with ID 5622-44585-0006-90525-0_sp1.1 from training. Duration: 23.4318125
2022-12-23 14:45:09,776 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 2.301e+02 3.375e+02 4.196e+02 5.009e+02 1.030e+03, threshold=8.392e+02, percent-clipped=5.0
2022-12-23 14:45:30,577 INFO [scaling.py:679] (2/4) Whitening: num_groups=1, num_channels=384, metric=2.53 vs. limit=5.0
2022-12-23 14:45:31,782 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.6273, 2.0251, 1.6854, 2.3325, 2.8587, 1.6202, 1.6186, 1.4127],
       device='cuda:2'), covar=tensor([0.1931, 0.1829, 0.1602, 0.1044, 0.1086, 0.1221, 0.2156, 0.1580],
       device='cuda:2'), in_proj_covar=tensor([0.0241, 0.0220, 0.0211, 0.0195, 0.0255, 0.0192, 0.0219, 0.0193],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-23 14:45:39,716 INFO [zipformer.py:660] (2/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=66957.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 14:45:47,457 INFO [zipformer.py:660] (2/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=66962.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 14:45:55,245 INFO [train.py:894] (2/4) Epoch 20, batch 350, loss[loss=0.1422, simple_loss=0.2294, pruned_loss=0.02746, over 18673.00 frames. ], tot_loss[loss=0.1744, simple_loss=0.2627, pruned_loss=0.043, over 3075202.37 frames. ], batch size: 46, lr: 5.81e-03, grad_scale: 16.0
2022-12-23 14:46:39,883 WARNING [train.py:1060] (2/4) Exclude cut with ID 4278-13270-0009-59344-0_sp1.1 from training. Duration: 20.82275
2022-12-23 14:46:41,273 WARNING [train.py:1060] (2/4) Exclude cut with ID 4278-13270-0009-59344-0_sp0.9 from training. Duration: 25.45
2022-12-23 14:46:44,773 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.9023, 1.9430, 1.4792, 1.9771, 2.0728, 1.8599, 2.5885, 2.0210],
       device='cuda:2'), covar=tensor([0.0904, 0.1605, 0.2630, 0.1813, 0.1719, 0.0870, 0.0963, 0.1256],
       device='cuda:2'), in_proj_covar=tensor([0.0177, 0.0207, 0.0248, 0.0289, 0.0236, 0.0189, 0.0207, 0.0202],
       device='cuda:2'), out_proj_covar=tensor([0.0001, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-23 14:47:07,776 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.5712, 1.4530, 1.4504, 0.7976, 1.7681, 1.6048, 1.5132, 1.3026],
       device='cuda:2'), covar=tensor([0.0370, 0.0501, 0.0453, 0.0743, 0.0379, 0.0367, 0.0431, 0.0937],
       device='cuda:2'), in_proj_covar=tensor([0.0122, 0.0124, 0.0128, 0.0118, 0.0096, 0.0119, 0.0132, 0.0154],
       device='cuda:2'), out_proj_covar=tensor([0.0001, 0.0002, 0.0002, 0.0001, 0.0001, 0.0001, 0.0002, 0.0002],
       device='cuda:2')
2022-12-23 14:47:11,577 INFO [train.py:894] (2/4) Epoch 20, batch 400, loss[loss=0.1672, simple_loss=0.2616, pruned_loss=0.03638, over 18432.00 frames. ], tot_loss[loss=0.1757, simple_loss=0.2639, pruned_loss=0.04374, over 3216511.26 frames. ], batch size: 48, lr: 5.80e-03, grad_scale: 16.0
2022-12-23 14:47:13,253 INFO [zipformer.py:660] (2/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=67018.0, num_to_drop=1, layers_to_drop={0}
2022-12-23 14:47:42,053 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 2.086e+02 3.535e+02 4.114e+02 5.180e+02 1.001e+03, threshold=8.228e+02, percent-clipped=4.0
2022-12-23 14:47:43,602 WARNING [train.py:1060] (2/4) Exclude cut with ID 5622-44585-0006-90525-0 from training. Duration: 25.775
2022-12-23 14:47:54,558 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([2.2158, 2.2629, 1.6585, 2.5411, 2.4590, 2.1598, 3.0986, 2.2961],
       device='cuda:2'), covar=tensor([0.0825, 0.1639, 0.2685, 0.1764, 0.1600, 0.0826, 0.0890, 0.1239],
       device='cuda:2'), in_proj_covar=tensor([0.0177, 0.0207, 0.0249, 0.0289, 0.0236, 0.0190, 0.0207, 0.0203],
       device='cuda:2'), out_proj_covar=tensor([0.0001, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-23 14:48:05,704 WARNING [train.py:1060] (2/4) Exclude cut with ID 2929-85685-0071-62375-0_sp0.9 from training. Duration: 22.25
2022-12-23 14:48:17,063 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.7423, 1.9387, 2.1019, 1.0935, 2.0512, 2.1253, 1.6137, 2.4964],
       device='cuda:2'), covar=tensor([0.1114, 0.1583, 0.1026, 0.1725, 0.0653, 0.0964, 0.2133, 0.0467],
       device='cuda:2'), in_proj_covar=tensor([0.0197, 0.0210, 0.0205, 0.0192, 0.0174, 0.0215, 0.0215, 0.0189],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-23 14:48:26,122 INFO [train.py:894] (2/4) Epoch 20, batch 450, loss[loss=0.1729, simple_loss=0.2551, pruned_loss=0.04532, over 18630.00 frames. ], tot_loss[loss=0.1764, simple_loss=0.2643, pruned_loss=0.04425, over 3327457.81 frames. ], batch size: 45, lr: 5.80e-03, grad_scale: 16.0
2022-12-23 14:48:34,698 WARNING [train.py:1060] (2/4) Exclude cut with ID 3972-170212-0014-23379-0 from training. Duration: 26.205
2022-12-23 14:48:50,566 INFO [zipformer.py:660] (2/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=67083.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 14:48:51,774 WARNING [train.py:1060] (2/4) Exclude cut with ID 5239-32139-0047-9341-0_sp0.9 from training. Duration: 30.1555625
2022-12-23 14:48:56,063 INFO [zipformer.py:660] (2/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=67087.0, num_to_drop=1, layers_to_drop={0}
2022-12-23 14:48:58,809 WARNING [train.py:1060] (2/4) Exclude cut with ID 1265-135635-0050-6781-0_sp0.9 from training. Duration: 21.8333125
2022-12-23 14:49:06,680 WARNING [train.py:1060] (2/4) Exclude cut with ID 1914-133440-0024-94914-0_sp1.1 from training. Duration: 20.6545625
2022-12-23 14:49:42,236 INFO [train.py:894] (2/4) Epoch 20, batch 500, loss[loss=0.1943, simple_loss=0.2783, pruned_loss=0.05516, over 18648.00 frames. ], tot_loss[loss=0.1774, simple_loss=0.2652, pruned_loss=0.0448, over 3414593.95 frames. ], batch size: 183, lr: 5.80e-03, grad_scale: 16.0
2022-12-23 14:49:46,782 WARNING [train.py:1060] (2/4) Exclude cut with ID 7395-89880-0045-39920-0_sp0.9 from training. Duration: 20.52225
2022-12-23 14:49:59,961 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.7884, 1.7601, 1.5133, 1.7469, 1.9974, 2.0828, 2.0017, 1.5562],
       device='cuda:2'), covar=tensor([0.0364, 0.0300, 0.0477, 0.0221, 0.0207, 0.0339, 0.0248, 0.0309],
       device='cuda:2'), in_proj_covar=tensor([0.0095, 0.0127, 0.0154, 0.0125, 0.0117, 0.0120, 0.0098, 0.0128],
       device='cuda:2'), out_proj_covar=tensor([7.6231e-05, 1.0128e-04, 1.2752e-04, 1.0006e-04, 9.5215e-05, 9.2592e-05,
        7.7151e-05, 1.0083e-04], device='cuda:2')
2022-12-23 14:50:06,691 WARNING [train.py:1060] (2/4) Exclude cut with ID 3972-170212-0014-23379-0_sp0.9 from training. Duration: 29.1166875
2022-12-23 14:50:12,593 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 2.117e+02 3.572e+02 4.412e+02 5.360e+02 1.378e+03, threshold=8.823e+02, percent-clipped=5.0
2022-12-23 14:50:22,401 INFO [zipformer.py:660] (2/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=67144.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 14:50:56,641 INFO [train.py:894] (2/4) Epoch 20, batch 550, loss[loss=0.2438, simple_loss=0.315, pruned_loss=0.08632, over 18491.00 frames. ], tot_loss[loss=0.1789, simple_loss=0.2667, pruned_loss=0.04559, over 3481664.26 frames. ], batch size: 52, lr: 5.80e-03, grad_scale: 16.0
2022-12-23 14:51:05,143 WARNING [train.py:1060] (2/4) Exclude cut with ID 543-133211-0007-59831-0_sp0.9 from training. Duration: 21.388875
2022-12-23 14:51:11,869 INFO [zipformer.py:660] (2/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=67177.0, num_to_drop=1, layers_to_drop={0}
2022-12-23 14:51:40,397 WARNING [train.py:1060] (2/4) Exclude cut with ID 1914-133440-0024-94914-0 from training. Duration: 22.72
2022-12-23 14:51:41,884 WARNING [train.py:1060] (2/4) Exclude cut with ID 1914-133440-0031-94921-0_sp0.9 from training. Duration: 22.7444375
2022-12-23 14:51:43,586 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.6279, 1.6242, 1.7035, 1.6714, 1.2366, 3.8068, 1.7731, 2.1158],
       device='cuda:2'), covar=tensor([0.3733, 0.2430, 0.2262, 0.2278, 0.1679, 0.0195, 0.1654, 0.0944],
       device='cuda:2'), in_proj_covar=tensor([0.0132, 0.0116, 0.0125, 0.0120, 0.0102, 0.0096, 0.0091, 0.0088],
       device='cuda:2'), out_proj_covar=tensor([0.0005, 0.0005, 0.0005, 0.0005, 0.0004, 0.0003, 0.0004, 0.0004],
       device='cuda:2')
2022-12-23 14:52:05,505 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([3.1469, 2.3674, 1.9312, 2.8772, 2.1975, 2.3722, 2.4009, 3.4246],
       device='cuda:2'), covar=tensor([0.1747, 0.3404, 0.1779, 0.2891, 0.3742, 0.0977, 0.3005, 0.0679],
       device='cuda:2'), in_proj_covar=tensor([0.0291, 0.0287, 0.0242, 0.0344, 0.0267, 0.0226, 0.0285, 0.0210],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0003, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-23 14:52:10,860 INFO [train.py:894] (2/4) Epoch 20, batch 600, loss[loss=0.1538, simple_loss=0.2366, pruned_loss=0.03548, over 18696.00 frames. ], tot_loss[loss=0.1796, simple_loss=0.2674, pruned_loss=0.04588, over 3533747.24 frames. ], batch size: 46, lr: 5.80e-03, grad_scale: 16.0
2022-12-23 14:52:17,846 INFO [scaling.py:679] (2/4) Whitening: num_groups=8, num_channels=96, metric=1.34 vs. limit=2.0
2022-12-23 14:52:22,872 INFO [zipformer.py:660] (2/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=67225.0, num_to_drop=1, layers_to_drop={1}
2022-12-23 14:52:24,110 WARNING [train.py:1060] (2/4) Exclude cut with ID 4133-6541-0027-40495-0_sp1.1 from training. Duration: 0.9681875
2022-12-23 14:52:28,523 WARNING [train.py:1060] (2/4) Exclude cut with ID 6330-62851-0022-91297-0_sp0.9 from training. Duration: 22.3166875
2022-12-23 14:52:34,070 WARNING [train.py:1060] (2/4) Exclude cut with ID 543-133212-0015-59917-0_sp0.9 from training. Duration: 21.8166875
2022-12-23 14:52:41,649 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 2.645e+02 3.735e+02 4.222e+02 4.991e+02 1.201e+03, threshold=8.444e+02, percent-clipped=1.0
2022-12-23 14:52:41,974 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([5.9048, 5.0074, 5.2025, 5.9488, 5.4781, 5.2981, 6.0129, 1.5424],
       device='cuda:2'), covar=tensor([0.0582, 0.0594, 0.0509, 0.0602, 0.1293, 0.0988, 0.0368, 0.5239],
       device='cuda:2'), in_proj_covar=tensor([0.0333, 0.0219, 0.0229, 0.0260, 0.0315, 0.0262, 0.0280, 0.0273],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0001, 0.0001, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-23 14:53:25,781 INFO [train.py:894] (2/4) Epoch 20, batch 650, loss[loss=0.1776, simple_loss=0.2678, pruned_loss=0.04371, over 18601.00 frames. ], tot_loss[loss=0.1798, simple_loss=0.2676, pruned_loss=0.04596, over 3572942.39 frames. ], batch size: 56, lr: 5.79e-03, grad_scale: 16.0
2022-12-23 14:54:16,839 WARNING [train.py:1060] (2/4) Exclude cut with ID 4957-30119-0041-23990-0_sp0.9 from training. Duration: 20.22775
2022-12-23 14:54:26,717 INFO [scaling.py:679] (2/4) Whitening: num_groups=1, num_channels=384, metric=2.63 vs. limit=5.0
2022-12-23 14:54:28,757 INFO [scaling.py:679] (2/4) Whitening: num_groups=8, num_channels=96, metric=1.22 vs. limit=2.0
2022-12-23 14:54:32,705 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([3.0002, 3.0322, 2.1473, 1.6750, 3.5653, 3.5608, 3.1777, 2.3882],
       device='cuda:2'), covar=tensor([0.0388, 0.0341, 0.0532, 0.0746, 0.0196, 0.0331, 0.0388, 0.0722],
       device='cuda:2'), in_proj_covar=tensor([0.0123, 0.0124, 0.0127, 0.0119, 0.0096, 0.0120, 0.0133, 0.0154],
       device='cuda:2'), out_proj_covar=tensor([0.0001, 0.0002, 0.0002, 0.0001, 0.0001, 0.0001, 0.0002, 0.0002],
       device='cuda:2')
2022-12-23 14:54:33,989 INFO [zipformer.py:660] (2/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=67313.0, num_to_drop=1, layers_to_drop={1}
2022-12-23 14:54:39,647 INFO [train.py:894] (2/4) Epoch 20, batch 700, loss[loss=0.2238, simple_loss=0.3058, pruned_loss=0.07093, over 18664.00 frames. ], tot_loss[loss=0.1802, simple_loss=0.2683, pruned_loss=0.04607, over 3605637.30 frames. ], batch size: 60, lr: 5.79e-03, grad_scale: 8.0
2022-12-23 14:55:00,176 WARNING [train.py:1060] (2/4) Exclude cut with ID 5239-32139-0047-9341-0_sp1.1 from training. Duration: 24.67275
2022-12-23 14:55:12,710 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 2.463e+02 3.535e+02 4.247e+02 5.530e+02 8.059e+02, threshold=8.494e+02, percent-clipped=0.0
2022-12-23 14:55:28,755 WARNING [train.py:1060] (2/4) Exclude cut with ID 3082-165428-0081-50734-0_sp0.9 from training. Duration: 21.8055625
2022-12-23 14:55:54,233 INFO [train.py:894] (2/4) Epoch 20, batch 750, loss[loss=0.1429, simple_loss=0.2306, pruned_loss=0.02765, over 18561.00 frames. ], tot_loss[loss=0.1783, simple_loss=0.2665, pruned_loss=0.04502, over 3628581.57 frames. ], batch size: 41, lr: 5.79e-03, grad_scale: 8.0
2022-12-23 14:56:05,649 WARNING [train.py:1060] (2/4) Exclude cut with ID 3340-169293-0054-76830-0_sp0.9 from training. Duration: 22.6666875
2022-12-23 14:56:23,986 INFO [zipformer.py:660] (2/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=67387.0, num_to_drop=1, layers_to_drop={0}
2022-12-23 14:57:08,166 INFO [train.py:894] (2/4) Epoch 20, batch 800, loss[loss=0.2052, simple_loss=0.2883, pruned_loss=0.0611, over 18594.00 frames. ], tot_loss[loss=0.1786, simple_loss=0.2669, pruned_loss=0.04511, over 3647079.89 frames. ], batch size: 175, lr: 5.79e-03, grad_scale: 8.0
2022-12-23 14:57:09,674 WARNING [train.py:1060] (2/4) Exclude cut with ID 2411-132532-0017-82279-0_sp1.1 from training. Duration: 0.9681875
2022-12-23 14:57:35,170 WARNING [train.py:1060] (2/4) Exclude cut with ID 6330-62850-0007-91323-0 from training. Duration: 22.485
2022-12-23 14:57:35,321 INFO [zipformer.py:660] (2/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=67435.0, num_to_drop=1, layers_to_drop={0}
2022-12-23 14:57:37,163 INFO [scaling.py:679] (2/4) Whitening: num_groups=8, num_channels=96, metric=1.51 vs. limit=2.0
2022-12-23 14:57:40,535 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 2.476e+02 3.510e+02 4.303e+02 5.243e+02 9.947e+02, threshold=8.606e+02, percent-clipped=5.0
2022-12-23 14:57:40,789 INFO [zipformer.py:660] (2/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=67439.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 14:58:10,495 WARNING [train.py:1060] (2/4) Exclude cut with ID 3972-170212-0014-23379-0_sp1.1 from training. Duration: 23.82275
2022-12-23 14:58:15,337 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([3.2825, 2.8662, 2.8496, 3.2435, 2.9822, 2.8553, 3.3746, 1.0913],
       device='cuda:2'), covar=tensor([0.1018, 0.0852, 0.0858, 0.0977, 0.1756, 0.1416, 0.0961, 0.4772],
       device='cuda:2'), in_proj_covar=tensor([0.0340, 0.0224, 0.0234, 0.0264, 0.0321, 0.0268, 0.0285, 0.0279],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0001, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-23 14:58:22,212 INFO [train.py:894] (2/4) Epoch 20, batch 850, loss[loss=0.1553, simple_loss=0.241, pruned_loss=0.03483, over 18525.00 frames. ], tot_loss[loss=0.1788, simple_loss=0.2671, pruned_loss=0.04529, over 3662157.80 frames. ], batch size: 47, lr: 5.79e-03, grad_scale: 8.0
2022-12-23 14:58:22,226 WARNING [train.py:1060] (2/4) Exclude cut with ID 4860-13185-0032-76709-0 from training. Duration: 20.77
2022-12-23 14:58:29,379 WARNING [train.py:1060] (2/4) Exclude cut with ID 6426-64292-0017-15984-0_sp0.9 from training. Duration: 24.088875
2022-12-23 14:58:56,640 WARNING [train.py:1060] (2/4) Exclude cut with ID 6330-62850-0007-91323-0_sp1.1 from training. Duration: 20.4409375
2022-12-23 14:59:32,027 INFO [zipformer.py:660] (2/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=67513.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 14:59:37,804 INFO [train.py:894] (2/4) Epoch 20, batch 900, loss[loss=0.2017, simple_loss=0.2857, pruned_loss=0.05882, over 18614.00 frames. ], tot_loss[loss=0.1783, simple_loss=0.2662, pruned_loss=0.04516, over 3673314.07 frames. ], batch size: 181, lr: 5.78e-03, grad_scale: 8.0
2022-12-23 15:00:11,358 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 2.522e+02 3.439e+02 4.457e+02 5.643e+02 1.323e+03, threshold=8.915e+02, percent-clipped=3.0
2022-12-23 15:00:11,437 WARNING [train.py:1060] (2/4) Exclude cut with ID 2929-85685-0062-62366-0_sp0.9 from training. Duration: 22.511125
2022-12-23 15:00:12,731 WARNING [train.py:1060] (2/4) Exclude cut with ID 7395-89880-0031-39906-0 from training. Duration: 20.675
2022-12-23 15:00:49,670 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([3.3047, 2.4838, 1.9456, 3.1659, 2.2800, 2.5884, 2.5946, 3.4806],
       device='cuda:2'), covar=tensor([0.1775, 0.3249, 0.1850, 0.2675, 0.3754, 0.0956, 0.3159, 0.0713],
       device='cuda:2'), in_proj_covar=tensor([0.0293, 0.0288, 0.0244, 0.0347, 0.0269, 0.0227, 0.0288, 0.0212],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0003, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-23 15:00:53,645 INFO [train.py:894] (2/4) Epoch 20, batch 950, loss[loss=0.1721, simple_loss=0.2559, pruned_loss=0.04419, over 18566.00 frames. ], tot_loss[loss=0.178, simple_loss=0.2658, pruned_loss=0.04507, over 3681698.81 frames. ], batch size: 49, lr: 5.78e-03, grad_scale: 8.0
2022-12-23 15:01:04,522 INFO [zipformer.py:660] (2/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=67574.0, num_to_drop=1, layers_to_drop={2}
2022-12-23 15:01:48,783 WARNING [train.py:1060] (2/4) Exclude cut with ID 6330-62850-0007-91323-0_sp0.9 from training. Duration: 24.9833125
2022-12-23 15:02:03,411 INFO [zipformer.py:660] (2/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=67613.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 15:02:05,044 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.9175, 1.7906, 1.6456, 1.0198, 2.2028, 2.0298, 1.8216, 1.4955],
       device='cuda:2'), covar=tensor([0.0399, 0.0478, 0.0497, 0.0833, 0.0302, 0.0389, 0.0467, 0.0903],
       device='cuda:2'), in_proj_covar=tensor([0.0123, 0.0124, 0.0127, 0.0119, 0.0096, 0.0120, 0.0132, 0.0153],
       device='cuda:2'), out_proj_covar=tensor([0.0001, 0.0002, 0.0002, 0.0001, 0.0001, 0.0001, 0.0002, 0.0002],
       device='cuda:2')
2022-12-23 15:02:08,867 INFO [train.py:894] (2/4) Epoch 20, batch 1000, loss[loss=0.1669, simple_loss=0.2551, pruned_loss=0.03934, over 18431.00 frames. ], tot_loss[loss=0.1768, simple_loss=0.2645, pruned_loss=0.04455, over 3686874.55 frames. ], batch size: 48, lr: 5.78e-03, grad_scale: 8.0
2022-12-23 15:02:22,296 WARNING [train.py:1060] (2/4) Exclude cut with ID 5239-32139-0047-9341-0 from training. Duration: 27.14
2022-12-23 15:02:36,319 WARNING [train.py:1060] (2/4) Exclude cut with ID 2929-85685-0044-62348-0 from training. Duration: 22.44
2022-12-23 15:02:42,943 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 2.353e+02 3.421e+02 4.027e+02 4.985e+02 1.255e+03, threshold=8.054e+02, percent-clipped=1.0
2022-12-23 15:03:16,103 INFO [zipformer.py:660] (2/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=67661.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 15:03:24,770 INFO [train.py:894] (2/4) Epoch 20, batch 1050, loss[loss=0.173, simple_loss=0.271, pruned_loss=0.03751, over 18510.00 frames. ], tot_loss[loss=0.177, simple_loss=0.2651, pruned_loss=0.04442, over 3693127.17 frames. ], batch size: 52, lr: 5.78e-03, grad_scale: 8.0
2022-12-23 15:03:44,624 INFO [zipformer.py:660] (2/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=67680.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 15:03:54,971 WARNING [train.py:1060] (2/4) Exclude cut with ID 2929-85685-0060-62364-0_sp0.9 from training. Duration: 21.361125
2022-12-23 15:03:58,482 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([3.1293, 2.7770, 2.7105, 3.1220, 2.8266, 2.7446, 3.2508, 1.0725],
       device='cuda:2'), covar=tensor([0.1073, 0.0809, 0.0878, 0.1125, 0.1776, 0.1352, 0.0919, 0.4721],
       device='cuda:2'), in_proj_covar=tensor([0.0338, 0.0222, 0.0233, 0.0263, 0.0319, 0.0267, 0.0284, 0.0277],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0001, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-23 15:03:59,681 WARNING [train.py:1060] (2/4) Exclude cut with ID 2929-85685-0079-62383-0_sp1.1 from training. Duration: 27.0318125
2022-12-23 15:04:03,101 INFO [zipformer.py:660] (2/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=67692.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 15:04:11,984 WARNING [train.py:1060] (2/4) Exclude cut with ID 5622-44585-0006-90525-0_sp0.9 from training. Duration: 28.638875
2022-12-23 15:04:19,725 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([2.2896, 1.9878, 1.8186, 1.2553, 2.4675, 2.3777, 2.2320, 1.5955],
       device='cuda:2'), covar=tensor([0.0318, 0.0435, 0.0479, 0.0710, 0.0264, 0.0328, 0.0382, 0.0854],
       device='cuda:2'), in_proj_covar=tensor([0.0122, 0.0124, 0.0127, 0.0118, 0.0095, 0.0120, 0.0132, 0.0153],
       device='cuda:2'), out_proj_covar=tensor([0.0001, 0.0002, 0.0002, 0.0001, 0.0001, 0.0001, 0.0002, 0.0002],
       device='cuda:2')
2022-12-23 15:04:28,206 WARNING [train.py:1060] (2/4) Exclude cut with ID 3340-169293-0054-76830-0 from training. Duration: 20.4
2022-12-23 15:04:40,536 INFO [train.py:894] (2/4) Epoch 20, batch 1100, loss[loss=0.166, simple_loss=0.2619, pruned_loss=0.03499, over 18724.00 frames. ], tot_loss[loss=0.1769, simple_loss=0.265, pruned_loss=0.04438, over 3698845.12 frames. ], batch size: 69, lr: 5.77e-03, grad_scale: 8.0
2022-12-23 15:04:59,974 WARNING [train.py:1060] (2/4) Exclude cut with ID 2929-85685-0071-62375-0 from training. Duration: 20.025
2022-12-23 15:05:01,395 WARNING [train.py:1060] (2/4) Exclude cut with ID 2364-131735-0112-64612-0_sp0.9 from training. Duration: 20.488875
2022-12-23 15:05:04,663 WARNING [train.py:1060] (2/4) Exclude cut with ID 2929-85685-0079-62383-0 from training. Duration: 29.735
2022-12-23 15:05:14,310 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 2.101e+02 3.349e+02 4.308e+02 5.408e+02 2.073e+03, threshold=8.617e+02, percent-clipped=3.0
2022-12-23 15:05:14,607 INFO [zipformer.py:660] (2/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=67739.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 15:05:17,691 INFO [zipformer.py:660] (2/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=67741.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 15:05:34,851 INFO [zipformer.py:660] (2/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=67753.0, num_to_drop=1, layers_to_drop={3}
2022-12-23 15:05:55,179 INFO [train.py:894] (2/4) Epoch 20, batch 1150, loss[loss=0.1783, simple_loss=0.2736, pruned_loss=0.04152, over 18585.00 frames. ], tot_loss[loss=0.177, simple_loss=0.265, pruned_loss=0.04452, over 3701907.72 frames. ], batch size: 56, lr: 5.77e-03, grad_scale: 8.0
2022-12-23 15:06:25,330 WARNING [train.py:1060] (2/4) Exclude cut with ID 7276-92427-0014-12983-0_sp0.9 from training. Duration: 21.3055625
2022-12-23 15:06:25,458 INFO [zipformer.py:660] (2/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=67787.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 15:06:26,770 WARNING [train.py:1060] (2/4) Exclude cut with ID 1025-75365-0008-79168-0_sp0.9 from training. Duration: 22.0666875
2022-12-23 15:07:10,099 INFO [train.py:894] (2/4) Epoch 20, batch 1200, loss[loss=0.1878, simple_loss=0.2752, pruned_loss=0.05024, over 18591.00 frames. ], tot_loss[loss=0.1777, simple_loss=0.2658, pruned_loss=0.04483, over 3704352.05 frames. ], batch size: 51, lr: 5.77e-03, grad_scale: 8.0
2022-12-23 15:07:32,198 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.5351, 1.2871, 1.2316, 1.3151, 1.5872, 1.5275, 1.5343, 1.1321],
       device='cuda:2'), covar=tensor([0.0273, 0.0248, 0.0523, 0.0211, 0.0199, 0.0388, 0.0290, 0.0306],
       device='cuda:2'), in_proj_covar=tensor([0.0095, 0.0127, 0.0154, 0.0125, 0.0116, 0.0120, 0.0098, 0.0126],
       device='cuda:2'), out_proj_covar=tensor([7.5607e-05, 1.0091e-04, 1.2720e-04, 1.0013e-04, 9.3935e-05, 9.2291e-05,
        7.6570e-05, 9.9228e-05], device='cuda:2')
2022-12-23 15:07:43,800 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 2.358e+02 3.704e+02 4.374e+02 5.678e+02 1.404e+03, threshold=8.749e+02, percent-clipped=7.0
2022-12-23 15:08:14,428 WARNING [train.py:1060] (2/4) Exclude cut with ID 2929-85685-0062-62366-0 from training. Duration: 20.26
2022-12-23 15:08:20,914 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.4716, 1.7071, 1.3319, 1.9669, 2.5314, 1.5429, 1.3797, 1.2318],
       device='cuda:2'), covar=tensor([0.1935, 0.1852, 0.1771, 0.1155, 0.1162, 0.1160, 0.2166, 0.1603],
       device='cuda:2'), in_proj_covar=tensor([0.0243, 0.0224, 0.0212, 0.0198, 0.0259, 0.0194, 0.0221, 0.0195],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-23 15:08:24,716 INFO [train.py:894] (2/4) Epoch 20, batch 1250, loss[loss=0.1841, simple_loss=0.2798, pruned_loss=0.04413, over 18664.00 frames. ], tot_loss[loss=0.1773, simple_loss=0.2656, pruned_loss=0.0445, over 3706110.53 frames. ], batch size: 78, lr: 5.77e-03, grad_scale: 8.0
2022-12-23 15:08:26,051 WARNING [train.py:1060] (2/4) Exclude cut with ID 5239-32139-0030-9324-0_sp0.9 from training. Duration: 21.3444375
2022-12-23 15:08:27,779 INFO [zipformer.py:660] (2/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=67869.0, num_to_drop=1, layers_to_drop={3}
2022-12-23 15:09:00,964 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.2802, 1.2978, 1.4718, 0.9907, 1.4332, 1.4899, 1.1526, 1.7373],
       device='cuda:2'), covar=tensor([0.1030, 0.1963, 0.1193, 0.1436, 0.0712, 0.0956, 0.2514, 0.0530],
       device='cuda:2'), in_proj_covar=tensor([0.0197, 0.0211, 0.0207, 0.0193, 0.0175, 0.0215, 0.0213, 0.0190],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-23 15:09:22,797 WARNING [train.py:1060] (2/4) Exclude cut with ID 497-129325-0061-62254-0_sp1.1 from training. Duration: 0.97725
2022-12-23 15:09:40,763 INFO [train.py:894] (2/4) Epoch 20, batch 1300, loss[loss=0.1791, simple_loss=0.2658, pruned_loss=0.04619, over 18580.00 frames. ], tot_loss[loss=0.1776, simple_loss=0.2659, pruned_loss=0.04471, over 3707961.55 frames. ], batch size: 49, lr: 5.77e-03, grad_scale: 8.0
2022-12-23 15:10:05,599 WARNING [train.py:1060] (2/4) Exclude cut with ID 7395-89880-0031-39906-0_sp0.9 from training. Duration: 22.97225
2022-12-23 15:10:15,142 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 2.448e+02 3.360e+02 4.182e+02 5.041e+02 9.316e+02, threshold=8.363e+02, percent-clipped=1.0
2022-12-23 15:10:38,170 WARNING [train.py:1060] (2/4) Exclude cut with ID 7395-89880-0047-39922-0_sp0.9 from training. Duration: 21.97775
2022-12-23 15:10:50,526 WARNING [train.py:1060] (2/4) Exclude cut with ID 1112-1043-0006-89194-0_sp0.9 from training. Duration: 21.8333125
2022-12-23 15:10:56,993 INFO [train.py:894] (2/4) Epoch 20, batch 1350, loss[loss=0.2198, simple_loss=0.2956, pruned_loss=0.07202, over 18562.00 frames. ], tot_loss[loss=0.178, simple_loss=0.2664, pruned_loss=0.0448, over 3709815.32 frames. ], batch size: 174, lr: 5.76e-03, grad_scale: 8.0
2022-12-23 15:11:01,554 WARNING [train.py:1060] (2/4) Exclude cut with ID 1914-133440-0031-94921-0 from training. Duration: 20.47
2022-12-23 15:11:18,902 INFO [zipformer.py:660] (2/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=67981.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 15:12:06,097 INFO [scaling.py:679] (2/4) Whitening: num_groups=1, num_channels=384, metric=2.99 vs. limit=5.0
2022-12-23 15:12:06,973 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([2.6551, 3.8445, 3.7569, 1.4305, 4.0185, 3.0346, 0.7744, 2.4560],
       device='cuda:2'), covar=tensor([0.1925, 0.1077, 0.1229, 0.3596, 0.0834, 0.0869, 0.4878, 0.1546],
       device='cuda:2'), in_proj_covar=tensor([0.0142, 0.0138, 0.0154, 0.0122, 0.0141, 0.0112, 0.0142, 0.0112],
       device='cuda:2'), out_proj_covar=tensor([0.0003, 0.0002, 0.0003, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-23 15:12:09,904 WARNING [train.py:1060] (2/4) Exclude cut with ID 7395-89880-0037-39912-0_sp0.9 from training. Duration: 20.67225
2022-12-23 15:12:15,615 INFO [train.py:894] (2/4) Epoch 20, batch 1400, loss[loss=0.2154, simple_loss=0.2971, pruned_loss=0.06683, over 18467.00 frames. ], tot_loss[loss=0.1776, simple_loss=0.2662, pruned_loss=0.04453, over 3710941.60 frames. ], batch size: 64, lr: 5.76e-03, grad_scale: 8.0
2022-12-23 15:12:16,338 INFO [scaling.py:679] (2/4) Whitening: num_groups=8, num_channels=192, metric=1.62 vs. limit=2.0
2022-12-23 15:12:28,935 WARNING [train.py:1060] (2/4) Exclude cut with ID 1914-133440-0024-94914-0_sp0.9 from training. Duration: 25.2444375
2022-12-23 15:12:44,185 INFO [zipformer.py:660] (2/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=68036.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 15:12:48,512 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 2.229e+02 4.016e+02 4.665e+02 5.451e+02 1.427e+03, threshold=9.330e+02, percent-clipped=6.0
2022-12-23 15:12:52,760 WARNING [train.py:1060] (2/4) Exclude cut with ID 3340-169293-0021-76797-0_sp0.9 from training. Duration: 21.1445
2022-12-23 15:12:53,079 INFO [zipformer.py:660] (2/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=68042.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 15:13:01,445 INFO [zipformer.py:660] (2/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=68048.0, num_to_drop=1, layers_to_drop={0}
2022-12-23 15:13:30,352 INFO [train.py:894] (2/4) Epoch 20, batch 1450, loss[loss=0.1584, simple_loss=0.2434, pruned_loss=0.03669, over 18696.00 frames. ], tot_loss[loss=0.1771, simple_loss=0.2655, pruned_loss=0.04434, over 3711521.38 frames. ], batch size: 46, lr: 5.76e-03, grad_scale: 8.0
2022-12-23 15:13:37,864 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.4737, 1.1288, 0.7618, 1.1873, 1.6353, 1.0123, 1.3260, 1.3989],
       device='cuda:2'), covar=tensor([0.1358, 0.1693, 0.1892, 0.1304, 0.1697, 0.1726, 0.1186, 0.1379],
       device='cuda:2'), in_proj_covar=tensor([0.0095, 0.0099, 0.0117, 0.0098, 0.0117, 0.0092, 0.0099, 0.0095],
       device='cuda:2'), out_proj_covar=tensor([0.0003, 0.0003, 0.0004, 0.0003, 0.0004, 0.0003, 0.0003, 0.0003],
       device='cuda:2')
2022-12-23 15:14:00,909 INFO [zipformer.py:660] (2/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=68087.0, num_to_drop=1, layers_to_drop={0}
2022-12-23 15:14:03,849 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([2.5817, 2.2249, 1.6916, 2.5251, 1.9466, 2.1207, 2.0446, 2.4646],
       device='cuda:2'), covar=tensor([0.2015, 0.3210, 0.1957, 0.2445, 0.3600, 0.1115, 0.3052, 0.0961],
       device='cuda:2'), in_proj_covar=tensor([0.0295, 0.0291, 0.0247, 0.0348, 0.0272, 0.0229, 0.0290, 0.0215],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0003, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-23 15:14:10,488 WARNING [train.py:1060] (2/4) Exclude cut with ID 2929-85685-0079-62383-0_sp0.9 from training. Duration: 33.038875
2022-12-23 15:14:27,310 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.7143, 1.7284, 1.8890, 1.7157, 1.1749, 3.8949, 1.7274, 2.0995],
       device='cuda:2'), covar=tensor([0.2970, 0.1949, 0.1801, 0.1998, 0.1403, 0.0139, 0.1508, 0.0822],
       device='cuda:2'), in_proj_covar=tensor([0.0131, 0.0117, 0.0125, 0.0120, 0.0103, 0.0096, 0.0091, 0.0088],
       device='cuda:2'), out_proj_covar=tensor([0.0005, 0.0005, 0.0005, 0.0005, 0.0004, 0.0003, 0.0004, 0.0004],
       device='cuda:2')
2022-12-23 15:14:45,604 INFO [train.py:894] (2/4) Epoch 20, batch 1500, loss[loss=0.1581, simple_loss=0.2377, pruned_loss=0.03923, over 18427.00 frames. ], tot_loss[loss=0.1766, simple_loss=0.265, pruned_loss=0.04412, over 3711629.91 frames. ], batch size: 42, lr: 5.76e-03, grad_scale: 8.0
2022-12-23 15:14:49,950 WARNING [train.py:1060] (2/4) Exclude cut with ID 6426-64291-0000-16059-0_sp0.9 from training. Duration: 20.0944375
2022-12-23 15:15:05,355 WARNING [train.py:1060] (2/4) Exclude cut with ID 2929-85685-0044-62348-0_sp1.1 from training. Duration: 20.4
2022-12-23 15:15:12,730 WARNING [train.py:1060] (2/4) Exclude cut with ID 6330-62851-0022-91297-0 from training. Duration: 20.085
2022-12-23 15:15:18,479 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 2.391e+02 3.554e+02 4.254e+02 5.043e+02 9.798e+02, threshold=8.509e+02, percent-clipped=1.0
2022-12-23 15:15:22,830 WARNING [train.py:1060] (2/4) Exclude cut with ID 4860-13185-0032-76709-0_sp0.9 from training. Duration: 23.07775
2022-12-23 15:15:32,109 INFO [zipformer.py:660] (2/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=68148.0, num_to_drop=1, layers_to_drop={1}
2022-12-23 15:16:00,685 INFO [train.py:894] (2/4) Epoch 20, batch 1550, loss[loss=0.164, simple_loss=0.2472, pruned_loss=0.04036, over 18679.00 frames. ], tot_loss[loss=0.1766, simple_loss=0.2649, pruned_loss=0.04415, over 3713417.64 frames. ], batch size: 46, lr: 5.76e-03, grad_scale: 8.0
2022-12-23 15:16:03,842 INFO [zipformer.py:660] (2/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=68169.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 15:16:10,363 WARNING [train.py:1060] (2/4) Exclude cut with ID 2929-85685-0044-62348-0_sp0.9 from training. Duration: 24.9333125
2022-12-23 15:16:56,174 WARNING [train.py:1060] (2/4) Exclude cut with ID 5118-111612-0016-124680-0_sp0.9 from training. Duration: 20.388875
2022-12-23 15:17:03,348 WARNING [train.py:1060] (2/4) Exclude cut with ID 432-122774-0017-62487-0_sp1.1 from training. Duration: 20.3590625
2022-12-23 15:17:15,386 INFO [train.py:894] (2/4) Epoch 20, batch 1600, loss[loss=0.1785, simple_loss=0.2635, pruned_loss=0.04678, over 18677.00 frames. ], tot_loss[loss=0.1756, simple_loss=0.2637, pruned_loss=0.0437, over 3713939.37 frames. ], batch size: 46, lr: 5.75e-03, grad_scale: 8.0
2022-12-23 15:17:15,566 INFO [zipformer.py:660] (2/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=68217.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 15:17:43,178 INFO [scaling.py:679] (2/4) Whitening: num_groups=8, num_channels=96, metric=1.31 vs. limit=2.0
2022-12-23 15:17:43,840 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.4649, 1.0807, 1.7271, 2.7750, 2.0225, 2.4479, 1.1253, 2.0816],
       device='cuda:2'), covar=tensor([0.2164, 0.2393, 0.1712, 0.0935, 0.1229, 0.1268, 0.2262, 0.1397],
       device='cuda:2'), in_proj_covar=tensor([0.0103, 0.0117, 0.0135, 0.0147, 0.0105, 0.0141, 0.0130, 0.0114],
       device='cuda:2'), out_proj_covar=tensor([0.0003, 0.0003, 0.0004, 0.0004, 0.0003, 0.0004, 0.0003, 0.0003],
       device='cuda:2')
2022-12-23 15:17:49,162 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 2.373e+02 3.652e+02 4.480e+02 5.377e+02 1.144e+03, threshold=8.960e+02, percent-clipped=5.0
2022-12-23 15:18:08,637 WARNING [train.py:1060] (2/4) Exclude cut with ID 3557-8342-0013-54691-0_sp1.1 from training. Duration: 0.836375
2022-12-23 15:18:30,942 INFO [train.py:894] (2/4) Epoch 20, batch 1650, loss[loss=0.1806, simple_loss=0.2692, pruned_loss=0.04603, over 18536.00 frames. ], tot_loss[loss=0.1772, simple_loss=0.265, pruned_loss=0.04471, over 3713556.93 frames. ], batch size: 58, lr: 5.75e-03, grad_scale: 8.0
2022-12-23 15:18:41,632 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([5.7481, 4.8184, 5.0372, 5.6948, 5.2497, 5.2056, 5.8339, 1.5858],
       device='cuda:2'), covar=tensor([0.0656, 0.0711, 0.0564, 0.0812, 0.1480, 0.1020, 0.0427, 0.5339],
       device='cuda:2'), in_proj_covar=tensor([0.0335, 0.0219, 0.0230, 0.0258, 0.0316, 0.0264, 0.0281, 0.0275],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0001, 0.0001, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-23 15:18:54,962 WARNING [train.py:1060] (2/4) Exclude cut with ID 8565-290391-0049-67394-0_sp0.9 from training. Duration: 21.3166875
2022-12-23 15:19:24,372 WARNING [train.py:1060] (2/4) Exclude cut with ID 6533-399-0029-104863-0_sp0.9 from training. Duration: 22.1055625
2022-12-23 15:19:26,224 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([2.9390, 2.8769, 2.1841, 1.8697, 3.3456, 3.4272, 2.8802, 2.4260],
       device='cuda:2'), covar=tensor([0.0362, 0.0319, 0.0553, 0.0695, 0.0238, 0.0289, 0.0400, 0.0619],
       device='cuda:2'), in_proj_covar=tensor([0.0122, 0.0125, 0.0128, 0.0119, 0.0096, 0.0121, 0.0134, 0.0154],
       device='cuda:2'), out_proj_covar=tensor([0.0001, 0.0002, 0.0002, 0.0001, 0.0001, 0.0001, 0.0002, 0.0002],
       device='cuda:2')
2022-12-23 15:19:35,116 WARNING [train.py:1060] (2/4) Exclude cut with ID 7699-105389-0094-26379-0_sp1.1 from training. Duration: 21.77725
2022-12-23 15:19:42,604 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([2.6857, 1.8260, 0.6460, 2.0198, 2.7306, 1.7025, 2.2357, 2.5085],
       device='cuda:2'), covar=tensor([0.1338, 0.1851, 0.2322, 0.1323, 0.1415, 0.1490, 0.1274, 0.1410],
       device='cuda:2'), in_proj_covar=tensor([0.0095, 0.0099, 0.0117, 0.0097, 0.0117, 0.0092, 0.0099, 0.0094],
       device='cuda:2'), out_proj_covar=tensor([0.0003, 0.0003, 0.0004, 0.0003, 0.0004, 0.0003, 0.0003, 0.0003],
       device='cuda:2')
2022-12-23 15:19:45,558 INFO [train.py:894] (2/4) Epoch 20, batch 1700, loss[loss=0.1695, simple_loss=0.2534, pruned_loss=0.04277, over 18665.00 frames. ], tot_loss[loss=0.1789, simple_loss=0.2663, pruned_loss=0.0458, over 3713991.81 frames. ], batch size: 48, lr: 5.75e-03, grad_scale: 8.0
2022-12-23 15:19:57,118 WARNING [train.py:1060] (2/4) Exclude cut with ID 7255-291500-0005-134304-0_sp0.9 from training. Duration: 27.8166875
2022-12-23 15:20:14,640 INFO [zipformer.py:660] (2/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=68336.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 15:20:15,829 INFO [zipformer.py:660] (2/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=68337.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 15:20:18,385 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 2.300e+02 4.261e+02 4.999e+02 6.266e+02 1.831e+03, threshold=9.998e+02, percent-clipped=7.0
2022-12-23 15:20:21,316 WARNING [train.py:1060] (2/4) Exclude cut with ID 7357-94126-0021-15852-0_sp1.1 from training. Duration: 22.5090625
2022-12-23 15:20:28,584 WARNING [train.py:1060] (2/4) Exclude cut with ID 7255-291500-0005-134304-0 from training. Duration: 25.035
2022-12-23 15:20:31,680 INFO [zipformer.py:660] (2/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=68348.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 15:20:46,700 WARNING [train.py:1060] (2/4) Exclude cut with ID 774-127930-0014-10412-0_sp1.1 from training. Duration: 0.95
2022-12-23 15:20:52,816 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.6276, 1.3638, 0.9867, 0.3129, 1.0982, 1.5173, 1.2946, 1.4011],
       device='cuda:2'), covar=tensor([0.0618, 0.0558, 0.0987, 0.1502, 0.1084, 0.1432, 0.1627, 0.0681],
       device='cuda:2'), in_proj_covar=tensor([0.0171, 0.0183, 0.0204, 0.0190, 0.0209, 0.0200, 0.0214, 0.0202],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-23 15:20:59,560 INFO [train.py:894] (2/4) Epoch 20, batch 1750, loss[loss=0.19, simple_loss=0.2767, pruned_loss=0.05166, over 18729.00 frames. ], tot_loss[loss=0.1799, simple_loss=0.2662, pruned_loss=0.0468, over 3713851.05 frames. ], batch size: 52, lr: 5.75e-03, grad_scale: 8.0
2022-12-23 15:21:05,876 WARNING [train.py:1060] (2/4) Exclude cut with ID 3033-130750-0096-55598-0_sp0.9 from training. Duration: 0.92225
2022-12-23 15:21:17,205 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.3376, 1.7521, 1.3357, 2.1285, 2.2869, 1.4595, 1.3924, 1.1058],
       device='cuda:2'), covar=tensor([0.2346, 0.2096, 0.1928, 0.1136, 0.1499, 0.1419, 0.2394, 0.1916],
       device='cuda:2'), in_proj_covar=tensor([0.0242, 0.0222, 0.0211, 0.0197, 0.0258, 0.0194, 0.0220, 0.0194],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-23 15:21:25,499 INFO [zipformer.py:660] (2/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=68384.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 15:21:32,755 WARNING [train.py:1060] (2/4) Exclude cut with ID 4511-76322-0006-80011-0 from training. Duration: 21.97
2022-12-23 15:21:43,008 INFO [zipformer.py:660] (2/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=68396.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 15:21:48,145 INFO [zipformer.py:660] (2/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=68399.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 15:21:53,956 WARNING [train.py:1060] (2/4) Exclude cut with ID 7492-105653-0055-62765-0_sp0.9 from training. Duration: 21.97225
2022-12-23 15:21:55,392 WARNING [train.py:1060] (2/4) Exclude cut with ID 453-131332-0000-47844-0_sp0.9 from training. Duration: 25.3333125
2022-12-23 15:22:06,814 WARNING [train.py:1060] (2/4) Exclude cut with ID 5172-29468-0015-19128-0_sp0.9 from training. Duration: 21.5055625
2022-12-23 15:22:14,871 INFO [train.py:894] (2/4) Epoch 20, batch 1800, loss[loss=0.1686, simple_loss=0.2437, pruned_loss=0.04673, over 18489.00 frames. ], tot_loss[loss=0.1819, simple_loss=0.2667, pruned_loss=0.04856, over 3712885.99 frames. ], batch size: 43, lr: 5.75e-03, grad_scale: 8.0
2022-12-23 15:22:16,419 WARNING [train.py:1060] (2/4) Exclude cut with ID 453-131332-0000-47844-0_sp1.1 from training. Duration: 20.72725
2022-12-23 15:22:28,467 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([2.0725, 2.0864, 1.9817, 2.0568, 1.6412, 5.1652, 2.0720, 2.3946],
       device='cuda:2'), covar=tensor([0.3051, 0.1801, 0.1822, 0.1948, 0.1281, 0.0100, 0.1468, 0.0799],
       device='cuda:2'), in_proj_covar=tensor([0.0132, 0.0117, 0.0126, 0.0121, 0.0103, 0.0096, 0.0092, 0.0088],
       device='cuda:2'), out_proj_covar=tensor([0.0005, 0.0005, 0.0005, 0.0005, 0.0004, 0.0003, 0.0004, 0.0004],
       device='cuda:2')
2022-12-23 15:22:46,889 WARNING [train.py:1060] (2/4) Exclude cut with ID 8631-249866-0030-130156-0_sp0.9 from training. Duration: 26.32775
2022-12-23 15:22:48,147 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 2.646e+02 4.362e+02 5.315e+02 6.425e+02 1.643e+03, threshold=1.063e+03, percent-clipped=4.0
2022-12-23 15:22:54,412 INFO [zipformer.py:660] (2/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=68443.0, num_to_drop=1, layers_to_drop={0}
2022-12-23 15:23:20,188 WARNING [train.py:1060] (2/4) Exclude cut with ID 3867-173237-0077-144769-0 from training. Duration: 20.025
2022-12-23 15:23:20,537 INFO [zipformer.py:660] (2/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=68460.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 15:23:24,902 WARNING [train.py:1060] (2/4) Exclude cut with ID 6709-74022-0004-86860-0_sp1.1 from training. Duration: 0.9409375
2022-12-23 15:23:24,913 WARNING [train.py:1060] (2/4) Exclude cut with ID 4757-1811-0023-62229-0_sp0.9 from training. Duration: 21.37775
2022-12-23 15:23:30,995 INFO [train.py:894] (2/4) Epoch 20, batch 1850, loss[loss=0.176, simple_loss=0.2441, pruned_loss=0.0539, over 18386.00 frames. ], tot_loss[loss=0.1847, simple_loss=0.2687, pruned_loss=0.05041, over 3713232.66 frames. ], batch size: 42, lr: 5.74e-03, grad_scale: 8.0
2022-12-23 15:23:45,458 WARNING [train.py:1060] (2/4) Exclude cut with ID 1250-135782-0004-25974-0_sp0.9 from training. Duration: 21.17225
2022-12-23 15:23:45,469 WARNING [train.py:1060] (2/4) Exclude cut with ID 7357-94126-0021-15852-0_sp0.9 from training. Duration: 27.511125
2022-12-23 15:24:00,314 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.7119, 1.4673, 1.1071, 0.2981, 1.2122, 1.5988, 1.4387, 1.5404],
       device='cuda:2'), covar=tensor([0.0646, 0.0540, 0.0975, 0.1687, 0.1072, 0.1565, 0.1622, 0.0630],
       device='cuda:2'), in_proj_covar=tensor([0.0172, 0.0184, 0.0205, 0.0191, 0.0210, 0.0202, 0.0216, 0.0202],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-23 15:24:19,220 WARNING [train.py:1060] (2/4) Exclude cut with ID 453-131332-0000-47844-0 from training. Duration: 22.8
2022-12-23 15:24:23,754 WARNING [train.py:1060] (2/4) Exclude cut with ID 4964-30587-0040-44509-0 from training. Duration: 22.585
2022-12-23 15:24:46,965 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.8112, 1.7973, 1.6277, 1.6029, 1.8921, 1.9360, 1.9992, 1.3655],
       device='cuda:2'), covar=tensor([0.0306, 0.0223, 0.0448, 0.0198, 0.0196, 0.0391, 0.0241, 0.0291],
       device='cuda:2'), in_proj_covar=tensor([0.0095, 0.0127, 0.0153, 0.0126, 0.0117, 0.0120, 0.0098, 0.0126],
       device='cuda:2'), out_proj_covar=tensor([7.5708e-05, 1.0130e-04, 1.2632e-04, 1.0095e-04, 9.5200e-05, 9.2500e-05,
        7.6433e-05, 9.9467e-05], device='cuda:2')
2022-12-23 15:24:48,438 INFO [train.py:894] (2/4) Epoch 20, batch 1900, loss[loss=0.181, simple_loss=0.2466, pruned_loss=0.05769, over 18394.00 frames. ], tot_loss[loss=0.1865, simple_loss=0.2692, pruned_loss=0.05193, over 3712825.91 frames. ], batch size: 42, lr: 5.74e-03, grad_scale: 8.0
2022-12-23 15:24:55,734 WARNING [train.py:1060] (2/4) Exclude cut with ID 6978-92210-0001-146967-0_sp0.9 from training. Duration: 22.0166875
2022-12-23 15:25:12,868 WARNING [train.py:1060] (2/4) Exclude cut with ID 7255-291500-0003-134302-0_sp1.1 from training. Duration: 24.395375
2022-12-23 15:25:16,941 WARNING [train.py:1060] (2/4) Exclude cut with ID 3488-85273-0017-41203-0_sp0.9 from training. Duration: 27.47775
2022-12-23 15:25:19,915 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 3.101e+02 4.390e+02 5.434e+02 6.998e+02 1.389e+03, threshold=1.087e+03, percent-clipped=7.0
2022-12-23 15:25:20,574 WARNING [train.py:1060] (2/4) Exclude cut with ID 432-122774-0017-62487-0_sp0.9 from training. Duration: 24.8833125
2022-12-23 15:25:23,729 WARNING [train.py:1060] (2/4) Exclude cut with ID 6758-72288-0033-108368-0 from training. Duration: 23.39
2022-12-23 15:25:29,739 WARNING [train.py:1060] (2/4) Exclude cut with ID 8291-282929-0007-12994-0_sp0.9 from training. Duration: 28.72225
2022-12-23 15:25:39,439 WARNING [train.py:1060] (2/4) Exclude cut with ID 585-294811-0110-133686-0_sp0.9 from training. Duration: 20.8944375
2022-12-23 15:25:55,786 WARNING [train.py:1060] (2/4) Exclude cut with ID 5796-66357-0007-116447-0_sp0.9 from training. Duration: 23.8444375
2022-12-23 15:26:04,395 INFO [train.py:894] (2/4) Epoch 20, batch 1950, loss[loss=0.1675, simple_loss=0.2543, pruned_loss=0.04033, over 18669.00 frames. ], tot_loss[loss=0.1872, simple_loss=0.2696, pruned_loss=0.05244, over 3712542.00 frames. ], batch size: 48, lr: 5.74e-03, grad_scale: 8.0
2022-12-23 15:26:20,158 WARNING [train.py:1060] (2/4) Exclude cut with ID 8291-282929-0007-12994-0 from training. Duration: 25.85
2022-12-23 15:26:20,170 WARNING [train.py:1060] (2/4) Exclude cut with ID 8291-282929-0023-13010-0 from training. Duration: 21.39
2022-12-23 15:26:20,582 INFO [zipformer.py:660] (2/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=68578.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 15:26:31,397 WARNING [train.py:1060] (2/4) Exclude cut with ID 7357-94126-0014-15845-0 from training. Duration: 27.92
2022-12-23 15:26:57,730 WARNING [train.py:1060] (2/4) Exclude cut with ID 8631-249866-0039-130165-0_sp0.9 from training. Duration: 20.661125
2022-12-23 15:27:10,833 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([2.5395, 2.1094, 2.1118, 2.1911, 2.4007, 2.3850, 2.4189, 1.9569],
       device='cuda:2'), covar=tensor([0.2020, 0.2974, 0.2338, 0.2944, 0.1833, 0.0905, 0.2899, 0.1188],
       device='cuda:2'), in_proj_covar=tensor([0.0265, 0.0297, 0.0275, 0.0312, 0.0302, 0.0249, 0.0340, 0.0238],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-23 15:27:19,094 INFO [train.py:894] (2/4) Epoch 20, batch 2000, loss[loss=0.1762, simple_loss=0.256, pruned_loss=0.04824, over 18398.00 frames. ], tot_loss[loss=0.1876, simple_loss=0.2696, pruned_loss=0.05283, over 3713022.77 frames. ], batch size: 46, lr: 5.74e-03, grad_scale: 8.0
2022-12-23 15:27:22,121 WARNING [train.py:1060] (2/4) Exclude cut with ID 7357-94126-0043-15874-0_sp0.9 from training. Duration: 20.07225
2022-12-23 15:27:29,608 WARNING [train.py:1060] (2/4) Exclude cut with ID 1085-156170-0017-128270-0 from training. Duration: 21.01
2022-12-23 15:27:50,166 INFO [zipformer.py:660] (2/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=68637.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 15:27:52,675 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 3.139e+02 4.584e+02 5.306e+02 6.807e+02 1.087e+03, threshold=1.061e+03, percent-clipped=0.0
2022-12-23 15:27:53,166 INFO [zipformer.py:660] (2/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=68639.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 15:28:00,872 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.7163, 1.7294, 1.7215, 1.6823, 1.1864, 3.6401, 1.5596, 2.0907],
       device='cuda:2'), covar=tensor([0.3349, 0.2126, 0.1974, 0.2087, 0.1494, 0.0201, 0.1470, 0.0813],
       device='cuda:2'), in_proj_covar=tensor([0.0132, 0.0117, 0.0125, 0.0120, 0.0102, 0.0096, 0.0091, 0.0088],
       device='cuda:2'), out_proj_covar=tensor([0.0005, 0.0005, 0.0005, 0.0005, 0.0004, 0.0003, 0.0004, 0.0004],
       device='cuda:2')
2022-12-23 15:28:35,210 INFO [train.py:894] (2/4) Epoch 20, batch 2050, loss[loss=0.1997, simple_loss=0.2812, pruned_loss=0.05912, over 18658.00 frames. ], tot_loss[loss=0.1882, simple_loss=0.2701, pruned_loss=0.05313, over 3714009.52 frames. ], batch size: 62, lr: 5.74e-03, grad_scale: 8.0
2022-12-23 15:28:38,182 WARNING [train.py:1060] (2/4) Exclude cut with ID 2195-150901-0045-59933-0 from training. Duration: 20.65
2022-12-23 15:28:40,036 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.7946, 1.3100, 2.2183, 3.4228, 2.4377, 2.4852, 1.0918, 2.4803],
       device='cuda:2'), covar=tensor([0.1771, 0.1754, 0.1301, 0.0647, 0.1052, 0.1093, 0.1964, 0.0978],
       device='cuda:2'), in_proj_covar=tensor([0.0102, 0.0118, 0.0135, 0.0148, 0.0106, 0.0141, 0.0130, 0.0113],
       device='cuda:2'), out_proj_covar=tensor([0.0003, 0.0003, 0.0004, 0.0004, 0.0003, 0.0004, 0.0003, 0.0003],
       device='cuda:2')
2022-12-23 15:28:45,654 WARNING [train.py:1060] (2/4) Exclude cut with ID 5796-66357-0007-116447-0 from training. Duration: 21.46
2022-12-23 15:29:03,105 INFO [zipformer.py:660] (2/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=68685.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 15:29:03,372 INFO [zipformer.py:660] (2/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=68685.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 15:29:28,001 INFO [scaling.py:679] (2/4) Whitening: num_groups=8, num_channels=192, metric=1.62 vs. limit=2.0
2022-12-23 15:29:30,056 WARNING [train.py:1060] (2/4) Exclude cut with ID 3557-8342-0013-54691-0 from training. Duration: 0.92
2022-12-23 15:29:36,637 WARNING [train.py:1060] (2/4) Exclude cut with ID 8291-282929-0023-13010-0_sp0.9 from training. Duration: 23.7666875
2022-12-23 15:29:47,270 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.7343, 1.7515, 1.7530, 1.7669, 1.4037, 3.9078, 1.7471, 2.1294],
       device='cuda:2'), covar=tensor([0.3089, 0.1955, 0.1821, 0.1922, 0.1363, 0.0165, 0.1450, 0.0819],
       device='cuda:2'), in_proj_covar=tensor([0.0132, 0.0117, 0.0125, 0.0120, 0.0103, 0.0097, 0.0092, 0.0089],
       device='cuda:2'), out_proj_covar=tensor([0.0005, 0.0005, 0.0005, 0.0005, 0.0004, 0.0003, 0.0004, 0.0004],
       device='cuda:2')
2022-12-23 15:29:51,479 INFO [train.py:894] (2/4) Epoch 20, batch 2100, loss[loss=0.1861, simple_loss=0.2667, pruned_loss=0.05274, over 18417.00 frames. ], tot_loss[loss=0.1876, simple_loss=0.2689, pruned_loss=0.05314, over 3713837.49 frames. ], batch size: 48, lr: 5.73e-03, grad_scale: 8.0
2022-12-23 15:29:53,295 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.8668, 1.1924, 0.8700, 1.3188, 2.1336, 1.3337, 1.5515, 1.6871],
       device='cuda:2'), covar=tensor([0.1647, 0.2262, 0.2320, 0.1635, 0.1831, 0.1771, 0.1489, 0.1722],
       device='cuda:2'), in_proj_covar=tensor([0.0096, 0.0100, 0.0119, 0.0098, 0.0119, 0.0093, 0.0099, 0.0095],
       device='cuda:2'), out_proj_covar=tensor([0.0003, 0.0003, 0.0004, 0.0003, 0.0004, 0.0003, 0.0003, 0.0003],
       device='cuda:2')
2022-12-23 15:30:13,761 WARNING [train.py:1060] (2/4) Exclude cut with ID 8544-281189-0060-101339-0_sp0.9 from training. Duration: 20.861125
2022-12-23 15:30:22,653 WARNING [train.py:1060] (2/4) Exclude cut with ID 3488-65654-0031-41259-0_sp0.9 from training. Duration: 22.711125
2022-12-23 15:30:24,147 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 2.984e+02 4.264e+02 4.924e+02 6.073e+02 1.296e+03, threshold=9.848e+02, percent-clipped=2.0
2022-12-23 15:30:30,242 INFO [zipformer.py:660] (2/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=68743.0, num_to_drop=1, layers_to_drop={0}
2022-12-23 15:30:35,315 INFO [zipformer.py:660] (2/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=68746.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 15:30:49,298 INFO [zipformer.py:660] (2/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=68755.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 15:31:02,346 WARNING [train.py:1060] (2/4) Exclude cut with ID 6951-79737-0043-132310-0_sp1.1 from training. Duration: 22.986375
2022-12-23 15:31:06,590 INFO [train.py:894] (2/4) Epoch 20, batch 2150, loss[loss=0.197, simple_loss=0.2813, pruned_loss=0.0564, over 18725.00 frames. ], tot_loss[loss=0.1881, simple_loss=0.2696, pruned_loss=0.05333, over 3714806.29 frames. ], batch size: 52, lr: 5.73e-03, grad_scale: 8.0
2022-12-23 15:31:18,252 WARNING [train.py:1060] (2/4) Exclude cut with ID 8040-260924-0003-80960-0_sp0.9 from training. Duration: 22.07225
2022-12-23 15:31:21,242 WARNING [train.py:1060] (2/4) Exclude cut with ID 7699-105389-0045-26330-0_sp0.9 from training. Duration: 20.3055625
2022-12-23 15:31:24,194 WARNING [train.py:1060] (2/4) Exclude cut with ID 6356-271890-0060-94317-0_sp0.9 from training. Duration: 20.72225
2022-12-23 15:31:24,787 INFO [scaling.py:679] (2/4) Whitening: num_groups=8, num_channels=192, metric=1.62 vs. limit=2.0
2022-12-23 15:31:42,098 WARNING [train.py:1060] (2/4) Exclude cut with ID 3488-85273-0017-41203-0_sp1.1 from training. Duration: 22.4818125
2022-12-23 15:31:44,317 INFO [zipformer.py:660] (2/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=68791.0, num_to_drop=1, layers_to_drop={0}
2022-12-23 15:32:02,439 INFO [scaling.py:679] (2/4) Whitening: num_groups=8, num_channels=192, metric=1.64 vs. limit=2.0
2022-12-23 15:32:10,105 WARNING [train.py:1060] (2/4) Exclude cut with ID 4964-30587-0040-44509-0_sp0.9 from training. Duration: 25.0944375
2022-12-23 15:32:13,086 WARNING [train.py:1060] (2/4) Exclude cut with ID 6533-399-0047-104881-0 from training. Duration: 21.515
2022-12-23 15:32:20,575 WARNING [train.py:1060] (2/4) Exclude cut with ID 7357-94126-0009-15840-0_sp0.9 from training. Duration: 27.02225
2022-12-23 15:32:23,501 INFO [train.py:894] (2/4) Epoch 20, batch 2200, loss[loss=0.1945, simple_loss=0.2663, pruned_loss=0.06139, over 18701.00 frames. ], tot_loss[loss=0.1889, simple_loss=0.2703, pruned_loss=0.05373, over 3714804.20 frames. ], batch size: 46, lr: 5.73e-03, grad_scale: 8.0
2022-12-23 15:32:24,207 WARNING [train.py:1060] (2/4) Exclude cut with ID 432-122774-0010-62480-0_sp0.9 from training. Duration: 22.22225
2022-12-23 15:32:33,548 WARNING [train.py:1060] (2/4) Exclude cut with ID 4964-30587-0085-44554-0_sp0.9 from training. Duration: 20.85
2022-12-23 15:32:34,297 INFO [scaling.py:679] (2/4) Whitening: num_groups=8, num_channels=96, metric=1.41 vs. limit=2.0
2022-12-23 15:32:57,620 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 2.534e+02 4.308e+02 5.383e+02 7.138e+02 1.577e+03, threshold=1.077e+03, percent-clipped=5.0
2022-12-23 15:33:03,730 WARNING [train.py:1060] (2/4) Exclude cut with ID 4295-39940-0007-92567-0 from training. Duration: 21.54
2022-12-23 15:33:10,667 WARNING [train.py:1060] (2/4) Exclude cut with ID 4964-30587-0040-44509-0_sp1.1 from training. Duration: 20.5318125
2022-12-23 15:33:20,758 WARNING [train.py:1060] (2/4) Exclude cut with ID 7255-291500-0012-134311-0_sp0.9 from training. Duration: 21.9333125
2022-12-23 15:33:40,880 INFO [train.py:894] (2/4) Epoch 20, batch 2250, loss[loss=0.1719, simple_loss=0.2592, pruned_loss=0.04226, over 18717.00 frames. ], tot_loss[loss=0.1881, simple_loss=0.2697, pruned_loss=0.05327, over 3715811.40 frames. ], batch size: 54, lr: 5.73e-03, grad_scale: 8.0
2022-12-23 15:34:08,707 WARNING [train.py:1060] (2/4) Exclude cut with ID 8631-249866-0025-130151-0_sp0.9 from training. Duration: 21.7944375
2022-12-23 15:34:20,871 WARNING [train.py:1060] (2/4) Exclude cut with ID 8291-282929-0002-12989-0_sp0.9 from training. Duration: 22.4666875
2022-12-23 15:34:26,940 WARNING [train.py:1060] (2/4) Exclude cut with ID 6121-9014-0076-24124-0 from training. Duration: 21.635
2022-12-23 15:34:33,059 WARNING [train.py:1060] (2/4) Exclude cut with ID 6121-9014-0076-24124-0_sp0.9 from training. Duration: 24.038875
2022-12-23 15:34:57,026 INFO [train.py:894] (2/4) Epoch 20, batch 2300, loss[loss=0.1922, simple_loss=0.2715, pruned_loss=0.05643, over 18665.00 frames. ], tot_loss[loss=0.1889, simple_loss=0.2702, pruned_loss=0.05382, over 3714913.98 frames. ], batch size: 48, lr: 5.72e-03, grad_scale: 8.0
2022-12-23 15:34:57,919 INFO [scaling.py:679] (2/4) Whitening: num_groups=8, num_channels=96, metric=1.21 vs. limit=2.0
2022-12-23 15:35:15,176 WARNING [train.py:1060] (2/4) Exclude cut with ID 7255-291500-0009-134308-0_sp1.1 from training. Duration: 21.786375
2022-12-23 15:35:22,747 INFO [zipformer.py:660] (2/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=68934.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 15:35:26,331 WARNING [train.py:1060] (2/4) Exclude cut with ID 8291-282929-0002-12989-0 from training. Duration: 20.22
2022-12-23 15:35:30,638 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 3.085e+02 4.536e+02 5.472e+02 6.588e+02 1.270e+03, threshold=1.094e+03, percent-clipped=1.0
2022-12-23 15:36:13,471 INFO [train.py:894] (2/4) Epoch 20, batch 2350, loss[loss=0.1946, simple_loss=0.2761, pruned_loss=0.05658, over 18578.00 frames. ], tot_loss[loss=0.1884, simple_loss=0.2696, pruned_loss=0.05364, over 3714283.75 frames. ], batch size: 51, lr: 5.72e-03, grad_scale: 8.0
2022-12-23 15:36:29,232 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.8534, 1.8583, 1.8608, 1.9253, 1.5150, 4.7130, 2.0815, 2.4611],
       device='cuda:2'), covar=tensor([0.2961, 0.1919, 0.1852, 0.1924, 0.1359, 0.0124, 0.1481, 0.0810],
       device='cuda:2'), in_proj_covar=tensor([0.0132, 0.0118, 0.0126, 0.0121, 0.0103, 0.0097, 0.0092, 0.0089],
       device='cuda:2'), out_proj_covar=tensor([0.0005, 0.0005, 0.0005, 0.0005, 0.0004, 0.0003, 0.0004, 0.0004],
       device='cuda:2')
2022-12-23 15:37:22,357 WARNING [train.py:1060] (2/4) Exclude cut with ID 6951-79737-0043-132310-0 from training. Duration: 25.285
2022-12-23 15:37:27,096 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.2925, 1.2428, 0.9326, 1.3294, 1.3004, 2.4595, 1.2388, 1.4298],
       device='cuda:2'), covar=tensor([0.0861, 0.1945, 0.1070, 0.0922, 0.1686, 0.0338, 0.1464, 0.1545],
       device='cuda:2'), in_proj_covar=tensor([0.0073, 0.0083, 0.0073, 0.0075, 0.0091, 0.0076, 0.0085, 0.0078],
       device='cuda:2'), out_proj_covar=tensor([0.0004, 0.0004, 0.0004, 0.0004, 0.0004, 0.0003, 0.0004, 0.0004],
       device='cuda:2')
2022-12-23 15:37:30,428 INFO [train.py:894] (2/4) Epoch 20, batch 2400, loss[loss=0.1722, simple_loss=0.2615, pruned_loss=0.04145, over 18494.00 frames. ], tot_loss[loss=0.1888, simple_loss=0.2696, pruned_loss=0.05397, over 3715181.61 frames. ], batch size: 52, lr: 5.72e-03, grad_scale: 8.0
2022-12-23 15:38:02,921 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 2.677e+02 4.455e+02 5.012e+02 6.194e+02 1.142e+03, threshold=1.002e+03, percent-clipped=1.0
2022-12-23 15:38:05,975 INFO [zipformer.py:660] (2/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=69041.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 15:38:09,056 INFO [zipformer.py:660] (2/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=69043.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 15:38:10,658 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([2.1021, 1.9205, 2.2091, 1.3814, 2.2793, 2.3021, 1.6801, 2.6958],
       device='cuda:2'), covar=tensor([0.1120, 0.1766, 0.1257, 0.1870, 0.0705, 0.1094, 0.2168, 0.0491],
       device='cuda:2'), in_proj_covar=tensor([0.0199, 0.0215, 0.0212, 0.0197, 0.0178, 0.0219, 0.0219, 0.0194],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-23 15:38:27,593 INFO [zipformer.py:660] (2/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=69055.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 15:38:30,168 WARNING [train.py:1060] (2/4) Exclude cut with ID 811-130148-0001-63453-0_sp0.9 from training. Duration: 20.861125
2022-12-23 15:38:45,648 INFO [train.py:894] (2/4) Epoch 20, batch 2450, loss[loss=0.17, simple_loss=0.2456, pruned_loss=0.04714, over 18533.00 frames. ], tot_loss[loss=0.188, simple_loss=0.2688, pruned_loss=0.05363, over 3714635.25 frames. ], batch size: 44, lr: 5.72e-03, grad_scale: 8.0
2022-12-23 15:38:51,253 WARNING [train.py:1060] (2/4) Exclude cut with ID 6010-56788-0055-90261-0 from training. Duration: 20.88
2022-12-23 15:39:26,638 WARNING [train.py:1060] (2/4) Exclude cut with ID 7357-94126-0045-15876-0_sp0.9 from training. Duration: 23.4166875
2022-12-23 15:39:34,112 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([2.1703, 2.2429, 1.6741, 2.4429, 2.3832, 2.0254, 2.8248, 2.2560],
       device='cuda:2'), covar=tensor([0.0814, 0.1515, 0.2525, 0.1713, 0.1539, 0.0840, 0.1019, 0.1091],
       device='cuda:2'), in_proj_covar=tensor([0.0177, 0.0208, 0.0250, 0.0290, 0.0236, 0.0189, 0.0209, 0.0202],
       device='cuda:2'), out_proj_covar=tensor([0.0001, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-23 15:39:39,596 INFO [zipformer.py:660] (2/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=69103.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 15:39:41,336 INFO [zipformer.py:660] (2/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=69104.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 15:40:00,763 INFO [train.py:894] (2/4) Epoch 20, batch 2500, loss[loss=0.1791, simple_loss=0.2602, pruned_loss=0.04903, over 18581.00 frames. ], tot_loss[loss=0.189, simple_loss=0.2698, pruned_loss=0.05415, over 3715099.01 frames. ], batch size: 49, lr: 5.72e-03, grad_scale: 8.0
2022-12-23 15:40:33,620 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 2.874e+02 4.273e+02 5.084e+02 6.265e+02 1.046e+03, threshold=1.017e+03, percent-clipped=1.0
2022-12-23 15:40:41,399 WARNING [train.py:1060] (2/4) Exclude cut with ID 7255-291500-0006-134305-0 from training. Duration: 21.24
2022-12-23 15:40:41,417 WARNING [train.py:1060] (2/4) Exclude cut with ID 6533-399-0047-104881-0_sp0.9 from training. Duration: 23.9055625
2022-12-23 15:41:09,445 INFO [zipformer.py:660] (2/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=69162.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 15:41:15,253 WARNING [train.py:1060] (2/4) Exclude cut with ID 6758-72288-0033-108368-0_sp0.9 from training. Duration: 25.988875
2022-12-23 15:41:16,787 INFO [train.py:894] (2/4) Epoch 20, batch 2550, loss[loss=0.2072, simple_loss=0.284, pruned_loss=0.06517, over 18661.00 frames. ], tot_loss[loss=0.1883, simple_loss=0.2692, pruned_loss=0.05373, over 3714668.09 frames. ], batch size: 180, lr: 5.71e-03, grad_scale: 8.0
2022-12-23 15:41:19,445 INFO [zipformer.py:660] (2/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=69168.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 15:41:22,094 WARNING [train.py:1060] (2/4) Exclude cut with ID 7255-291500-0001-134300-0_sp0.9 from training. Duration: 20.67225
2022-12-23 15:42:05,889 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([0.6232, 1.9069, 2.1128, 1.1517, 1.5085, 2.2715, 2.0228, 1.7806],
       device='cuda:2'), covar=tensor([0.0815, 0.0344, 0.0308, 0.0401, 0.0351, 0.0409, 0.0251, 0.0745],
       device='cuda:2'), in_proj_covar=tensor([0.0151, 0.0171, 0.0126, 0.0140, 0.0148, 0.0143, 0.0163, 0.0173],
       device='cuda:2'), out_proj_covar=tensor([1.1576e-04, 1.3101e-04, 9.5349e-05, 1.0464e-04, 1.1131e-04, 1.0974e-04,
        1.2573e-04, 1.3190e-04], device='cuda:2')
2022-12-23 15:42:11,692 WARNING [train.py:1060] (2/4) Exclude cut with ID 3488-85273-0038-41224-0 from training. Duration: 20.34
2022-12-23 15:42:29,801 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.6425, 1.4274, 1.4815, 1.8713, 1.6505, 3.4035, 1.4036, 1.6714],
       device='cuda:2'), covar=tensor([0.0894, 0.1916, 0.1156, 0.0950, 0.1570, 0.0251, 0.1430, 0.1528],
       device='cuda:2'), in_proj_covar=tensor([0.0073, 0.0083, 0.0073, 0.0074, 0.0091, 0.0076, 0.0085, 0.0077],
       device='cuda:2'), out_proj_covar=tensor([0.0004, 0.0004, 0.0004, 0.0004, 0.0004, 0.0003, 0.0004, 0.0004],
       device='cuda:2')
2022-12-23 15:42:33,139 INFO [train.py:894] (2/4) Epoch 20, batch 2600, loss[loss=0.188, simple_loss=0.2802, pruned_loss=0.0479, over 18579.00 frames. ], tot_loss[loss=0.1886, simple_loss=0.2696, pruned_loss=0.05382, over 3714359.30 frames. ], batch size: 57, lr: 5.71e-03, grad_scale: 8.0
2022-12-23 15:42:42,262 INFO [zipformer.py:660] (2/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=69223.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 15:42:45,300 INFO [zipformer.py:660] (2/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=69225.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 15:42:51,333 INFO [zipformer.py:660] (2/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=69229.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 15:42:59,377 INFO [zipformer.py:660] (2/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=69234.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 15:43:06,597 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 2.608e+02 4.487e+02 5.367e+02 6.345e+02 1.020e+03, threshold=1.073e+03, percent-clipped=1.0
2022-12-23 15:43:15,745 INFO [zipformer.py:660] (2/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=69245.0, num_to_drop=1, layers_to_drop={0}
2022-12-23 15:43:17,220 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([2.4541, 2.5462, 2.8592, 1.5981, 3.0197, 2.8571, 2.1187, 3.3477],
       device='cuda:2'), covar=tensor([0.1225, 0.1651, 0.1484, 0.2101, 0.0687, 0.1290, 0.2074, 0.0506],
       device='cuda:2'), in_proj_covar=tensor([0.0200, 0.0215, 0.0211, 0.0198, 0.0178, 0.0221, 0.0220, 0.0195],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-23 15:43:21,636 WARNING [train.py:1060] (2/4) Exclude cut with ID 7357-94126-0026-15857-0_sp0.9 from training. Duration: 25.061125
2022-12-23 15:43:35,412 WARNING [train.py:1060] (2/4) Exclude cut with ID 3033-130750-0096-55598-0 from training. Duration: 0.83
2022-12-23 15:43:48,923 INFO [train.py:894] (2/4) Epoch 20, batch 2650, loss[loss=0.1963, simple_loss=0.2769, pruned_loss=0.05783, over 18568.00 frames. ], tot_loss[loss=0.1891, simple_loss=0.27, pruned_loss=0.05408, over 3714250.14 frames. ], batch size: 99, lr: 5.71e-03, grad_scale: 8.0
2022-12-23 15:43:58,044 INFO [zipformer.py:660] (2/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=69273.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 15:44:01,043 WARNING [train.py:1060] (2/4) Exclude cut with ID 3488-85273-0017-41203-0 from training. Duration: 24.73
2022-12-23 15:44:12,003 INFO [zipformer.py:660] (2/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=69282.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 15:44:13,309 WARNING [train.py:1060] (2/4) Exclude cut with ID 7255-291500-0009-134308-0 from training. Duration: 23.965
2022-12-23 15:44:18,294 INFO [zipformer.py:660] (2/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=69286.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 15:44:20,985 WARNING [train.py:1060] (2/4) Exclude cut with ID 6978-92210-0030-146996-0_sp0.9 from training. Duration: 22.088875
2022-12-23 15:44:27,595 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.3002, 1.1494, 1.4906, 2.1791, 1.5956, 2.1776, 0.8241, 1.6375],
       device='cuda:2'), covar=tensor([0.2003, 0.1689, 0.1341, 0.0847, 0.1275, 0.1014, 0.1967, 0.1303],
       device='cuda:2'), in_proj_covar=tensor([0.0103, 0.0118, 0.0136, 0.0149, 0.0107, 0.0141, 0.0131, 0.0114],
       device='cuda:2'), out_proj_covar=tensor([0.0003, 0.0003, 0.0004, 0.0004, 0.0003, 0.0004, 0.0004, 0.0003],
       device='cuda:2')
2022-12-23 15:44:36,677 WARNING [train.py:1060] (2/4) Exclude cut with ID 7255-291500-0006-134305-0_sp0.9 from training. Duration: 23.6
2022-12-23 15:44:49,797 INFO [zipformer.py:660] (2/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=69306.0, num_to_drop=1, layers_to_drop={0}
2022-12-23 15:44:57,293 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([2.6992, 4.2243, 4.0595, 1.8621, 4.3266, 3.1973, 0.8090, 2.9836],
       device='cuda:2'), covar=tensor([0.2015, 0.0954, 0.1249, 0.3351, 0.0784, 0.0872, 0.4999, 0.1245],
       device='cuda:2'), in_proj_covar=tensor([0.0147, 0.0145, 0.0159, 0.0126, 0.0146, 0.0115, 0.0147, 0.0115],
       device='cuda:2'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0002, 0.0003, 0.0002, 0.0003, 0.0002],
       device='cuda:2')
2022-12-23 15:45:06,093 INFO [train.py:894] (2/4) Epoch 20, batch 2700, loss[loss=0.2114, simple_loss=0.2906, pruned_loss=0.06609, over 18671.00 frames. ], tot_loss[loss=0.1888, simple_loss=0.2697, pruned_loss=0.05394, over 3713820.41 frames. ], batch size: 99, lr: 5.71e-03, grad_scale: 16.0
2022-12-23 15:45:11,000 INFO [zipformer.py:660] (2/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=69320.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 15:45:32,386 INFO [zipformer.py:660] (2/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=69334.0, num_to_drop=1, layers_to_drop={3}
2022-12-23 15:45:39,090 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 2.969e+02 4.616e+02 5.478e+02 6.566e+02 1.187e+03, threshold=1.096e+03, percent-clipped=2.0
2022-12-23 15:45:42,669 INFO [zipformer.py:660] (2/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=69341.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 15:46:17,865 WARNING [train.py:1060] (2/4) Exclude cut with ID 8291-282929-0024-13011-0 from training. Duration: 23.795
2022-12-23 15:46:20,602 INFO [train.py:894] (2/4) Epoch 20, batch 2750, loss[loss=0.2056, simple_loss=0.2853, pruned_loss=0.0629, over 18729.00 frames. ], tot_loss[loss=0.1887, simple_loss=0.2692, pruned_loss=0.05412, over 3714003.91 frames. ], batch size: 54, lr: 5.71e-03, grad_scale: 16.0
2022-12-23 15:46:22,370 INFO [zipformer.py:660] (2/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=69368.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 15:46:35,051 WARNING [train.py:1060] (2/4) Exclude cut with ID 8631-249866-0030-130156-0_sp1.1 from training. Duration: 21.5409375
2022-12-23 15:46:38,243 WARNING [train.py:1060] (2/4) Exclude cut with ID 6978-92210-0019-146985-0_sp0.9 from training. Duration: 24.97775
2022-12-23 15:46:42,735 INFO [zipformer.py:660] (2/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=69381.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 15:46:49,426 WARNING [train.py:1060] (2/4) Exclude cut with ID 1085-156170-0017-128270-0_sp0.9 from training. Duration: 23.3444375
2022-12-23 15:46:53,923 INFO [zipformer.py:660] (2/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=69389.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 15:46:56,637 INFO [scaling.py:679] (2/4) Whitening: num_groups=8, num_channels=96, metric=1.20 vs. limit=2.0
2022-12-23 15:47:01,016 INFO [scaling.py:679] (2/4) Whitening: num_groups=8, num_channels=96, metric=1.27 vs. limit=2.0
2022-12-23 15:47:01,955 INFO [zipformer.py:660] (2/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=69394.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 15:47:08,953 INFO [zipformer.py:660] (2/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=69399.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 15:47:18,645 WARNING [train.py:1060] (2/4) Exclude cut with ID 6010-56788-0055-90261-0_sp0.9 from training. Duration: 23.2
2022-12-23 15:47:23,117 WARNING [train.py:1060] (2/4) Exclude cut with ID 5653-46179-0060-117930-0_sp0.9 from training. Duration: 21.17225
2022-12-23 15:47:36,458 INFO [train.py:894] (2/4) Epoch 20, batch 2800, loss[loss=0.1837, simple_loss=0.2652, pruned_loss=0.05113, over 18668.00 frames. ], tot_loss[loss=0.188, simple_loss=0.2684, pruned_loss=0.05376, over 3714151.11 frames. ], batch size: 48, lr: 5.70e-03, grad_scale: 16.0
2022-12-23 15:47:43,249 WARNING [train.py:1060] (2/4) Exclude cut with ID 7255-291500-0008-134307-0_sp0.9 from training. Duration: 24.6555625
2022-12-23 15:47:54,930 INFO [zipformer.py:660] (2/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=69429.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 15:48:09,260 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 2.532e+02 4.423e+02 5.251e+02 6.577e+02 9.250e+02, threshold=1.050e+03, percent-clipped=0.0
2022-12-23 15:48:23,261 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([2.0117, 2.0819, 2.2948, 1.2912, 2.3590, 2.4524, 1.7448, 2.7817],
       device='cuda:2'), covar=tensor([0.1124, 0.1643, 0.1269, 0.1986, 0.0675, 0.1074, 0.2159, 0.0484],
       device='cuda:2'), in_proj_covar=tensor([0.0201, 0.0216, 0.0211, 0.0199, 0.0178, 0.0221, 0.0221, 0.0195],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-23 15:48:33,264 INFO [zipformer.py:660] (2/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=69455.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 15:48:38,895 WARNING [train.py:1060] (2/4) Exclude cut with ID 3488-65654-0031-41259-0 from training. Duration: 20.44
2022-12-23 15:48:50,735 INFO [train.py:894] (2/4) Epoch 20, batch 2850, loss[loss=0.1896, simple_loss=0.2707, pruned_loss=0.05427, over 18718.00 frames. ], tot_loss[loss=0.1879, simple_loss=0.2687, pruned_loss=0.05349, over 3714187.57 frames. ], batch size: 52, lr: 5.70e-03, grad_scale: 16.0
2022-12-23 15:48:54,350 WARNING [train.py:1060] (2/4) Exclude cut with ID 6951-79737-0018-132285-0_sp0.9 from training. Duration: 23.45
2022-12-23 15:49:24,416 WARNING [train.py:1060] (2/4) Exclude cut with ID 6945-60535-0076-12784-0_sp0.9 from training. Duration: 20.52225
2022-12-23 15:49:31,604 WARNING [train.py:1060] (2/4) Exclude cut with ID 7255-291500-0008-134307-0 from training. Duration: 22.19
2022-12-23 15:49:41,303 WARNING [train.py:1060] (2/4) Exclude cut with ID 7357-94126-0014-15845-0_sp1.1 from training. Duration: 25.3818125
2022-12-23 15:49:58,822 WARNING [train.py:1060] (2/4) Exclude cut with ID 6951-79737-0043-132310-0_sp0.9 from training. Duration: 28.0944375
2022-12-23 15:50:05,622 WARNING [train.py:1060] (2/4) Exclude cut with ID 2195-150901-0045-59933-0_sp0.9 from training. Duration: 22.9444375
2022-12-23 15:50:07,295 INFO [train.py:894] (2/4) Epoch 20, batch 2900, loss[loss=0.1933, simple_loss=0.2798, pruned_loss=0.05341, over 18721.00 frames. ], tot_loss[loss=0.1884, simple_loss=0.2692, pruned_loss=0.0538, over 3713973.91 frames. ], batch size: 65, lr: 5.70e-03, grad_scale: 16.0
2022-12-23 15:50:08,947 INFO [zipformer.py:660] (2/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=69518.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 15:50:12,332 WARNING [train.py:1060] (2/4) Exclude cut with ID 8291-282929-0024-13011-0_sp1.1 from training. Duration: 21.6318125
2022-12-23 15:50:18,167 INFO [zipformer.py:660] (2/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=69524.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 15:50:31,987 WARNING [train.py:1060] (2/4) Exclude cut with ID 8631-249866-0030-130156-0 from training. Duration: 23.695
2022-12-23 15:50:40,705 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 3.134e+02 4.802e+02 5.771e+02 6.924e+02 1.235e+03, threshold=1.154e+03, percent-clipped=4.0
2022-12-23 15:50:56,237 WARNING [train.py:1060] (2/4) Exclude cut with ID 7699-105389-0094-26379-0 from training. Duration: 23.955
2022-12-23 15:51:24,254 INFO [train.py:894] (2/4) Epoch 20, batch 2950, loss[loss=0.1562, simple_loss=0.2349, pruned_loss=0.03882, over 18542.00 frames. ], tot_loss[loss=0.1885, simple_loss=0.2695, pruned_loss=0.05371, over 3714082.18 frames. ], batch size: 44, lr: 5.70e-03, grad_scale: 16.0
2022-12-23 15:51:29,909 WARNING [train.py:1060] (2/4) Exclude cut with ID 8291-282929-0024-13011-0_sp0.9 from training. Duration: 26.438875
2022-12-23 15:51:44,965 INFO [zipformer.py:660] (2/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=69581.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 15:52:14,218 INFO [zipformer.py:660] (2/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=69601.0, num_to_drop=1, layers_to_drop={2}
2022-12-23 15:52:15,540 WARNING [train.py:1060] (2/4) Exclude cut with ID 7699-105389-0021-26306-0_sp0.9 from training. Duration: 21.2444375
2022-12-23 15:52:16,976 WARNING [train.py:1060] (2/4) Exclude cut with ID 7357-94126-0014-15845-0_sp0.9 from training. Duration: 31.02225
2022-12-23 15:52:27,187 WARNING [train.py:1060] (2/4) Exclude cut with ID 432-122774-0017-62487-0 from training. Duration: 22.395
2022-12-23 15:52:38,568 INFO [train.py:894] (2/4) Epoch 20, batch 3000, loss[loss=0.1549, simple_loss=0.2343, pruned_loss=0.03777, over 18410.00 frames. ], tot_loss[loss=0.1873, simple_loss=0.2681, pruned_loss=0.05319, over 3713637.00 frames. ], batch size: 42, lr: 5.70e-03, grad_scale: 16.0
2022-12-23 15:52:38,569 INFO [train.py:919] (2/4) Computing validation loss
2022-12-23 15:52:49,540 INFO [train.py:928] (2/4) Epoch 20, validation: loss=0.1661, simple_loss=0.2631, pruned_loss=0.03452, over 944034.00 frames. 
2022-12-23 15:52:49,541 INFO [train.py:929] (2/4) Maximum memory allocated so far is 24680MB
2022-12-23 15:52:55,189 WARNING [train.py:1060] (2/4) Exclude cut with ID 7357-94126-0045-15876-0 from training. Duration: 21.075
2022-12-23 15:53:01,735 WARNING [train.py:1060] (2/4) Exclude cut with ID 6482-98857-0025-147532-0_sp0.9 from training. Duration: 20.0055625
2022-12-23 15:53:01,746 WARNING [train.py:1060] (2/4) Exclude cut with ID 6951-79737-0037-132304-0_sp0.9 from training. Duration: 22.05
2022-12-23 15:53:01,757 WARNING [train.py:1060] (2/4) Exclude cut with ID 7255-291500-0003-134302-0 from training. Duration: 26.8349375
2022-12-23 15:53:04,729 WARNING [train.py:1060] (2/4) Exclude cut with ID 7357-94126-0009-15840-0_sp1.1 from training. Duration: 22.1090625
2022-12-23 15:53:07,656 INFO [zipformer.py:660] (2/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=69629.0, num_to_drop=1, layers_to_drop={0}
2022-12-23 15:53:11,754 WARNING [train.py:1060] (2/4) Exclude cut with ID 7699-105389-0094-26379-0_sp0.9 from training. Duration: 26.6166875
2022-12-23 15:53:22,194 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 3.117e+02 4.417e+02 5.463e+02 6.727e+02 1.419e+03, threshold=1.093e+03, percent-clipped=3.0
2022-12-23 15:53:30,675 WARNING [train.py:1060] (2/4) Exclude cut with ID 2046-178027-0000-53705-0_sp0.9 from training. Duration: 20.3055625
2022-12-23 15:53:37,122 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([2.3079, 2.1844, 1.8385, 1.1752, 2.5936, 2.4445, 2.2127, 1.7251],
       device='cuda:2'), covar=tensor([0.0409, 0.0434, 0.0562, 0.0835, 0.0294, 0.0387, 0.0444, 0.0905],
       device='cuda:2'), in_proj_covar=tensor([0.0123, 0.0129, 0.0129, 0.0121, 0.0099, 0.0123, 0.0136, 0.0158],
       device='cuda:2'), out_proj_covar=tensor([0.0001, 0.0002, 0.0002, 0.0001, 0.0001, 0.0001, 0.0002, 0.0002],
       device='cuda:2')
2022-12-23 15:53:53,486 WARNING [train.py:1060] (2/4) Exclude cut with ID 7205-50138-0008-5373-0_sp0.9 from training. Duration: 20.7
2022-12-23 15:54:05,302 INFO [train.py:894] (2/4) Epoch 20, batch 3050, loss[loss=0.1982, simple_loss=0.2842, pruned_loss=0.05605, over 18702.00 frames. ], tot_loss[loss=0.1863, simple_loss=0.2674, pruned_loss=0.05259, over 3713208.49 frames. ], batch size: 62, lr: 5.69e-03, grad_scale: 16.0
2022-12-23 15:54:16,300 INFO [zipformer.py:660] (2/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=69674.0, num_to_drop=1, layers_to_drop={1}
2022-12-23 15:54:18,875 INFO [zipformer.py:660] (2/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=69676.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 15:54:26,600 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([4.1927, 3.6151, 3.5939, 4.1332, 3.7733, 3.6963, 4.3165, 1.3092],
       device='cuda:2'), covar=tensor([0.0752, 0.0681, 0.0695, 0.0862, 0.1562, 0.1207, 0.0641, 0.4921],
       device='cuda:2'), in_proj_covar=tensor([0.0345, 0.0226, 0.0238, 0.0268, 0.0326, 0.0272, 0.0288, 0.0283],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0001, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-23 15:54:33,844 WARNING [train.py:1060] (2/4) Exclude cut with ID 6978-92210-0019-146985-0 from training. Duration: 22.48
2022-12-23 15:54:49,365 WARNING [train.py:1060] (2/4) Exclude cut with ID 7255-291500-0003-134302-0_sp0.9 from training. Duration: 29.816625
2022-12-23 15:54:53,921 INFO [zipformer.py:660] (2/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=69699.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 15:55:09,425 WARNING [train.py:1060] (2/4) Exclude cut with ID 7255-291500-0005-134304-0_sp1.1 from training. Duration: 22.7590625
2022-12-23 15:55:13,778 WARNING [train.py:1060] (2/4) Exclude cut with ID 7357-94126-0026-15857-0 from training. Duration: 22.555
2022-12-23 15:55:21,876 INFO [train.py:894] (2/4) Epoch 20, batch 3100, loss[loss=0.1771, simple_loss=0.2654, pruned_loss=0.04438, over 18634.00 frames. ], tot_loss[loss=0.1864, simple_loss=0.2677, pruned_loss=0.05258, over 3713296.52 frames. ], batch size: 77, lr: 5.69e-03, grad_scale: 16.0
2022-12-23 15:55:31,910 INFO [zipformer.py:660] (2/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=69724.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 15:55:33,581 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([2.6306, 2.9869, 3.0708, 1.5420, 3.2102, 3.1829, 2.1513, 3.5686],
       device='cuda:2'), covar=tensor([0.1153, 0.1467, 0.1393, 0.2342, 0.0625, 0.1154, 0.2132, 0.0440],
       device='cuda:2'), in_proj_covar=tensor([0.0199, 0.0216, 0.0209, 0.0196, 0.0176, 0.0219, 0.0217, 0.0193],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-23 15:55:34,725 WARNING [train.py:1060] (2/4) Exclude cut with ID 1250-135782-0005-25975-0_sp0.9 from training. Duration: 21.688875
2022-12-23 15:55:49,035 INFO [zipformer.py:660] (2/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=69735.0, num_to_drop=1, layers_to_drop={3}
2022-12-23 15:55:53,829 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([2.1748, 2.3916, 2.1354, 2.6365, 2.7727, 2.0363, 2.1643, 1.9214],
       device='cuda:2'), covar=tensor([0.1372, 0.1206, 0.1142, 0.0724, 0.1074, 0.0904, 0.1620, 0.1104],
       device='cuda:2'), in_proj_covar=tensor([0.0245, 0.0223, 0.0211, 0.0197, 0.0259, 0.0195, 0.0222, 0.0197],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-23 15:55:54,820 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 2.420e+02 4.377e+02 5.437e+02 7.323e+02 2.028e+03, threshold=1.087e+03, percent-clipped=6.0
2022-12-23 15:56:06,106 WARNING [train.py:1060] (2/4) Exclude cut with ID 3488-85273-0038-41224-0_sp0.9 from training. Duration: 22.6
2022-12-23 15:56:07,624 INFO [zipformer.py:660] (2/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=69747.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 15:56:08,299 INFO [scaling.py:679] (2/4) Whitening: num_groups=8, num_channels=96, metric=1.59 vs. limit=2.0
2022-12-23 15:56:12,459 INFO [zipformer.py:660] (2/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=69750.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 15:56:38,091 INFO [train.py:894] (2/4) Epoch 20, batch 3150, loss[loss=0.2064, simple_loss=0.2881, pruned_loss=0.06229, over 18703.00 frames. ], tot_loss[loss=0.1872, simple_loss=0.2685, pruned_loss=0.05292, over 3712378.98 frames. ], batch size: 54, lr: 5.69e-03, grad_scale: 16.0
2022-12-23 15:56:45,332 WARNING [train.py:1060] (2/4) Exclude cut with ID 7357-94126-0009-15840-0 from training. Duration: 24.32
2022-12-23 15:57:06,205 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([2.1213, 1.7650, 2.2683, 2.4374, 2.1119, 4.1583, 1.8176, 2.0566],
       device='cuda:2'), covar=tensor([0.0934, 0.1778, 0.1028, 0.0968, 0.1392, 0.0267, 0.1434, 0.1439],
       device='cuda:2'), in_proj_covar=tensor([0.0073, 0.0082, 0.0073, 0.0074, 0.0091, 0.0075, 0.0084, 0.0077],
       device='cuda:2'), out_proj_covar=tensor([0.0003, 0.0004, 0.0004, 0.0004, 0.0004, 0.0003, 0.0004, 0.0004],
       device='cuda:2')
2022-12-23 15:57:46,008 WARNING [train.py:1060] (2/4) Exclude cut with ID 8291-276745-0093-13116-0_sp0.9 from training. Duration: 21.061125
2022-12-23 15:57:55,010 INFO [train.py:894] (2/4) Epoch 20, batch 3200, loss[loss=0.1713, simple_loss=0.2588, pruned_loss=0.04186, over 18652.00 frames. ], tot_loss[loss=0.1865, simple_loss=0.2677, pruned_loss=0.05268, over 3712438.20 frames. ], batch size: 48, lr: 5.69e-03, grad_scale: 16.0
2022-12-23 15:57:56,703 INFO [zipformer.py:660] (2/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=69818.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 15:58:00,890 WARNING [train.py:1060] (2/4) Exclude cut with ID 7357-94126-0024-15855-0_sp0.9 from training. Duration: 20.32225
2022-12-23 15:58:05,294 INFO [zipformer.py:660] (2/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=69824.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 15:58:12,799 WARNING [train.py:1060] (2/4) Exclude cut with ID 3033-130750-0096-55598-0_sp1.1 from training. Duration: 0.7545625
2022-12-23 15:58:27,640 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 2.717e+02 4.162e+02 5.372e+02 7.071e+02 1.448e+03, threshold=1.074e+03, percent-clipped=3.0
2022-12-23 15:58:29,403 WARNING [train.py:1060] (2/4) Exclude cut with ID 4295-39940-0007-92567-0_sp0.9 from training. Duration: 23.9333125
2022-12-23 15:59:01,144 WARNING [train.py:1060] (2/4) Exclude cut with ID 7255-291500-0008-134307-0_sp1.1 from training. Duration: 20.17275
2022-12-23 15:59:02,793 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.9426, 1.5841, 1.8510, 2.3132, 2.0840, 4.6534, 1.5762, 2.0108],
       device='cuda:2'), covar=tensor([0.0838, 0.1821, 0.1092, 0.0930, 0.1344, 0.0178, 0.1412, 0.1509],
       device='cuda:2'), in_proj_covar=tensor([0.0072, 0.0082, 0.0073, 0.0074, 0.0091, 0.0075, 0.0084, 0.0077],
       device='cuda:2'), out_proj_covar=tensor([0.0003, 0.0004, 0.0004, 0.0004, 0.0004, 0.0003, 0.0004, 0.0004],
       device='cuda:2')
2022-12-23 15:59:08,732 WARNING [train.py:1060] (2/4) Exclude cut with ID 6978-92210-0019-146985-0_sp1.1 from training. Duration: 20.436375
2022-12-23 15:59:08,891 INFO [zipformer.py:660] (2/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=69866.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 15:59:10,196 INFO [train.py:894] (2/4) Epoch 20, batch 3250, loss[loss=0.1586, simple_loss=0.2454, pruned_loss=0.03584, over 18516.00 frames. ], tot_loss[loss=0.1854, simple_loss=0.2668, pruned_loss=0.05201, over 3712262.34 frames. ], batch size: 47, lr: 5.69e-03, grad_scale: 16.0
2022-12-23 15:59:18,249 INFO [zipformer.py:660] (2/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=69872.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 15:59:18,333 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([2.7042, 4.0454, 3.9302, 2.0095, 4.1783, 3.0329, 0.8398, 2.8743],
       device='cuda:2'), covar=tensor([0.2028, 0.1165, 0.1429, 0.3223, 0.0842, 0.0993, 0.5318, 0.1429],
       device='cuda:2'), in_proj_covar=tensor([0.0150, 0.0146, 0.0162, 0.0126, 0.0147, 0.0117, 0.0148, 0.0117],
       device='cuda:2'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0002, 0.0003, 0.0002, 0.0003, 0.0002],
       device='cuda:2')
2022-12-23 15:59:32,431 INFO [zipformer.py:660] (2/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=69881.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 15:59:37,217 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([2.5319, 2.1684, 1.6803, 2.2400, 1.9027, 2.0716, 2.0519, 2.4103],
       device='cuda:2'), covar=tensor([0.2006, 0.3175, 0.1905, 0.2675, 0.3534, 0.1063, 0.2907, 0.0933],
       device='cuda:2'), in_proj_covar=tensor([0.0293, 0.0288, 0.0243, 0.0347, 0.0269, 0.0227, 0.0287, 0.0213],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0003, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-23 16:00:02,325 INFO [zipformer.py:660] (2/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=69901.0, num_to_drop=1, layers_to_drop={2}
2022-12-23 16:00:25,362 INFO [train.py:894] (2/4) Epoch 20, batch 3300, loss[loss=0.183, simple_loss=0.2693, pruned_loss=0.04838, over 18460.00 frames. ], tot_loss[loss=0.1864, simple_loss=0.2673, pruned_loss=0.05274, over 3712770.33 frames. ], batch size: 50, lr: 5.68e-03, grad_scale: 16.0
2022-12-23 16:00:25,471 WARNING [train.py:1060] (2/4) Exclude cut with ID 4234-40345-0022-142709-0_sp0.9 from training. Duration: 23.1055625
2022-12-23 16:00:28,832 WARNING [train.py:1060] (2/4) Exclude cut with ID 8291-282929-0007-12994-0_sp1.1 from training. Duration: 23.5
2022-12-23 16:00:41,293 WARNING [train.py:1060] (2/4) Exclude cut with ID 7255-291500-0009-134308-0_sp0.9 from training. Duration: 26.62775
2022-12-23 16:00:45,010 INFO [zipformer.py:660] (2/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=69929.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 16:00:45,135 INFO [zipformer.py:660] (2/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=69929.0, num_to_drop=1, layers_to_drop={1}
2022-12-23 16:00:51,203 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([0.0494, 0.8806, 1.0546, 0.6102, 0.5147, 1.0944, 1.0906, 1.0430],
       device='cuda:2'), covar=tensor([0.0709, 0.0334, 0.0351, 0.0398, 0.0426, 0.0483, 0.0259, 0.0569],
       device='cuda:2'), in_proj_covar=tensor([0.0150, 0.0170, 0.0127, 0.0141, 0.0148, 0.0142, 0.0163, 0.0172],
       device='cuda:2'), out_proj_covar=tensor([1.1471e-04, 1.3053e-04, 9.5828e-05, 1.0490e-04, 1.1152e-04, 1.0918e-04,
        1.2553e-04, 1.3136e-04], device='cuda:2')
2022-12-23 16:00:55,326 WARNING [train.py:1060] (2/4) Exclude cut with ID 6951-79737-0018-132285-0 from training. Duration: 21.105
2022-12-23 16:00:59,527 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 3.216e+02 4.161e+02 4.947e+02 6.395e+02 1.283e+03, threshold=9.894e+02, percent-clipped=2.0
2022-12-23 16:00:59,626 WARNING [train.py:1060] (2/4) Exclude cut with ID 4511-76322-0006-80011-0_sp0.9 from training. Duration: 24.411125
2022-12-23 16:01:14,917 INFO [zipformer.py:660] (2/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=69949.0, num_to_drop=1, layers_to_drop={1}
2022-12-23 16:01:25,425 WARNING [train.py:1060] (2/4) Exclude cut with ID 6758-72288-0033-108368-0_sp1.1 from training. Duration: 21.263625
2022-12-23 16:01:42,436 INFO [train.py:894] (2/4) Epoch 20, batch 3350, loss[loss=0.158, simple_loss=0.2343, pruned_loss=0.04088, over 18428.00 frames. ], tot_loss[loss=0.1853, simple_loss=0.2667, pruned_loss=0.05196, over 3712644.65 frames. ], batch size: 42, lr: 5.68e-03, grad_scale: 16.0
2022-12-23 16:01:57,070 INFO [zipformer.py:660] (2/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=69976.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 16:01:58,273 INFO [zipformer.py:660] (2/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=69977.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 16:02:01,009 WARNING [train.py:1060] (2/4) Exclude cut with ID 4234-40345-0022-142709-0 from training. Duration: 20.795
2022-12-23 16:02:11,152 WARNING [train.py:1060] (2/4) Exclude cut with ID 7357-94126-0021-15852-0 from training. Duration: 24.76
2022-12-23 16:02:11,163 WARNING [train.py:1060] (2/4) Exclude cut with ID 3867-173237-0077-144769-0_sp0.9 from training. Duration: 22.25
2022-12-23 16:02:38,214 WARNING [train.py:1060] (2/4) Exclude cut with ID 7357-94126-0026-15857-0_sp1.1 from training. Duration: 20.5045625
2022-12-23 16:03:03,974 INFO [train.py:894] (2/4) Epoch 20, batch 3400, loss[loss=0.2019, simple_loss=0.2849, pruned_loss=0.05944, over 18500.00 frames. ], tot_loss[loss=0.1851, simple_loss=0.2666, pruned_loss=0.05178, over 3713961.87 frames. ], batch size: 58, lr: 5.68e-03, grad_scale: 16.0
2022-12-23 16:03:14,441 INFO [zipformer.py:660] (2/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=70024.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 16:03:14,569 INFO [zipformer.py:660] (2/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=70024.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 16:03:23,151 INFO [zipformer.py:660] (2/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=70030.0, num_to_drop=1, layers_to_drop={2}
2022-12-23 16:03:32,177 INFO [scaling.py:679] (2/4) Whitening: num_groups=8, num_channels=192, metric=1.51 vs. limit=2.0
2022-12-23 16:03:36,037 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 2.836e+02 4.387e+02 5.410e+02 6.591e+02 1.191e+03, threshold=1.082e+03, percent-clipped=3.0
2022-12-23 16:03:52,284 INFO [zipformer.py:660] (2/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=70050.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 16:04:13,555 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([3.1565, 2.4314, 1.9183, 3.0974, 2.2020, 2.3350, 2.3897, 3.3850],
       device='cuda:2'), covar=tensor([0.1935, 0.3357, 0.1821, 0.2804, 0.3852, 0.1022, 0.3276, 0.0774],
       device='cuda:2'), in_proj_covar=tensor([0.0297, 0.0291, 0.0245, 0.0352, 0.0272, 0.0230, 0.0291, 0.0215],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0003, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-23 16:04:15,718 INFO [train.py:894] (2/4) Epoch 20, batch 3450, loss[loss=0.18, simple_loss=0.2707, pruned_loss=0.04468, over 18512.00 frames. ], tot_loss[loss=0.1858, simple_loss=0.2671, pruned_loss=0.05223, over 3715072.21 frames. ], batch size: 77, lr: 5.68e-03, grad_scale: 16.0
2022-12-23 16:04:22,932 INFO [zipformer.py:660] (2/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=70072.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 16:05:01,362 INFO [zipformer.py:660] (2/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=70098.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 16:05:18,895 INFO [zipformer.py:660] (2/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=70110.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 16:05:30,002 INFO [train.py:894] (2/4) Epoch 20, batch 3500, loss[loss=0.1986, simple_loss=0.2776, pruned_loss=0.05979, over 18568.00 frames. ], tot_loss[loss=0.1864, simple_loss=0.2675, pruned_loss=0.0526, over 3714520.67 frames. ], batch size: 179, lr: 5.68e-03, grad_scale: 16.0
2022-12-23 16:05:50,383 WARNING [train.py:1060] (2/4) Exclude cut with ID 7859-102521-0017-7548-0_sp1.1 from training. Duration: 22.2954375
2022-12-23 16:06:02,624 INFO [train.py:894] (2/4) Epoch 21, batch 0, loss[loss=0.1776, simple_loss=0.2661, pruned_loss=0.04455, over 18671.00 frames. ], tot_loss[loss=0.1776, simple_loss=0.2661, pruned_loss=0.04455, over 18671.00 frames. ], batch size: 48, lr: 5.54e-03, grad_scale: 16.0
2022-12-23 16:06:02,624 INFO [train.py:919] (2/4) Computing validation loss
2022-12-23 16:06:13,593 INFO [train.py:928] (2/4) Epoch 21, validation: loss=0.1647, simple_loss=0.2623, pruned_loss=0.03359, over 944034.00 frames. 
2022-12-23 16:06:13,593 INFO [train.py:929] (2/4) Maximum memory allocated so far is 24680MB
2022-12-23 16:06:36,622 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 2.766e+02 4.209e+02 5.633e+02 7.242e+02 1.131e+03, threshold=1.127e+03, percent-clipped=2.0
2022-12-23 16:07:04,030 WARNING [train.py:1060] (2/4) Exclude cut with ID 298-126791-0067-24026-0_sp0.9 from training. Duration: 21.438875
2022-12-23 16:07:09,089 WARNING [train.py:1060] (2/4) Exclude cut with ID 5652-39938-0025-23684-0_sp0.9 from training. Duration: 22.2055625
2022-12-23 16:07:26,272 INFO [zipformer.py:660] (2/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=70171.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 16:07:29,007 INFO [train.py:894] (2/4) Epoch 21, batch 50, loss[loss=0.169, simple_loss=0.2618, pruned_loss=0.03813, over 18562.00 frames. ], tot_loss[loss=0.1784, simple_loss=0.2663, pruned_loss=0.04524, over 838295.81 frames. ], batch size: 51, lr: 5.53e-03, grad_scale: 16.0
2022-12-23 16:08:44,899 INFO [train.py:894] (2/4) Epoch 21, batch 100, loss[loss=0.159, simple_loss=0.2373, pruned_loss=0.0403, over 18400.00 frames. ], tot_loss[loss=0.175, simple_loss=0.2615, pruned_loss=0.04422, over 1474693.46 frames. ], batch size: 42, lr: 5.53e-03, grad_scale: 16.0
2022-12-23 16:09:07,225 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 2.079e+02 3.457e+02 4.152e+02 4.912e+02 1.134e+03, threshold=8.304e+02, percent-clipped=1.0
2022-12-23 16:09:10,559 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.9326, 1.2653, 0.6837, 1.4725, 2.0369, 1.4424, 1.5658, 1.6582],
       device='cuda:2'), covar=tensor([0.1578, 0.2183, 0.2460, 0.1466, 0.1813, 0.1695, 0.1471, 0.1711],
       device='cuda:2'), in_proj_covar=tensor([0.0094, 0.0098, 0.0116, 0.0094, 0.0117, 0.0091, 0.0097, 0.0093],
       device='cuda:2'), out_proj_covar=tensor([0.0003, 0.0003, 0.0004, 0.0003, 0.0004, 0.0003, 0.0003, 0.0003],
       device='cuda:2')
2022-12-23 16:09:14,313 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.6293, 1.5609, 1.6043, 1.5647, 1.1689, 3.0384, 1.3104, 1.9290],
       device='cuda:2'), covar=tensor([0.3093, 0.2112, 0.2021, 0.2020, 0.1452, 0.0228, 0.1656, 0.0793],
       device='cuda:2'), in_proj_covar=tensor([0.0133, 0.0118, 0.0126, 0.0121, 0.0104, 0.0098, 0.0092, 0.0090],
       device='cuda:2'), out_proj_covar=tensor([0.0005, 0.0005, 0.0005, 0.0005, 0.0004, 0.0003, 0.0004, 0.0004],
       device='cuda:2')
2022-12-23 16:09:40,111 INFO [zipformer.py:660] (2/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=70260.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 16:09:59,306 INFO [train.py:894] (2/4) Epoch 21, batch 150, loss[loss=0.1554, simple_loss=0.2407, pruned_loss=0.03507, over 18684.00 frames. ], tot_loss[loss=0.1735, simple_loss=0.2607, pruned_loss=0.04319, over 1971542.10 frames. ], batch size: 46, lr: 5.53e-03, grad_scale: 16.0
2022-12-23 16:10:08,363 WARNING [train.py:1060] (2/4) Exclude cut with ID 7859-102521-0017-7548-0 from training. Duration: 24.525
2022-12-23 16:10:10,204 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.7294, 1.6817, 1.9757, 1.1433, 1.9553, 2.0084, 1.5591, 2.3785],
       device='cuda:2'), covar=tensor([0.1202, 0.2261, 0.1405, 0.2084, 0.0831, 0.1271, 0.2434, 0.0601],
       device='cuda:2'), in_proj_covar=tensor([0.0197, 0.0215, 0.0207, 0.0195, 0.0176, 0.0216, 0.0216, 0.0192],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-23 16:10:41,542 WARNING [train.py:1060] (2/4) Exclude cut with ID 3699-47246-0007-3408-0_sp0.9 from training. Duration: 20.26675
2022-12-23 16:10:54,098 WARNING [train.py:1060] (2/4) Exclude cut with ID 7859-102521-0017-7548-0_sp0.9 from training. Duration: 27.25
2022-12-23 16:11:13,232 INFO [zipformer.py:660] (2/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=70321.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 16:11:15,743 INFO [train.py:894] (2/4) Epoch 21, batch 200, loss[loss=0.167, simple_loss=0.2467, pruned_loss=0.04369, over 18586.00 frames. ], tot_loss[loss=0.1722, simple_loss=0.259, pruned_loss=0.04272, over 2357186.40 frames. ], batch size: 45, lr: 5.53e-03, grad_scale: 16.0
2022-12-23 16:11:25,762 INFO [zipformer.py:660] (2/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=70330.0, num_to_drop=1, layers_to_drop={1}
2022-12-23 16:11:39,251 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 2.182e+02 3.326e+02 4.110e+02 5.021e+02 8.767e+02, threshold=8.220e+02, percent-clipped=2.0
2022-12-23 16:11:41,678 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.6176, 1.6472, 1.9364, 1.0837, 1.8180, 1.7996, 1.5083, 2.2631],
       device='cuda:2'), covar=tensor([0.1050, 0.1969, 0.1027, 0.1659, 0.0770, 0.1161, 0.2237, 0.0507],
       device='cuda:2'), in_proj_covar=tensor([0.0197, 0.0214, 0.0207, 0.0195, 0.0175, 0.0217, 0.0215, 0.0192],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-23 16:12:06,349 WARNING [train.py:1060] (2/4) Exclude cut with ID 6426-64292-0017-15984-0 from training. Duration: 21.68
2022-12-23 16:12:16,371 WARNING [train.py:1060] (2/4) Exclude cut with ID 4278-13270-0007-59342-0 from training. Duration: 21.6300625
2022-12-23 16:12:29,192 INFO [train.py:894] (2/4) Epoch 21, batch 250, loss[loss=0.1764, simple_loss=0.2659, pruned_loss=0.04346, over 18575.00 frames. ], tot_loss[loss=0.1729, simple_loss=0.2602, pruned_loss=0.04278, over 2657633.41 frames. ], batch size: 51, lr: 5.53e-03, grad_scale: 16.0
2022-12-23 16:12:36,713 INFO [zipformer.py:660] (2/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=70378.0, num_to_drop=1, layers_to_drop={1}
2022-12-23 16:12:39,385 WARNING [train.py:1060] (2/4) Exclude cut with ID 4278-13270-0007-59342-0_sp0.9 from training. Duration: 24.033375
2022-12-23 16:13:37,961 WARNING [train.py:1060] (2/4) Exclude cut with ID 4278-13270-0009-59344-0 from training. Duration: 22.905
2022-12-23 16:13:39,536 WARNING [train.py:1060] (2/4) Exclude cut with ID 5622-44585-0006-90525-0_sp1.1 from training. Duration: 23.4318125
2022-12-23 16:13:44,119 INFO [train.py:894] (2/4) Epoch 21, batch 300, loss[loss=0.1705, simple_loss=0.2485, pruned_loss=0.04624, over 18666.00 frames. ], tot_loss[loss=0.1732, simple_loss=0.2607, pruned_loss=0.04283, over 2891139.05 frames. ], batch size: 41, lr: 5.52e-03, grad_scale: 16.0
2022-12-23 16:14:08,535 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 2.163e+02 3.548e+02 4.548e+02 5.706e+02 1.035e+03, threshold=9.095e+02, percent-clipped=5.0
2022-12-23 16:14:49,704 INFO [zipformer.py:660] (2/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=70466.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 16:14:59,949 INFO [train.py:894] (2/4) Epoch 21, batch 350, loss[loss=0.211, simple_loss=0.2936, pruned_loss=0.06421, over 18596.00 frames. ], tot_loss[loss=0.1742, simple_loss=0.2619, pruned_loss=0.04324, over 3073363.40 frames. ], batch size: 56, lr: 5.52e-03, grad_scale: 16.0
2022-12-23 16:15:10,699 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([2.7560, 3.9461, 3.7017, 1.6276, 3.9823, 3.1299, 0.5227, 2.3959],
       device='cuda:2'), covar=tensor([0.1902, 0.0994, 0.1370, 0.3709, 0.0791, 0.0789, 0.5286, 0.1692],
       device='cuda:2'), in_proj_covar=tensor([0.0146, 0.0142, 0.0158, 0.0124, 0.0143, 0.0114, 0.0144, 0.0116],
       device='cuda:2'), out_proj_covar=tensor([0.0003, 0.0002, 0.0003, 0.0002, 0.0003, 0.0002, 0.0003, 0.0002],
       device='cuda:2')
2022-12-23 16:15:37,878 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.7070, 1.6142, 1.9046, 1.0967, 1.8686, 1.8924, 1.4409, 2.2833],
       device='cuda:2'), covar=tensor([0.1000, 0.1923, 0.1101, 0.1699, 0.0684, 0.1098, 0.2294, 0.0490],
       device='cuda:2'), in_proj_covar=tensor([0.0196, 0.0214, 0.0206, 0.0195, 0.0176, 0.0216, 0.0215, 0.0193],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-23 16:15:40,643 WARNING [train.py:1060] (2/4) Exclude cut with ID 4278-13270-0009-59344-0_sp1.1 from training. Duration: 20.82275
2022-12-23 16:15:40,686 WARNING [train.py:1060] (2/4) Exclude cut with ID 4278-13270-0009-59344-0_sp0.9 from training. Duration: 25.45
2022-12-23 16:16:16,688 INFO [train.py:894] (2/4) Epoch 21, batch 400, loss[loss=0.1966, simple_loss=0.2825, pruned_loss=0.05535, over 18508.00 frames. ], tot_loss[loss=0.1759, simple_loss=0.2638, pruned_loss=0.044, over 3214958.83 frames. ], batch size: 52, lr: 5.52e-03, grad_scale: 16.0
2022-12-23 16:16:40,729 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 2.333e+02 3.461e+02 4.220e+02 5.034e+02 8.271e+02, threshold=8.440e+02, percent-clipped=0.0
2022-12-23 16:16:40,773 WARNING [train.py:1060] (2/4) Exclude cut with ID 5622-44585-0006-90525-0 from training. Duration: 25.775
2022-12-23 16:17:02,591 WARNING [train.py:1060] (2/4) Exclude cut with ID 2929-85685-0071-62375-0_sp0.9 from training. Duration: 22.25
2022-12-23 16:17:29,947 WARNING [train.py:1060] (2/4) Exclude cut with ID 3972-170212-0014-23379-0 from training. Duration: 26.205
2022-12-23 16:17:31,363 INFO [train.py:894] (2/4) Epoch 21, batch 450, loss[loss=0.2284, simple_loss=0.3138, pruned_loss=0.07147, over 18583.00 frames. ], tot_loss[loss=0.1761, simple_loss=0.2643, pruned_loss=0.04396, over 3325763.55 frames. ], batch size: 57, lr: 5.52e-03, grad_scale: 16.0
2022-12-23 16:17:47,463 WARNING [train.py:1060] (2/4) Exclude cut with ID 5239-32139-0047-9341-0_sp0.9 from training. Duration: 30.1555625
2022-12-23 16:17:51,977 WARNING [train.py:1060] (2/4) Exclude cut with ID 1265-135635-0050-6781-0_sp0.9 from training. Duration: 21.8333125
2022-12-23 16:18:01,237 WARNING [train.py:1060] (2/4) Exclude cut with ID 1914-133440-0024-94914-0_sp1.1 from training. Duration: 20.6545625
2022-12-23 16:18:33,150 INFO [scaling.py:679] (2/4) Whitening: num_groups=8, num_channels=96, metric=1.12 vs. limit=2.0
2022-12-23 16:18:38,265 INFO [zipformer.py:660] (2/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=70616.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 16:18:45,288 WARNING [train.py:1060] (2/4) Exclude cut with ID 7395-89880-0045-39920-0_sp0.9 from training. Duration: 20.52225
2022-12-23 16:18:48,353 INFO [train.py:894] (2/4) Epoch 21, batch 500, loss[loss=0.2017, simple_loss=0.2923, pruned_loss=0.05555, over 18634.00 frames. ], tot_loss[loss=0.1765, simple_loss=0.2647, pruned_loss=0.04416, over 3411183.32 frames. ], batch size: 65, lr: 5.52e-03, grad_scale: 16.0
2022-12-23 16:18:57,035 INFO [scaling.py:679] (2/4) Whitening: num_groups=8, num_channels=192, metric=1.98 vs. limit=2.0
2022-12-23 16:19:07,511 WARNING [train.py:1060] (2/4) Exclude cut with ID 3972-170212-0014-23379-0_sp0.9 from training. Duration: 29.1166875
2022-12-23 16:19:13,338 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 2.177e+02 3.381e+02 4.074e+02 5.237e+02 1.211e+03, threshold=8.147e+02, percent-clipped=2.0
2022-12-23 16:19:31,665 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([2.4369, 1.9320, 2.0395, 2.1646, 2.3636, 2.4606, 2.3478, 1.9414],
       device='cuda:2'), covar=tensor([0.2104, 0.3308, 0.2487, 0.2789, 0.1979, 0.0913, 0.3386, 0.1280],
       device='cuda:2'), in_proj_covar=tensor([0.0267, 0.0297, 0.0278, 0.0317, 0.0306, 0.0251, 0.0342, 0.0241],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-23 16:19:58,525 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.6449, 1.5872, 1.5513, 1.4251, 1.8307, 1.7666, 1.7789, 1.3362],
       device='cuda:2'), covar=tensor([0.0309, 0.0228, 0.0433, 0.0225, 0.0183, 0.0377, 0.0253, 0.0303],
       device='cuda:2'), in_proj_covar=tensor([0.0095, 0.0128, 0.0151, 0.0125, 0.0116, 0.0120, 0.0098, 0.0127],
       device='cuda:2'), out_proj_covar=tensor([7.5639e-05, 1.0162e-04, 1.2484e-04, 9.9706e-05, 9.3773e-05, 9.2173e-05,
        7.6509e-05, 9.9993e-05], device='cuda:2')
2022-12-23 16:20:05,319 INFO [train.py:894] (2/4) Epoch 21, batch 550, loss[loss=0.1523, simple_loss=0.2507, pruned_loss=0.02694, over 18394.00 frames. ], tot_loss[loss=0.1763, simple_loss=0.2644, pruned_loss=0.04412, over 3477482.63 frames. ], batch size: 53, lr: 5.51e-03, grad_scale: 16.0
2022-12-23 16:20:05,387 WARNING [train.py:1060] (2/4) Exclude cut with ID 543-133211-0007-59831-0_sp0.9 from training. Duration: 21.388875
2022-12-23 16:20:42,382 WARNING [train.py:1060] (2/4) Exclude cut with ID 1914-133440-0024-94914-0 from training. Duration: 22.72
2022-12-23 16:20:43,788 WARNING [train.py:1060] (2/4) Exclude cut with ID 1914-133440-0031-94921-0_sp0.9 from training. Duration: 22.7444375
2022-12-23 16:20:45,473 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.6016, 1.2536, 2.1014, 3.2500, 2.4943, 2.5827, 0.5438, 2.2728],
       device='cuda:2'), covar=tensor([0.1773, 0.1637, 0.1291, 0.0539, 0.0896, 0.1011, 0.2281, 0.1028],
       device='cuda:2'), in_proj_covar=tensor([0.0101, 0.0115, 0.0134, 0.0146, 0.0104, 0.0138, 0.0128, 0.0112],
       device='cuda:2'), out_proj_covar=tensor([0.0003, 0.0003, 0.0004, 0.0004, 0.0003, 0.0004, 0.0003, 0.0003],
       device='cuda:2')
2022-12-23 16:20:51,688 INFO [scaling.py:679] (2/4) Whitening: num_groups=8, num_channels=192, metric=1.86 vs. limit=2.0
2022-12-23 16:21:07,992 INFO [scaling.py:679] (2/4) Whitening: num_groups=1, num_channels=384, metric=4.54 vs. limit=5.0
2022-12-23 16:21:20,560 INFO [train.py:894] (2/4) Epoch 21, batch 600, loss[loss=0.2006, simple_loss=0.2938, pruned_loss=0.05376, over 18711.00 frames. ], tot_loss[loss=0.1779, simple_loss=0.2656, pruned_loss=0.04505, over 3529690.17 frames. ], batch size: 60, lr: 5.51e-03, grad_scale: 16.0
2022-12-23 16:21:25,176 WARNING [train.py:1060] (2/4) Exclude cut with ID 4133-6541-0027-40495-0_sp1.1 from training. Duration: 0.9681875
2022-12-23 16:21:27,925 WARNING [train.py:1060] (2/4) Exclude cut with ID 6330-62851-0022-91297-0_sp0.9 from training. Duration: 22.3166875
2022-12-23 16:21:33,886 WARNING [train.py:1060] (2/4) Exclude cut with ID 543-133212-0015-59917-0_sp0.9 from training. Duration: 21.8166875
2022-12-23 16:21:44,620 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 2.038e+02 3.325e+02 3.856e+02 4.616e+02 9.252e+02, threshold=7.711e+02, percent-clipped=1.0
2022-12-23 16:22:25,003 INFO [zipformer.py:660] (2/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=70766.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 16:22:35,291 INFO [train.py:894] (2/4) Epoch 21, batch 650, loss[loss=0.2021, simple_loss=0.2961, pruned_loss=0.05405, over 18535.00 frames. ], tot_loss[loss=0.1773, simple_loss=0.2654, pruned_loss=0.04465, over 3570341.34 frames. ], batch size: 77, lr: 5.51e-03, grad_scale: 16.0
2022-12-23 16:23:16,787 WARNING [train.py:1060] (2/4) Exclude cut with ID 4957-30119-0041-23990-0_sp0.9 from training. Duration: 20.22775
2022-12-23 16:23:37,709 INFO [zipformer.py:660] (2/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=70814.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 16:23:45,972 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.8874, 1.9545, 1.7915, 2.0263, 1.3093, 4.9877, 1.9019, 2.4607],
       device='cuda:2'), covar=tensor([0.2967, 0.1891, 0.1883, 0.1857, 0.1411, 0.0096, 0.1460, 0.0791],
       device='cuda:2'), in_proj_covar=tensor([0.0133, 0.0118, 0.0125, 0.0121, 0.0104, 0.0097, 0.0092, 0.0089],
       device='cuda:2'), out_proj_covar=tensor([0.0005, 0.0005, 0.0005, 0.0005, 0.0004, 0.0003, 0.0004, 0.0004],
       device='cuda:2')
2022-12-23 16:23:47,174 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([2.5222, 3.6769, 3.4913, 1.3980, 3.8795, 2.9652, 0.5753, 2.3348],
       device='cuda:2'), covar=tensor([0.2056, 0.0938, 0.1525, 0.3746, 0.0712, 0.0839, 0.5056, 0.1491],
       device='cuda:2'), in_proj_covar=tensor([0.0147, 0.0143, 0.0158, 0.0124, 0.0143, 0.0114, 0.0144, 0.0115],
       device='cuda:2'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0002, 0.0003, 0.0002, 0.0003, 0.0002],
       device='cuda:2')
2022-12-23 16:23:51,365 INFO [train.py:894] (2/4) Epoch 21, batch 700, loss[loss=0.1769, simple_loss=0.2731, pruned_loss=0.04035, over 18720.00 frames. ], tot_loss[loss=0.176, simple_loss=0.2641, pruned_loss=0.04389, over 3602698.11 frames. ], batch size: 54, lr: 5.51e-03, grad_scale: 16.0
2022-12-23 16:23:59,435 WARNING [train.py:1060] (2/4) Exclude cut with ID 5239-32139-0047-9341-0_sp1.1 from training. Duration: 24.67275
2022-12-23 16:24:16,537 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 1.979e+02 3.538e+02 4.225e+02 5.093e+02 8.209e+02, threshold=8.450e+02, percent-clipped=1.0
2022-12-23 16:24:29,929 WARNING [train.py:1060] (2/4) Exclude cut with ID 3082-165428-0081-50734-0_sp0.9 from training. Duration: 21.8055625
2022-12-23 16:25:06,021 WARNING [train.py:1060] (2/4) Exclude cut with ID 3340-169293-0054-76830-0_sp0.9 from training. Duration: 22.6666875
2022-12-23 16:25:07,412 INFO [train.py:894] (2/4) Epoch 21, batch 750, loss[loss=0.1559, simple_loss=0.2436, pruned_loss=0.03407, over 18364.00 frames. ], tot_loss[loss=0.1759, simple_loss=0.2642, pruned_loss=0.04381, over 3626748.65 frames. ], batch size: 46, lr: 5.51e-03, grad_scale: 16.0
2022-12-23 16:26:08,046 WARNING [train.py:1060] (2/4) Exclude cut with ID 2411-132532-0017-82279-0_sp1.1 from training. Duration: 0.9681875
2022-12-23 16:26:12,752 INFO [zipformer.py:660] (2/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=70916.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 16:26:23,421 INFO [train.py:894] (2/4) Epoch 21, batch 800, loss[loss=0.1699, simple_loss=0.2553, pruned_loss=0.04225, over 18419.00 frames. ], tot_loss[loss=0.1764, simple_loss=0.2646, pruned_loss=0.04411, over 3645498.32 frames. ], batch size: 48, lr: 5.51e-03, grad_scale: 16.0
2022-12-23 16:26:32,260 WARNING [train.py:1060] (2/4) Exclude cut with ID 6330-62850-0007-91323-0 from training. Duration: 22.485
2022-12-23 16:26:48,150 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 2.388e+02 3.554e+02 4.225e+02 4.915e+02 1.041e+03, threshold=8.450e+02, percent-clipped=3.0
2022-12-23 16:27:09,852 WARNING [train.py:1060] (2/4) Exclude cut with ID 3972-170212-0014-23379-0_sp1.1 from training. Duration: 23.82275
2022-12-23 16:27:23,634 WARNING [train.py:1060] (2/4) Exclude cut with ID 4860-13185-0032-76709-0 from training. Duration: 20.77
2022-12-23 16:27:25,191 INFO [zipformer.py:660] (2/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=70964.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 16:27:32,396 WARNING [train.py:1060] (2/4) Exclude cut with ID 6426-64292-0017-15984-0_sp0.9 from training. Duration: 24.088875
2022-12-23 16:27:38,672 INFO [train.py:894] (2/4) Epoch 21, batch 850, loss[loss=0.1621, simple_loss=0.2561, pruned_loss=0.03408, over 18455.00 frames. ], tot_loss[loss=0.176, simple_loss=0.2644, pruned_loss=0.04376, over 3660496.74 frames. ], batch size: 50, lr: 5.50e-03, grad_scale: 16.0
2022-12-23 16:28:01,266 WARNING [train.py:1060] (2/4) Exclude cut with ID 6330-62850-0007-91323-0_sp1.1 from training. Duration: 20.4409375
2022-12-23 16:28:07,715 INFO [zipformer.py:660] (2/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=70992.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 16:28:20,694 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.8550, 1.2317, 0.7456, 1.3672, 1.9943, 1.0852, 1.4786, 1.5336],
       device='cuda:2'), covar=tensor([0.1512, 0.2141, 0.2224, 0.1479, 0.1840, 0.1793, 0.1437, 0.1785],
       device='cuda:2'), in_proj_covar=tensor([0.0094, 0.0098, 0.0117, 0.0096, 0.0118, 0.0092, 0.0098, 0.0094],
       device='cuda:2'), out_proj_covar=tensor([0.0003, 0.0003, 0.0004, 0.0003, 0.0004, 0.0003, 0.0003, 0.0003],
       device='cuda:2')
2022-12-23 16:28:53,475 INFO [train.py:894] (2/4) Epoch 21, batch 900, loss[loss=0.1827, simple_loss=0.2719, pruned_loss=0.04682, over 18507.00 frames. ], tot_loss[loss=0.1761, simple_loss=0.2647, pruned_loss=0.04372, over 3672922.48 frames. ], batch size: 52, lr: 5.50e-03, grad_scale: 16.0
2022-12-23 16:29:05,264 INFO [zipformer.py:660] (2/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=71030.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 16:29:18,053 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 2.311e+02 3.474e+02 4.263e+02 5.002e+02 9.201e+02, threshold=8.526e+02, percent-clipped=2.0
2022-12-23 16:29:18,074 WARNING [train.py:1060] (2/4) Exclude cut with ID 2929-85685-0062-62366-0_sp0.9 from training. Duration: 22.511125
2022-12-23 16:29:18,094 WARNING [train.py:1060] (2/4) Exclude cut with ID 7395-89880-0031-39906-0 from training. Duration: 20.675
2022-12-23 16:29:38,856 INFO [zipformer.py:660] (2/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=71053.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 16:30:09,402 INFO [train.py:894] (2/4) Epoch 21, batch 950, loss[loss=0.1964, simple_loss=0.2864, pruned_loss=0.05315, over 18710.00 frames. ], tot_loss[loss=0.1766, simple_loss=0.2653, pruned_loss=0.04392, over 3682054.88 frames. ], batch size: 65, lr: 5.50e-03, grad_scale: 16.0
2022-12-23 16:30:36,355 INFO [zipformer.py:660] (2/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=71091.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 16:30:56,421 WARNING [train.py:1060] (2/4) Exclude cut with ID 6330-62850-0007-91323-0_sp0.9 from training. Duration: 24.9833125
2022-12-23 16:31:24,680 INFO [train.py:894] (2/4) Epoch 21, batch 1000, loss[loss=0.1681, simple_loss=0.2557, pruned_loss=0.04022, over 18703.00 frames. ], tot_loss[loss=0.1763, simple_loss=0.2653, pruned_loss=0.04369, over 3689033.69 frames. ], batch size: 50, lr: 5.50e-03, grad_scale: 16.0
2022-12-23 16:31:27,897 WARNING [train.py:1060] (2/4) Exclude cut with ID 5239-32139-0047-9341-0 from training. Duration: 27.14
2022-12-23 16:31:45,525 WARNING [train.py:1060] (2/4) Exclude cut with ID 2929-85685-0044-62348-0 from training. Duration: 22.44
2022-12-23 16:31:48,412 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 1.909e+02 3.318e+02 3.818e+02 4.635e+02 9.263e+02, threshold=7.636e+02, percent-clipped=2.0
2022-12-23 16:32:37,323 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.9547, 1.3689, 2.4326, 4.2959, 3.3072, 2.8459, 0.7541, 3.1305],
       device='cuda:2'), covar=tensor([0.1879, 0.1783, 0.1454, 0.0454, 0.0860, 0.1156, 0.2395, 0.0850],
       device='cuda:2'), in_proj_covar=tensor([0.0101, 0.0115, 0.0133, 0.0147, 0.0105, 0.0139, 0.0129, 0.0112],
       device='cuda:2'), out_proj_covar=tensor([0.0003, 0.0003, 0.0004, 0.0004, 0.0003, 0.0004, 0.0003, 0.0003],
       device='cuda:2')
2022-12-23 16:32:40,053 INFO [train.py:894] (2/4) Epoch 21, batch 1050, loss[loss=0.1864, simple_loss=0.262, pruned_loss=0.05542, over 18544.00 frames. ], tot_loss[loss=0.1764, simple_loss=0.2653, pruned_loss=0.04373, over 3694577.34 frames. ], batch size: 44, lr: 5.50e-03, grad_scale: 16.0
2022-12-23 16:33:03,676 WARNING [train.py:1060] (2/4) Exclude cut with ID 2929-85685-0060-62364-0_sp0.9 from training. Duration: 21.361125
2022-12-23 16:33:09,717 WARNING [train.py:1060] (2/4) Exclude cut with ID 2929-85685-0079-62383-0_sp1.1 from training. Duration: 27.0318125
2022-12-23 16:33:09,937 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([2.6308, 3.6670, 3.5056, 1.2626, 3.8872, 2.9065, 0.4954, 2.3051],
       device='cuda:2'), covar=tensor([0.1831, 0.1018, 0.1335, 0.3846, 0.0698, 0.0824, 0.4965, 0.1600],
       device='cuda:2'), in_proj_covar=tensor([0.0146, 0.0143, 0.0157, 0.0124, 0.0143, 0.0114, 0.0143, 0.0114],
       device='cuda:2'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0002, 0.0002, 0.0002, 0.0003, 0.0002],
       device='cuda:2')
2022-12-23 16:33:20,142 WARNING [train.py:1060] (2/4) Exclude cut with ID 5622-44585-0006-90525-0_sp0.9 from training. Duration: 28.638875
2022-12-23 16:33:34,529 WARNING [train.py:1060] (2/4) Exclude cut with ID 3340-169293-0054-76830-0 from training. Duration: 20.4
2022-12-23 16:33:56,813 INFO [train.py:894] (2/4) Epoch 21, batch 1100, loss[loss=0.1583, simple_loss=0.2537, pruned_loss=0.03146, over 18458.00 frames. ], tot_loss[loss=0.1765, simple_loss=0.2654, pruned_loss=0.04375, over 3697824.78 frames. ], batch size: 50, lr: 5.49e-03, grad_scale: 16.0
2022-12-23 16:34:07,116 WARNING [train.py:1060] (2/4) Exclude cut with ID 2929-85685-0071-62375-0 from training. Duration: 20.025
2022-12-23 16:34:07,127 WARNING [train.py:1060] (2/4) Exclude cut with ID 2364-131735-0112-64612-0_sp0.9 from training. Duration: 20.488875
2022-12-23 16:34:14,475 WARNING [train.py:1060] (2/4) Exclude cut with ID 2929-85685-0079-62383-0 from training. Duration: 29.735
2022-12-23 16:34:20,018 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 2.240e+02 3.581e+02 4.240e+02 5.301e+02 9.691e+02, threshold=8.480e+02, percent-clipped=4.0
2022-12-23 16:35:00,314 INFO [scaling.py:679] (2/4) Whitening: num_groups=8, num_channels=192, metric=1.89 vs. limit=2.0
2022-12-23 16:35:11,397 INFO [train.py:894] (2/4) Epoch 21, batch 1150, loss[loss=0.1846, simple_loss=0.2675, pruned_loss=0.05083, over 18506.00 frames. ], tot_loss[loss=0.176, simple_loss=0.2648, pruned_loss=0.0436, over 3701842.56 frames. ], batch size: 52, lr: 5.49e-03, grad_scale: 16.0
2022-12-23 16:35:29,533 INFO [scaling.py:679] (2/4) Whitening: num_groups=1, num_channels=384, metric=3.14 vs. limit=5.0
2022-12-23 16:35:33,393 WARNING [train.py:1060] (2/4) Exclude cut with ID 7276-92427-0014-12983-0_sp0.9 from training. Duration: 21.3055625
2022-12-23 16:35:35,060 WARNING [train.py:1060] (2/4) Exclude cut with ID 1025-75365-0008-79168-0_sp0.9 from training. Duration: 22.0666875
2022-12-23 16:36:26,490 INFO [train.py:894] (2/4) Epoch 21, batch 1200, loss[loss=0.1999, simple_loss=0.2827, pruned_loss=0.05854, over 18627.00 frames. ], tot_loss[loss=0.1762, simple_loss=0.2652, pruned_loss=0.0436, over 3704498.07 frames. ], batch size: 173, lr: 5.49e-03, grad_scale: 32.0
2022-12-23 16:36:48,756 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 2.215e+02 3.270e+02 4.020e+02 4.826e+02 8.816e+02, threshold=8.041e+02, percent-clipped=1.0
2022-12-23 16:37:03,419 INFO [zipformer.py:660] (2/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=71348.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 16:37:24,222 WARNING [train.py:1060] (2/4) Exclude cut with ID 2929-85685-0062-62366-0 from training. Duration: 20.26
2022-12-23 16:37:38,373 WARNING [train.py:1060] (2/4) Exclude cut with ID 5239-32139-0030-9324-0_sp0.9 from training. Duration: 21.3444375
2022-12-23 16:37:41,230 INFO [train.py:894] (2/4) Epoch 21, batch 1250, loss[loss=0.1469, simple_loss=0.2392, pruned_loss=0.0273, over 18424.00 frames. ], tot_loss[loss=0.1758, simple_loss=0.2649, pruned_loss=0.0434, over 3706662.88 frames. ], batch size: 48, lr: 5.49e-03, grad_scale: 32.0
2022-12-23 16:38:00,385 INFO [zipformer.py:660] (2/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=71386.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 16:38:33,469 WARNING [train.py:1060] (2/4) Exclude cut with ID 497-129325-0061-62254-0_sp1.1 from training. Duration: 0.97725
2022-12-23 16:38:56,755 INFO [train.py:894] (2/4) Epoch 21, batch 1300, loss[loss=0.1702, simple_loss=0.2504, pruned_loss=0.04496, over 18378.00 frames. ], tot_loss[loss=0.1747, simple_loss=0.2635, pruned_loss=0.04297, over 3707892.49 frames. ], batch size: 46, lr: 5.49e-03, grad_scale: 32.0
2022-12-23 16:39:16,404 INFO [scaling.py:679] (2/4) Whitening: num_groups=8, num_channels=96, metric=1.12 vs. limit=2.0
2022-12-23 16:39:18,556 WARNING [train.py:1060] (2/4) Exclude cut with ID 7395-89880-0031-39906-0_sp0.9 from training. Duration: 22.97225
2022-12-23 16:39:19,917 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 1.912e+02 3.273e+02 4.091e+02 5.101e+02 9.665e+02, threshold=8.182e+02, percent-clipped=2.0
2022-12-23 16:39:48,941 WARNING [train.py:1060] (2/4) Exclude cut with ID 7395-89880-0047-39922-0_sp0.9 from training. Duration: 21.97775
2022-12-23 16:40:04,188 WARNING [train.py:1060] (2/4) Exclude cut with ID 1112-1043-0006-89194-0_sp0.9 from training. Duration: 21.8333125
2022-12-23 16:40:06,124 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.5979, 1.6110, 1.6508, 1.6073, 1.2354, 3.9005, 1.6976, 2.1048],
       device='cuda:2'), covar=tensor([0.3166, 0.2075, 0.2000, 0.2150, 0.1508, 0.0136, 0.1650, 0.0894],
       device='cuda:2'), in_proj_covar=tensor([0.0132, 0.0117, 0.0124, 0.0121, 0.0104, 0.0096, 0.0091, 0.0089],
       device='cuda:2'), out_proj_covar=tensor([0.0005, 0.0005, 0.0005, 0.0005, 0.0004, 0.0003, 0.0004, 0.0004],
       device='cuda:2')
2022-12-23 16:40:13,525 INFO [train.py:894] (2/4) Epoch 21, batch 1350, loss[loss=0.1528, simple_loss=0.2419, pruned_loss=0.03185, over 18394.00 frames. ], tot_loss[loss=0.1749, simple_loss=0.2639, pruned_loss=0.043, over 3709683.26 frames. ], batch size: 46, lr: 5.48e-03, grad_scale: 32.0
2022-12-23 16:40:13,566 WARNING [train.py:1060] (2/4) Exclude cut with ID 1914-133440-0031-94921-0 from training. Duration: 20.47
2022-12-23 16:41:19,036 WARNING [train.py:1060] (2/4) Exclude cut with ID 7395-89880-0037-39912-0_sp0.9 from training. Duration: 20.67225
2022-12-23 16:41:27,516 INFO [train.py:894] (2/4) Epoch 21, batch 1400, loss[loss=0.2117, simple_loss=0.3013, pruned_loss=0.06102, over 18705.00 frames. ], tot_loss[loss=0.1752, simple_loss=0.2639, pruned_loss=0.04326, over 3710104.19 frames. ], batch size: 65, lr: 5.48e-03, grad_scale: 32.0
2022-12-23 16:41:38,874 WARNING [train.py:1060] (2/4) Exclude cut with ID 1914-133440-0024-94914-0_sp0.9 from training. Duration: 25.2444375
2022-12-23 16:41:51,593 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 2.246e+02 3.506e+02 4.118e+02 5.338e+02 9.015e+02, threshold=8.236e+02, percent-clipped=4.0
2022-12-23 16:42:00,567 WARNING [train.py:1060] (2/4) Exclude cut with ID 3340-169293-0021-76797-0_sp0.9 from training. Duration: 21.1445
2022-12-23 16:42:21,825 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([2.4127, 2.2613, 1.9059, 1.1695, 2.7451, 2.5132, 2.2300, 1.7956],
       device='cuda:2'), covar=tensor([0.0380, 0.0451, 0.0554, 0.0878, 0.0272, 0.0389, 0.0521, 0.0908],
       device='cuda:2'), in_proj_covar=tensor([0.0124, 0.0129, 0.0129, 0.0120, 0.0100, 0.0123, 0.0137, 0.0159],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0001, 0.0001, 0.0001, 0.0002, 0.0002],
       device='cuda:2')
2022-12-23 16:42:40,865 INFO [train.py:894] (2/4) Epoch 21, batch 1450, loss[loss=0.1594, simple_loss=0.2549, pruned_loss=0.03199, over 18603.00 frames. ], tot_loss[loss=0.1749, simple_loss=0.2636, pruned_loss=0.04313, over 3711475.04 frames. ], batch size: 53, lr: 5.48e-03, grad_scale: 16.0
2022-12-23 16:43:07,658 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([0.7279, 1.8859, 2.2445, 1.0447, 1.5360, 2.4335, 2.1545, 1.7038],
       device='cuda:2'), covar=tensor([0.0802, 0.0327, 0.0280, 0.0462, 0.0362, 0.0383, 0.0248, 0.0734],
       device='cuda:2'), in_proj_covar=tensor([0.0148, 0.0170, 0.0128, 0.0140, 0.0149, 0.0142, 0.0162, 0.0173],
       device='cuda:2'), out_proj_covar=tensor([1.1288e-04, 1.3067e-04, 9.6200e-05, 1.0458e-04, 1.1136e-04, 1.0881e-04,
        1.2501e-04, 1.3188e-04], device='cuda:2')
2022-12-23 16:43:15,979 WARNING [train.py:1060] (2/4) Exclude cut with ID 2929-85685-0079-62383-0_sp0.9 from training. Duration: 33.038875
2022-12-23 16:43:46,328 INFO [zipformer.py:660] (2/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=71616.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 16:43:51,771 WARNING [train.py:1060] (2/4) Exclude cut with ID 6426-64291-0000-16059-0_sp0.9 from training. Duration: 20.0944375
2022-12-23 16:43:54,995 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([3.3770, 4.8984, 4.4769, 2.4955, 4.8457, 3.8370, 1.1000, 3.3643],
       device='cuda:2'), covar=tensor([0.1678, 0.0905, 0.1257, 0.3050, 0.0661, 0.0703, 0.4817, 0.1305],
       device='cuda:2'), in_proj_covar=tensor([0.0143, 0.0140, 0.0155, 0.0123, 0.0142, 0.0113, 0.0142, 0.0112],
       device='cuda:2'), out_proj_covar=tensor([0.0003, 0.0002, 0.0003, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-23 16:43:56,084 INFO [train.py:894] (2/4) Epoch 21, batch 1500, loss[loss=0.1717, simple_loss=0.2682, pruned_loss=0.03764, over 18691.00 frames. ], tot_loss[loss=0.1756, simple_loss=0.2645, pruned_loss=0.04333, over 3711635.25 frames. ], batch size: 98, lr: 5.48e-03, grad_scale: 16.0
2022-12-23 16:43:56,389 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.3471, 1.0861, 1.5862, 2.6781, 1.8525, 2.2713, 0.6273, 1.8840],
       device='cuda:2'), covar=tensor([0.1985, 0.2066, 0.1675, 0.0830, 0.1183, 0.1185, 0.2568, 0.1328],
       device='cuda:2'), in_proj_covar=tensor([0.0102, 0.0116, 0.0134, 0.0147, 0.0105, 0.0139, 0.0130, 0.0112],
       device='cuda:2'), out_proj_covar=tensor([0.0003, 0.0003, 0.0004, 0.0004, 0.0003, 0.0004, 0.0003, 0.0003],
       device='cuda:2')
2022-12-23 16:44:07,013 WARNING [train.py:1060] (2/4) Exclude cut with ID 2929-85685-0044-62348-0_sp1.1 from training. Duration: 20.4
2022-12-23 16:44:15,915 WARNING [train.py:1060] (2/4) Exclude cut with ID 6330-62851-0022-91297-0 from training. Duration: 20.085
2022-12-23 16:44:22,491 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 2.435e+02 3.355e+02 3.963e+02 5.477e+02 1.143e+03, threshold=7.926e+02, percent-clipped=5.0
2022-12-23 16:44:26,753 WARNING [train.py:1060] (2/4) Exclude cut with ID 4860-13185-0032-76709-0_sp0.9 from training. Duration: 23.07775
2022-12-23 16:44:34,307 INFO [zipformer.py:660] (2/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=71648.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 16:45:06,468 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.5656, 1.4716, 1.5317, 1.5310, 1.2348, 3.3678, 1.5104, 1.8856],
       device='cuda:2'), covar=tensor([0.3225, 0.2252, 0.2044, 0.2124, 0.1455, 0.0178, 0.1591, 0.0859],
       device='cuda:2'), in_proj_covar=tensor([0.0133, 0.0118, 0.0125, 0.0121, 0.0104, 0.0097, 0.0092, 0.0089],
       device='cuda:2'), out_proj_covar=tensor([0.0005, 0.0005, 0.0005, 0.0005, 0.0004, 0.0003, 0.0004, 0.0004],
       device='cuda:2')
2022-12-23 16:45:11,907 INFO [train.py:894] (2/4) Epoch 21, batch 1550, loss[loss=0.1771, simple_loss=0.2672, pruned_loss=0.04353, over 18513.00 frames. ], tot_loss[loss=0.1745, simple_loss=0.2632, pruned_loss=0.04287, over 3711929.77 frames. ], batch size: 52, lr: 5.48e-03, grad_scale: 16.0
2022-12-23 16:45:14,922 WARNING [train.py:1060] (2/4) Exclude cut with ID 2929-85685-0044-62348-0_sp0.9 from training. Duration: 24.9333125
2022-12-23 16:45:18,060 INFO [zipformer.py:660] (2/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=71677.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 16:45:30,961 INFO [zipformer.py:660] (2/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=71686.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 16:45:45,731 INFO [zipformer.py:660] (2/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=71696.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 16:45:47,578 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([3.3209, 3.2088, 2.2115, 1.7862, 3.7373, 3.7377, 3.1810, 2.4274],
       device='cuda:2'), covar=tensor([0.0363, 0.0388, 0.0575, 0.0732, 0.0213, 0.0308, 0.0457, 0.0842],
       device='cuda:2'), in_proj_covar=tensor([0.0125, 0.0129, 0.0130, 0.0120, 0.0101, 0.0124, 0.0137, 0.0159],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0001, 0.0001, 0.0001, 0.0002, 0.0002],
       device='cuda:2')
2022-12-23 16:45:57,909 WARNING [train.py:1060] (2/4) Exclude cut with ID 5118-111612-0016-124680-0_sp0.9 from training. Duration: 20.388875
2022-12-23 16:46:04,389 WARNING [train.py:1060] (2/4) Exclude cut with ID 432-122774-0017-62487-0_sp1.1 from training. Duration: 20.3590625
2022-12-23 16:46:18,589 INFO [scaling.py:679] (2/4) Whitening: num_groups=1, num_channels=384, metric=4.76 vs. limit=5.0
2022-12-23 16:46:25,799 INFO [train.py:894] (2/4) Epoch 21, batch 1600, loss[loss=0.1657, simple_loss=0.252, pruned_loss=0.03972, over 18580.00 frames. ], tot_loss[loss=0.1741, simple_loss=0.2627, pruned_loss=0.04269, over 3711669.96 frames. ], batch size: 49, lr: 5.47e-03, grad_scale: 16.0
2022-12-23 16:46:41,430 INFO [zipformer.py:660] (2/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=71734.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 16:46:50,328 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 2.261e+02 3.265e+02 3.938e+02 4.745e+02 1.111e+03, threshold=7.876e+02, percent-clipped=1.0
2022-12-23 16:47:13,751 WARNING [train.py:1060] (2/4) Exclude cut with ID 3557-8342-0013-54691-0_sp1.1 from training. Duration: 0.836375
2022-12-23 16:47:40,614 INFO [train.py:894] (2/4) Epoch 21, batch 1650, loss[loss=0.1927, simple_loss=0.2821, pruned_loss=0.05159, over 18724.00 frames. ], tot_loss[loss=0.1748, simple_loss=0.2634, pruned_loss=0.04311, over 3710952.65 frames. ], batch size: 52, lr: 5.47e-03, grad_scale: 16.0
2022-12-23 16:47:57,350 WARNING [train.py:1060] (2/4) Exclude cut with ID 8565-290391-0049-67394-0_sp0.9 from training. Duration: 21.3166875
2022-12-23 16:48:25,810 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.6552, 1.7239, 1.5896, 1.7785, 1.2782, 3.7765, 1.6229, 2.2107],
       device='cuda:2'), covar=tensor([0.4438, 0.2805, 0.2440, 0.2767, 0.1586, 0.0261, 0.1640, 0.0878],
       device='cuda:2'), in_proj_covar=tensor([0.0133, 0.0118, 0.0125, 0.0121, 0.0104, 0.0096, 0.0092, 0.0089],
       device='cuda:2'), out_proj_covar=tensor([0.0005, 0.0005, 0.0005, 0.0005, 0.0004, 0.0003, 0.0004, 0.0004],
       device='cuda:2')
2022-12-23 16:48:28,784 WARNING [train.py:1060] (2/4) Exclude cut with ID 6533-399-0029-104863-0_sp0.9 from training. Duration: 22.1055625
2022-12-23 16:48:39,024 WARNING [train.py:1060] (2/4) Exclude cut with ID 7699-105389-0094-26379-0_sp1.1 from training. Duration: 21.77725
2022-12-23 16:48:56,723 INFO [train.py:894] (2/4) Epoch 21, batch 1700, loss[loss=0.1889, simple_loss=0.2766, pruned_loss=0.05056, over 18528.00 frames. ], tot_loss[loss=0.176, simple_loss=0.2633, pruned_loss=0.04435, over 3711677.02 frames. ], batch size: 58, lr: 5.47e-03, grad_scale: 16.0
2022-12-23 16:48:59,624 WARNING [train.py:1060] (2/4) Exclude cut with ID 7255-291500-0005-134304-0_sp0.9 from training. Duration: 27.8166875
2022-12-23 16:49:21,107 WARNING [train.py:1060] (2/4) Exclude cut with ID 7357-94126-0021-15852-0_sp1.1 from training. Duration: 22.5090625
2022-12-23 16:49:22,519 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 2.287e+02 4.126e+02 5.487e+02 6.780e+02 1.150e+03, threshold=1.097e+03, percent-clipped=15.0
2022-12-23 16:49:27,525 WARNING [train.py:1060] (2/4) Exclude cut with ID 7255-291500-0005-134304-0 from training. Duration: 25.035
2022-12-23 16:49:47,132 WARNING [train.py:1060] (2/4) Exclude cut with ID 774-127930-0014-10412-0_sp1.1 from training. Duration: 0.95
2022-12-23 16:50:02,287 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([0.6223, 1.8049, 2.1164, 1.1563, 1.4061, 2.2619, 2.0953, 1.6643],
       device='cuda:2'), covar=tensor([0.0771, 0.0333, 0.0289, 0.0404, 0.0362, 0.0411, 0.0221, 0.0694],
       device='cuda:2'), in_proj_covar=tensor([0.0149, 0.0171, 0.0128, 0.0140, 0.0149, 0.0142, 0.0163, 0.0174],
       device='cuda:2'), out_proj_covar=tensor([1.1359e-04, 1.3102e-04, 9.6489e-05, 1.0431e-04, 1.1146e-04, 1.0896e-04,
        1.2587e-04, 1.3236e-04], device='cuda:2')
2022-12-23 16:50:04,660 WARNING [train.py:1060] (2/4) Exclude cut with ID 3033-130750-0096-55598-0_sp0.9 from training. Duration: 0.92225
2022-12-23 16:50:13,433 INFO [train.py:894] (2/4) Epoch 21, batch 1750, loss[loss=0.1586, simple_loss=0.2362, pruned_loss=0.04049, over 18484.00 frames. ], tot_loss[loss=0.1787, simple_loss=0.2647, pruned_loss=0.0464, over 3712323.22 frames. ], batch size: 43, lr: 5.47e-03, grad_scale: 16.0
2022-12-23 16:50:32,221 WARNING [train.py:1060] (2/4) Exclude cut with ID 4511-76322-0006-80011-0 from training. Duration: 21.97
2022-12-23 16:50:52,940 WARNING [train.py:1060] (2/4) Exclude cut with ID 7492-105653-0055-62765-0_sp0.9 from training. Duration: 21.97225
2022-12-23 16:50:54,401 WARNING [train.py:1060] (2/4) Exclude cut with ID 453-131332-0000-47844-0_sp0.9 from training. Duration: 25.3333125
2022-12-23 16:51:02,746 WARNING [train.py:1060] (2/4) Exclude cut with ID 5172-29468-0015-19128-0_sp0.9 from training. Duration: 21.5055625
2022-12-23 16:51:13,436 WARNING [train.py:1060] (2/4) Exclude cut with ID 453-131332-0000-47844-0_sp1.1 from training. Duration: 20.72725
2022-12-23 16:51:23,858 INFO [zipformer.py:660] (2/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=71919.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 16:51:29,044 INFO [train.py:894] (2/4) Epoch 21, batch 1800, loss[loss=0.2018, simple_loss=0.2851, pruned_loss=0.05926, over 18604.00 frames. ], tot_loss[loss=0.1811, simple_loss=0.2661, pruned_loss=0.04807, over 3713851.32 frames. ], batch size: 99, lr: 5.47e-03, grad_scale: 16.0
2022-12-23 16:51:45,384 WARNING [train.py:1060] (2/4) Exclude cut with ID 8631-249866-0030-130156-0_sp0.9 from training. Duration: 26.32775
2022-12-23 16:51:54,252 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 3.190e+02 4.570e+02 5.531e+02 6.951e+02 1.422e+03, threshold=1.106e+03, percent-clipped=2.0
2022-12-23 16:52:12,828 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([2.2519, 2.8723, 3.0447, 1.6626, 3.3374, 3.1889, 2.1432, 3.4447],
       device='cuda:2'), covar=tensor([0.1425, 0.1658, 0.1426, 0.2277, 0.0681, 0.1189, 0.2254, 0.0565],
       device='cuda:2'), in_proj_covar=tensor([0.0199, 0.0214, 0.0207, 0.0195, 0.0176, 0.0218, 0.0214, 0.0194],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-23 16:52:19,611 WARNING [train.py:1060] (2/4) Exclude cut with ID 3867-173237-0077-144769-0 from training. Duration: 20.025
2022-12-23 16:52:23,995 WARNING [train.py:1060] (2/4) Exclude cut with ID 6709-74022-0004-86860-0_sp1.1 from training. Duration: 0.9409375
2022-12-23 16:52:24,005 WARNING [train.py:1060] (2/4) Exclude cut with ID 4757-1811-0023-62229-0_sp0.9 from training. Duration: 21.37775
2022-12-23 16:52:41,871 INFO [zipformer.py:660] (2/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=71972.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 16:52:43,123 INFO [train.py:894] (2/4) Epoch 21, batch 1850, loss[loss=0.2086, simple_loss=0.2928, pruned_loss=0.06222, over 18622.00 frames. ], tot_loss[loss=0.1834, simple_loss=0.2674, pruned_loss=0.0497, over 3713773.62 frames. ], batch size: 78, lr: 5.47e-03, grad_scale: 16.0
2022-12-23 16:52:46,896 WARNING [train.py:1060] (2/4) Exclude cut with ID 1250-135782-0004-25974-0_sp0.9 from training. Duration: 21.17225
2022-12-23 16:52:46,908 WARNING [train.py:1060] (2/4) Exclude cut with ID 7357-94126-0021-15852-0_sp0.9 from training. Duration: 27.511125
2022-12-23 16:52:54,586 INFO [zipformer.py:660] (2/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=71980.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 16:52:56,099 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.9778, 1.9564, 1.4866, 1.9602, 2.0509, 1.8354, 2.6084, 2.0339],
       device='cuda:2'), covar=tensor([0.0897, 0.1606, 0.2755, 0.1840, 0.1922, 0.0912, 0.0916, 0.1213],
       device='cuda:2'), in_proj_covar=tensor([0.0178, 0.0209, 0.0252, 0.0290, 0.0238, 0.0192, 0.0208, 0.0204],
       device='cuda:2'), out_proj_covar=tensor([0.0001, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-23 16:53:20,214 WARNING [train.py:1060] (2/4) Exclude cut with ID 453-131332-0000-47844-0 from training. Duration: 22.8
2022-12-23 16:53:23,290 WARNING [train.py:1060] (2/4) Exclude cut with ID 4964-30587-0040-44509-0 from training. Duration: 22.585
2022-12-23 16:53:58,070 WARNING [train.py:1060] (2/4) Exclude cut with ID 6978-92210-0001-146967-0_sp0.9 from training. Duration: 22.0166875
2022-12-23 16:54:02,865 INFO [train.py:894] (2/4) Epoch 21, batch 1900, loss[loss=0.1929, simple_loss=0.2795, pruned_loss=0.05311, over 18601.00 frames. ], tot_loss[loss=0.1844, simple_loss=0.2678, pruned_loss=0.05051, over 3714566.26 frames. ], batch size: 69, lr: 5.46e-03, grad_scale: 16.0
2022-12-23 16:54:14,575 WARNING [train.py:1060] (2/4) Exclude cut with ID 7255-291500-0003-134302-0_sp1.1 from training. Duration: 24.395375
2022-12-23 16:54:19,244 WARNING [train.py:1060] (2/4) Exclude cut with ID 3488-85273-0017-41203-0_sp0.9 from training. Duration: 27.47775
2022-12-23 16:54:23,841 WARNING [train.py:1060] (2/4) Exclude cut with ID 432-122774-0017-62487-0_sp0.9 from training. Duration: 24.8833125
2022-12-23 16:54:27,499 WARNING [train.py:1060] (2/4) Exclude cut with ID 6758-72288-0033-108368-0 from training. Duration: 23.39
2022-12-23 16:54:28,687 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 2.768e+02 4.620e+02 5.475e+02 6.791e+02 1.324e+03, threshold=1.095e+03, percent-clipped=3.0
2022-12-23 16:54:32,060 WARNING [train.py:1060] (2/4) Exclude cut with ID 8291-282929-0007-12994-0_sp0.9 from training. Duration: 28.72225
2022-12-23 16:54:42,242 WARNING [train.py:1060] (2/4) Exclude cut with ID 585-294811-0110-133686-0_sp0.9 from training. Duration: 20.8944375
2022-12-23 16:54:57,073 WARNING [train.py:1060] (2/4) Exclude cut with ID 5796-66357-0007-116447-0_sp0.9 from training. Duration: 23.8444375
2022-12-23 16:55:18,376 INFO [train.py:894] (2/4) Epoch 21, batch 1950, loss[loss=0.1936, simple_loss=0.2775, pruned_loss=0.05485, over 18586.00 frames. ], tot_loss[loss=0.1848, simple_loss=0.2678, pruned_loss=0.05095, over 3715194.87 frames. ], batch size: 98, lr: 5.46e-03, grad_scale: 16.0
2022-12-23 16:55:22,696 WARNING [train.py:1060] (2/4) Exclude cut with ID 8291-282929-0007-12994-0 from training. Duration: 25.85
2022-12-23 16:55:23,970 WARNING [train.py:1060] (2/4) Exclude cut with ID 8291-282929-0023-13010-0 from training. Duration: 21.39
2022-12-23 16:55:34,509 WARNING [train.py:1060] (2/4) Exclude cut with ID 7357-94126-0014-15845-0 from training. Duration: 27.92
2022-12-23 16:56:01,713 WARNING [train.py:1060] (2/4) Exclude cut with ID 8631-249866-0039-130165-0_sp0.9 from training. Duration: 20.661125
2022-12-23 16:56:24,256 WARNING [train.py:1060] (2/4) Exclude cut with ID 7357-94126-0043-15874-0_sp0.9 from training. Duration: 20.07225
2022-12-23 16:56:31,944 WARNING [train.py:1060] (2/4) Exclude cut with ID 1085-156170-0017-128270-0 from training. Duration: 21.01
2022-12-23 16:56:33,407 INFO [train.py:894] (2/4) Epoch 21, batch 2000, loss[loss=0.2013, simple_loss=0.2852, pruned_loss=0.05872, over 18532.00 frames. ], tot_loss[loss=0.1859, simple_loss=0.2685, pruned_loss=0.05162, over 3714835.79 frames. ], batch size: 55, lr: 5.46e-03, grad_scale: 16.0
2022-12-23 16:56:42,989 INFO [zipformer.py:660] (2/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=72129.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 16:57:01,403 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 2.763e+02 4.237e+02 5.051e+02 6.066e+02 1.490e+03, threshold=1.010e+03, percent-clipped=3.0
2022-12-23 16:57:04,885 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([3.2199, 3.1888, 2.0258, 1.5926, 3.5573, 3.5206, 2.9300, 2.1880],
       device='cuda:2'), covar=tensor([0.0367, 0.0374, 0.0617, 0.0770, 0.0253, 0.0344, 0.0475, 0.0894],
       device='cuda:2'), in_proj_covar=tensor([0.0124, 0.0127, 0.0129, 0.0119, 0.0099, 0.0123, 0.0135, 0.0157],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0001, 0.0001, 0.0001, 0.0002, 0.0002],
       device='cuda:2')
2022-12-23 16:57:38,839 WARNING [train.py:1060] (2/4) Exclude cut with ID 2195-150901-0045-59933-0 from training. Duration: 20.65
2022-12-23 16:57:47,574 WARNING [train.py:1060] (2/4) Exclude cut with ID 5796-66357-0007-116447-0 from training. Duration: 21.46
2022-12-23 16:57:50,427 INFO [train.py:894] (2/4) Epoch 21, batch 2050, loss[loss=0.1781, simple_loss=0.2594, pruned_loss=0.04838, over 18536.00 frames. ], tot_loss[loss=0.1864, simple_loss=0.2686, pruned_loss=0.05206, over 3714800.20 frames. ], batch size: 47, lr: 5.46e-03, grad_scale: 16.0
2022-12-23 16:58:08,837 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.3971, 2.4983, 3.2610, 0.9322, 2.6861, 3.4391, 2.4710, 2.5430],
       device='cuda:2'), covar=tensor([0.0822, 0.0451, 0.0249, 0.0582, 0.0445, 0.0442, 0.0403, 0.0824],
       device='cuda:2'), in_proj_covar=tensor([0.0148, 0.0170, 0.0128, 0.0140, 0.0148, 0.0142, 0.0164, 0.0173],
       device='cuda:2'), out_proj_covar=tensor([1.1300e-04, 1.3026e-04, 9.6539e-05, 1.0425e-04, 1.1047e-04, 1.0861e-04,
        1.2604e-04, 1.3217e-04], device='cuda:2')
2022-12-23 16:58:16,464 INFO [zipformer.py:660] (2/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=72190.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 16:58:33,044 WARNING [train.py:1060] (2/4) Exclude cut with ID 3557-8342-0013-54691-0 from training. Duration: 0.92
2022-12-23 16:58:39,071 WARNING [train.py:1060] (2/4) Exclude cut with ID 8291-282929-0023-13010-0_sp0.9 from training. Duration: 23.7666875
2022-12-23 16:59:00,833 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.7837, 1.9952, 2.2748, 1.3468, 2.1795, 2.2101, 1.5936, 2.6289],
       device='cuda:2'), covar=tensor([0.1343, 0.1905, 0.1358, 0.1957, 0.0792, 0.1185, 0.2420, 0.0547],
       device='cuda:2'), in_proj_covar=tensor([0.0197, 0.0214, 0.0207, 0.0193, 0.0176, 0.0217, 0.0215, 0.0192],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-23 16:59:04,644 INFO [train.py:894] (2/4) Epoch 21, batch 2100, loss[loss=0.1829, simple_loss=0.2539, pruned_loss=0.05595, over 18462.00 frames. ], tot_loss[loss=0.1874, simple_loss=0.2693, pruned_loss=0.05275, over 3715522.99 frames. ], batch size: 43, lr: 5.46e-03, grad_scale: 16.0
2022-12-23 16:59:16,095 WARNING [train.py:1060] (2/4) Exclude cut with ID 8544-281189-0060-101339-0_sp0.9 from training. Duration: 20.861125
2022-12-23 16:59:26,223 WARNING [train.py:1060] (2/4) Exclude cut with ID 3488-65654-0031-41259-0_sp0.9 from training. Duration: 22.711125
2022-12-23 16:59:30,366 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 3.195e+02 4.359e+02 5.318e+02 6.615e+02 1.286e+03, threshold=1.064e+03, percent-clipped=4.0
2022-12-23 16:59:45,655 INFO [scaling.py:679] (2/4) Whitening: num_groups=1, num_channels=384, metric=3.04 vs. limit=5.0
2022-12-23 16:59:54,555 INFO [scaling.py:679] (2/4) Whitening: num_groups=8, num_channels=96, metric=1.33 vs. limit=2.0
2022-12-23 17:00:07,799 WARNING [train.py:1060] (2/4) Exclude cut with ID 6951-79737-0043-132310-0_sp1.1 from training. Duration: 22.986375
2022-12-23 17:00:11,409 INFO [scaling.py:679] (2/4) Whitening: num_groups=8, num_channels=192, metric=1.59 vs. limit=2.0
2022-12-23 17:00:18,267 INFO [zipformer.py:660] (2/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=72272.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 17:00:19,408 INFO [train.py:894] (2/4) Epoch 21, batch 2150, loss[loss=0.2208, simple_loss=0.2993, pruned_loss=0.07116, over 18571.00 frames. ], tot_loss[loss=0.1885, simple_loss=0.27, pruned_loss=0.05353, over 3715087.45 frames. ], batch size: 56, lr: 5.45e-03, grad_scale: 16.0
2022-12-23 17:00:23,088 INFO [zipformer.py:660] (2/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=72275.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 17:00:24,505 WARNING [train.py:1060] (2/4) Exclude cut with ID 8040-260924-0003-80960-0_sp0.9 from training. Duration: 22.07225
2022-12-23 17:00:30,523 WARNING [train.py:1060] (2/4) Exclude cut with ID 7699-105389-0045-26330-0_sp0.9 from training. Duration: 20.3055625
2022-12-23 17:00:32,051 WARNING [train.py:1060] (2/4) Exclude cut with ID 6356-271890-0060-94317-0_sp0.9 from training. Duration: 20.72225
2022-12-23 17:00:50,050 WARNING [train.py:1060] (2/4) Exclude cut with ID 3488-85273-0017-41203-0_sp1.1 from training. Duration: 22.4818125
2022-12-23 17:01:14,447 WARNING [train.py:1060] (2/4) Exclude cut with ID 4964-30587-0040-44509-0_sp0.9 from training. Duration: 25.0944375
2022-12-23 17:01:17,597 WARNING [train.py:1060] (2/4) Exclude cut with ID 6533-399-0047-104881-0 from training. Duration: 21.515
2022-12-23 17:01:22,532 WARNING [train.py:1060] (2/4) Exclude cut with ID 7357-94126-0009-15840-0_sp0.9 from training. Duration: 27.02225
2022-12-23 17:01:27,306 WARNING [train.py:1060] (2/4) Exclude cut with ID 432-122774-0010-62480-0_sp0.9 from training. Duration: 22.22225
2022-12-23 17:01:33,673 INFO [zipformer.py:660] (2/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=72320.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 17:01:37,997 INFO [train.py:894] (2/4) Epoch 21, batch 2200, loss[loss=0.1787, simple_loss=0.2626, pruned_loss=0.0474, over 18450.00 frames. ], tot_loss[loss=0.1884, simple_loss=0.2698, pruned_loss=0.05352, over 3714786.20 frames. ], batch size: 50, lr: 5.45e-03, grad_scale: 16.0
2022-12-23 17:01:38,055 WARNING [train.py:1060] (2/4) Exclude cut with ID 4964-30587-0085-44554-0_sp0.9 from training. Duration: 20.85
2022-12-23 17:01:43,159 INFO [zipformer.py:660] (2/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=72326.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 17:02:04,135 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 2.723e+02 4.275e+02 5.311e+02 6.247e+02 1.700e+03, threshold=1.062e+03, percent-clipped=3.0
2022-12-23 17:02:10,065 WARNING [train.py:1060] (2/4) Exclude cut with ID 4295-39940-0007-92567-0 from training. Duration: 21.54
2022-12-23 17:02:13,232 WARNING [train.py:1060] (2/4) Exclude cut with ID 4964-30587-0040-44509-0_sp1.1 from training. Duration: 20.5318125
2022-12-23 17:02:22,021 WARNING [train.py:1060] (2/4) Exclude cut with ID 7255-291500-0012-134311-0_sp0.9 from training. Duration: 21.9333125
2022-12-23 17:02:48,627 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([2.2638, 1.9631, 1.5699, 0.5979, 1.4713, 1.9012, 1.6320, 1.8409],
       device='cuda:2'), covar=tensor([0.0615, 0.0543, 0.1113, 0.1586, 0.1092, 0.1491, 0.1562, 0.0644],
       device='cuda:2'), in_proj_covar=tensor([0.0171, 0.0185, 0.0205, 0.0191, 0.0208, 0.0199, 0.0213, 0.0201],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-23 17:02:54,652 INFO [train.py:894] (2/4) Epoch 21, batch 2250, loss[loss=0.1907, simple_loss=0.2821, pruned_loss=0.04963, over 18507.00 frames. ], tot_loss[loss=0.1865, simple_loss=0.2681, pruned_loss=0.05247, over 3712936.47 frames. ], batch size: 52, lr: 5.45e-03, grad_scale: 16.0
2022-12-23 17:03:13,043 WARNING [train.py:1060] (2/4) Exclude cut with ID 8631-249866-0025-130151-0_sp0.9 from training. Duration: 21.7944375
2022-12-23 17:03:16,241 INFO [zipformer.py:660] (2/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=72387.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 17:03:26,978 WARNING [train.py:1060] (2/4) Exclude cut with ID 8291-282929-0002-12989-0_sp0.9 from training. Duration: 22.4666875
2022-12-23 17:03:34,255 WARNING [train.py:1060] (2/4) Exclude cut with ID 6121-9014-0076-24124-0 from training. Duration: 21.635
2022-12-23 17:03:40,362 WARNING [train.py:1060] (2/4) Exclude cut with ID 6121-9014-0076-24124-0_sp0.9 from training. Duration: 24.038875
2022-12-23 17:04:10,546 INFO [train.py:894] (2/4) Epoch 21, batch 2300, loss[loss=0.1803, simple_loss=0.2638, pruned_loss=0.04845, over 18531.00 frames. ], tot_loss[loss=0.1858, simple_loss=0.2672, pruned_loss=0.05218, over 3714133.21 frames. ], batch size: 98, lr: 5.45e-03, grad_scale: 16.0
2022-12-23 17:04:24,328 WARNING [train.py:1060] (2/4) Exclude cut with ID 7255-291500-0009-134308-0_sp1.1 from training. Duration: 21.786375
2022-12-23 17:04:35,980 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 2.221e+02 4.085e+02 4.933e+02 6.235e+02 1.106e+03, threshold=9.865e+02, percent-clipped=2.0
2022-12-23 17:04:36,046 WARNING [train.py:1060] (2/4) Exclude cut with ID 8291-282929-0002-12989-0 from training. Duration: 20.22
2022-12-23 17:05:25,821 INFO [train.py:894] (2/4) Epoch 21, batch 2350, loss[loss=0.1794, simple_loss=0.2721, pruned_loss=0.0434, over 18655.00 frames. ], tot_loss[loss=0.1866, simple_loss=0.268, pruned_loss=0.05253, over 3714858.49 frames. ], batch size: 62, lr: 5.45e-03, grad_scale: 16.0
2022-12-23 17:05:38,404 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([2.0319, 2.3656, 1.9729, 2.5628, 2.9836, 1.9327, 2.1330, 1.7320],
       device='cuda:2'), covar=tensor([0.1734, 0.1569, 0.1432, 0.0908, 0.1360, 0.1051, 0.1884, 0.1344],
       device='cuda:2'), in_proj_covar=tensor([0.0245, 0.0225, 0.0213, 0.0196, 0.0258, 0.0196, 0.0222, 0.0197],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-23 17:05:43,966 INFO [zipformer.py:660] (2/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=72485.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 17:06:35,799 WARNING [train.py:1060] (2/4) Exclude cut with ID 6951-79737-0043-132310-0 from training. Duration: 25.285
2022-12-23 17:06:41,687 INFO [train.py:894] (2/4) Epoch 21, batch 2400, loss[loss=0.2226, simple_loss=0.2972, pruned_loss=0.074, over 18592.00 frames. ], tot_loss[loss=0.1855, simple_loss=0.2669, pruned_loss=0.05201, over 3715236.61 frames. ], batch size: 179, lr: 5.44e-03, grad_scale: 8.0
2022-12-23 17:06:56,746 INFO [scaling.py:679] (2/4) Whitening: num_groups=8, num_channels=96, metric=1.34 vs. limit=2.0
2022-12-23 17:07:08,714 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 2.768e+02 4.305e+02 5.500e+02 6.750e+02 1.440e+03, threshold=1.100e+03, percent-clipped=3.0
2022-12-23 17:07:43,888 WARNING [train.py:1060] (2/4) Exclude cut with ID 811-130148-0001-63453-0_sp0.9 from training. Duration: 20.861125
2022-12-23 17:07:58,621 INFO [train.py:894] (2/4) Epoch 21, batch 2450, loss[loss=0.2159, simple_loss=0.2915, pruned_loss=0.0702, over 18682.00 frames. ], tot_loss[loss=0.1848, simple_loss=0.2662, pruned_loss=0.05173, over 3716127.31 frames. ], batch size: 62, lr: 5.44e-03, grad_scale: 8.0
2022-12-23 17:08:02,302 INFO [zipformer.py:660] (2/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=72575.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 17:08:05,062 WARNING [train.py:1060] (2/4) Exclude cut with ID 6010-56788-0055-90261-0 from training. Duration: 20.88
2022-12-23 17:08:36,165 WARNING [train.py:1060] (2/4) Exclude cut with ID 7357-94126-0045-15876-0_sp0.9 from training. Duration: 23.4166875
2022-12-23 17:09:05,510 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([2.5186, 1.8961, 2.0792, 2.1201, 2.3600, 2.3875, 2.2464, 1.9044],
       device='cuda:2'), covar=tensor([0.2305, 0.3476, 0.2589, 0.3093, 0.2041, 0.0964, 0.3517, 0.1354],
       device='cuda:2'), in_proj_covar=tensor([0.0268, 0.0299, 0.0277, 0.0314, 0.0306, 0.0251, 0.0342, 0.0240],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-23 17:09:11,902 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([2.1955, 2.2916, 1.6441, 2.6251, 2.4123, 2.0939, 2.9947, 2.2957],
       device='cuda:2'), covar=tensor([0.0841, 0.1526, 0.2619, 0.1563, 0.1582, 0.0882, 0.0860, 0.1062],
       device='cuda:2'), in_proj_covar=tensor([0.0177, 0.0208, 0.0250, 0.0289, 0.0236, 0.0190, 0.0206, 0.0202],
       device='cuda:2'), out_proj_covar=tensor([0.0001, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-23 17:09:15,641 INFO [train.py:894] (2/4) Epoch 21, batch 2500, loss[loss=0.2094, simple_loss=0.2904, pruned_loss=0.06419, over 18696.00 frames. ], tot_loss[loss=0.1843, simple_loss=0.2656, pruned_loss=0.05149, over 3714831.11 frames. ], batch size: 78, lr: 5.44e-03, grad_scale: 8.0
2022-12-23 17:09:15,796 INFO [zipformer.py:660] (2/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=72623.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 17:09:42,409 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 3.111e+02 4.322e+02 5.104e+02 6.362e+02 1.624e+03, threshold=1.021e+03, percent-clipped=3.0
2022-12-23 17:09:53,106 WARNING [train.py:1060] (2/4) Exclude cut with ID 7255-291500-0006-134305-0 from training. Duration: 21.24
2022-12-23 17:09:53,122 WARNING [train.py:1060] (2/4) Exclude cut with ID 6533-399-0047-104881-0_sp0.9 from training. Duration: 23.9055625
2022-12-23 17:10:18,008 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.7289, 1.6558, 1.7643, 1.6511, 1.4154, 3.9203, 1.6883, 2.1138],
       device='cuda:2'), covar=tensor([0.3143, 0.2068, 0.1880, 0.2074, 0.1362, 0.0191, 0.1582, 0.0842],
       device='cuda:2'), in_proj_covar=tensor([0.0134, 0.0118, 0.0125, 0.0122, 0.0104, 0.0097, 0.0092, 0.0089],
       device='cuda:2'), out_proj_covar=tensor([0.0005, 0.0005, 0.0005, 0.0005, 0.0004, 0.0003, 0.0004, 0.0004],
       device='cuda:2')
2022-12-23 17:10:29,144 WARNING [train.py:1060] (2/4) Exclude cut with ID 6758-72288-0033-108368-0_sp0.9 from training. Duration: 25.988875
2022-12-23 17:10:32,085 INFO [train.py:894] (2/4) Epoch 21, batch 2550, loss[loss=0.1801, simple_loss=0.2673, pruned_loss=0.04644, over 18515.00 frames. ], tot_loss[loss=0.184, simple_loss=0.2652, pruned_loss=0.05143, over 3715000.04 frames. ], batch size: 78, lr: 5.44e-03, grad_scale: 8.0
2022-12-23 17:10:37,606 WARNING [train.py:1060] (2/4) Exclude cut with ID 7255-291500-0001-134300-0_sp0.9 from training. Duration: 20.67225
2022-12-23 17:10:40,993 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([2.4842, 2.3051, 2.0298, 1.2101, 2.8716, 2.5987, 2.4314, 1.7544],
       device='cuda:2'), covar=tensor([0.0411, 0.0436, 0.0528, 0.0850, 0.0254, 0.0396, 0.0415, 0.0925],
       device='cuda:2'), in_proj_covar=tensor([0.0125, 0.0128, 0.0129, 0.0120, 0.0099, 0.0124, 0.0135, 0.0159],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0001, 0.0001, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-23 17:10:44,997 INFO [zipformer.py:660] (2/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=72682.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 17:11:15,172 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.9939, 1.3927, 1.7110, 1.7471, 1.9286, 1.9909, 1.8500, 1.7060],
       device='cuda:2'), covar=tensor([0.2560, 0.3666, 0.3094, 0.3050, 0.2543, 0.1286, 0.3515, 0.1586],
       device='cuda:2'), in_proj_covar=tensor([0.0268, 0.0299, 0.0278, 0.0314, 0.0306, 0.0251, 0.0343, 0.0241],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-23 17:11:26,821 WARNING [train.py:1060] (2/4) Exclude cut with ID 3488-85273-0038-41224-0 from training. Duration: 20.34
2022-12-23 17:11:48,020 INFO [train.py:894] (2/4) Epoch 21, batch 2600, loss[loss=0.1579, simple_loss=0.2366, pruned_loss=0.03958, over 18592.00 frames. ], tot_loss[loss=0.1833, simple_loss=0.2646, pruned_loss=0.05102, over 3714596.34 frames. ], batch size: 45, lr: 5.44e-03, grad_scale: 8.0
2022-12-23 17:12:16,331 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 2.887e+02 4.261e+02 5.084e+02 5.993e+02 1.354e+03, threshold=1.017e+03, percent-clipped=3.0
2022-12-23 17:12:35,724 INFO [scaling.py:679] (2/4) Whitening: num_groups=1, num_channels=384, metric=2.76 vs. limit=5.0
2022-12-23 17:12:36,339 WARNING [train.py:1060] (2/4) Exclude cut with ID 7357-94126-0026-15857-0_sp0.9 from training. Duration: 25.061125
2022-12-23 17:12:46,779 INFO [scaling.py:679] (2/4) Whitening: num_groups=8, num_channels=192, metric=1.93 vs. limit=2.0
2022-12-23 17:12:48,900 WARNING [train.py:1060] (2/4) Exclude cut with ID 3033-130750-0096-55598-0 from training. Duration: 0.83
2022-12-23 17:13:04,940 INFO [train.py:894] (2/4) Epoch 21, batch 2650, loss[loss=0.1982, simple_loss=0.2699, pruned_loss=0.06323, over 18650.00 frames. ], tot_loss[loss=0.1842, simple_loss=0.2654, pruned_loss=0.05145, over 3714476.19 frames. ], batch size: 48, lr: 5.44e-03, grad_scale: 8.0
2022-12-23 17:13:13,739 WARNING [train.py:1060] (2/4) Exclude cut with ID 3488-85273-0017-41203-0 from training. Duration: 24.73
2022-12-23 17:13:23,647 INFO [zipformer.py:660] (2/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=72785.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 17:13:28,236 WARNING [train.py:1060] (2/4) Exclude cut with ID 7255-291500-0009-134308-0 from training. Duration: 23.965
2022-12-23 17:13:35,774 WARNING [train.py:1060] (2/4) Exclude cut with ID 6978-92210-0030-146996-0_sp0.9 from training. Duration: 22.088875
2022-12-23 17:13:43,849 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([2.6594, 2.1825, 1.5627, 2.3803, 1.9734, 2.0563, 2.0746, 2.4973],
       device='cuda:2'), covar=tensor([0.1994, 0.3117, 0.2087, 0.2793, 0.3503, 0.1163, 0.3096, 0.0945],
       device='cuda:2'), in_proj_covar=tensor([0.0294, 0.0292, 0.0246, 0.0350, 0.0273, 0.0227, 0.0288, 0.0215],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0003, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-23 17:13:51,741 WARNING [train.py:1060] (2/4) Exclude cut with ID 7255-291500-0006-134305-0_sp0.9 from training. Duration: 23.6
2022-12-23 17:14:20,768 INFO [train.py:894] (2/4) Epoch 21, batch 2700, loss[loss=0.178, simple_loss=0.2455, pruned_loss=0.05525, over 18472.00 frames. ], tot_loss[loss=0.1846, simple_loss=0.266, pruned_loss=0.05164, over 3713836.46 frames. ], batch size: 43, lr: 5.43e-03, grad_scale: 8.0
2022-12-23 17:14:24,858 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.6946, 1.2351, 0.5981, 1.2987, 2.1104, 1.1331, 1.5023, 1.6700],
       device='cuda:2'), covar=tensor([0.1740, 0.2178, 0.2439, 0.1703, 0.1808, 0.1857, 0.1498, 0.1728],
       device='cuda:2'), in_proj_covar=tensor([0.0093, 0.0097, 0.0115, 0.0096, 0.0117, 0.0090, 0.0097, 0.0093],
       device='cuda:2'), out_proj_covar=tensor([0.0003, 0.0003, 0.0004, 0.0003, 0.0004, 0.0003, 0.0003, 0.0003],
       device='cuda:2')
2022-12-23 17:14:37,559 INFO [zipformer.py:660] (2/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=72833.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 17:14:49,642 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 3.034e+02 4.615e+02 5.963e+02 7.996e+02 1.786e+03, threshold=1.193e+03, percent-clipped=8.0
2022-12-23 17:14:59,680 INFO [scaling.py:679] (2/4) Whitening: num_groups=8, num_channels=192, metric=1.79 vs. limit=2.0
2022-12-23 17:15:27,600 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([2.8776, 4.1723, 3.9648, 1.9766, 4.2953, 3.0710, 0.8906, 2.8706],
       device='cuda:2'), covar=tensor([0.1720, 0.1115, 0.1422, 0.3270, 0.0871, 0.0963, 0.4910, 0.1453],
       device='cuda:2'), in_proj_covar=tensor([0.0147, 0.0145, 0.0161, 0.0124, 0.0147, 0.0115, 0.0146, 0.0115],
       device='cuda:2'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0002, 0.0003, 0.0002, 0.0003, 0.0002],
       device='cuda:2')
2022-12-23 17:15:35,300 WARNING [train.py:1060] (2/4) Exclude cut with ID 8291-282929-0024-13011-0 from training. Duration: 23.795
2022-12-23 17:15:38,028 INFO [train.py:894] (2/4) Epoch 21, batch 2750, loss[loss=0.1709, simple_loss=0.2606, pruned_loss=0.04063, over 18599.00 frames. ], tot_loss[loss=0.1839, simple_loss=0.2654, pruned_loss=0.0512, over 3713012.82 frames. ], batch size: 51, lr: 5.43e-03, grad_scale: 8.0
2022-12-23 17:15:51,737 WARNING [train.py:1060] (2/4) Exclude cut with ID 8631-249866-0030-130156-0_sp1.1 from training. Duration: 21.5409375
2022-12-23 17:15:55,308 WARNING [train.py:1060] (2/4) Exclude cut with ID 6978-92210-0019-146985-0_sp0.9 from training. Duration: 24.97775
2022-12-23 17:16:06,885 WARNING [train.py:1060] (2/4) Exclude cut with ID 1085-156170-0017-128270-0_sp0.9 from training. Duration: 23.3444375
2022-12-23 17:16:10,673 INFO [zipformer.py:660] (2/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=72894.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 17:16:35,177 WARNING [train.py:1060] (2/4) Exclude cut with ID 6010-56788-0055-90261-0_sp0.9 from training. Duration: 23.2
2022-12-23 17:16:39,689 WARNING [train.py:1060] (2/4) Exclude cut with ID 5653-46179-0060-117930-0_sp0.9 from training. Duration: 21.17225
2022-12-23 17:16:53,398 INFO [train.py:894] (2/4) Epoch 21, batch 2800, loss[loss=0.2375, simple_loss=0.3058, pruned_loss=0.08454, over 18573.00 frames. ], tot_loss[loss=0.1848, simple_loss=0.2662, pruned_loss=0.05169, over 3713376.26 frames. ], batch size: 56, lr: 5.43e-03, grad_scale: 8.0
2022-12-23 17:17:04,249 WARNING [train.py:1060] (2/4) Exclude cut with ID 7255-291500-0008-134307-0_sp0.9 from training. Duration: 24.6555625
2022-12-23 17:17:21,263 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 2.700e+02 4.162e+02 4.894e+02 6.036e+02 1.354e+03, threshold=9.787e+02, percent-clipped=2.0
2022-12-23 17:17:41,890 INFO [zipformer.py:660] (2/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=72955.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 17:17:55,217 INFO [zipformer.py:660] (2/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=72964.0, num_to_drop=1, layers_to_drop={0}
2022-12-23 17:17:59,399 WARNING [train.py:1060] (2/4) Exclude cut with ID 3488-65654-0031-41259-0 from training. Duration: 20.44
2022-12-23 17:18:07,924 INFO [train.py:894] (2/4) Epoch 21, batch 2850, loss[loss=0.2012, simple_loss=0.282, pruned_loss=0.06017, over 18706.00 frames. ], tot_loss[loss=0.185, simple_loss=0.2667, pruned_loss=0.05166, over 3713395.61 frames. ], batch size: 60, lr: 5.43e-03, grad_scale: 8.0
2022-12-23 17:18:14,408 WARNING [train.py:1060] (2/4) Exclude cut with ID 6951-79737-0018-132285-0_sp0.9 from training. Duration: 23.45
2022-12-23 17:18:23,054 INFO [zipformer.py:660] (2/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=72982.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 17:18:43,796 WARNING [train.py:1060] (2/4) Exclude cut with ID 6945-60535-0076-12784-0_sp0.9 from training. Duration: 20.52225
2022-12-23 17:18:46,274 INFO [scaling.py:679] (2/4) Whitening: num_groups=8, num_channels=96, metric=1.26 vs. limit=2.0
2022-12-23 17:18:51,588 WARNING [train.py:1060] (2/4) Exclude cut with ID 7255-291500-0008-134307-0 from training. Duration: 22.19
2022-12-23 17:19:01,927 WARNING [train.py:1060] (2/4) Exclude cut with ID 7357-94126-0014-15845-0_sp1.1 from training. Duration: 25.3818125
2022-12-23 17:19:17,420 WARNING [train.py:1060] (2/4) Exclude cut with ID 6951-79737-0043-132310-0_sp0.9 from training. Duration: 28.0944375
2022-12-23 17:19:23,509 WARNING [train.py:1060] (2/4) Exclude cut with ID 2195-150901-0045-59933-0_sp0.9 from training. Duration: 22.9444375
2022-12-23 17:19:24,835 INFO [train.py:894] (2/4) Epoch 21, batch 2900, loss[loss=0.1629, simple_loss=0.2444, pruned_loss=0.04074, over 18420.00 frames. ], tot_loss[loss=0.1847, simple_loss=0.2666, pruned_loss=0.05139, over 3714235.25 frames. ], batch size: 48, lr: 5.43e-03, grad_scale: 8.0
2022-12-23 17:19:29,376 INFO [zipformer.py:660] (2/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=73025.0, num_to_drop=1, layers_to_drop={2}
2022-12-23 17:19:33,486 WARNING [train.py:1060] (2/4) Exclude cut with ID 8291-282929-0024-13011-0_sp1.1 from training. Duration: 21.6318125
2022-12-23 17:19:36,565 INFO [zipformer.py:660] (2/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=73030.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 17:19:41,474 INFO [scaling.py:679] (2/4) Whitening: num_groups=8, num_channels=192, metric=1.52 vs. limit=2.0
2022-12-23 17:19:50,193 WARNING [train.py:1060] (2/4) Exclude cut with ID 8631-249866-0030-130156-0 from training. Duration: 23.695
2022-12-23 17:19:53,305 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 2.702e+02 4.963e+02 5.828e+02 7.417e+02 1.430e+03, threshold=1.166e+03, percent-clipped=7.0
2022-12-23 17:20:14,981 WARNING [train.py:1060] (2/4) Exclude cut with ID 7699-105389-0094-26379-0 from training. Duration: 23.955
2022-12-23 17:20:41,613 INFO [train.py:894] (2/4) Epoch 21, batch 2950, loss[loss=0.1867, simple_loss=0.2764, pruned_loss=0.04853, over 18663.00 frames. ], tot_loss[loss=0.1848, simple_loss=0.2664, pruned_loss=0.05156, over 3713725.10 frames. ], batch size: 69, lr: 5.42e-03, grad_scale: 8.0
2022-12-23 17:20:50,705 WARNING [train.py:1060] (2/4) Exclude cut with ID 8291-282929-0024-13011-0_sp0.9 from training. Duration: 26.438875
2022-12-23 17:21:34,847 WARNING [train.py:1060] (2/4) Exclude cut with ID 7699-105389-0021-26306-0_sp0.9 from training. Duration: 21.2444375
2022-12-23 17:21:34,879 WARNING [train.py:1060] (2/4) Exclude cut with ID 7357-94126-0014-15845-0_sp0.9 from training. Duration: 31.02225
2022-12-23 17:21:44,489 WARNING [train.py:1060] (2/4) Exclude cut with ID 432-122774-0017-62487-0 from training. Duration: 22.395
2022-12-23 17:21:48,113 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.6484, 1.6114, 1.4541, 1.5315, 1.8583, 1.7488, 1.8163, 1.2397],
       device='cuda:2'), covar=tensor([0.0297, 0.0213, 0.0433, 0.0182, 0.0174, 0.0359, 0.0250, 0.0286],
       device='cuda:2'), in_proj_covar=tensor([0.0092, 0.0124, 0.0148, 0.0122, 0.0114, 0.0117, 0.0096, 0.0124],
       device='cuda:2'), out_proj_covar=tensor([7.3652e-05, 9.8658e-05, 1.2230e-04, 9.6982e-05, 9.2016e-05, 8.9826e-05,
        7.5048e-05, 9.7704e-05], device='cuda:2')
2022-12-23 17:21:57,239 INFO [train.py:894] (2/4) Epoch 21, batch 3000, loss[loss=0.1748, simple_loss=0.2628, pruned_loss=0.04337, over 18575.00 frames. ], tot_loss[loss=0.185, simple_loss=0.2665, pruned_loss=0.05176, over 3713688.09 frames. ], batch size: 57, lr: 5.42e-03, grad_scale: 8.0
2022-12-23 17:21:57,240 INFO [train.py:919] (2/4) Computing validation loss
2022-12-23 17:22:08,277 INFO [train.py:928] (2/4) Epoch 21, validation: loss=0.1664, simple_loss=0.2634, pruned_loss=0.03471, over 944034.00 frames. 
2022-12-23 17:22:08,277 INFO [train.py:929] (2/4) Maximum memory allocated so far is 24680MB
2022-12-23 17:22:12,877 WARNING [train.py:1060] (2/4) Exclude cut with ID 7357-94126-0045-15876-0 from training. Duration: 21.075
2022-12-23 17:22:19,064 WARNING [train.py:1060] (2/4) Exclude cut with ID 6482-98857-0025-147532-0_sp0.9 from training. Duration: 20.0055625
2022-12-23 17:22:19,072 WARNING [train.py:1060] (2/4) Exclude cut with ID 6951-79737-0037-132304-0_sp0.9 from training. Duration: 22.05
2022-12-23 17:22:19,088 WARNING [train.py:1060] (2/4) Exclude cut with ID 7255-291500-0003-134302-0 from training. Duration: 26.8349375
2022-12-23 17:22:23,723 WARNING [train.py:1060] (2/4) Exclude cut with ID 7357-94126-0009-15840-0_sp1.1 from training. Duration: 22.1090625
2022-12-23 17:22:29,461 WARNING [train.py:1060] (2/4) Exclude cut with ID 7699-105389-0094-26379-0_sp0.9 from training. Duration: 26.6166875
2022-12-23 17:22:35,633 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 3.005e+02 4.313e+02 5.237e+02 6.690e+02 1.463e+03, threshold=1.047e+03, percent-clipped=1.0
2022-12-23 17:22:46,001 WARNING [train.py:1060] (2/4) Exclude cut with ID 2046-178027-0000-53705-0_sp0.9 from training. Duration: 20.3055625
2022-12-23 17:23:10,760 WARNING [train.py:1060] (2/4) Exclude cut with ID 7205-50138-0008-5373-0_sp0.9 from training. Duration: 20.7
2022-12-23 17:23:24,048 INFO [train.py:894] (2/4) Epoch 21, batch 3050, loss[loss=0.1676, simple_loss=0.2483, pruned_loss=0.04345, over 18578.00 frames. ], tot_loss[loss=0.1854, simple_loss=0.2669, pruned_loss=0.05197, over 3713420.34 frames. ], batch size: 41, lr: 5.42e-03, grad_scale: 8.0
2022-12-23 17:23:47,206 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([4.6196, 3.9287, 3.9583, 4.5150, 4.2096, 4.0935, 4.7590, 1.3503],
       device='cuda:2'), covar=tensor([0.0672, 0.0708, 0.0612, 0.0745, 0.1308, 0.1109, 0.0522, 0.5247],
       device='cuda:2'), in_proj_covar=tensor([0.0343, 0.0224, 0.0234, 0.0269, 0.0325, 0.0269, 0.0288, 0.0283],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0001, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-23 17:23:51,658 WARNING [train.py:1060] (2/4) Exclude cut with ID 6978-92210-0019-146985-0 from training. Duration: 22.48
2022-12-23 17:24:07,624 WARNING [train.py:1060] (2/4) Exclude cut with ID 7255-291500-0003-134302-0_sp0.9 from training. Duration: 29.816625
2022-12-23 17:24:15,972 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.6987, 1.4617, 1.0896, 0.3213, 1.1483, 1.6078, 1.3980, 1.4404],
       device='cuda:2'), covar=tensor([0.0670, 0.0608, 0.1044, 0.1587, 0.1102, 0.1622, 0.1700, 0.0712],
       device='cuda:2'), in_proj_covar=tensor([0.0175, 0.0188, 0.0209, 0.0194, 0.0213, 0.0204, 0.0218, 0.0204],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-23 17:24:27,378 WARNING [train.py:1060] (2/4) Exclude cut with ID 7255-291500-0005-134304-0_sp1.1 from training. Duration: 22.7590625
2022-12-23 17:24:33,248 WARNING [train.py:1060] (2/4) Exclude cut with ID 7357-94126-0026-15857-0 from training. Duration: 22.555
2022-12-23 17:24:41,058 INFO [train.py:894] (2/4) Epoch 21, batch 3100, loss[loss=0.2114, simple_loss=0.2916, pruned_loss=0.06558, over 18625.00 frames. ], tot_loss[loss=0.1848, simple_loss=0.2666, pruned_loss=0.0515, over 3713709.65 frames. ], batch size: 78, lr: 5.42e-03, grad_scale: 8.0
2022-12-23 17:24:53,248 WARNING [train.py:1060] (2/4) Exclude cut with ID 1250-135782-0005-25975-0_sp0.9 from training. Duration: 21.688875
2022-12-23 17:25:08,303 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 2.440e+02 3.995e+02 5.110e+02 7.002e+02 1.344e+03, threshold=1.022e+03, percent-clipped=3.0
2022-12-23 17:25:22,316 INFO [zipformer.py:660] (2/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=73250.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 17:25:26,720 WARNING [train.py:1060] (2/4) Exclude cut with ID 3488-85273-0038-41224-0_sp0.9 from training. Duration: 22.6
2022-12-23 17:25:55,601 INFO [train.py:894] (2/4) Epoch 21, batch 3150, loss[loss=0.1634, simple_loss=0.2399, pruned_loss=0.04348, over 18410.00 frames. ], tot_loss[loss=0.1844, simple_loss=0.2663, pruned_loss=0.0512, over 3713334.08 frames. ], batch size: 42, lr: 5.42e-03, grad_scale: 8.0
2022-12-23 17:26:00,712 INFO [scaling.py:679] (2/4) Whitening: num_groups=8, num_channels=96, metric=1.23 vs. limit=2.0
2022-12-23 17:26:08,025 WARNING [train.py:1060] (2/4) Exclude cut with ID 7357-94126-0009-15840-0 from training. Duration: 24.32
2022-12-23 17:26:27,015 INFO [zipformer.py:660] (2/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=73293.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 17:27:04,372 WARNING [train.py:1060] (2/4) Exclude cut with ID 8291-276745-0093-13116-0_sp0.9 from training. Duration: 21.061125
2022-12-23 17:27:07,391 INFO [zipformer.py:660] (2/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=73320.0, num_to_drop=1, layers_to_drop={0}
2022-12-23 17:27:11,712 INFO [train.py:894] (2/4) Epoch 21, batch 3200, loss[loss=0.1776, simple_loss=0.2703, pruned_loss=0.0425, over 18605.00 frames. ], tot_loss[loss=0.1843, simple_loss=0.2662, pruned_loss=0.0512, over 3712711.67 frames. ], batch size: 56, lr: 5.41e-03, grad_scale: 8.0
2022-12-23 17:27:17,758 WARNING [train.py:1060] (2/4) Exclude cut with ID 7357-94126-0024-15855-0_sp0.9 from training. Duration: 20.32225
2022-12-23 17:27:30,137 WARNING [train.py:1060] (2/4) Exclude cut with ID 3033-130750-0096-55598-0_sp1.1 from training. Duration: 0.7545625
2022-12-23 17:27:39,054 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 2.630e+02 4.267e+02 5.062e+02 6.321e+02 9.305e+02, threshold=1.012e+03, percent-clipped=0.0
2022-12-23 17:27:41,335 INFO [scaling.py:679] (2/4) Whitening: num_groups=8, num_channels=96, metric=1.21 vs. limit=2.0
2022-12-23 17:27:45,485 WARNING [train.py:1060] (2/4) Exclude cut with ID 4295-39940-0007-92567-0_sp0.9 from training. Duration: 23.9333125
2022-12-23 17:27:59,106 INFO [zipformer.py:660] (2/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=73354.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 17:28:19,124 WARNING [train.py:1060] (2/4) Exclude cut with ID 7255-291500-0008-134307-0_sp1.1 from training. Duration: 20.17275
2022-12-23 17:28:23,975 WARNING [train.py:1060] (2/4) Exclude cut with ID 6978-92210-0019-146985-0_sp1.1 from training. Duration: 20.436375
2022-12-23 17:28:27,516 INFO [train.py:894] (2/4) Epoch 21, batch 3250, loss[loss=0.2015, simple_loss=0.2795, pruned_loss=0.06179, over 18601.00 frames. ], tot_loss[loss=0.1834, simple_loss=0.2655, pruned_loss=0.05068, over 3712770.04 frames. ], batch size: 182, lr: 5.41e-03, grad_scale: 8.0
2022-12-23 17:29:42,410 INFO [train.py:894] (2/4) Epoch 21, batch 3300, loss[loss=0.1802, simple_loss=0.2607, pruned_loss=0.04986, over 18614.00 frames. ], tot_loss[loss=0.1839, simple_loss=0.266, pruned_loss=0.05093, over 3713747.92 frames. ], batch size: 53, lr: 5.41e-03, grad_scale: 8.0
2022-12-23 17:29:42,458 WARNING [train.py:1060] (2/4) Exclude cut with ID 4234-40345-0022-142709-0_sp0.9 from training. Duration: 23.1055625
2022-12-23 17:29:44,083 WARNING [train.py:1060] (2/4) Exclude cut with ID 8291-282929-0007-12994-0_sp1.1 from training. Duration: 23.5
2022-12-23 17:29:56,992 WARNING [train.py:1060] (2/4) Exclude cut with ID 7255-291500-0009-134308-0_sp0.9 from training. Duration: 26.62775
2022-12-23 17:30:10,519 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 2.729e+02 4.288e+02 5.181e+02 6.375e+02 1.469e+03, threshold=1.036e+03, percent-clipped=5.0
2022-12-23 17:30:10,585 WARNING [train.py:1060] (2/4) Exclude cut with ID 6951-79737-0018-132285-0 from training. Duration: 21.105
2022-12-23 17:30:14,504 WARNING [train.py:1060] (2/4) Exclude cut with ID 4511-76322-0006-80011-0_sp0.9 from training. Duration: 24.411125
2022-12-23 17:30:38,649 WARNING [train.py:1060] (2/4) Exclude cut with ID 6758-72288-0033-108368-0_sp1.1 from training. Duration: 21.263625
2022-12-23 17:30:58,757 INFO [train.py:894] (2/4) Epoch 21, batch 3350, loss[loss=0.1742, simple_loss=0.2557, pruned_loss=0.04634, over 18566.00 frames. ], tot_loss[loss=0.1831, simple_loss=0.265, pruned_loss=0.05064, over 3714171.72 frames. ], batch size: 49, lr: 5.41e-03, grad_scale: 8.0
2022-12-23 17:31:00,538 INFO [scaling.py:679] (2/4) Whitening: num_groups=8, num_channels=96, metric=1.37 vs. limit=2.0
2022-12-23 17:31:13,928 WARNING [train.py:1060] (2/4) Exclude cut with ID 4234-40345-0022-142709-0 from training. Duration: 20.795
2022-12-23 17:31:22,524 WARNING [train.py:1060] (2/4) Exclude cut with ID 7357-94126-0021-15852-0 from training. Duration: 24.76
2022-12-23 17:31:22,539 WARNING [train.py:1060] (2/4) Exclude cut with ID 3867-173237-0077-144769-0_sp0.9 from training. Duration: 22.25
2022-12-23 17:31:47,978 WARNING [train.py:1060] (2/4) Exclude cut with ID 7357-94126-0026-15857-0_sp1.1 from training. Duration: 20.5045625
2022-12-23 17:32:17,085 INFO [train.py:894] (2/4) Epoch 21, batch 3400, loss[loss=0.1951, simple_loss=0.2785, pruned_loss=0.05584, over 18633.00 frames. ], tot_loss[loss=0.1823, simple_loss=0.2642, pruned_loss=0.05014, over 3714139.97 frames. ], batch size: 53, lr: 5.41e-03, grad_scale: 8.0
2022-12-23 17:32:43,248 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 2.477e+02 4.102e+02 5.173e+02 6.286e+02 1.561e+03, threshold=1.035e+03, percent-clipped=1.0
2022-12-23 17:32:56,694 INFO [zipformer.py:660] (2/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=73550.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 17:33:29,499 INFO [train.py:894] (2/4) Epoch 21, batch 3450, loss[loss=0.1741, simple_loss=0.2617, pruned_loss=0.04331, over 18463.00 frames. ], tot_loss[loss=0.1831, simple_loss=0.265, pruned_loss=0.05065, over 3714645.36 frames. ], batch size: 68, lr: 5.41e-03, grad_scale: 8.0
2022-12-23 17:33:53,940 INFO [zipformer.py:660] (2/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=73590.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 17:34:05,256 INFO [zipformer.py:660] (2/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=73598.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 17:34:14,406 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.9810, 1.7319, 2.2007, 1.2226, 2.0379, 2.0701, 1.4954, 2.3949],
       device='cuda:2'), covar=tensor([0.1062, 0.1911, 0.0995, 0.1704, 0.0814, 0.1086, 0.2365, 0.0537],
       device='cuda:2'), in_proj_covar=tensor([0.0197, 0.0211, 0.0207, 0.0194, 0.0176, 0.0218, 0.0215, 0.0193],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-23 17:34:37,731 INFO [zipformer.py:660] (2/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=73620.0, num_to_drop=1, layers_to_drop={1}
2022-12-23 17:34:42,262 INFO [train.py:894] (2/4) Epoch 21, batch 3500, loss[loss=0.2065, simple_loss=0.2789, pruned_loss=0.06703, over 18634.00 frames. ], tot_loss[loss=0.1836, simple_loss=0.2653, pruned_loss=0.05093, over 3714918.33 frames. ], batch size: 182, lr: 5.40e-03, grad_scale: 8.0
2022-12-23 17:35:03,766 WARNING [train.py:1060] (2/4) Exclude cut with ID 7859-102521-0017-7548-0_sp1.1 from training. Duration: 22.2954375
2022-12-23 17:35:14,810 INFO [train.py:894] (2/4) Epoch 22, batch 0, loss[loss=0.159, simple_loss=0.244, pruned_loss=0.03699, over 18524.00 frames. ], tot_loss[loss=0.159, simple_loss=0.244, pruned_loss=0.03699, over 18524.00 frames. ], batch size: 47, lr: 5.28e-03, grad_scale: 8.0
2022-12-23 17:35:14,811 INFO [train.py:919] (2/4) Computing validation loss
2022-12-23 17:35:26,081 INFO [train.py:928] (2/4) Epoch 22, validation: loss=0.1635, simple_loss=0.2611, pruned_loss=0.03298, over 944034.00 frames. 
2022-12-23 17:35:26,081 INFO [train.py:929] (2/4) Maximum memory allocated so far is 24680MB
2022-12-23 17:35:44,484 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 2.826e+02 4.412e+02 5.413e+02 6.975e+02 1.595e+03, threshold=1.083e+03, percent-clipped=3.0
2022-12-23 17:35:57,125 INFO [zipformer.py:660] (2/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=73649.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 17:36:00,369 INFO [zipformer.py:660] (2/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=73651.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 17:36:18,839 WARNING [train.py:1060] (2/4) Exclude cut with ID 298-126791-0067-24026-0_sp0.9 from training. Duration: 21.438875
2022-12-23 17:36:23,468 WARNING [train.py:1060] (2/4) Exclude cut with ID 5652-39938-0025-23684-0_sp0.9 from training. Duration: 22.2055625
2022-12-23 17:36:26,666 INFO [zipformer.py:660] (2/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=73668.0, num_to_drop=1, layers_to_drop={0}
2022-12-23 17:36:43,282 INFO [train.py:894] (2/4) Epoch 22, batch 50, loss[loss=0.1556, simple_loss=0.2427, pruned_loss=0.03425, over 18585.00 frames. ], tot_loss[loss=0.171, simple_loss=0.2593, pruned_loss=0.04141, over 837975.54 frames. ], batch size: 45, lr: 5.28e-03, grad_scale: 8.0
2022-12-23 17:37:33,361 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.2756, 1.0423, 1.4519, 2.2110, 1.4985, 2.2027, 0.7334, 1.6159],
       device='cuda:2'), covar=tensor([0.1944, 0.1779, 0.1294, 0.0744, 0.1211, 0.0931, 0.1850, 0.1240],
       device='cuda:2'), in_proj_covar=tensor([0.0102, 0.0117, 0.0135, 0.0148, 0.0106, 0.0140, 0.0129, 0.0112],
       device='cuda:2'), out_proj_covar=tensor([0.0003, 0.0003, 0.0004, 0.0004, 0.0003, 0.0004, 0.0003, 0.0003],
       device='cuda:2')
2022-12-23 17:37:59,107 INFO [train.py:894] (2/4) Epoch 22, batch 100, loss[loss=0.1676, simple_loss=0.255, pruned_loss=0.04007, over 18680.00 frames. ], tot_loss[loss=0.1734, simple_loss=0.261, pruned_loss=0.04293, over 1476217.02 frames. ], batch size: 48, lr: 5.27e-03, grad_scale: 8.0
2022-12-23 17:38:03,884 INFO [zipformer.py:660] (2/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=73732.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 17:38:16,524 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 2.153e+02 3.091e+02 3.725e+02 4.885e+02 1.063e+03, threshold=7.451e+02, percent-clipped=0.0
2022-12-23 17:39:15,704 INFO [train.py:894] (2/4) Epoch 22, batch 150, loss[loss=0.1927, simple_loss=0.2792, pruned_loss=0.05307, over 18704.00 frames. ], tot_loss[loss=0.1727, simple_loss=0.261, pruned_loss=0.04222, over 1971924.53 frames. ], batch size: 60, lr: 5.27e-03, grad_scale: 8.0
2022-12-23 17:39:24,323 WARNING [train.py:1060] (2/4) Exclude cut with ID 7859-102521-0017-7548-0 from training. Duration: 24.525
2022-12-23 17:39:37,158 INFO [zipformer.py:660] (2/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=73793.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 17:40:00,492 WARNING [train.py:1060] (2/4) Exclude cut with ID 3699-47246-0007-3408-0_sp0.9 from training. Duration: 20.26675
2022-12-23 17:40:14,087 WARNING [train.py:1060] (2/4) Exclude cut with ID 7859-102521-0017-7548-0_sp0.9 from training. Duration: 27.25
2022-12-23 17:40:32,535 INFO [train.py:894] (2/4) Epoch 22, batch 200, loss[loss=0.1484, simple_loss=0.2336, pruned_loss=0.03157, over 18449.00 frames. ], tot_loss[loss=0.1715, simple_loss=0.2598, pruned_loss=0.04163, over 2358639.69 frames. ], batch size: 48, lr: 5.27e-03, grad_scale: 8.0
2022-12-23 17:40:49,758 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 2.045e+02 3.217e+02 3.912e+02 4.713e+02 9.147e+02, threshold=7.824e+02, percent-clipped=3.0
2022-12-23 17:41:20,423 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.8508, 1.4555, 0.6473, 1.3666, 2.1171, 1.2304, 1.5810, 1.7142],
       device='cuda:2'), covar=tensor([0.1537, 0.1812, 0.2261, 0.1504, 0.1713, 0.1691, 0.1339, 0.1641],
       device='cuda:2'), in_proj_covar=tensor([0.0093, 0.0097, 0.0115, 0.0095, 0.0116, 0.0090, 0.0097, 0.0093],
       device='cuda:2'), out_proj_covar=tensor([0.0003, 0.0003, 0.0004, 0.0003, 0.0004, 0.0003, 0.0003, 0.0003],
       device='cuda:2')
2022-12-23 17:41:27,648 WARNING [train.py:1060] (2/4) Exclude cut with ID 6426-64292-0017-15984-0 from training. Duration: 21.68
2022-12-23 17:41:40,347 WARNING [train.py:1060] (2/4) Exclude cut with ID 4278-13270-0007-59342-0 from training. Duration: 21.6300625
2022-12-23 17:41:48,828 INFO [train.py:894] (2/4) Epoch 22, batch 250, loss[loss=0.1862, simple_loss=0.2759, pruned_loss=0.0483, over 18729.00 frames. ], tot_loss[loss=0.1713, simple_loss=0.2596, pruned_loss=0.04148, over 2659938.42 frames. ], batch size: 52, lr: 5.27e-03, grad_scale: 8.0
2022-12-23 17:42:05,114 WARNING [train.py:1060] (2/4) Exclude cut with ID 4278-13270-0007-59342-0_sp0.9 from training. Duration: 24.033375
2022-12-23 17:42:21,749 INFO [scaling.py:679] (2/4) Whitening: num_groups=8, num_channels=96, metric=1.28 vs. limit=2.0
2022-12-23 17:42:50,794 INFO [scaling.py:679] (2/4) Whitening: num_groups=8, num_channels=96, metric=1.19 vs. limit=2.0
2022-12-23 17:42:55,244 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.7428, 1.5500, 1.7396, 1.5706, 1.2025, 3.5798, 1.5930, 2.0885],
       device='cuda:2'), covar=tensor([0.3127, 0.2090, 0.1911, 0.2081, 0.1487, 0.0176, 0.1568, 0.0845],
       device='cuda:2'), in_proj_covar=tensor([0.0133, 0.0118, 0.0125, 0.0121, 0.0104, 0.0097, 0.0091, 0.0089],
       device='cuda:2'), out_proj_covar=tensor([0.0005, 0.0005, 0.0005, 0.0005, 0.0004, 0.0003, 0.0004, 0.0004],
       device='cuda:2')
2022-12-23 17:43:02,003 WARNING [train.py:1060] (2/4) Exclude cut with ID 4278-13270-0009-59344-0 from training. Duration: 22.905
2022-12-23 17:43:03,547 INFO [train.py:894] (2/4) Epoch 22, batch 300, loss[loss=0.1825, simple_loss=0.2769, pruned_loss=0.0441, over 18604.00 frames. ], tot_loss[loss=0.1705, simple_loss=0.2587, pruned_loss=0.04118, over 2893722.25 frames. ], batch size: 57, lr: 5.27e-03, grad_scale: 8.0
2022-12-23 17:43:03,622 WARNING [train.py:1060] (2/4) Exclude cut with ID 5622-44585-0006-90525-0_sp1.1 from training. Duration: 23.4318125
2022-12-23 17:43:22,881 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 2.285e+02 3.196e+02 4.010e+02 5.060e+02 9.017e+02, threshold=8.020e+02, percent-clipped=3.0
2022-12-23 17:43:30,107 INFO [zipformer.py:660] (2/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=73946.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 17:43:34,830 INFO [zipformer.py:660] (2/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=73949.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 17:43:35,038 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.5848, 1.8741, 1.5904, 2.2509, 2.3436, 1.6906, 1.5802, 1.3573],
       device='cuda:2'), covar=tensor([0.1746, 0.1696, 0.1506, 0.0923, 0.1087, 0.1045, 0.1994, 0.1499],
       device='cuda:2'), in_proj_covar=tensor([0.0244, 0.0223, 0.0214, 0.0197, 0.0256, 0.0195, 0.0223, 0.0198],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-23 17:43:43,513 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.9446, 1.3688, 0.9397, 1.4019, 2.1337, 1.3288, 1.6160, 1.6610],
       device='cuda:2'), covar=tensor([0.1490, 0.2007, 0.2179, 0.1532, 0.1840, 0.1857, 0.1479, 0.1763],
       device='cuda:2'), in_proj_covar=tensor([0.0094, 0.0098, 0.0115, 0.0096, 0.0116, 0.0091, 0.0098, 0.0093],
       device='cuda:2'), out_proj_covar=tensor([0.0003, 0.0003, 0.0004, 0.0003, 0.0004, 0.0003, 0.0003, 0.0003],
       device='cuda:2')
2022-12-23 17:44:20,094 INFO [train.py:894] (2/4) Epoch 22, batch 350, loss[loss=0.1689, simple_loss=0.2626, pruned_loss=0.03761, over 18670.00 frames. ], tot_loss[loss=0.1707, simple_loss=0.259, pruned_loss=0.04123, over 3076127.64 frames. ], batch size: 69, lr: 5.26e-03, grad_scale: 8.0
2022-12-23 17:44:48,067 INFO [zipformer.py:660] (2/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=73997.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 17:45:10,565 WARNING [train.py:1060] (2/4) Exclude cut with ID 4278-13270-0009-59344-0_sp1.1 from training. Duration: 20.82275
2022-12-23 17:45:11,862 WARNING [train.py:1060] (2/4) Exclude cut with ID 4278-13270-0009-59344-0_sp0.9 from training. Duration: 25.45
2022-12-23 17:45:15,228 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.4918, 1.9370, 1.5486, 2.2520, 2.4640, 1.6433, 1.5905, 1.3051],
       device='cuda:2'), covar=tensor([0.1890, 0.1698, 0.1564, 0.0964, 0.1197, 0.1081, 0.2057, 0.1539],
       device='cuda:2'), in_proj_covar=tensor([0.0245, 0.0224, 0.0214, 0.0198, 0.0256, 0.0195, 0.0223, 0.0199],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-23 17:45:40,510 INFO [train.py:894] (2/4) Epoch 22, batch 400, loss[loss=0.192, simple_loss=0.2855, pruned_loss=0.04924, over 18544.00 frames. ], tot_loss[loss=0.171, simple_loss=0.2594, pruned_loss=0.04133, over 3216988.60 frames. ], batch size: 55, lr: 5.26e-03, grad_scale: 8.0
2022-12-23 17:45:57,385 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 2.190e+02 3.712e+02 4.382e+02 4.971e+02 1.685e+03, threshold=8.765e+02, percent-clipped=4.0
2022-12-23 17:46:11,692 WARNING [train.py:1060] (2/4) Exclude cut with ID 5622-44585-0006-90525-0 from training. Duration: 25.775
2022-12-23 17:46:32,048 WARNING [train.py:1060] (2/4) Exclude cut with ID 2929-85685-0071-62375-0_sp0.9 from training. Duration: 22.25
2022-12-23 17:46:32,383 INFO [zipformer.py:660] (2/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=74064.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 17:46:52,551 INFO [scaling.py:679] (2/4) Whitening: num_groups=1, num_channels=384, metric=3.75 vs. limit=5.0
2022-12-23 17:46:53,853 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([0.0948, 0.8459, 1.0527, 0.6302, 0.5021, 1.0459, 1.0742, 1.0427],
       device='cuda:2'), covar=tensor([0.0724, 0.0340, 0.0334, 0.0377, 0.0430, 0.0532, 0.0264, 0.0606],
       device='cuda:2'), in_proj_covar=tensor([0.0149, 0.0171, 0.0127, 0.0141, 0.0150, 0.0143, 0.0165, 0.0173],
       device='cuda:2'), out_proj_covar=tensor([1.1311e-04, 1.3081e-04, 9.5502e-05, 1.0485e-04, 1.1187e-04, 1.0930e-04,
        1.2637e-04, 1.3150e-04], device='cuda:2')
2022-12-23 17:46:54,831 INFO [train.py:894] (2/4) Epoch 22, batch 450, loss[loss=0.2112, simple_loss=0.2983, pruned_loss=0.06212, over 18530.00 frames. ], tot_loss[loss=0.173, simple_loss=0.2616, pruned_loss=0.0422, over 3326708.32 frames. ], batch size: 98, lr: 5.26e-03, grad_scale: 8.0
2022-12-23 17:46:59,079 WARNING [train.py:1060] (2/4) Exclude cut with ID 3972-170212-0014-23379-0 from training. Duration: 26.205
2022-12-23 17:47:08,177 INFO [zipformer.py:660] (2/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=74088.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 17:47:15,337 WARNING [train.py:1060] (2/4) Exclude cut with ID 5239-32139-0047-9341-0_sp0.9 from training. Duration: 30.1555625
2022-12-23 17:47:20,948 WARNING [train.py:1060] (2/4) Exclude cut with ID 1265-135635-0050-6781-0_sp0.9 from training. Duration: 21.8333125
2022-12-23 17:47:30,960 WARNING [train.py:1060] (2/4) Exclude cut with ID 1914-133440-0024-94914-0_sp1.1 from training. Duration: 20.6545625
2022-12-23 17:48:04,664 INFO [zipformer.py:660] (2/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=74125.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 17:48:09,816 INFO [train.py:894] (2/4) Epoch 22, batch 500, loss[loss=0.1724, simple_loss=0.2629, pruned_loss=0.04094, over 18620.00 frames. ], tot_loss[loss=0.173, simple_loss=0.2618, pruned_loss=0.04212, over 3413004.34 frames. ], batch size: 53, lr: 5.26e-03, grad_scale: 8.0
2022-12-23 17:48:12,578 WARNING [train.py:1060] (2/4) Exclude cut with ID 7395-89880-0045-39920-0_sp0.9 from training. Duration: 20.52225
2022-12-23 17:48:27,461 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 2.598e+02 3.456e+02 4.144e+02 5.225e+02 1.116e+03, threshold=8.288e+02, percent-clipped=3.0
2022-12-23 17:48:31,335 INFO [scaling.py:679] (2/4) Whitening: num_groups=8, num_channels=192, metric=1.94 vs. limit=2.0
2022-12-23 17:48:31,651 WARNING [train.py:1060] (2/4) Exclude cut with ID 3972-170212-0014-23379-0_sp0.9 from training. Duration: 29.1166875
2022-12-23 17:49:25,520 INFO [train.py:894] (2/4) Epoch 22, batch 550, loss[loss=0.177, simple_loss=0.2646, pruned_loss=0.04465, over 18585.00 frames. ], tot_loss[loss=0.1736, simple_loss=0.2625, pruned_loss=0.04235, over 3480026.77 frames. ], batch size: 49, lr: 5.26e-03, grad_scale: 8.0
2022-12-23 17:49:33,224 WARNING [train.py:1060] (2/4) Exclude cut with ID 543-133211-0007-59831-0_sp0.9 from training. Duration: 21.388875
2022-12-23 17:50:08,144 WARNING [train.py:1060] (2/4) Exclude cut with ID 1914-133440-0024-94914-0 from training. Duration: 22.72
2022-12-23 17:50:09,641 WARNING [train.py:1060] (2/4) Exclude cut with ID 1914-133440-0031-94921-0_sp0.9 from training. Duration: 22.7444375
2022-12-23 17:50:42,992 INFO [train.py:894] (2/4) Epoch 22, batch 600, loss[loss=0.1715, simple_loss=0.2706, pruned_loss=0.03621, over 18543.00 frames. ], tot_loss[loss=0.1741, simple_loss=0.2634, pruned_loss=0.04242, over 3532954.38 frames. ], batch size: 55, lr: 5.26e-03, grad_scale: 8.0
2022-12-23 17:50:53,991 WARNING [train.py:1060] (2/4) Exclude cut with ID 4133-6541-0027-40495-0_sp1.1 from training. Duration: 0.9681875
2022-12-23 17:50:56,936 WARNING [train.py:1060] (2/4) Exclude cut with ID 6330-62851-0022-91297-0_sp0.9 from training. Duration: 22.3166875
2022-12-23 17:51:00,095 INFO [zipformer.py:660] (2/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=74240.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 17:51:01,176 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 2.128e+02 3.307e+02 4.178e+02 5.054e+02 1.095e+03, threshold=8.355e+02, percent-clipped=2.0
2022-12-23 17:51:02,653 WARNING [train.py:1060] (2/4) Exclude cut with ID 543-133212-0015-59917-0_sp0.9 from training. Duration: 21.8166875
2022-12-23 17:51:09,562 INFO [zipformer.py:660] (2/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=74246.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 17:51:59,078 INFO [train.py:894] (2/4) Epoch 22, batch 650, loss[loss=0.1714, simple_loss=0.265, pruned_loss=0.03893, over 18585.00 frames. ], tot_loss[loss=0.1738, simple_loss=0.2633, pruned_loss=0.0422, over 3573159.04 frames. ], batch size: 51, lr: 5.25e-03, grad_scale: 8.0
2022-12-23 17:52:23,006 INFO [zipformer.py:660] (2/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=74294.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 17:52:33,853 INFO [zipformer.py:660] (2/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=74301.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 17:52:46,684 WARNING [train.py:1060] (2/4) Exclude cut with ID 4957-30119-0041-23990-0_sp0.9 from training. Duration: 20.22775
2022-12-23 17:53:14,485 INFO [train.py:894] (2/4) Epoch 22, batch 700, loss[loss=0.1673, simple_loss=0.2606, pruned_loss=0.03695, over 18466.00 frames. ], tot_loss[loss=0.1737, simple_loss=0.2629, pruned_loss=0.04223, over 3603843.13 frames. ], batch size: 54, lr: 5.25e-03, grad_scale: 8.0
2022-12-23 17:53:31,850 WARNING [train.py:1060] (2/4) Exclude cut with ID 5239-32139-0047-9341-0_sp1.1 from training. Duration: 24.67275
2022-12-23 17:53:33,148 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 2.504e+02 3.513e+02 4.127e+02 5.298e+02 1.016e+03, threshold=8.254e+02, percent-clipped=2.0
2022-12-23 17:54:00,329 WARNING [train.py:1060] (2/4) Exclude cut with ID 3082-165428-0081-50734-0_sp0.9 from training. Duration: 21.8055625
2022-12-23 17:54:02,653 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.6497, 1.5480, 1.6782, 1.9146, 1.8161, 3.7217, 1.3920, 1.6643],
       device='cuda:2'), covar=tensor([0.0776, 0.1724, 0.0971, 0.0897, 0.1311, 0.0184, 0.1362, 0.1431],
       device='cuda:2'), in_proj_covar=tensor([0.0071, 0.0082, 0.0071, 0.0074, 0.0090, 0.0075, 0.0085, 0.0077],
       device='cuda:2'), out_proj_covar=tensor([0.0003, 0.0004, 0.0004, 0.0004, 0.0004, 0.0003, 0.0004, 0.0004],
       device='cuda:2')
2022-12-23 17:54:28,471 INFO [scaling.py:679] (2/4) Whitening: num_groups=1, num_channels=384, metric=2.91 vs. limit=5.0
2022-12-23 17:54:30,478 INFO [train.py:894] (2/4) Epoch 22, batch 750, loss[loss=0.1781, simple_loss=0.2735, pruned_loss=0.04136, over 18454.00 frames. ], tot_loss[loss=0.1736, simple_loss=0.2628, pruned_loss=0.04218, over 3628959.04 frames. ], batch size: 54, lr: 5.25e-03, grad_scale: 8.0
2022-12-23 17:54:37,784 WARNING [train.py:1060] (2/4) Exclude cut with ID 3340-169293-0054-76830-0_sp0.9 from training. Duration: 22.6666875
2022-12-23 17:54:44,521 INFO [zipformer.py:660] (2/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=74388.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 17:55:32,266 INFO [zipformer.py:660] (2/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=74420.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 17:55:42,083 WARNING [train.py:1060] (2/4) Exclude cut with ID 2411-132532-0017-82279-0_sp1.1 from training. Duration: 0.9681875
2022-12-23 17:55:44,838 INFO [train.py:894] (2/4) Epoch 22, batch 800, loss[loss=0.1858, simple_loss=0.2728, pruned_loss=0.04944, over 18382.00 frames. ], tot_loss[loss=0.1739, simple_loss=0.2633, pruned_loss=0.04224, over 3647341.57 frames. ], batch size: 51, lr: 5.25e-03, grad_scale: 8.0
2022-12-23 17:55:55,880 INFO [zipformer.py:660] (2/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=74436.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 17:56:02,822 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 2.088e+02 3.453e+02 4.133e+02 5.100e+02 8.129e+02, threshold=8.265e+02, percent-clipped=0.0
2022-12-23 17:56:06,280 WARNING [train.py:1060] (2/4) Exclude cut with ID 6330-62850-0007-91323-0 from training. Duration: 22.485
2022-12-23 17:56:08,084 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.7602, 1.3138, 0.7807, 1.3749, 2.0723, 1.0820, 1.4806, 1.5793],
       device='cuda:2'), covar=tensor([0.1556, 0.1910, 0.2278, 0.1468, 0.1828, 0.1937, 0.1432, 0.1701],
       device='cuda:2'), in_proj_covar=tensor([0.0094, 0.0097, 0.0116, 0.0096, 0.0117, 0.0091, 0.0098, 0.0093],
       device='cuda:2'), out_proj_covar=tensor([0.0003, 0.0003, 0.0004, 0.0003, 0.0004, 0.0003, 0.0003, 0.0003],
       device='cuda:2')
2022-12-23 17:56:43,761 WARNING [train.py:1060] (2/4) Exclude cut with ID 3972-170212-0014-23379-0_sp1.1 from training. Duration: 23.82275
2022-12-23 17:56:58,641 WARNING [train.py:1060] (2/4) Exclude cut with ID 4860-13185-0032-76709-0 from training. Duration: 20.77
2022-12-23 17:56:59,897 INFO [train.py:894] (2/4) Epoch 22, batch 850, loss[loss=0.1809, simple_loss=0.2708, pruned_loss=0.04551, over 18720.00 frames. ], tot_loss[loss=0.1737, simple_loss=0.2632, pruned_loss=0.0421, over 3661153.56 frames. ], batch size: 52, lr: 5.25e-03, grad_scale: 16.0
2022-12-23 17:57:05,405 WARNING [train.py:1060] (2/4) Exclude cut with ID 6426-64292-0017-15984-0_sp0.9 from training. Duration: 24.088875
2022-12-23 17:57:36,036 WARNING [train.py:1060] (2/4) Exclude cut with ID 6330-62850-0007-91323-0_sp1.1 from training. Duration: 20.4409375
2022-12-23 17:58:15,038 INFO [train.py:894] (2/4) Epoch 22, batch 900, loss[loss=0.1577, simple_loss=0.239, pruned_loss=0.03824, over 18575.00 frames. ], tot_loss[loss=0.1736, simple_loss=0.263, pruned_loss=0.04212, over 3672503.62 frames. ], batch size: 41, lr: 5.25e-03, grad_scale: 16.0
2022-12-23 17:58:33,730 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 2.249e+02 3.429e+02 4.111e+02 5.297e+02 1.524e+03, threshold=8.223e+02, percent-clipped=3.0
2022-12-23 17:58:54,592 WARNING [train.py:1060] (2/4) Exclude cut with ID 2929-85685-0062-62366-0_sp0.9 from training. Duration: 22.511125
2022-12-23 17:58:54,619 WARNING [train.py:1060] (2/4) Exclude cut with ID 7395-89880-0031-39906-0 from training. Duration: 20.675
2022-12-23 17:59:22,440 INFO [scaling.py:679] (2/4) Whitening: num_groups=1, num_channels=384, metric=4.89 vs. limit=5.0
2022-12-23 17:59:30,701 INFO [train.py:894] (2/4) Epoch 22, batch 950, loss[loss=0.1427, simple_loss=0.2266, pruned_loss=0.02941, over 18434.00 frames. ], tot_loss[loss=0.1737, simple_loss=0.2632, pruned_loss=0.04212, over 3680378.95 frames. ], batch size: 42, lr: 5.24e-03, grad_scale: 16.0
2022-12-23 17:59:56,778 INFO [zipformer.py:660] (2/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=74595.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 17:59:58,038 INFO [zipformer.py:660] (2/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=74596.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 18:00:31,567 WARNING [train.py:1060] (2/4) Exclude cut with ID 6330-62850-0007-91323-0_sp0.9 from training. Duration: 24.9833125
2022-12-23 18:00:48,655 INFO [train.py:894] (2/4) Epoch 22, batch 1000, loss[loss=0.1815, simple_loss=0.2808, pruned_loss=0.04112, over 18508.00 frames. ], tot_loss[loss=0.1735, simple_loss=0.2631, pruned_loss=0.04194, over 3688273.39 frames. ], batch size: 52, lr: 5.24e-03, grad_scale: 16.0
2022-12-23 18:01:03,972 WARNING [train.py:1060] (2/4) Exclude cut with ID 5239-32139-0047-9341-0 from training. Duration: 27.14
2022-12-23 18:01:05,938 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([3.8867, 3.2045, 2.8637, 1.2011, 2.7457, 2.7861, 2.3878, 2.8176],
       device='cuda:2'), covar=tensor([0.0597, 0.0630, 0.1575, 0.2036, 0.1589, 0.1191, 0.1502, 0.1038],
       device='cuda:2'), in_proj_covar=tensor([0.0173, 0.0186, 0.0207, 0.0192, 0.0211, 0.0201, 0.0217, 0.0201],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-23 18:01:06,918 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 2.184e+02 3.436e+02 4.038e+02 4.889e+02 8.968e+02, threshold=8.077e+02, percent-clipped=3.0
2022-12-23 18:01:19,186 WARNING [train.py:1060] (2/4) Exclude cut with ID 2929-85685-0044-62348-0 from training. Duration: 22.44
2022-12-23 18:01:30,462 INFO [zipformer.py:660] (2/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=74656.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 18:02:05,192 INFO [train.py:894] (2/4) Epoch 22, batch 1050, loss[loss=0.1616, simple_loss=0.2546, pruned_loss=0.03431, over 18586.00 frames. ], tot_loss[loss=0.1739, simple_loss=0.2635, pruned_loss=0.04214, over 3693139.93 frames. ], batch size: 56, lr: 5.24e-03, grad_scale: 16.0
2022-12-23 18:02:37,336 WARNING [train.py:1060] (2/4) Exclude cut with ID 2929-85685-0060-62364-0_sp0.9 from training. Duration: 21.361125
2022-12-23 18:02:43,463 WARNING [train.py:1060] (2/4) Exclude cut with ID 2929-85685-0079-62383-0_sp1.1 from training. Duration: 27.0318125
2022-12-23 18:02:53,575 WARNING [train.py:1060] (2/4) Exclude cut with ID 5622-44585-0006-90525-0_sp0.9 from training. Duration: 28.638875
2022-12-23 18:03:07,424 INFO [zipformer.py:660] (2/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=74720.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 18:03:08,748 WARNING [train.py:1060] (2/4) Exclude cut with ID 3340-169293-0054-76830-0 from training. Duration: 20.4
2022-12-23 18:03:20,743 INFO [train.py:894] (2/4) Epoch 22, batch 1100, loss[loss=0.1733, simple_loss=0.2656, pruned_loss=0.04048, over 18557.00 frames. ], tot_loss[loss=0.1743, simple_loss=0.2637, pruned_loss=0.04241, over 3697846.73 frames. ], batch size: 55, lr: 5.24e-03, grad_scale: 16.0
2022-12-23 18:03:21,118 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.8594, 1.6638, 1.9987, 2.2379, 1.9564, 4.4127, 1.5328, 1.6577],
       device='cuda:2'), covar=tensor([0.0794, 0.1731, 0.0914, 0.0922, 0.1394, 0.0167, 0.1423, 0.1547],
       device='cuda:2'), in_proj_covar=tensor([0.0072, 0.0083, 0.0071, 0.0075, 0.0091, 0.0076, 0.0086, 0.0078],
       device='cuda:2'), out_proj_covar=tensor([0.0003, 0.0004, 0.0004, 0.0004, 0.0004, 0.0003, 0.0004, 0.0004],
       device='cuda:2')
2022-12-23 18:03:39,050 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 1.975e+02 3.285e+02 3.886e+02 4.761e+02 1.198e+03, threshold=7.773e+02, percent-clipped=4.0
2022-12-23 18:03:41,273 WARNING [train.py:1060] (2/4) Exclude cut with ID 2929-85685-0071-62375-0 from training. Duration: 20.025
2022-12-23 18:03:41,282 WARNING [train.py:1060] (2/4) Exclude cut with ID 2364-131735-0112-64612-0_sp0.9 from training. Duration: 20.488875
2022-12-23 18:03:45,554 WARNING [train.py:1060] (2/4) Exclude cut with ID 2929-85685-0079-62383-0 from training. Duration: 29.735
2022-12-23 18:03:59,528 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([3.6965, 3.2512, 3.3267, 3.5013, 3.0937, 2.9982, 3.8584, 1.2926],
       device='cuda:2'), covar=tensor([0.1255, 0.1340, 0.1198, 0.1771, 0.2363, 0.2183, 0.1258, 0.6477],
       device='cuda:2'), in_proj_covar=tensor([0.0342, 0.0224, 0.0234, 0.0269, 0.0323, 0.0268, 0.0288, 0.0283],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0001, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-23 18:04:15,422 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([2.9178, 2.5067, 1.9980, 0.9783, 1.9848, 2.4267, 2.0381, 2.2957],
       device='cuda:2'), covar=tensor([0.0544, 0.0526, 0.1231, 0.1751, 0.1239, 0.1263, 0.1478, 0.0717],
       device='cuda:2'), in_proj_covar=tensor([0.0171, 0.0184, 0.0205, 0.0191, 0.0208, 0.0199, 0.0214, 0.0199],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-23 18:04:20,905 INFO [zipformer.py:660] (2/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=74768.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 18:04:37,413 INFO [train.py:894] (2/4) Epoch 22, batch 1150, loss[loss=0.1901, simple_loss=0.2829, pruned_loss=0.04871, over 18466.00 frames. ], tot_loss[loss=0.1747, simple_loss=0.2637, pruned_loss=0.0428, over 3701123.54 frames. ], batch size: 54, lr: 5.24e-03, grad_scale: 16.0
2022-12-23 18:04:58,754 INFO [scaling.py:679] (2/4) Whitening: num_groups=1, num_channels=384, metric=2.97 vs. limit=5.0
2022-12-23 18:05:08,143 WARNING [train.py:1060] (2/4) Exclude cut with ID 7276-92427-0014-12983-0_sp0.9 from training. Duration: 21.3055625
2022-12-23 18:05:09,602 WARNING [train.py:1060] (2/4) Exclude cut with ID 1025-75365-0008-79168-0_sp0.9 from training. Duration: 22.0666875
2022-12-23 18:05:53,485 INFO [train.py:894] (2/4) Epoch 22, batch 1200, loss[loss=0.1675, simple_loss=0.2584, pruned_loss=0.03833, over 18637.00 frames. ], tot_loss[loss=0.1748, simple_loss=0.2636, pruned_loss=0.04299, over 3704591.20 frames. ], batch size: 53, lr: 5.23e-03, grad_scale: 16.0
2022-12-23 18:06:10,684 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 2.171e+02 3.301e+02 3.924e+02 4.920e+02 7.003e+02, threshold=7.847e+02, percent-clipped=0.0
2022-12-23 18:06:56,755 WARNING [train.py:1060] (2/4) Exclude cut with ID 2929-85685-0062-62366-0 from training. Duration: 20.26
2022-12-23 18:07:07,533 INFO [train.py:894] (2/4) Epoch 22, batch 1250, loss[loss=0.2032, simple_loss=0.2933, pruned_loss=0.05656, over 18674.00 frames. ], tot_loss[loss=0.1744, simple_loss=0.2632, pruned_loss=0.04277, over 3706876.88 frames. ], batch size: 176, lr: 5.23e-03, grad_scale: 16.0
2022-12-23 18:07:10,303 WARNING [train.py:1060] (2/4) Exclude cut with ID 5239-32139-0030-9324-0_sp0.9 from training. Duration: 21.3444375
2022-12-23 18:07:26,476 INFO [scaling.py:679] (2/4) Whitening: num_groups=1, num_channels=384, metric=4.11 vs. limit=5.0
2022-12-23 18:07:28,897 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([2.9382, 3.6813, 3.3810, 2.0999, 3.6689, 2.8799, 1.4225, 2.6204],
       device='cuda:2'), covar=tensor([0.2584, 0.1262, 0.1453, 0.2919, 0.0948, 0.0906, 0.3899, 0.1513],
       device='cuda:2'), in_proj_covar=tensor([0.0146, 0.0142, 0.0158, 0.0125, 0.0146, 0.0115, 0.0144, 0.0115],
       device='cuda:2'), out_proj_covar=tensor([0.0003, 0.0002, 0.0003, 0.0002, 0.0003, 0.0002, 0.0003, 0.0002],
       device='cuda:2')
2022-12-23 18:07:33,770 INFO [zipformer.py:660] (2/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=74896.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 18:07:53,254 INFO [zipformer.py:660] (2/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=74909.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 18:08:06,608 WARNING [train.py:1060] (2/4) Exclude cut with ID 497-129325-0061-62254-0_sp1.1 from training. Duration: 0.97725
2022-12-23 18:08:22,525 INFO [train.py:894] (2/4) Epoch 22, batch 1300, loss[loss=0.1681, simple_loss=0.2428, pruned_loss=0.04674, over 18519.00 frames. ], tot_loss[loss=0.1735, simple_loss=0.2621, pruned_loss=0.04241, over 3707160.99 frames. ], batch size: 41, lr: 5.23e-03, grad_scale: 16.0
2022-12-23 18:08:40,173 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 2.220e+02 3.437e+02 4.231e+02 5.337e+02 9.980e+02, threshold=8.462e+02, percent-clipped=5.0
2022-12-23 18:08:44,637 INFO [zipformer.py:660] (2/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=74944.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 18:08:50,290 WARNING [train.py:1060] (2/4) Exclude cut with ID 7395-89880-0031-39906-0_sp0.9 from training. Duration: 22.97225
2022-12-23 18:08:55,008 INFO [zipformer.py:660] (2/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=74951.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 18:08:59,889 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([2.1955, 1.9294, 2.2797, 1.4246, 2.3839, 2.2546, 1.5866, 2.6062],
       device='cuda:2'), covar=tensor([0.1165, 0.1966, 0.1430, 0.2079, 0.0730, 0.1332, 0.2392, 0.0581],
       device='cuda:2'), in_proj_covar=tensor([0.0194, 0.0208, 0.0205, 0.0191, 0.0171, 0.0213, 0.0210, 0.0191],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-23 18:09:14,736 INFO [zipformer.py:660] (2/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=74964.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 18:09:22,553 WARNING [train.py:1060] (2/4) Exclude cut with ID 7395-89880-0047-39922-0_sp0.9 from training. Duration: 21.97775
2022-12-23 18:09:24,295 INFO [zipformer.py:660] (2/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=74970.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 18:09:37,115 INFO [train.py:894] (2/4) Epoch 22, batch 1350, loss[loss=0.1845, simple_loss=0.2755, pruned_loss=0.04673, over 18504.00 frames. ], tot_loss[loss=0.173, simple_loss=0.2617, pruned_loss=0.04212, over 3707946.00 frames. ], batch size: 52, lr: 5.23e-03, grad_scale: 16.0
2022-12-23 18:09:37,161 WARNING [train.py:1060] (2/4) Exclude cut with ID 1112-1043-0006-89194-0_sp0.9 from training. Duration: 21.8333125
2022-12-23 18:09:49,474 WARNING [train.py:1060] (2/4) Exclude cut with ID 1914-133440-0031-94921-0 from training. Duration: 20.47
2022-12-23 18:10:38,482 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.6366, 1.5587, 1.5872, 1.9527, 1.7252, 3.7543, 1.3400, 1.6355],
       device='cuda:2'), covar=tensor([0.0794, 0.1694, 0.1024, 0.0865, 0.1348, 0.0186, 0.1324, 0.1378],
       device='cuda:2'), in_proj_covar=tensor([0.0072, 0.0082, 0.0071, 0.0074, 0.0091, 0.0076, 0.0085, 0.0077],
       device='cuda:2'), out_proj_covar=tensor([0.0003, 0.0004, 0.0004, 0.0004, 0.0004, 0.0003, 0.0004, 0.0004],
       device='cuda:2')
2022-12-23 18:10:47,874 INFO [zipformer.py:660] (2/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=75025.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 18:10:53,383 INFO [train.py:894] (2/4) Epoch 22, batch 1400, loss[loss=0.2035, simple_loss=0.2812, pruned_loss=0.06285, over 18400.00 frames. ], tot_loss[loss=0.1726, simple_loss=0.2612, pruned_loss=0.04199, over 3709029.90 frames. ], batch size: 46, lr: 5.23e-03, grad_scale: 16.0
2022-12-23 18:10:53,428 WARNING [train.py:1060] (2/4) Exclude cut with ID 7395-89880-0037-39912-0_sp0.9 from training. Duration: 20.67225
2022-12-23 18:11:11,918 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 2.292e+02 3.487e+02 4.292e+02 5.613e+02 1.451e+03, threshold=8.583e+02, percent-clipped=4.0
2022-12-23 18:11:11,954 WARNING [train.py:1060] (2/4) Exclude cut with ID 1914-133440-0024-94914-0_sp0.9 from training. Duration: 25.2444375
2022-12-23 18:11:34,020 WARNING [train.py:1060] (2/4) Exclude cut with ID 3340-169293-0021-76797-0_sp0.9 from training. Duration: 21.1445
2022-12-23 18:12:07,222 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([2.2210, 2.4080, 1.9188, 1.6826, 2.4245, 2.8172, 2.4862, 2.0691],
       device='cuda:2'), covar=tensor([0.0419, 0.0256, 0.0426, 0.0280, 0.0240, 0.0311, 0.0390, 0.0310],
       device='cuda:2'), in_proj_covar=tensor([0.0093, 0.0125, 0.0151, 0.0123, 0.0114, 0.0119, 0.0097, 0.0126],
       device='cuda:2'), out_proj_covar=tensor([7.4396e-05, 9.9274e-05, 1.2435e-04, 9.7729e-05, 9.2236e-05, 9.1478e-05,
        7.6170e-05, 9.9560e-05], device='cuda:2')
2022-12-23 18:12:08,228 INFO [train.py:894] (2/4) Epoch 22, batch 1450, loss[loss=0.1709, simple_loss=0.2652, pruned_loss=0.03824, over 18570.00 frames. ], tot_loss[loss=0.1729, simple_loss=0.2614, pruned_loss=0.04221, over 3708963.53 frames. ], batch size: 69, lr: 5.23e-03, grad_scale: 16.0
2022-12-23 18:12:48,720 WARNING [train.py:1060] (2/4) Exclude cut with ID 2929-85685-0079-62383-0_sp0.9 from training. Duration: 33.038875
2022-12-23 18:13:25,048 INFO [train.py:894] (2/4) Epoch 22, batch 1500, loss[loss=0.16, simple_loss=0.2427, pruned_loss=0.03867, over 18690.00 frames. ], tot_loss[loss=0.174, simple_loss=0.2624, pruned_loss=0.04283, over 3710301.02 frames. ], batch size: 46, lr: 5.22e-03, grad_scale: 16.0
2022-12-23 18:13:25,114 WARNING [train.py:1060] (2/4) Exclude cut with ID 6426-64291-0000-16059-0_sp0.9 from training. Duration: 20.0944375
2022-12-23 18:13:39,578 WARNING [train.py:1060] (2/4) Exclude cut with ID 2929-85685-0044-62348-0_sp1.1 from training. Duration: 20.4
2022-12-23 18:13:42,477 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 2.040e+02 3.361e+02 3.944e+02 4.833e+02 7.905e+02, threshold=7.889e+02, percent-clipped=0.0
2022-12-23 18:13:48,159 WARNING [train.py:1060] (2/4) Exclude cut with ID 6330-62851-0022-91297-0 from training. Duration: 20.085
2022-12-23 18:14:00,966 WARNING [train.py:1060] (2/4) Exclude cut with ID 4860-13185-0032-76709-0_sp0.9 from training. Duration: 23.07775
2022-12-23 18:14:21,157 INFO [scaling.py:679] (2/4) Whitening: num_groups=1, num_channels=384, metric=4.30 vs. limit=5.0
2022-12-23 18:14:40,482 INFO [train.py:894] (2/4) Epoch 22, batch 1550, loss[loss=0.1902, simple_loss=0.2766, pruned_loss=0.05186, over 18720.00 frames. ], tot_loss[loss=0.1738, simple_loss=0.2621, pruned_loss=0.04272, over 3711829.72 frames. ], batch size: 54, lr: 5.22e-03, grad_scale: 16.0
2022-12-23 18:14:46,246 WARNING [train.py:1060] (2/4) Exclude cut with ID 2929-85685-0044-62348-0_sp0.9 from training. Duration: 24.9333125
2022-12-23 18:15:27,838 WARNING [train.py:1060] (2/4) Exclude cut with ID 5118-111612-0016-124680-0_sp0.9 from training. Duration: 20.388875
2022-12-23 18:15:32,148 WARNING [train.py:1060] (2/4) Exclude cut with ID 432-122774-0017-62487-0_sp1.1 from training. Duration: 20.3590625
2022-12-23 18:15:55,868 INFO [zipformer.py:660] (2/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=75228.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 18:15:56,903 INFO [train.py:894] (2/4) Epoch 22, batch 1600, loss[loss=0.1581, simple_loss=0.2352, pruned_loss=0.04049, over 18423.00 frames. ], tot_loss[loss=0.1743, simple_loss=0.2628, pruned_loss=0.04289, over 3712702.34 frames. ], batch size: 42, lr: 5.22e-03, grad_scale: 16.0
2022-12-23 18:16:14,678 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 2.171e+02 3.261e+02 3.774e+02 4.790e+02 7.917e+02, threshold=7.548e+02, percent-clipped=1.0
2022-12-23 18:16:30,220 INFO [zipformer.py:660] (2/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=75251.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 18:16:42,686 WARNING [train.py:1060] (2/4) Exclude cut with ID 3557-8342-0013-54691-0_sp1.1 from training. Duration: 0.836375
2022-12-23 18:16:51,731 INFO [zipformer.py:660] (2/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=75265.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 18:16:59,785 INFO [zipformer.py:660] (2/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=75270.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 18:17:13,153 INFO [train.py:894] (2/4) Epoch 22, batch 1650, loss[loss=0.1487, simple_loss=0.2289, pruned_loss=0.03425, over 18534.00 frames. ], tot_loss[loss=0.1743, simple_loss=0.2627, pruned_loss=0.04299, over 3713051.55 frames. ], batch size: 44, lr: 5.22e-03, grad_scale: 16.0
2022-12-23 18:17:25,141 WARNING [train.py:1060] (2/4) Exclude cut with ID 8565-290391-0049-67394-0_sp0.9 from training. Duration: 21.3166875
2022-12-23 18:17:28,475 INFO [zipformer.py:660] (2/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=75289.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 18:17:42,022 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([2.2085, 2.5800, 3.0421, 1.3219, 2.7648, 2.9212, 1.9759, 2.9849],
       device='cuda:2'), covar=tensor([0.1563, 0.1803, 0.1270, 0.2306, 0.0935, 0.1463, 0.2164, 0.0799],
       device='cuda:2'), in_proj_covar=tensor([0.0197, 0.0212, 0.0208, 0.0195, 0.0174, 0.0216, 0.0213, 0.0194],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-23 18:17:43,251 INFO [zipformer.py:660] (2/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=75299.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 18:17:55,566 WARNING [train.py:1060] (2/4) Exclude cut with ID 6533-399-0029-104863-0_sp0.9 from training. Duration: 22.1055625
2022-12-23 18:17:58,232 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([2.1858, 1.5578, 1.8400, 1.8690, 2.1227, 2.1597, 1.9546, 1.7964],
       device='cuda:2'), covar=tensor([0.2222, 0.3327, 0.2711, 0.3028, 0.2161, 0.1059, 0.3410, 0.1387],
       device='cuda:2'), in_proj_covar=tensor([0.0264, 0.0295, 0.0277, 0.0313, 0.0305, 0.0249, 0.0341, 0.0240],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-23 18:18:06,193 WARNING [train.py:1060] (2/4) Exclude cut with ID 7699-105389-0094-26379-0_sp1.1 from training. Duration: 21.77725
2022-12-23 18:18:15,421 INFO [zipformer.py:660] (2/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=75320.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 18:18:25,494 WARNING [train.py:1060] (2/4) Exclude cut with ID 7255-291500-0005-134304-0_sp0.9 from training. Duration: 27.8166875
2022-12-23 18:18:28,722 INFO [train.py:894] (2/4) Epoch 22, batch 1700, loss[loss=0.1606, simple_loss=0.241, pruned_loss=0.0401, over 18478.00 frames. ], tot_loss[loss=0.1741, simple_loss=0.2619, pruned_loss=0.04312, over 3712750.37 frames. ], batch size: 43, lr: 5.22e-03, grad_scale: 16.0
2022-12-23 18:18:32,230 INFO [zipformer.py:660] (2/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=75331.0, num_to_drop=1, layers_to_drop={0}
2022-12-23 18:18:46,420 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 2.494e+02 3.900e+02 4.543e+02 5.903e+02 1.187e+03, threshold=9.086e+02, percent-clipped=10.0
2022-12-23 18:18:51,124 WARNING [train.py:1060] (2/4) Exclude cut with ID 7357-94126-0021-15852-0_sp1.1 from training. Duration: 22.5090625
2022-12-23 18:18:55,708 WARNING [train.py:1060] (2/4) Exclude cut with ID 7255-291500-0005-134304-0 from training. Duration: 25.035
2022-12-23 18:19:14,657 WARNING [train.py:1060] (2/4) Exclude cut with ID 774-127930-0014-10412-0_sp1.1 from training. Duration: 0.95
2022-12-23 18:19:33,760 WARNING [train.py:1060] (2/4) Exclude cut with ID 3033-130750-0096-55598-0_sp0.9 from training. Duration: 0.92225
2022-12-23 18:19:44,169 INFO [train.py:894] (2/4) Epoch 22, batch 1750, loss[loss=0.1579, simple_loss=0.2462, pruned_loss=0.0348, over 18684.00 frames. ], tot_loss[loss=0.1765, simple_loss=0.2633, pruned_loss=0.04483, over 3713135.60 frames. ], batch size: 48, lr: 5.22e-03, grad_scale: 16.0
2022-12-23 18:20:00,900 WARNING [train.py:1060] (2/4) Exclude cut with ID 4511-76322-0006-80011-0 from training. Duration: 21.97
2022-12-23 18:20:07,091 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([2.6668, 2.2132, 1.7597, 2.4385, 1.9678, 2.1168, 2.0625, 2.5716],
       device='cuda:2'), covar=tensor([0.1896, 0.2963, 0.1902, 0.2535, 0.3519, 0.1073, 0.2843, 0.0898],
       device='cuda:2'), in_proj_covar=tensor([0.0292, 0.0291, 0.0245, 0.0345, 0.0270, 0.0227, 0.0286, 0.0215],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-23 18:20:19,429 WARNING [train.py:1060] (2/4) Exclude cut with ID 7492-105653-0055-62765-0_sp0.9 from training. Duration: 21.97225
2022-12-23 18:20:19,457 WARNING [train.py:1060] (2/4) Exclude cut with ID 453-131332-0000-47844-0_sp0.9 from training. Duration: 25.3333125
2022-12-23 18:20:30,600 WARNING [train.py:1060] (2/4) Exclude cut with ID 5172-29468-0015-19128-0_sp0.9 from training. Duration: 21.5055625
2022-12-23 18:20:39,861 WARNING [train.py:1060] (2/4) Exclude cut with ID 453-131332-0000-47844-0_sp1.1 from training. Duration: 20.72725
2022-12-23 18:21:00,858 INFO [train.py:894] (2/4) Epoch 22, batch 1800, loss[loss=0.1602, simple_loss=0.2425, pruned_loss=0.03895, over 18692.00 frames. ], tot_loss[loss=0.1793, simple_loss=0.2649, pruned_loss=0.04687, over 3714100.55 frames. ], batch size: 46, lr: 5.21e-03, grad_scale: 16.0
2022-12-23 18:21:10,332 WARNING [train.py:1060] (2/4) Exclude cut with ID 8631-249866-0030-130156-0_sp0.9 from training. Duration: 26.32775
2022-12-23 18:21:19,343 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 2.815e+02 4.099e+02 5.163e+02 6.501e+02 1.367e+03, threshold=1.033e+03, percent-clipped=5.0
2022-12-23 18:21:41,015 WARNING [train.py:1060] (2/4) Exclude cut with ID 3867-173237-0077-144769-0 from training. Duration: 20.025
2022-12-23 18:21:46,212 WARNING [train.py:1060] (2/4) Exclude cut with ID 6709-74022-0004-86860-0_sp1.1 from training. Duration: 0.9409375
2022-12-23 18:21:47,554 WARNING [train.py:1060] (2/4) Exclude cut with ID 4757-1811-0023-62229-0_sp0.9 from training. Duration: 21.37775
2022-12-23 18:22:07,014 WARNING [train.py:1060] (2/4) Exclude cut with ID 1250-135782-0004-25974-0_sp0.9 from training. Duration: 21.17225
2022-12-23 18:22:07,361 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([2.3326, 2.1924, 2.0212, 1.4097, 2.6511, 2.5452, 2.2547, 1.8136],
       device='cuda:2'), covar=tensor([0.0408, 0.0432, 0.0456, 0.0728, 0.0304, 0.0347, 0.0442, 0.0896],
       device='cuda:2'), in_proj_covar=tensor([0.0125, 0.0126, 0.0128, 0.0118, 0.0101, 0.0123, 0.0133, 0.0158],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0001, 0.0001, 0.0001, 0.0002, 0.0002],
       device='cuda:2')
2022-12-23 18:22:08,350 WARNING [train.py:1060] (2/4) Exclude cut with ID 7357-94126-0021-15852-0_sp0.9 from training. Duration: 27.511125
2022-12-23 18:22:17,418 INFO [train.py:894] (2/4) Epoch 22, batch 1850, loss[loss=0.1628, simple_loss=0.2476, pruned_loss=0.03904, over 18388.00 frames. ], tot_loss[loss=0.1805, simple_loss=0.2652, pruned_loss=0.04787, over 3712641.44 frames. ], batch size: 46, lr: 5.21e-03, grad_scale: 16.0
2022-12-23 18:22:40,782 WARNING [train.py:1060] (2/4) Exclude cut with ID 453-131332-0000-47844-0 from training. Duration: 22.8
2022-12-23 18:22:45,987 WARNING [train.py:1060] (2/4) Exclude cut with ID 4964-30587-0040-44509-0 from training. Duration: 22.585
2022-12-23 18:23:17,374 WARNING [train.py:1060] (2/4) Exclude cut with ID 6978-92210-0001-146967-0_sp0.9 from training. Duration: 22.0166875
2022-12-23 18:23:32,327 INFO [train.py:894] (2/4) Epoch 22, batch 1900, loss[loss=0.2007, simple_loss=0.2772, pruned_loss=0.06208, over 18453.00 frames. ], tot_loss[loss=0.1822, simple_loss=0.2663, pruned_loss=0.04908, over 3713123.45 frames. ], batch size: 68, lr: 5.21e-03, grad_scale: 16.0
2022-12-23 18:23:33,773 WARNING [train.py:1060] (2/4) Exclude cut with ID 7255-291500-0003-134302-0_sp1.1 from training. Duration: 24.395375
2022-12-23 18:23:37,183 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.8310, 1.4115, 0.7155, 1.4577, 2.2291, 1.2643, 1.5611, 1.8773],
       device='cuda:2'), covar=tensor([0.1548, 0.1918, 0.2348, 0.1547, 0.1628, 0.1703, 0.1433, 0.1496],
       device='cuda:2'), in_proj_covar=tensor([0.0093, 0.0097, 0.0115, 0.0096, 0.0117, 0.0091, 0.0097, 0.0092],
       device='cuda:2'), out_proj_covar=tensor([0.0003, 0.0003, 0.0004, 0.0003, 0.0004, 0.0003, 0.0003, 0.0003],
       device='cuda:2')
2022-12-23 18:23:41,103 WARNING [train.py:1060] (2/4) Exclude cut with ID 3488-85273-0017-41203-0_sp0.9 from training. Duration: 27.47775
2022-12-23 18:23:45,181 WARNING [train.py:1060] (2/4) Exclude cut with ID 432-122774-0017-62487-0_sp0.9 from training. Duration: 24.8833125
2022-12-23 18:23:47,918 WARNING [train.py:1060] (2/4) Exclude cut with ID 6758-72288-0033-108368-0 from training. Duration: 23.39
2022-12-23 18:23:49,338 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 2.656e+02 4.293e+02 5.268e+02 6.591e+02 1.165e+03, threshold=1.054e+03, percent-clipped=3.0
2022-12-23 18:23:53,642 WARNING [train.py:1060] (2/4) Exclude cut with ID 8291-282929-0007-12994-0_sp0.9 from training. Duration: 28.72225
2022-12-23 18:24:03,156 WARNING [train.py:1060] (2/4) Exclude cut with ID 585-294811-0110-133686-0_sp0.9 from training. Duration: 20.8944375
2022-12-23 18:24:20,418 WARNING [train.py:1060] (2/4) Exclude cut with ID 5796-66357-0007-116447-0_sp0.9 from training. Duration: 23.8444375
2022-12-23 18:24:26,542 INFO [zipformer.py:660] (2/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=75565.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 18:24:43,071 WARNING [train.py:1060] (2/4) Exclude cut with ID 8291-282929-0007-12994-0 from training. Duration: 25.85
2022-12-23 18:24:43,079 WARNING [train.py:1060] (2/4) Exclude cut with ID 8291-282929-0023-13010-0 from training. Duration: 21.39
2022-12-23 18:24:47,517 INFO [train.py:894] (2/4) Epoch 22, batch 1950, loss[loss=0.1654, simple_loss=0.2441, pruned_loss=0.0433, over 18542.00 frames. ], tot_loss[loss=0.1833, simple_loss=0.2667, pruned_loss=0.04999, over 3712873.17 frames. ], batch size: 44, lr: 5.21e-03, grad_scale: 16.0
2022-12-23 18:24:53,731 WARNING [train.py:1060] (2/4) Exclude cut with ID 7357-94126-0014-15845-0 from training. Duration: 27.92
2022-12-23 18:24:55,329 INFO [zipformer.py:660] (2/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=75584.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 18:25:23,646 WARNING [train.py:1060] (2/4) Exclude cut with ID 8631-249866-0039-130165-0_sp0.9 from training. Duration: 20.661125
2022-12-23 18:25:39,998 INFO [zipformer.py:660] (2/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=75613.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 18:25:41,655 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([3.2668, 4.1696, 4.0164, 2.7318, 4.2620, 3.3307, 1.7903, 3.0701],
       device='cuda:2'), covar=tensor([0.2070, 0.1204, 0.1317, 0.2465, 0.0893, 0.0776, 0.3686, 0.1257],
       device='cuda:2'), in_proj_covar=tensor([0.0148, 0.0143, 0.0160, 0.0126, 0.0147, 0.0115, 0.0146, 0.0116],
       device='cuda:2'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0002, 0.0003, 0.0002, 0.0003, 0.0002],
       device='cuda:2')
2022-12-23 18:25:45,888 WARNING [train.py:1060] (2/4) Exclude cut with ID 7357-94126-0043-15874-0_sp0.9 from training. Duration: 20.07225
2022-12-23 18:25:50,801 INFO [zipformer.py:660] (2/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=75620.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 18:25:53,506 WARNING [train.py:1060] (2/4) Exclude cut with ID 1085-156170-0017-128270-0 from training. Duration: 21.01
2022-12-23 18:26:00,012 INFO [zipformer.py:660] (2/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=75626.0, num_to_drop=1, layers_to_drop={3}
2022-12-23 18:26:04,785 INFO [train.py:894] (2/4) Epoch 22, batch 2000, loss[loss=0.1639, simple_loss=0.244, pruned_loss=0.04195, over 18595.00 frames. ], tot_loss[loss=0.184, simple_loss=0.2666, pruned_loss=0.05073, over 3713565.41 frames. ], batch size: 51, lr: 5.21e-03, grad_scale: 16.0
2022-12-23 18:26:14,185 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.7817, 1.8220, 2.0360, 1.2239, 2.0406, 2.0305, 1.5967, 2.3160],
       device='cuda:2'), covar=tensor([0.1131, 0.1891, 0.1269, 0.1892, 0.0739, 0.1135, 0.2308, 0.0569],
       device='cuda:2'), in_proj_covar=tensor([0.0197, 0.0211, 0.0208, 0.0196, 0.0174, 0.0216, 0.0213, 0.0193],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-23 18:26:15,955 INFO [scaling.py:679] (2/4) Whitening: num_groups=8, num_channels=96, metric=1.17 vs. limit=2.0
2022-12-23 18:26:23,205 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 3.177e+02 4.476e+02 5.612e+02 7.026e+02 1.347e+03, threshold=1.122e+03, percent-clipped=1.0
2022-12-23 18:26:53,985 INFO [scaling.py:679] (2/4) Whitening: num_groups=8, num_channels=96, metric=1.18 vs. limit=2.0
2022-12-23 18:27:03,394 WARNING [train.py:1060] (2/4) Exclude cut with ID 2195-150901-0045-59933-0 from training. Duration: 20.65
2022-12-23 18:27:04,173 INFO [zipformer.py:660] (2/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=75668.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 18:27:11,542 WARNING [train.py:1060] (2/4) Exclude cut with ID 5796-66357-0007-116447-0 from training. Duration: 21.46
2022-12-23 18:27:20,871 INFO [train.py:894] (2/4) Epoch 22, batch 2050, loss[loss=0.1858, simple_loss=0.2785, pruned_loss=0.04657, over 18466.00 frames. ], tot_loss[loss=0.1844, simple_loss=0.2667, pruned_loss=0.05102, over 3713696.69 frames. ], batch size: 64, lr: 5.21e-03, grad_scale: 16.0
2022-12-23 18:27:27,014 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.8236, 1.7683, 1.8536, 1.7802, 1.3684, 3.8989, 1.7652, 2.3675],
       device='cuda:2'), covar=tensor([0.2945, 0.1908, 0.1786, 0.1894, 0.1381, 0.0167, 0.1477, 0.0733],
       device='cuda:2'), in_proj_covar=tensor([0.0133, 0.0118, 0.0126, 0.0122, 0.0105, 0.0097, 0.0091, 0.0090],
       device='cuda:2'), out_proj_covar=tensor([0.0005, 0.0005, 0.0005, 0.0005, 0.0004, 0.0003, 0.0004, 0.0004],
       device='cuda:2')
2022-12-23 18:27:28,821 INFO [scaling.py:679] (2/4) Whitening: num_groups=8, num_channels=96, metric=1.12 vs. limit=2.0
2022-12-23 18:27:56,650 WARNING [train.py:1060] (2/4) Exclude cut with ID 3557-8342-0013-54691-0 from training. Duration: 0.92
2022-12-23 18:28:03,741 WARNING [train.py:1060] (2/4) Exclude cut with ID 8291-282929-0023-13010-0_sp0.9 from training. Duration: 23.7666875
2022-12-23 18:28:34,389 INFO [train.py:894] (2/4) Epoch 22, batch 2100, loss[loss=0.1846, simple_loss=0.2639, pruned_loss=0.05263, over 18376.00 frames. ], tot_loss[loss=0.1848, simple_loss=0.2666, pruned_loss=0.05151, over 3713694.22 frames. ], batch size: 46, lr: 5.20e-03, grad_scale: 16.0
2022-12-23 18:28:40,252 WARNING [train.py:1060] (2/4) Exclude cut with ID 8544-281189-0060-101339-0_sp0.9 from training. Duration: 20.861125
2022-12-23 18:28:53,128 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 2.853e+02 4.059e+02 5.294e+02 7.011e+02 1.474e+03, threshold=1.059e+03, percent-clipped=5.0
2022-12-23 18:28:53,175 WARNING [train.py:1060] (2/4) Exclude cut with ID 3488-65654-0031-41259-0_sp0.9 from training. Duration: 22.711125
2022-12-23 18:28:59,590 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.8119, 1.8703, 1.6009, 1.6690, 2.0039, 2.0409, 2.0195, 1.3982],
       device='cuda:2'), covar=tensor([0.0326, 0.0248, 0.0447, 0.0209, 0.0189, 0.0338, 0.0287, 0.0312],
       device='cuda:2'), in_proj_covar=tensor([0.0094, 0.0127, 0.0153, 0.0124, 0.0116, 0.0121, 0.0098, 0.0127],
       device='cuda:2'), out_proj_covar=tensor([7.4724e-05, 1.0044e-04, 1.2563e-04, 9.8832e-05, 9.3765e-05, 9.3314e-05,
        7.6560e-05, 1.0054e-04], device='cuda:2')
2022-12-23 18:29:34,898 WARNING [train.py:1060] (2/4) Exclude cut with ID 6951-79737-0043-132310-0_sp1.1 from training. Duration: 22.986375
2022-12-23 18:29:50,876 INFO [train.py:894] (2/4) Epoch 22, batch 2150, loss[loss=0.1745, simple_loss=0.2701, pruned_loss=0.03945, over 18682.00 frames. ], tot_loss[loss=0.1842, simple_loss=0.2662, pruned_loss=0.05109, over 3714493.54 frames. ], batch size: 60, lr: 5.20e-03, grad_scale: 8.0
2022-12-23 18:29:52,304 WARNING [train.py:1060] (2/4) Exclude cut with ID 8040-260924-0003-80960-0_sp0.9 from training. Duration: 22.07225
2022-12-23 18:29:56,493 WARNING [train.py:1060] (2/4) Exclude cut with ID 7699-105389-0045-26330-0_sp0.9 from training. Duration: 20.3055625
2022-12-23 18:29:57,820 WARNING [train.py:1060] (2/4) Exclude cut with ID 6356-271890-0060-94317-0_sp0.9 from training. Duration: 20.72225
2022-12-23 18:30:18,096 WARNING [train.py:1060] (2/4) Exclude cut with ID 3488-85273-0017-41203-0_sp1.1 from training. Duration: 22.4818125
2022-12-23 18:30:45,459 WARNING [train.py:1060] (2/4) Exclude cut with ID 4964-30587-0040-44509-0_sp0.9 from training. Duration: 25.0944375
2022-12-23 18:30:49,988 WARNING [train.py:1060] (2/4) Exclude cut with ID 6533-399-0047-104881-0 from training. Duration: 21.515
2022-12-23 18:30:56,328 WARNING [train.py:1060] (2/4) Exclude cut with ID 7357-94126-0009-15840-0_sp0.9 from training. Duration: 27.02225
2022-12-23 18:31:00,779 WARNING [train.py:1060] (2/4) Exclude cut with ID 432-122774-0010-62480-0_sp0.9 from training. Duration: 22.22225
2022-12-23 18:31:06,948 INFO [train.py:894] (2/4) Epoch 22, batch 2200, loss[loss=0.2138, simple_loss=0.2902, pruned_loss=0.06869, over 18620.00 frames. ], tot_loss[loss=0.1847, simple_loss=0.2666, pruned_loss=0.0514, over 3714872.74 frames. ], batch size: 53, lr: 5.20e-03, grad_scale: 8.0
2022-12-23 18:31:08,382 WARNING [train.py:1060] (2/4) Exclude cut with ID 4964-30587-0085-44554-0_sp0.9 from training. Duration: 20.85
2022-12-23 18:31:27,606 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 2.539e+02 4.148e+02 5.070e+02 6.126e+02 9.714e+02, threshold=1.014e+03, percent-clipped=0.0
2022-12-23 18:31:34,106 INFO [scaling.py:679] (2/4) Whitening: num_groups=1, num_channels=384, metric=2.98 vs. limit=5.0
2022-12-23 18:31:41,015 WARNING [train.py:1060] (2/4) Exclude cut with ID 4295-39940-0007-92567-0 from training. Duration: 21.54
2022-12-23 18:31:44,193 WARNING [train.py:1060] (2/4) Exclude cut with ID 4964-30587-0040-44509-0_sp1.1 from training. Duration: 20.5318125
2022-12-23 18:31:54,935 WARNING [train.py:1060] (2/4) Exclude cut with ID 7255-291500-0012-134311-0_sp0.9 from training. Duration: 21.9333125
2022-12-23 18:32:07,868 INFO [scaling.py:679] (2/4) Whitening: num_groups=8, num_channels=96, metric=1.28 vs. limit=2.0
2022-12-23 18:32:22,614 INFO [train.py:894] (2/4) Epoch 22, batch 2250, loss[loss=0.1785, simple_loss=0.26, pruned_loss=0.04852, over 18524.00 frames. ], tot_loss[loss=0.1865, simple_loss=0.268, pruned_loss=0.05246, over 3714598.51 frames. ], batch size: 47, lr: 5.20e-03, grad_scale: 8.0
2022-12-23 18:32:31,526 INFO [zipformer.py:660] (2/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=75884.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 18:32:41,787 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([2.4423, 3.3174, 3.2087, 1.2578, 3.4903, 2.4369, 0.5361, 2.2439],
       device='cuda:2'), covar=tensor([0.2150, 0.1352, 0.1678, 0.3909, 0.1014, 0.1082, 0.5081, 0.1525],
       device='cuda:2'), in_proj_covar=tensor([0.0150, 0.0145, 0.0161, 0.0127, 0.0147, 0.0116, 0.0147, 0.0117],
       device='cuda:2'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0002, 0.0003, 0.0002, 0.0003, 0.0002],
       device='cuda:2')
2022-12-23 18:32:46,274 WARNING [train.py:1060] (2/4) Exclude cut with ID 8631-249866-0025-130151-0_sp0.9 from training. Duration: 21.7944375
2022-12-23 18:32:58,581 WARNING [train.py:1060] (2/4) Exclude cut with ID 8291-282929-0002-12989-0_sp0.9 from training. Duration: 22.4666875
2022-12-23 18:33:06,219 WARNING [train.py:1060] (2/4) Exclude cut with ID 6121-9014-0076-24124-0 from training. Duration: 21.635
2022-12-23 18:33:12,529 WARNING [train.py:1060] (2/4) Exclude cut with ID 6121-9014-0076-24124-0_sp0.9 from training. Duration: 24.038875
2022-12-23 18:33:36,569 INFO [zipformer.py:660] (2/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=75926.0, num_to_drop=1, layers_to_drop={1}
2022-12-23 18:33:41,668 INFO [train.py:894] (2/4) Epoch 22, batch 2300, loss[loss=0.1428, simple_loss=0.2244, pruned_loss=0.03062, over 18534.00 frames. ], tot_loss[loss=0.1851, simple_loss=0.2667, pruned_loss=0.05176, over 3715192.29 frames. ], batch size: 41, lr: 5.20e-03, grad_scale: 8.0
2022-12-23 18:33:46,488 INFO [zipformer.py:660] (2/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=75932.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 18:33:55,210 WARNING [train.py:1060] (2/4) Exclude cut with ID 7255-291500-0009-134308-0_sp1.1 from training. Duration: 21.786375
2022-12-23 18:34:01,092 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 2.457e+02 4.223e+02 4.994e+02 5.710e+02 1.631e+03, threshold=9.988e+02, percent-clipped=1.0
2022-12-23 18:34:05,487 WARNING [train.py:1060] (2/4) Exclude cut with ID 8291-282929-0002-12989-0 from training. Duration: 20.22
2022-12-23 18:34:50,788 INFO [zipformer.py:660] (2/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=75974.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 18:34:57,508 INFO [train.py:894] (2/4) Epoch 22, batch 2350, loss[loss=0.1675, simple_loss=0.2403, pruned_loss=0.04732, over 18684.00 frames. ], tot_loss[loss=0.1847, simple_loss=0.2663, pruned_loss=0.05154, over 3715391.93 frames. ], batch size: 46, lr: 5.20e-03, grad_scale: 8.0
2022-12-23 18:35:54,372 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([2.0721, 1.6533, 2.4695, 3.9874, 2.9623, 3.0368, 1.3362, 2.9729],
       device='cuda:2'), covar=tensor([0.1691, 0.1503, 0.1344, 0.0531, 0.0858, 0.1220, 0.1991, 0.0872],
       device='cuda:2'), in_proj_covar=tensor([0.0103, 0.0116, 0.0135, 0.0150, 0.0105, 0.0140, 0.0129, 0.0113],
       device='cuda:2'), out_proj_covar=tensor([0.0003, 0.0003, 0.0004, 0.0004, 0.0003, 0.0004, 0.0003, 0.0003],
       device='cuda:2')
2022-12-23 18:36:10,213 WARNING [train.py:1060] (2/4) Exclude cut with ID 6951-79737-0043-132310-0 from training. Duration: 25.285
2022-12-23 18:36:17,371 INFO [train.py:894] (2/4) Epoch 22, batch 2400, loss[loss=0.1694, simple_loss=0.2424, pruned_loss=0.04824, over 18382.00 frames. ], tot_loss[loss=0.1842, simple_loss=0.2659, pruned_loss=0.05121, over 3715896.89 frames. ], batch size: 42, lr: 5.19e-03, grad_scale: 8.0
2022-12-23 18:36:36,128 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 2.436e+02 4.375e+02 5.426e+02 6.772e+02 1.323e+03, threshold=1.085e+03, percent-clipped=1.0
2022-12-23 18:37:13,527 WARNING [train.py:1060] (2/4) Exclude cut with ID 811-130148-0001-63453-0_sp0.9 from training. Duration: 20.861125
2022-12-23 18:37:32,744 INFO [train.py:894] (2/4) Epoch 22, batch 2450, loss[loss=0.1948, simple_loss=0.2668, pruned_loss=0.06142, over 18663.00 frames. ], tot_loss[loss=0.1853, simple_loss=0.267, pruned_loss=0.05178, over 3715026.67 frames. ], batch size: 46, lr: 5.19e-03, grad_scale: 8.0
2022-12-23 18:37:37,535 WARNING [train.py:1060] (2/4) Exclude cut with ID 6010-56788-0055-90261-0 from training. Duration: 20.88
2022-12-23 18:38:10,759 WARNING [train.py:1060] (2/4) Exclude cut with ID 7357-94126-0045-15876-0_sp0.9 from training. Duration: 23.4166875
2022-12-23 18:38:24,489 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.8076, 1.8489, 1.5786, 1.7115, 1.8526, 2.1067, 2.0892, 1.4432],
       device='cuda:2'), covar=tensor([0.0325, 0.0240, 0.0473, 0.0214, 0.0213, 0.0321, 0.0251, 0.0312],
       device='cuda:2'), in_proj_covar=tensor([0.0095, 0.0127, 0.0153, 0.0124, 0.0117, 0.0122, 0.0099, 0.0128],
       device='cuda:2'), out_proj_covar=tensor([7.5703e-05, 1.0083e-04, 1.2577e-04, 9.8881e-05, 9.4565e-05, 9.3768e-05,
        7.7595e-05, 1.0105e-04], device='cuda:2')
2022-12-23 18:38:47,654 INFO [train.py:894] (2/4) Epoch 22, batch 2500, loss[loss=0.2056, simple_loss=0.2868, pruned_loss=0.06227, over 18587.00 frames. ], tot_loss[loss=0.1844, simple_loss=0.2662, pruned_loss=0.05125, over 3714100.33 frames. ], batch size: 99, lr: 5.19e-03, grad_scale: 8.0
2022-12-23 18:39:05,874 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 2.797e+02 4.308e+02 5.296e+02 6.059e+02 1.304e+03, threshold=1.059e+03, percent-clipped=2.0
2022-12-23 18:39:30,992 WARNING [train.py:1060] (2/4) Exclude cut with ID 7255-291500-0006-134305-0 from training. Duration: 21.24
2022-12-23 18:39:31,009 WARNING [train.py:1060] (2/4) Exclude cut with ID 6533-399-0047-104881-0_sp0.9 from training. Duration: 23.9055625
2022-12-23 18:39:52,094 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([3.0533, 2.5854, 1.9909, 3.0494, 2.2862, 2.4050, 2.4264, 3.2100],
       device='cuda:2'), covar=tensor([0.1897, 0.2960, 0.1815, 0.2679, 0.3565, 0.1030, 0.3062, 0.0822],
       device='cuda:2'), in_proj_covar=tensor([0.0295, 0.0292, 0.0245, 0.0349, 0.0272, 0.0229, 0.0288, 0.0215],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0003, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-23 18:39:55,214 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.5970, 1.8860, 1.5330, 2.1944, 2.5066, 1.6909, 1.5449, 1.3670],
       device='cuda:2'), covar=tensor([0.2000, 0.1754, 0.1701, 0.1035, 0.1146, 0.1123, 0.2102, 0.1611],
       device='cuda:2'), in_proj_covar=tensor([0.0246, 0.0225, 0.0215, 0.0197, 0.0258, 0.0195, 0.0224, 0.0200],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-23 18:39:58,954 INFO [scaling.py:679] (2/4) Whitening: num_groups=1, num_channels=384, metric=4.79 vs. limit=5.0
2022-12-23 18:40:02,975 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([0.5740, 1.9042, 2.1722, 1.1492, 1.3870, 2.3387, 2.0668, 1.7605],
       device='cuda:2'), covar=tensor([0.0810, 0.0355, 0.0302, 0.0405, 0.0400, 0.0460, 0.0261, 0.0672],
       device='cuda:2'), in_proj_covar=tensor([0.0150, 0.0171, 0.0128, 0.0140, 0.0148, 0.0143, 0.0166, 0.0175],
       device='cuda:2'), out_proj_covar=tensor([1.1421e-04, 1.3022e-04, 9.5306e-05, 1.0370e-04, 1.1035e-04, 1.0892e-04,
        1.2680e-04, 1.3287e-04], device='cuda:2')
2022-12-23 18:40:03,982 INFO [train.py:894] (2/4) Epoch 22, batch 2550, loss[loss=0.1731, simple_loss=0.2568, pruned_loss=0.04472, over 18589.00 frames. ], tot_loss[loss=0.1833, simple_loss=0.2651, pruned_loss=0.05072, over 3713170.68 frames. ], batch size: 51, lr: 5.19e-03, grad_scale: 8.0
2022-12-23 18:40:04,019 WARNING [train.py:1060] (2/4) Exclude cut with ID 6758-72288-0033-108368-0_sp0.9 from training. Duration: 25.988875
2022-12-23 18:40:12,673 WARNING [train.py:1060] (2/4) Exclude cut with ID 7255-291500-0001-134300-0_sp0.9 from training. Duration: 20.67225
2022-12-23 18:40:36,959 INFO [scaling.py:679] (2/4) Whitening: num_groups=8, num_channels=96, metric=1.22 vs. limit=2.0
2022-12-23 18:40:53,287 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([3.0490, 5.5622, 5.0943, 2.7980, 5.5700, 4.1598, 1.2461, 3.8830],
       device='cuda:2'), covar=tensor([0.1878, 0.0884, 0.1195, 0.2930, 0.0653, 0.0723, 0.4573, 0.1100],
       device='cuda:2'), in_proj_covar=tensor([0.0147, 0.0143, 0.0158, 0.0124, 0.0145, 0.0114, 0.0144, 0.0115],
       device='cuda:2'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0002, 0.0003, 0.0002, 0.0003, 0.0002],
       device='cuda:2')
2022-12-23 18:41:00,515 WARNING [train.py:1060] (2/4) Exclude cut with ID 3488-85273-0038-41224-0 from training. Duration: 20.34
2022-12-23 18:41:19,987 INFO [train.py:894] (2/4) Epoch 22, batch 2600, loss[loss=0.1574, simple_loss=0.2329, pruned_loss=0.04094, over 18479.00 frames. ], tot_loss[loss=0.1831, simple_loss=0.265, pruned_loss=0.05062, over 3714001.89 frames. ], batch size: 43, lr: 5.19e-03, grad_scale: 8.0
2022-12-23 18:41:40,542 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 3.086e+02 4.163e+02 5.127e+02 6.183e+02 1.258e+03, threshold=1.025e+03, percent-clipped=3.0
2022-12-23 18:41:47,279 INFO [scaling.py:679] (2/4) Whitening: num_groups=8, num_channels=96, metric=1.33 vs. limit=2.0
2022-12-23 18:42:16,460 WARNING [train.py:1060] (2/4) Exclude cut with ID 7357-94126-0026-15857-0_sp0.9 from training. Duration: 25.061125
2022-12-23 18:42:21,189 INFO [zipformer.py:660] (2/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=76269.0, num_to_drop=1, layers_to_drop={1}
2022-12-23 18:42:26,675 WARNING [train.py:1060] (2/4) Exclude cut with ID 3033-130750-0096-55598-0 from training. Duration: 0.83
2022-12-23 18:42:36,181 INFO [train.py:894] (2/4) Epoch 22, batch 2650, loss[loss=0.2247, simple_loss=0.2945, pruned_loss=0.07747, over 18646.00 frames. ], tot_loss[loss=0.1834, simple_loss=0.2651, pruned_loss=0.05084, over 3714377.82 frames. ], batch size: 187, lr: 5.18e-03, grad_scale: 8.0
2022-12-23 18:42:45,111 INFO [scaling.py:679] (2/4) Whitening: num_groups=8, num_channels=192, metric=1.88 vs. limit=2.0
2022-12-23 18:42:51,382 WARNING [train.py:1060] (2/4) Exclude cut with ID 3488-85273-0017-41203-0 from training. Duration: 24.73
2022-12-23 18:43:00,262 INFO [zipformer.py:660] (2/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=76294.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 18:43:04,003 WARNING [train.py:1060] (2/4) Exclude cut with ID 7255-291500-0009-134308-0 from training. Duration: 23.965
2022-12-23 18:43:09,771 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([0.7126, 2.0174, 2.2466, 1.1179, 1.6272, 2.4105, 2.2220, 1.8098],
       device='cuda:2'), covar=tensor([0.0768, 0.0362, 0.0365, 0.0408, 0.0387, 0.0408, 0.0242, 0.0696],
       device='cuda:2'), in_proj_covar=tensor([0.0150, 0.0170, 0.0127, 0.0139, 0.0147, 0.0143, 0.0164, 0.0174],
       device='cuda:2'), out_proj_covar=tensor([1.1390e-04, 1.2962e-04, 9.4699e-05, 1.0315e-04, 1.0943e-04, 1.0869e-04,
        1.2583e-04, 1.3201e-04], device='cuda:2')
2022-12-23 18:43:12,762 WARNING [train.py:1060] (2/4) Exclude cut with ID 6978-92210-0030-146996-0_sp0.9 from training. Duration: 22.088875
2022-12-23 18:43:28,547 WARNING [train.py:1060] (2/4) Exclude cut with ID 7255-291500-0006-134305-0_sp0.9 from training. Duration: 23.6
2022-12-23 18:43:43,995 INFO [zipformer.py:660] (2/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=76323.0, num_to_drop=1, layers_to_drop={0}
2022-12-23 18:43:46,018 INFO [scaling.py:679] (2/4) Whitening: num_groups=8, num_channels=192, metric=1.85 vs. limit=2.0
2022-12-23 18:43:52,451 INFO [train.py:894] (2/4) Epoch 22, batch 2700, loss[loss=0.1856, simple_loss=0.2683, pruned_loss=0.05148, over 18649.00 frames. ], tot_loss[loss=0.1828, simple_loss=0.2647, pruned_loss=0.0505, over 3714526.43 frames. ], batch size: 78, lr: 5.18e-03, grad_scale: 8.0
2022-12-23 18:43:54,201 INFO [zipformer.py:660] (2/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=76330.0, num_to_drop=1, layers_to_drop={0}
2022-12-23 18:43:56,066 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([6.0663, 5.2002, 5.3394, 6.0426, 5.7113, 5.4662, 6.1218, 1.6368],
       device='cuda:2'), covar=tensor([0.0476, 0.0530, 0.0506, 0.0653, 0.1001, 0.1045, 0.0373, 0.5296],
       device='cuda:2'), in_proj_covar=tensor([0.0344, 0.0226, 0.0236, 0.0271, 0.0325, 0.0269, 0.0290, 0.0284],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0001, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-23 18:44:13,386 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 3.173e+02 4.323e+02 5.325e+02 6.564e+02 1.409e+03, threshold=1.065e+03, percent-clipped=3.0
2022-12-23 18:44:34,652 INFO [zipformer.py:660] (2/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=76355.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 18:44:45,309 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([6.1049, 5.1882, 5.4580, 6.0785, 5.7539, 5.4698, 6.1207, 1.9968],
       device='cuda:2'), covar=tensor([0.0506, 0.0609, 0.0478, 0.0714, 0.1013, 0.1038, 0.0373, 0.4783],
       device='cuda:2'), in_proj_covar=tensor([0.0345, 0.0227, 0.0237, 0.0272, 0.0326, 0.0270, 0.0292, 0.0285],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0001, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-23 18:45:10,787 INFO [train.py:894] (2/4) Epoch 22, batch 2750, loss[loss=0.2015, simple_loss=0.2825, pruned_loss=0.06028, over 18692.00 frames. ], tot_loss[loss=0.1833, simple_loss=0.2649, pruned_loss=0.0508, over 3714254.61 frames. ], batch size: 60, lr: 5.18e-03, grad_scale: 8.0
2022-12-23 18:45:10,845 WARNING [train.py:1060] (2/4) Exclude cut with ID 8291-282929-0024-13011-0 from training. Duration: 23.795
2022-12-23 18:45:18,948 INFO [zipformer.py:660] (2/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=76384.0, num_to_drop=1, layers_to_drop={3}
2022-12-23 18:45:27,259 WARNING [train.py:1060] (2/4) Exclude cut with ID 8631-249866-0030-130156-0_sp1.1 from training. Duration: 21.5409375
2022-12-23 18:45:30,209 WARNING [train.py:1060] (2/4) Exclude cut with ID 6978-92210-0019-146985-0_sp0.9 from training. Duration: 24.97775
2022-12-23 18:45:43,339 WARNING [train.py:1060] (2/4) Exclude cut with ID 1085-156170-0017-128270-0_sp0.9 from training. Duration: 23.3444375
2022-12-23 18:45:44,497 INFO [scaling.py:679] (2/4) Whitening: num_groups=1, num_channels=384, metric=3.37 vs. limit=5.0
2022-12-23 18:46:09,338 WARNING [train.py:1060] (2/4) Exclude cut with ID 6010-56788-0055-90261-0_sp0.9 from training. Duration: 23.2
2022-12-23 18:46:15,907 WARNING [train.py:1060] (2/4) Exclude cut with ID 5653-46179-0060-117930-0_sp0.9 from training. Duration: 21.17225
2022-12-23 18:46:27,840 INFO [train.py:894] (2/4) Epoch 22, batch 2800, loss[loss=0.1735, simple_loss=0.2618, pruned_loss=0.04265, over 18719.00 frames. ], tot_loss[loss=0.1827, simple_loss=0.2647, pruned_loss=0.05036, over 3714274.33 frames. ], batch size: 52, lr: 5.18e-03, grad_scale: 8.0
2022-12-23 18:46:36,560 WARNING [train.py:1060] (2/4) Exclude cut with ID 7255-291500-0008-134307-0_sp0.9 from training. Duration: 24.6555625
2022-12-23 18:46:47,870 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 2.459e+02 4.216e+02 4.683e+02 5.845e+02 1.389e+03, threshold=9.366e+02, percent-clipped=1.0
2022-12-23 18:47:03,233 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.4965, 1.9441, 1.5419, 2.3725, 2.4082, 1.6148, 1.4136, 1.3290],
       device='cuda:2'), covar=tensor([0.1912, 0.1692, 0.1567, 0.0873, 0.1278, 0.1062, 0.2155, 0.1505],
       device='cuda:2'), in_proj_covar=tensor([0.0245, 0.0224, 0.0213, 0.0197, 0.0256, 0.0193, 0.0223, 0.0198],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-23 18:47:31,194 WARNING [train.py:1060] (2/4) Exclude cut with ID 3488-65654-0031-41259-0 from training. Duration: 20.44
2022-12-23 18:47:43,342 INFO [train.py:894] (2/4) Epoch 22, batch 2850, loss[loss=0.1744, simple_loss=0.2427, pruned_loss=0.05308, over 18417.00 frames. ], tot_loss[loss=0.1824, simple_loss=0.2642, pruned_loss=0.05026, over 3713779.80 frames. ], batch size: 42, lr: 5.18e-03, grad_scale: 8.0
2022-12-23 18:47:47,945 WARNING [train.py:1060] (2/4) Exclude cut with ID 6951-79737-0018-132285-0_sp0.9 from training. Duration: 23.45
2022-12-23 18:48:19,973 WARNING [train.py:1060] (2/4) Exclude cut with ID 6945-60535-0076-12784-0_sp0.9 from training. Duration: 20.52225
2022-12-23 18:48:26,970 WARNING [train.py:1060] (2/4) Exclude cut with ID 7255-291500-0008-134307-0 from training. Duration: 22.19
2022-12-23 18:48:37,759 WARNING [train.py:1060] (2/4) Exclude cut with ID 7357-94126-0014-15845-0_sp1.1 from training. Duration: 25.3818125
2022-12-23 18:48:53,278 WARNING [train.py:1060] (2/4) Exclude cut with ID 6951-79737-0043-132310-0_sp0.9 from training. Duration: 28.0944375
2022-12-23 18:48:59,554 INFO [train.py:894] (2/4) Epoch 22, batch 2900, loss[loss=0.1624, simple_loss=0.2513, pruned_loss=0.03672, over 18545.00 frames. ], tot_loss[loss=0.1823, simple_loss=0.2646, pruned_loss=0.05, over 3713951.65 frames. ], batch size: 47, lr: 5.18e-03, grad_scale: 8.0
2022-12-23 18:48:59,593 WARNING [train.py:1060] (2/4) Exclude cut with ID 2195-150901-0045-59933-0_sp0.9 from training. Duration: 22.9444375
2022-12-23 18:49:08,559 WARNING [train.py:1060] (2/4) Exclude cut with ID 8291-282929-0024-13011-0_sp1.1 from training. Duration: 21.6318125
2022-12-23 18:49:20,303 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 2.306e+02 4.171e+02 5.104e+02 6.624e+02 1.059e+03, threshold=1.021e+03, percent-clipped=3.0
2022-12-23 18:49:28,852 WARNING [train.py:1060] (2/4) Exclude cut with ID 8631-249866-0030-130156-0 from training. Duration: 23.695
2022-12-23 18:49:54,584 WARNING [train.py:1060] (2/4) Exclude cut with ID 7699-105389-0094-26379-0 from training. Duration: 23.955
2022-12-23 18:50:15,598 INFO [train.py:894] (2/4) Epoch 22, batch 2950, loss[loss=0.2033, simple_loss=0.2844, pruned_loss=0.06113, over 18530.00 frames. ], tot_loss[loss=0.1819, simple_loss=0.2641, pruned_loss=0.04989, over 3713797.29 frames. ], batch size: 98, lr: 5.17e-03, grad_scale: 8.0
2022-12-23 18:50:28,565 WARNING [train.py:1060] (2/4) Exclude cut with ID 8291-282929-0024-13011-0_sp0.9 from training. Duration: 26.438875
2022-12-23 18:51:12,769 WARNING [train.py:1060] (2/4) Exclude cut with ID 7699-105389-0021-26306-0_sp0.9 from training. Duration: 21.2444375
2022-12-23 18:51:12,794 WARNING [train.py:1060] (2/4) Exclude cut with ID 7357-94126-0014-15845-0_sp0.9 from training. Duration: 31.02225
2022-12-23 18:51:23,390 WARNING [train.py:1060] (2/4) Exclude cut with ID 432-122774-0017-62487-0 from training. Duration: 22.395
2022-12-23 18:51:27,881 INFO [zipformer.py:660] (2/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=76625.0, num_to_drop=1, layers_to_drop={0}
2022-12-23 18:51:34,307 INFO [train.py:894] (2/4) Epoch 22, batch 3000, loss[loss=0.1537, simple_loss=0.2373, pruned_loss=0.03509, over 18456.00 frames. ], tot_loss[loss=0.1833, simple_loss=0.2651, pruned_loss=0.05079, over 3713362.04 frames. ], batch size: 50, lr: 5.17e-03, grad_scale: 8.0
2022-12-23 18:51:34,307 INFO [train.py:919] (2/4) Computing validation loss
2022-12-23 18:51:43,913 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([2.0973, 1.6452, 1.9459, 1.6572, 2.0340, 2.0169, 1.8295, 1.8952],
       device='cuda:2'), covar=tensor([0.2133, 0.3015, 0.2233, 0.3169, 0.1964, 0.0983, 0.3147, 0.1194],
       device='cuda:2'), in_proj_covar=tensor([0.0271, 0.0301, 0.0282, 0.0319, 0.0310, 0.0254, 0.0347, 0.0243],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-23 18:51:45,374 INFO [train.py:928] (2/4) Epoch 22, validation: loss=0.1633, simple_loss=0.2609, pruned_loss=0.03279, over 944034.00 frames. 
2022-12-23 18:51:45,375 INFO [train.py:929] (2/4) Maximum memory allocated so far is 24680MB
2022-12-23 18:51:51,292 WARNING [train.py:1060] (2/4) Exclude cut with ID 7357-94126-0045-15876-0 from training. Duration: 21.075
2022-12-23 18:51:57,360 WARNING [train.py:1060] (2/4) Exclude cut with ID 6482-98857-0025-147532-0_sp0.9 from training. Duration: 20.0055625
2022-12-23 18:51:57,372 WARNING [train.py:1060] (2/4) Exclude cut with ID 6951-79737-0037-132304-0_sp0.9 from training. Duration: 22.05
2022-12-23 18:51:57,384 WARNING [train.py:1060] (2/4) Exclude cut with ID 7255-291500-0003-134302-0 from training. Duration: 26.8349375
2022-12-23 18:52:01,078 WARNING [train.py:1060] (2/4) Exclude cut with ID 7357-94126-0009-15840-0_sp1.1 from training. Duration: 22.1090625
2022-12-23 18:52:05,252 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 3.086e+02 4.293e+02 5.248e+02 6.560e+02 1.696e+03, threshold=1.050e+03, percent-clipped=5.0
2022-12-23 18:52:08,298 WARNING [train.py:1060] (2/4) Exclude cut with ID 7699-105389-0094-26379-0_sp0.9 from training. Duration: 26.6166875
2022-12-23 18:52:09,041 INFO [scaling.py:679] (2/4) Whitening: num_groups=1, num_channels=384, metric=2.86 vs. limit=5.0
2022-12-23 18:52:18,189 INFO [zipformer.py:660] (2/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=76650.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 18:52:26,841 WARNING [train.py:1060] (2/4) Exclude cut with ID 2046-178027-0000-53705-0_sp0.9 from training. Duration: 20.3055625
2022-12-23 18:52:48,275 WARNING [train.py:1060] (2/4) Exclude cut with ID 7205-50138-0008-5373-0_sp0.9 from training. Duration: 20.7
2022-12-23 18:53:02,179 INFO [train.py:894] (2/4) Epoch 22, batch 3050, loss[loss=0.2187, simple_loss=0.2859, pruned_loss=0.07574, over 18619.00 frames. ], tot_loss[loss=0.1841, simple_loss=0.2659, pruned_loss=0.05119, over 3713121.89 frames. ], batch size: 184, lr: 5.17e-03, grad_scale: 8.0
2022-12-23 18:53:02,359 INFO [zipformer.py:660] (2/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=76679.0, num_to_drop=1, layers_to_drop={1}
2022-12-23 18:53:28,771 WARNING [train.py:1060] (2/4) Exclude cut with ID 6978-92210-0019-146985-0 from training. Duration: 22.48
2022-12-23 18:53:46,408 WARNING [train.py:1060] (2/4) Exclude cut with ID 7255-291500-0003-134302-0_sp0.9 from training. Duration: 29.816625
2022-12-23 18:53:56,398 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.7155, 1.5633, 1.6669, 1.7010, 1.0826, 3.6613, 1.5514, 2.1585],
       device='cuda:2'), covar=tensor([0.3109, 0.2060, 0.1964, 0.2008, 0.1582, 0.0186, 0.1645, 0.0858],
       device='cuda:2'), in_proj_covar=tensor([0.0132, 0.0117, 0.0125, 0.0120, 0.0104, 0.0096, 0.0091, 0.0090],
       device='cuda:2'), out_proj_covar=tensor([0.0005, 0.0005, 0.0005, 0.0005, 0.0004, 0.0003, 0.0004, 0.0004],
       device='cuda:2')
2022-12-23 18:54:04,676 WARNING [train.py:1060] (2/4) Exclude cut with ID 7255-291500-0005-134304-0_sp1.1 from training. Duration: 22.7590625
2022-12-23 18:54:08,987 WARNING [train.py:1060] (2/4) Exclude cut with ID 7357-94126-0026-15857-0 from training. Duration: 22.555
2022-12-23 18:54:19,692 INFO [train.py:894] (2/4) Epoch 22, batch 3100, loss[loss=0.1569, simple_loss=0.231, pruned_loss=0.04142, over 18628.00 frames. ], tot_loss[loss=0.1842, simple_loss=0.2663, pruned_loss=0.05107, over 3713000.13 frames. ], batch size: 41, lr: 5.17e-03, grad_scale: 8.0
2022-12-23 18:54:31,479 WARNING [train.py:1060] (2/4) Exclude cut with ID 1250-135782-0005-25975-0_sp0.9 from training. Duration: 21.688875
2022-12-23 18:54:39,576 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 2.526e+02 4.110e+02 5.190e+02 6.058e+02 1.141e+03, threshold=1.038e+03, percent-clipped=2.0
2022-12-23 18:55:01,355 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([2.5671, 2.0608, 0.7107, 2.2623, 2.6261, 1.7911, 2.3025, 2.4943],
       device='cuda:2'), covar=tensor([0.1487, 0.1911, 0.2426, 0.1380, 0.1728, 0.1622, 0.1328, 0.1527],
       device='cuda:2'), in_proj_covar=tensor([0.0095, 0.0098, 0.0117, 0.0097, 0.0120, 0.0092, 0.0098, 0.0093],
       device='cuda:2'), out_proj_covar=tensor([0.0003, 0.0003, 0.0004, 0.0003, 0.0004, 0.0003, 0.0003, 0.0003],
       device='cuda:2')
2022-12-23 18:55:06,982 WARNING [train.py:1060] (2/4) Exclude cut with ID 3488-85273-0038-41224-0_sp0.9 from training. Duration: 22.6
2022-12-23 18:55:20,901 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([2.2431, 1.4814, 1.9153, 1.8009, 2.2191, 2.2110, 2.0333, 1.8907],
       device='cuda:2'), covar=tensor([0.2361, 0.3473, 0.2697, 0.2873, 0.2136, 0.0972, 0.3269, 0.1342],
       device='cuda:2'), in_proj_covar=tensor([0.0271, 0.0300, 0.0282, 0.0319, 0.0310, 0.0254, 0.0346, 0.0243],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-23 18:55:35,645 INFO [train.py:894] (2/4) Epoch 22, batch 3150, loss[loss=0.2162, simple_loss=0.3002, pruned_loss=0.06614, over 18568.00 frames. ], tot_loss[loss=0.1832, simple_loss=0.2652, pruned_loss=0.05061, over 3712283.37 frames. ], batch size: 57, lr: 5.17e-03, grad_scale: 8.0
2022-12-23 18:55:46,367 WARNING [train.py:1060] (2/4) Exclude cut with ID 7357-94126-0009-15840-0 from training. Duration: 24.32
2022-12-23 18:55:55,047 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.4155, 1.0888, 0.7714, 1.1724, 1.8617, 0.7103, 1.1808, 1.3877],
       device='cuda:2'), covar=tensor([0.1759, 0.2224, 0.1953, 0.1621, 0.1933, 0.1900, 0.1590, 0.1683],
       device='cuda:2'), in_proj_covar=tensor([0.0095, 0.0098, 0.0117, 0.0097, 0.0120, 0.0092, 0.0099, 0.0093],
       device='cuda:2'), out_proj_covar=tensor([0.0003, 0.0003, 0.0004, 0.0003, 0.0004, 0.0003, 0.0003, 0.0003],
       device='cuda:2')
2022-12-23 18:56:30,534 INFO [scaling.py:679] (2/4) Whitening: num_groups=8, num_channels=192, metric=1.55 vs. limit=2.0
2022-12-23 18:56:44,979 WARNING [train.py:1060] (2/4) Exclude cut with ID 8291-276745-0093-13116-0_sp0.9 from training. Duration: 21.061125
2022-12-23 18:56:50,982 INFO [train.py:894] (2/4) Epoch 22, batch 3200, loss[loss=0.1648, simple_loss=0.2613, pruned_loss=0.03415, over 18470.00 frames. ], tot_loss[loss=0.183, simple_loss=0.2649, pruned_loss=0.05055, over 3712674.34 frames. ], batch size: 54, lr: 5.17e-03, grad_scale: 8.0
2022-12-23 18:56:59,030 WARNING [train.py:1060] (2/4) Exclude cut with ID 7357-94126-0024-15855-0_sp0.9 from training. Duration: 20.32225
2022-12-23 18:57:06,597 INFO [zipformer.py:660] (2/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=76839.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 18:57:11,036 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 2.867e+02 4.150e+02 5.402e+02 6.828e+02 2.211e+03, threshold=1.080e+03, percent-clipped=5.0
2022-12-23 18:57:11,109 WARNING [train.py:1060] (2/4) Exclude cut with ID 3033-130750-0096-55598-0_sp1.1 from training. Duration: 0.7545625
2022-12-23 18:57:25,221 WARNING [train.py:1060] (2/4) Exclude cut with ID 4295-39940-0007-92567-0_sp0.9 from training. Duration: 23.9333125
2022-12-23 18:57:58,512 WARNING [train.py:1060] (2/4) Exclude cut with ID 7255-291500-0008-134307-0_sp1.1 from training. Duration: 20.17275
2022-12-23 18:58:04,242 WARNING [train.py:1060] (2/4) Exclude cut with ID 6978-92210-0019-146985-0_sp1.1 from training. Duration: 20.436375
2022-12-23 18:58:07,837 INFO [train.py:894] (2/4) Epoch 22, batch 3250, loss[loss=0.1563, simple_loss=0.2415, pruned_loss=0.0356, over 18514.00 frames. ], tot_loss[loss=0.1825, simple_loss=0.2643, pruned_loss=0.05035, over 3712850.84 frames. ], batch size: 52, lr: 5.16e-03, grad_scale: 8.0
2022-12-23 18:58:37,849 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([2.4099, 1.7913, 0.6833, 2.0993, 2.6057, 1.6797, 2.1873, 2.3740],
       device='cuda:2'), covar=tensor([0.1440, 0.1930, 0.2400, 0.1386, 0.1623, 0.1761, 0.1324, 0.1467],
       device='cuda:2'), in_proj_covar=tensor([0.0095, 0.0098, 0.0117, 0.0097, 0.0119, 0.0092, 0.0098, 0.0093],
       device='cuda:2'), out_proj_covar=tensor([0.0003, 0.0003, 0.0004, 0.0003, 0.0004, 0.0003, 0.0003, 0.0003],
       device='cuda:2')
2022-12-23 18:58:40,877 INFO [zipformer.py:660] (2/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=76900.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 18:59:19,217 INFO [zipformer.py:660] (2/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=76925.0, num_to_drop=1, layers_to_drop={1}
2022-12-23 18:59:24,924 INFO [train.py:894] (2/4) Epoch 22, batch 3300, loss[loss=0.185, simple_loss=0.2643, pruned_loss=0.05283, over 18411.00 frames. ], tot_loss[loss=0.1823, simple_loss=0.2644, pruned_loss=0.05011, over 3713697.91 frames. ], batch size: 46, lr: 5.16e-03, grad_scale: 8.0
2022-12-23 18:59:26,549 WARNING [train.py:1060] (2/4) Exclude cut with ID 4234-40345-0022-142709-0_sp0.9 from training. Duration: 23.1055625
2022-12-23 18:59:28,029 WARNING [train.py:1060] (2/4) Exclude cut with ID 8291-282929-0007-12994-0_sp1.1 from training. Duration: 23.5
2022-12-23 18:59:38,840 WARNING [train.py:1060] (2/4) Exclude cut with ID 7255-291500-0009-134308-0_sp0.9 from training. Duration: 26.62775
2022-12-23 18:59:40,636 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([2.2796, 2.0075, 1.5292, 1.8930, 1.8280, 1.9153, 1.8346, 2.2711],
       device='cuda:2'), covar=tensor([0.2303, 0.3076, 0.2112, 0.2824, 0.3327, 0.1210, 0.3081, 0.1004],
       device='cuda:2'), in_proj_covar=tensor([0.0297, 0.0295, 0.0247, 0.0349, 0.0273, 0.0230, 0.0290, 0.0216],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0003, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-23 18:59:44,627 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 2.429e+02 4.022e+02 4.936e+02 5.959e+02 1.401e+03, threshold=9.873e+02, percent-clipped=1.0
2022-12-23 18:59:52,297 WARNING [train.py:1060] (2/4) Exclude cut with ID 6951-79737-0018-132285-0 from training. Duration: 21.105
2022-12-23 18:59:55,366 WARNING [train.py:1060] (2/4) Exclude cut with ID 4511-76322-0006-80011-0_sp0.9 from training. Duration: 24.411125
2022-12-23 18:59:56,916 INFO [zipformer.py:660] (2/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=76950.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 19:00:18,116 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([2.0181, 1.9607, 1.5659, 2.0873, 2.1428, 1.8778, 2.5891, 2.1127],
       device='cuda:2'), covar=tensor([0.0953, 0.1712, 0.2830, 0.1712, 0.1972, 0.0999, 0.1039, 0.1246],
       device='cuda:2'), in_proj_covar=tensor([0.0179, 0.0212, 0.0254, 0.0291, 0.0239, 0.0193, 0.0210, 0.0205],
       device='cuda:2'), out_proj_covar=tensor([0.0001, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-23 19:00:23,296 WARNING [train.py:1060] (2/4) Exclude cut with ID 6758-72288-0033-108368-0_sp1.1 from training. Duration: 21.263625
2022-12-23 19:00:31,494 INFO [zipformer.py:660] (2/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=76973.0, num_to_drop=1, layers_to_drop={1}
2022-12-23 19:00:38,180 INFO [scaling.py:679] (2/4) Whitening: num_groups=8, num_channels=96, metric=1.55 vs. limit=2.0
2022-12-23 19:00:40,033 INFO [train.py:894] (2/4) Epoch 22, batch 3350, loss[loss=0.1924, simple_loss=0.2794, pruned_loss=0.05267, over 18583.00 frames. ], tot_loss[loss=0.1826, simple_loss=0.2646, pruned_loss=0.05031, over 3713841.55 frames. ], batch size: 57, lr: 5.16e-03, grad_scale: 8.0
2022-12-23 19:00:40,386 INFO [zipformer.py:660] (2/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=76979.0, num_to_drop=1, layers_to_drop={0}
2022-12-23 19:00:44,919 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([0.0495, 1.0394, 1.1529, 0.5273, 0.5705, 1.2515, 1.2389, 1.2150],
       device='cuda:2'), covar=tensor([0.0744, 0.0348, 0.0349, 0.0397, 0.0481, 0.0493, 0.0283, 0.0603],
       device='cuda:2'), in_proj_covar=tensor([0.0148, 0.0169, 0.0127, 0.0138, 0.0147, 0.0142, 0.0163, 0.0174],
       device='cuda:2'), out_proj_covar=tensor([1.1262e-04, 1.2867e-04, 9.4992e-05, 1.0210e-04, 1.0933e-04, 1.0839e-04,
        1.2467e-04, 1.3217e-04], device='cuda:2')
2022-12-23 19:00:58,084 WARNING [train.py:1060] (2/4) Exclude cut with ID 4234-40345-0022-142709-0 from training. Duration: 20.795
2022-12-23 19:01:07,330 WARNING [train.py:1060] (2/4) Exclude cut with ID 7357-94126-0021-15852-0 from training. Duration: 24.76
2022-12-23 19:01:07,347 WARNING [train.py:1060] (2/4) Exclude cut with ID 3867-173237-0077-144769-0_sp0.9 from training. Duration: 22.25
2022-12-23 19:01:08,561 INFO [zipformer.py:660] (2/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=76998.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 19:01:32,775 WARNING [train.py:1060] (2/4) Exclude cut with ID 7357-94126-0026-15857-0_sp1.1 from training. Duration: 20.5045625
2022-12-23 19:01:53,155 INFO [zipformer.py:660] (2/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=77027.0, num_to_drop=1, layers_to_drop={1}
2022-12-23 19:01:55,606 INFO [train.py:894] (2/4) Epoch 22, batch 3400, loss[loss=0.1787, simple_loss=0.2686, pruned_loss=0.04445, over 18666.00 frames. ], tot_loss[loss=0.1826, simple_loss=0.2644, pruned_loss=0.05037, over 3713521.89 frames. ], batch size: 60, lr: 5.16e-03, grad_scale: 8.0
2022-12-23 19:02:16,428 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 2.658e+02 4.239e+02 5.278e+02 6.844e+02 1.140e+03, threshold=1.056e+03, percent-clipped=4.0
2022-12-23 19:03:08,434 INFO [train.py:894] (2/4) Epoch 22, batch 3450, loss[loss=0.2252, simple_loss=0.302, pruned_loss=0.07418, over 18544.00 frames. ], tot_loss[loss=0.1832, simple_loss=0.2652, pruned_loss=0.05057, over 3712858.51 frames. ], batch size: 98, lr: 5.16e-03, grad_scale: 8.0
2022-12-23 19:03:32,290 INFO [zipformer.py:660] (2/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=77095.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 19:04:20,379 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([2.1594, 1.8611, 2.1547, 2.4122, 2.3004, 3.1655, 1.8878, 1.8861],
       device='cuda:2'), covar=tensor([0.0713, 0.1392, 0.1137, 0.0714, 0.1042, 0.0359, 0.1139, 0.1207],
       device='cuda:2'), in_proj_covar=tensor([0.0072, 0.0081, 0.0071, 0.0073, 0.0090, 0.0075, 0.0084, 0.0077],
       device='cuda:2'), out_proj_covar=tensor([0.0003, 0.0004, 0.0004, 0.0004, 0.0004, 0.0003, 0.0004, 0.0004],
       device='cuda:2')
2022-12-23 19:04:23,451 INFO [train.py:894] (2/4) Epoch 22, batch 3500, loss[loss=0.1843, simple_loss=0.2631, pruned_loss=0.05275, over 18655.00 frames. ], tot_loss[loss=0.1836, simple_loss=0.2657, pruned_loss=0.05078, over 3713710.27 frames. ], batch size: 177, lr: 5.16e-03, grad_scale: 8.0
2022-12-23 19:04:44,892 WARNING [train.py:1060] (2/4) Exclude cut with ID 7859-102521-0017-7548-0_sp1.1 from training. Duration: 22.2954375
2022-12-23 19:04:56,373 INFO [train.py:894] (2/4) Epoch 23, batch 0, loss[loss=0.1971, simple_loss=0.2825, pruned_loss=0.05579, over 18503.00 frames. ], tot_loss[loss=0.1971, simple_loss=0.2825, pruned_loss=0.05579, over 18503.00 frames. ], batch size: 58, lr: 5.04e-03, grad_scale: 8.0
2022-12-23 19:04:56,373 INFO [train.py:919] (2/4) Computing validation loss
2022-12-23 19:05:07,270 INFO [train.py:928] (2/4) Epoch 23, validation: loss=0.1628, simple_loss=0.2608, pruned_loss=0.03242, over 944034.00 frames. 
2022-12-23 19:05:07,271 INFO [train.py:929] (2/4) Maximum memory allocated so far is 24680MB
2022-12-23 19:05:17,280 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 2.568e+02 4.089e+02 5.198e+02 6.379e+02 2.209e+03, threshold=1.040e+03, percent-clipped=4.0
2022-12-23 19:05:38,662 INFO [zipformer.py:660] (2/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=77156.0, num_to_drop=1, layers_to_drop={2}
2022-12-23 19:05:52,326 INFO [scaling.py:679] (2/4) Whitening: num_groups=8, num_channels=192, metric=1.57 vs. limit=2.0
2022-12-23 19:05:55,671 WARNING [train.py:1060] (2/4) Exclude cut with ID 298-126791-0067-24026-0_sp0.9 from training. Duration: 21.438875
2022-12-23 19:06:01,457 WARNING [train.py:1060] (2/4) Exclude cut with ID 5652-39938-0025-23684-0_sp0.9 from training. Duration: 22.2055625
2022-12-23 19:06:21,062 INFO [train.py:894] (2/4) Epoch 23, batch 50, loss[loss=0.1743, simple_loss=0.271, pruned_loss=0.03875, over 18710.00 frames. ], tot_loss[loss=0.1734, simple_loss=0.2635, pruned_loss=0.04168, over 838013.11 frames. ], batch size: 60, lr: 5.04e-03, grad_scale: 8.0
2022-12-23 19:06:35,668 INFO [zipformer.py:660] (2/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=77195.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 19:07:01,057 INFO [zipformer.py:660] (2/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=77211.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 19:07:37,915 INFO [train.py:894] (2/4) Epoch 23, batch 100, loss[loss=0.1704, simple_loss=0.2709, pruned_loss=0.03494, over 18623.00 frames. ], tot_loss[loss=0.1716, simple_loss=0.2602, pruned_loss=0.04151, over 1474975.48 frames. ], batch size: 69, lr: 5.04e-03, grad_scale: 8.0
2022-12-23 19:07:48,675 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 2.031e+02 3.278e+02 3.906e+02 4.585e+02 8.522e+02, threshold=7.812e+02, percent-clipped=0.0
2022-12-23 19:08:34,832 INFO [zipformer.py:660] (2/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=77272.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 19:08:36,924 INFO [zipformer.py:660] (2/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=77273.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 19:08:42,341 INFO [zipformer.py:660] (2/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=77277.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 19:08:53,294 INFO [train.py:894] (2/4) Epoch 23, batch 150, loss[loss=0.1891, simple_loss=0.2606, pruned_loss=0.05883, over 18387.00 frames. ], tot_loss[loss=0.1704, simple_loss=0.2589, pruned_loss=0.04099, over 1972092.96 frames. ], batch size: 46, lr: 5.04e-03, grad_scale: 8.0
2022-12-23 19:09:04,309 WARNING [train.py:1060] (2/4) Exclude cut with ID 7859-102521-0017-7548-0 from training. Duration: 24.525
2022-12-23 19:09:22,157 INFO [scaling.py:679] (2/4) Whitening: num_groups=8, num_channels=96, metric=1.24 vs. limit=2.0
2022-12-23 19:09:37,418 WARNING [train.py:1060] (2/4) Exclude cut with ID 3699-47246-0007-3408-0_sp0.9 from training. Duration: 20.26675
2022-12-23 19:09:50,820 WARNING [train.py:1060] (2/4) Exclude cut with ID 7859-102521-0017-7548-0_sp0.9 from training. Duration: 27.25
2022-12-23 19:10:08,200 INFO [zipformer.py:660] (2/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=77334.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 19:10:09,216 INFO [train.py:894] (2/4) Epoch 23, batch 200, loss[loss=0.1503, simple_loss=0.2415, pruned_loss=0.02959, over 18395.00 frames. ], tot_loss[loss=0.1716, simple_loss=0.26, pruned_loss=0.04161, over 2358406.98 frames. ], batch size: 46, lr: 5.03e-03, grad_scale: 8.0
2022-12-23 19:10:14,907 INFO [zipformer.py:660] (2/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=77338.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 19:10:21,078 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 2.110e+02 3.066e+02 3.714e+02 4.513e+02 7.950e+02, threshold=7.428e+02, percent-clipped=1.0
2022-12-23 19:10:25,769 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.8605, 1.7208, 1.4615, 1.7401, 1.9243, 1.7806, 2.1510, 1.9322],
       device='cuda:2'), covar=tensor([0.0898, 0.1781, 0.2776, 0.1706, 0.1936, 0.0976, 0.1035, 0.1330],
       device='cuda:2'), in_proj_covar=tensor([0.0177, 0.0210, 0.0252, 0.0290, 0.0238, 0.0192, 0.0207, 0.0205],
       device='cuda:2'), out_proj_covar=tensor([0.0001, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-23 19:10:39,251 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.5817, 1.4494, 1.8431, 1.0596, 1.6146, 1.6576, 1.2411, 2.0727],
       device='cuda:2'), covar=tensor([0.1352, 0.2299, 0.1087, 0.1817, 0.1012, 0.1429, 0.2570, 0.0721],
       device='cuda:2'), in_proj_covar=tensor([0.0199, 0.0214, 0.0209, 0.0196, 0.0177, 0.0217, 0.0216, 0.0194],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-23 19:11:06,925 WARNING [train.py:1060] (2/4) Exclude cut with ID 6426-64292-0017-15984-0 from training. Duration: 21.68
2022-12-23 19:11:08,711 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([2.3248, 1.9705, 1.4737, 1.9352, 1.8586, 1.8963, 1.8274, 2.2530],
       device='cuda:2'), covar=tensor([0.2193, 0.3209, 0.2259, 0.2692, 0.3277, 0.1265, 0.2989, 0.0988],
       device='cuda:2'), in_proj_covar=tensor([0.0298, 0.0296, 0.0249, 0.0351, 0.0275, 0.0232, 0.0291, 0.0219],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0003, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-23 19:11:18,509 WARNING [train.py:1060] (2/4) Exclude cut with ID 4278-13270-0007-59342-0 from training. Duration: 21.6300625
2022-12-23 19:11:24,634 INFO [train.py:894] (2/4) Epoch 23, batch 250, loss[loss=0.1619, simple_loss=0.2607, pruned_loss=0.03156, over 18474.00 frames. ], tot_loss[loss=0.1699, simple_loss=0.2582, pruned_loss=0.04079, over 2659113.36 frames. ], batch size: 54, lr: 5.03e-03, grad_scale: 8.0
2022-12-23 19:11:41,202 WARNING [train.py:1060] (2/4) Exclude cut with ID 4278-13270-0007-59342-0_sp0.9 from training. Duration: 24.033375
2022-12-23 19:12:39,046 INFO [train.py:894] (2/4) Epoch 23, batch 300, loss[loss=0.1553, simple_loss=0.2462, pruned_loss=0.03226, over 18637.00 frames. ], tot_loss[loss=0.1692, simple_loss=0.2578, pruned_loss=0.04035, over 2893749.59 frames. ], batch size: 53, lr: 5.03e-03, grad_scale: 8.0
2022-12-23 19:12:41,935 WARNING [train.py:1060] (2/4) Exclude cut with ID 4278-13270-0009-59344-0 from training. Duration: 22.905
2022-12-23 19:12:43,878 WARNING [train.py:1060] (2/4) Exclude cut with ID 5622-44585-0006-90525-0_sp1.1 from training. Duration: 23.4318125
2022-12-23 19:12:49,724 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 2.384e+02 3.220e+02 3.955e+02 4.566e+02 1.223e+03, threshold=7.910e+02, percent-clipped=4.0
2022-12-23 19:13:02,775 INFO [zipformer.py:660] (2/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=77451.0, num_to_drop=1, layers_to_drop={3}
2022-12-23 19:13:09,094 INFO [scaling.py:679] (2/4) Whitening: num_groups=8, num_channels=96, metric=1.39 vs. limit=2.0
2022-12-23 19:13:53,143 INFO [train.py:894] (2/4) Epoch 23, batch 350, loss[loss=0.1725, simple_loss=0.2664, pruned_loss=0.03924, over 18446.00 frames. ], tot_loss[loss=0.1703, simple_loss=0.2585, pruned_loss=0.04105, over 3075884.99 frames. ], batch size: 50, lr: 5.03e-03, grad_scale: 8.0
2022-12-23 19:14:09,199 INFO [zipformer.py:660] (2/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=77495.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 19:14:35,986 WARNING [train.py:1060] (2/4) Exclude cut with ID 4278-13270-0009-59344-0_sp1.1 from training. Duration: 20.82275
2022-12-23 19:14:36,029 WARNING [train.py:1060] (2/4) Exclude cut with ID 4278-13270-0009-59344-0_sp0.9 from training. Duration: 25.45
2022-12-23 19:14:52,580 INFO [scaling.py:679] (2/4) Whitening: num_groups=8, num_channels=192, metric=1.94 vs. limit=2.0
2022-12-23 19:15:09,546 INFO [train.py:894] (2/4) Epoch 23, batch 400, loss[loss=0.1834, simple_loss=0.2754, pruned_loss=0.04569, over 18576.00 frames. ], tot_loss[loss=0.1718, simple_loss=0.2604, pruned_loss=0.0416, over 3217289.32 frames. ], batch size: 56, lr: 5.03e-03, grad_scale: 8.0
2022-12-23 19:15:20,430 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 2.174e+02 3.387e+02 4.125e+02 5.058e+02 9.204e+02, threshold=8.250e+02, percent-clipped=5.0
2022-12-23 19:15:22,078 INFO [zipformer.py:660] (2/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=77543.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 19:15:32,240 WARNING [train.py:1060] (2/4) Exclude cut with ID 5622-44585-0006-90525-0 from training. Duration: 25.775
2022-12-23 19:15:45,773 INFO [scaling.py:679] (2/4) Whitening: num_groups=1, num_channels=384, metric=4.88 vs. limit=5.0
2022-12-23 19:15:54,489 WARNING [train.py:1060] (2/4) Exclude cut with ID 2929-85685-0071-62375-0_sp0.9 from training. Duration: 22.25
2022-12-23 19:15:57,098 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([2.0472, 1.1610, 1.7768, 1.6231, 2.0876, 2.1070, 1.8159, 1.7963],
       device='cuda:2'), covar=tensor([0.2181, 0.3283, 0.2575, 0.2700, 0.2057, 0.0925, 0.3085, 0.1295],
       device='cuda:2'), in_proj_covar=tensor([0.0266, 0.0297, 0.0279, 0.0317, 0.0309, 0.0252, 0.0342, 0.0241],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-23 19:15:58,309 INFO [zipformer.py:660] (2/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=77567.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 19:16:03,794 INFO [scaling.py:679] (2/4) Whitening: num_groups=1, num_channels=384, metric=4.70 vs. limit=5.0
2022-12-23 19:16:24,144 WARNING [train.py:1060] (2/4) Exclude cut with ID 3972-170212-0014-23379-0 from training. Duration: 26.205
2022-12-23 19:16:25,498 INFO [train.py:894] (2/4) Epoch 23, batch 450, loss[loss=0.1813, simple_loss=0.2762, pruned_loss=0.04323, over 18471.00 frames. ], tot_loss[loss=0.1723, simple_loss=0.2614, pruned_loss=0.04161, over 3326218.71 frames. ], batch size: 54, lr: 5.03e-03, grad_scale: 8.0
2022-12-23 19:16:38,941 WARNING [train.py:1060] (2/4) Exclude cut with ID 5239-32139-0047-9341-0_sp0.9 from training. Duration: 30.1555625
2022-12-23 19:16:39,445 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.5722, 1.5663, 1.2420, 1.5866, 1.6971, 1.5234, 1.9384, 1.7012],
       device='cuda:2'), covar=tensor([0.0831, 0.1616, 0.2545, 0.1602, 0.1640, 0.0898, 0.0985, 0.1194],
       device='cuda:2'), in_proj_covar=tensor([0.0177, 0.0209, 0.0251, 0.0288, 0.0237, 0.0191, 0.0206, 0.0204],
       device='cuda:2'), out_proj_covar=tensor([0.0001, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-23 19:16:44,887 WARNING [train.py:1060] (2/4) Exclude cut with ID 1265-135635-0050-6781-0_sp0.9 from training. Duration: 21.8333125
2022-12-23 19:16:53,718 WARNING [train.py:1060] (2/4) Exclude cut with ID 1914-133440-0024-94914-0_sp1.1 from training. Duration: 20.6545625
2022-12-23 19:16:57,232 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([2.4239, 3.6920, 3.5229, 1.4703, 3.7105, 2.8981, 0.9749, 2.4116],
       device='cuda:2'), covar=tensor([0.2200, 0.1118, 0.1378, 0.3488, 0.0923, 0.0910, 0.4561, 0.1435],
       device='cuda:2'), in_proj_covar=tensor([0.0149, 0.0142, 0.0160, 0.0125, 0.0146, 0.0116, 0.0146, 0.0116],
       device='cuda:2'), out_proj_covar=tensor([0.0003, 0.0002, 0.0003, 0.0002, 0.0003, 0.0002, 0.0003, 0.0002],
       device='cuda:2')
2022-12-23 19:17:12,034 INFO [zipformer.py:660] (2/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=77616.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 19:17:31,632 INFO [zipformer.py:660] (2/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=77629.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 19:17:35,835 INFO [scaling.py:679] (2/4) Whitening: num_groups=8, num_channels=96, metric=1.29 vs. limit=2.0
2022-12-23 19:17:37,786 WARNING [train.py:1060] (2/4) Exclude cut with ID 7395-89880-0045-39920-0_sp0.9 from training. Duration: 20.52225
2022-12-23 19:17:37,914 INFO [zipformer.py:660] (2/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=77633.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 19:17:40,697 INFO [train.py:894] (2/4) Epoch 23, batch 500, loss[loss=0.204, simple_loss=0.2904, pruned_loss=0.05882, over 18646.00 frames. ], tot_loss[loss=0.1734, simple_loss=0.2622, pruned_loss=0.04227, over 3412322.07 frames. ], batch size: 53, lr: 5.02e-03, grad_scale: 8.0
2022-12-23 19:17:44,801 INFO [scaling.py:679] (2/4) Whitening: num_groups=8, num_channels=96, metric=1.61 vs. limit=2.0
2022-12-23 19:17:50,549 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 2.063e+02 3.373e+02 4.171e+02 5.386e+02 1.486e+03, threshold=8.341e+02, percent-clipped=6.0
2022-12-23 19:17:55,462 WARNING [train.py:1060] (2/4) Exclude cut with ID 3972-170212-0014-23379-0_sp0.9 from training. Duration: 29.1166875
2022-12-23 19:18:43,302 INFO [zipformer.py:660] (2/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=77677.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 19:18:43,357 INFO [zipformer.py:660] (2/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=77677.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 19:18:55,111 INFO [train.py:894] (2/4) Epoch 23, batch 550, loss[loss=0.163, simple_loss=0.2516, pruned_loss=0.03721, over 18705.00 frames. ], tot_loss[loss=0.174, simple_loss=0.2627, pruned_loss=0.04268, over 3478843.75 frames. ], batch size: 50, lr: 5.02e-03, grad_scale: 8.0
2022-12-23 19:18:55,170 WARNING [train.py:1060] (2/4) Exclude cut with ID 543-133211-0007-59831-0_sp0.9 from training. Duration: 21.388875
2022-12-23 19:19:31,894 WARNING [train.py:1060] (2/4) Exclude cut with ID 1914-133440-0024-94914-0 from training. Duration: 22.72
2022-12-23 19:19:31,937 WARNING [train.py:1060] (2/4) Exclude cut with ID 1914-133440-0031-94921-0_sp0.9 from training. Duration: 22.7444375
2022-12-23 19:20:02,379 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([2.6988, 2.3728, 2.0266, 0.8590, 1.9549, 2.0731, 1.7743, 2.0952],
       device='cuda:2'), covar=tensor([0.0653, 0.0637, 0.1342, 0.1820, 0.1425, 0.1553, 0.1715, 0.0902],
       device='cuda:2'), in_proj_covar=tensor([0.0171, 0.0184, 0.0204, 0.0188, 0.0209, 0.0199, 0.0213, 0.0199],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-23 19:20:09,202 INFO [train.py:894] (2/4) Epoch 23, batch 600, loss[loss=0.1626, simple_loss=0.24, pruned_loss=0.0426, over 18601.00 frames. ], tot_loss[loss=0.1741, simple_loss=0.263, pruned_loss=0.04263, over 3530536.27 frames. ], batch size: 41, lr: 5.02e-03, grad_scale: 8.0
2022-12-23 19:20:09,524 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([2.0517, 1.2369, 2.2403, 4.3870, 3.3284, 2.7426, 0.8729, 3.1547],
       device='cuda:2'), covar=tensor([0.1815, 0.1895, 0.1612, 0.0418, 0.0842, 0.1324, 0.2357, 0.0777],
       device='cuda:2'), in_proj_covar=tensor([0.0102, 0.0115, 0.0134, 0.0150, 0.0104, 0.0141, 0.0128, 0.0113],
       device='cuda:2'), out_proj_covar=tensor([0.0003, 0.0003, 0.0004, 0.0004, 0.0003, 0.0004, 0.0003, 0.0003],
       device='cuda:2')
2022-12-23 19:20:14,129 INFO [zipformer.py:660] (2/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=77738.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 19:20:16,353 WARNING [train.py:1060] (2/4) Exclude cut with ID 4133-6541-0027-40495-0_sp1.1 from training. Duration: 0.9681875
2022-12-23 19:20:19,173 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 2.410e+02 3.680e+02 4.304e+02 5.162e+02 1.245e+03, threshold=8.609e+02, percent-clipped=2.0
2022-12-23 19:20:19,256 WARNING [train.py:1060] (2/4) Exclude cut with ID 6330-62851-0022-91297-0_sp0.9 from training. Duration: 22.3166875
2022-12-23 19:20:24,808 WARNING [train.py:1060] (2/4) Exclude cut with ID 543-133212-0015-59917-0_sp0.9 from training. Duration: 21.8166875
2022-12-23 19:20:32,996 INFO [zipformer.py:660] (2/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=77751.0, num_to_drop=1, layers_to_drop={2}
2022-12-23 19:21:24,609 INFO [train.py:894] (2/4) Epoch 23, batch 650, loss[loss=0.1643, simple_loss=0.256, pruned_loss=0.03627, over 18413.00 frames. ], tot_loss[loss=0.1743, simple_loss=0.2629, pruned_loss=0.04287, over 3571799.56 frames. ], batch size: 48, lr: 5.02e-03, grad_scale: 16.0
2022-12-23 19:21:37,061 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([2.0416, 1.8766, 1.5579, 1.6638, 1.7279, 1.8875, 1.6906, 1.8881],
       device='cuda:2'), covar=tensor([0.2178, 0.3089, 0.2040, 0.2381, 0.3381, 0.1057, 0.2837, 0.1006],
       device='cuda:2'), in_proj_covar=tensor([0.0297, 0.0294, 0.0248, 0.0348, 0.0274, 0.0230, 0.0290, 0.0217],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0003, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-23 19:21:37,697 INFO [scaling.py:679] (2/4) Whitening: num_groups=1, num_channels=384, metric=5.00 vs. limit=5.0
2022-12-23 19:21:45,776 INFO [zipformer.py:660] (2/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=77799.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 19:22:07,349 WARNING [train.py:1060] (2/4) Exclude cut with ID 4957-30119-0041-23990-0_sp0.9 from training. Duration: 20.22775
2022-12-23 19:22:40,272 INFO [train.py:894] (2/4) Epoch 23, batch 700, loss[loss=0.1551, simple_loss=0.2381, pruned_loss=0.03605, over 18676.00 frames. ], tot_loss[loss=0.1741, simple_loss=0.2625, pruned_loss=0.04285, over 3602889.80 frames. ], batch size: 48, lr: 5.02e-03, grad_scale: 16.0
2022-12-23 19:22:50,401 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 2.138e+02 3.371e+02 3.749e+02 4.634e+02 8.391e+02, threshold=7.499e+02, percent-clipped=0.0
2022-12-23 19:22:50,425 WARNING [train.py:1060] (2/4) Exclude cut with ID 5239-32139-0047-9341-0_sp1.1 from training. Duration: 24.67275
2022-12-23 19:23:19,190 WARNING [train.py:1060] (2/4) Exclude cut with ID 3082-165428-0081-50734-0_sp0.9 from training. Duration: 21.8055625
2022-12-23 19:23:27,879 INFO [zipformer.py:660] (2/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=77867.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 19:23:54,213 INFO [train.py:894] (2/4) Epoch 23, batch 750, loss[loss=0.2058, simple_loss=0.2954, pruned_loss=0.05812, over 18652.00 frames. ], tot_loss[loss=0.1742, simple_loss=0.263, pruned_loss=0.04272, over 3627130.94 frames. ], batch size: 62, lr: 5.02e-03, grad_scale: 16.0
2022-12-23 19:23:55,717 WARNING [train.py:1060] (2/4) Exclude cut with ID 3340-169293-0054-76830-0_sp0.9 from training. Duration: 22.6666875
2022-12-23 19:24:12,038 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([2.1691, 1.5068, 1.8622, 1.7994, 2.1573, 2.1620, 1.9959, 1.8180],
       device='cuda:2'), covar=tensor([0.2260, 0.3286, 0.2576, 0.2938, 0.2140, 0.0976, 0.3359, 0.1314],
       device='cuda:2'), in_proj_covar=tensor([0.0267, 0.0297, 0.0279, 0.0317, 0.0309, 0.0252, 0.0343, 0.0240],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-23 19:24:16,638 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([2.5469, 2.0684, 2.1925, 2.2907, 2.4684, 2.5020, 2.4072, 1.9826],
       device='cuda:2'), covar=tensor([0.2185, 0.3262, 0.2502, 0.2892, 0.2121, 0.0933, 0.3548, 0.1303],
       device='cuda:2'), in_proj_covar=tensor([0.0267, 0.0297, 0.0279, 0.0317, 0.0308, 0.0252, 0.0342, 0.0240],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-23 19:24:39,713 INFO [zipformer.py:660] (2/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=77915.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 19:24:59,332 WARNING [train.py:1060] (2/4) Exclude cut with ID 2411-132532-0017-82279-0_sp1.1 from training. Duration: 0.9681875
2022-12-23 19:25:01,041 INFO [zipformer.py:660] (2/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=77929.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 19:25:06,790 INFO [zipformer.py:660] (2/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=77933.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 19:25:09,356 INFO [train.py:894] (2/4) Epoch 23, batch 800, loss[loss=0.1736, simple_loss=0.2699, pruned_loss=0.03861, over 18519.00 frames. ], tot_loss[loss=0.1743, simple_loss=0.263, pruned_loss=0.04275, over 3646758.35 frames. ], batch size: 52, lr: 5.01e-03, grad_scale: 16.0
2022-12-23 19:25:17,039 INFO [scaling.py:679] (2/4) Whitening: num_groups=1, num_channels=384, metric=3.07 vs. limit=5.0
2022-12-23 19:25:19,140 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 2.500e+02 3.367e+02 3.962e+02 4.709e+02 1.041e+03, threshold=7.923e+02, percent-clipped=5.0
2022-12-23 19:25:25,839 WARNING [train.py:1060] (2/4) Exclude cut with ID 6330-62850-0007-91323-0 from training. Duration: 22.485
2022-12-23 19:25:34,084 INFO [scaling.py:679] (2/4) Whitening: num_groups=8, num_channels=192, metric=1.52 vs. limit=2.0
2022-12-23 19:26:00,594 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.5489, 1.9333, 1.6201, 2.2454, 2.6139, 1.6302, 1.6180, 1.3264],
       device='cuda:2'), covar=tensor([0.1975, 0.1797, 0.1552, 0.1005, 0.1168, 0.1130, 0.2022, 0.1571],
       device='cuda:2'), in_proj_covar=tensor([0.0249, 0.0226, 0.0216, 0.0200, 0.0258, 0.0196, 0.0224, 0.0200],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-23 19:26:05,098 WARNING [train.py:1060] (2/4) Exclude cut with ID 3972-170212-0014-23379-0_sp1.1 from training. Duration: 23.82275
2022-12-23 19:26:05,204 INFO [zipformer.py:660] (2/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=77972.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 19:26:12,811 INFO [zipformer.py:660] (2/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=77977.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 19:26:18,636 INFO [zipformer.py:660] (2/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=77981.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 19:26:19,936 WARNING [train.py:1060] (2/4) Exclude cut with ID 4860-13185-0032-76709-0 from training. Duration: 20.77
2022-12-23 19:26:20,414 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.5463, 1.9708, 1.6064, 2.2235, 2.5272, 1.5808, 1.4635, 1.3300],
       device='cuda:2'), covar=tensor([0.1969, 0.1739, 0.1555, 0.1029, 0.1194, 0.1180, 0.2138, 0.1592],
       device='cuda:2'), in_proj_covar=tensor([0.0248, 0.0225, 0.0215, 0.0199, 0.0257, 0.0196, 0.0223, 0.0199],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-23 19:26:24,423 INFO [train.py:894] (2/4) Epoch 23, batch 850, loss[loss=0.1859, simple_loss=0.2807, pruned_loss=0.04555, over 18699.00 frames. ], tot_loss[loss=0.1742, simple_loss=0.2632, pruned_loss=0.04265, over 3661028.66 frames. ], batch size: 98, lr: 5.01e-03, grad_scale: 16.0
2022-12-23 19:26:27,341 WARNING [train.py:1060] (2/4) Exclude cut with ID 6426-64292-0017-15984-0_sp0.9 from training. Duration: 24.088875
2022-12-23 19:27:00,171 WARNING [train.py:1060] (2/4) Exclude cut with ID 6330-62850-0007-91323-0_sp1.1 from training. Duration: 20.4409375
2022-12-23 19:27:39,990 INFO [zipformer.py:660] (2/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=78033.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 19:27:42,667 INFO [train.py:894] (2/4) Epoch 23, batch 900, loss[loss=0.2039, simple_loss=0.2949, pruned_loss=0.05648, over 18487.00 frames. ], tot_loss[loss=0.1737, simple_loss=0.2628, pruned_loss=0.04228, over 3671695.18 frames. ], batch size: 64, lr: 5.01e-03, grad_scale: 16.0
2022-12-23 19:27:52,432 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 1.942e+02 3.202e+02 3.974e+02 4.965e+02 8.252e+02, threshold=7.949e+02, percent-clipped=1.0
2022-12-23 19:28:18,046 WARNING [train.py:1060] (2/4) Exclude cut with ID 2929-85685-0062-62366-0_sp0.9 from training. Duration: 22.511125
2022-12-23 19:28:18,069 WARNING [train.py:1060] (2/4) Exclude cut with ID 7395-89880-0031-39906-0 from training. Duration: 20.675
2022-12-23 19:28:46,108 INFO [scaling.py:679] (2/4) Whitening: num_groups=8, num_channels=96, metric=1.33 vs. limit=2.0
2022-12-23 19:28:58,720 INFO [train.py:894] (2/4) Epoch 23, batch 950, loss[loss=0.1796, simple_loss=0.2696, pruned_loss=0.04482, over 18588.00 frames. ], tot_loss[loss=0.174, simple_loss=0.2632, pruned_loss=0.04236, over 3681028.84 frames. ], batch size: 77, lr: 5.01e-03, grad_scale: 16.0
2022-12-23 19:29:12,791 INFO [scaling.py:679] (2/4) Whitening: num_groups=1, num_channels=384, metric=2.91 vs. limit=5.0
2022-12-23 19:29:57,238 WARNING [train.py:1060] (2/4) Exclude cut with ID 6330-62850-0007-91323-0_sp0.9 from training. Duration: 24.9833125
2022-12-23 19:30:13,047 INFO [train.py:894] (2/4) Epoch 23, batch 1000, loss[loss=0.1868, simple_loss=0.2846, pruned_loss=0.04453, over 18729.00 frames. ], tot_loss[loss=0.173, simple_loss=0.2623, pruned_loss=0.0419, over 3687803.98 frames. ], batch size: 54, lr: 5.01e-03, grad_scale: 16.0
2022-12-23 19:30:23,694 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 2.248e+02 3.311e+02 3.940e+02 4.673e+02 6.833e+02, threshold=7.881e+02, percent-clipped=0.0
2022-12-23 19:30:28,144 WARNING [train.py:1060] (2/4) Exclude cut with ID 5239-32139-0047-9341-0 from training. Duration: 27.14
2022-12-23 19:30:44,250 WARNING [train.py:1060] (2/4) Exclude cut with ID 2929-85685-0044-62348-0 from training. Duration: 22.44
2022-12-23 19:31:27,919 INFO [train.py:894] (2/4) Epoch 23, batch 1050, loss[loss=0.1658, simple_loss=0.2524, pruned_loss=0.03956, over 18440.00 frames. ], tot_loss[loss=0.1721, simple_loss=0.2612, pruned_loss=0.04151, over 3692965.31 frames. ], batch size: 48, lr: 5.01e-03, grad_scale: 8.0
2022-12-23 19:31:34,827 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([2.0031, 2.0174, 1.5407, 2.3360, 2.2253, 1.9443, 2.8041, 2.1014],
       device='cuda:2'), covar=tensor([0.0884, 0.1778, 0.2822, 0.1749, 0.1827, 0.0915, 0.0924, 0.1252],
       device='cuda:2'), in_proj_covar=tensor([0.0176, 0.0209, 0.0250, 0.0288, 0.0237, 0.0190, 0.0204, 0.0204],
       device='cuda:2'), out_proj_covar=tensor([0.0001, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-23 19:32:03,638 WARNING [train.py:1060] (2/4) Exclude cut with ID 2929-85685-0060-62364-0_sp0.9 from training. Duration: 21.361125
2022-12-23 19:32:10,736 WARNING [train.py:1060] (2/4) Exclude cut with ID 2929-85685-0079-62383-0_sp1.1 from training. Duration: 27.0318125
2022-12-23 19:32:20,804 WARNING [train.py:1060] (2/4) Exclude cut with ID 5622-44585-0006-90525-0_sp0.9 from training. Duration: 28.638875
2022-12-23 19:32:21,229 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([3.2266, 3.2776, 2.2486, 1.4740, 3.7033, 3.8369, 3.2274, 2.9357],
       device='cuda:2'), covar=tensor([0.0384, 0.0335, 0.0558, 0.0794, 0.0217, 0.0289, 0.0386, 0.0605],
       device='cuda:2'), in_proj_covar=tensor([0.0126, 0.0128, 0.0130, 0.0119, 0.0101, 0.0123, 0.0133, 0.0158],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0001, 0.0001, 0.0001, 0.0002, 0.0002],
       device='cuda:2')
2022-12-23 19:32:34,642 WARNING [train.py:1060] (2/4) Exclude cut with ID 3340-169293-0054-76830-0 from training. Duration: 20.4
2022-12-23 19:32:39,371 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.3045, 1.3421, 1.4749, 0.9147, 1.3027, 1.4048, 1.2466, 1.6193],
       device='cuda:2'), covar=tensor([0.1094, 0.2022, 0.1150, 0.1458, 0.0814, 0.1081, 0.2552, 0.0633],
       device='cuda:2'), in_proj_covar=tensor([0.0197, 0.0212, 0.0208, 0.0195, 0.0174, 0.0216, 0.0212, 0.0193],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-23 19:32:41,362 INFO [zipformer.py:660] (2/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=78232.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 19:32:45,133 INFO [train.py:894] (2/4) Epoch 23, batch 1100, loss[loss=0.1841, simple_loss=0.2738, pruned_loss=0.04719, over 18503.00 frames. ], tot_loss[loss=0.1721, simple_loss=0.261, pruned_loss=0.04158, over 3697777.49 frames. ], batch size: 58, lr: 5.01e-03, grad_scale: 8.0
2022-12-23 19:32:56,856 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 2.120e+02 3.350e+02 4.234e+02 4.822e+02 9.215e+02, threshold=8.468e+02, percent-clipped=3.0
2022-12-23 19:33:04,617 WARNING [train.py:1060] (2/4) Exclude cut with ID 2929-85685-0071-62375-0 from training. Duration: 20.025
2022-12-23 19:33:05,978 WARNING [train.py:1060] (2/4) Exclude cut with ID 2364-131735-0112-64612-0_sp0.9 from training. Duration: 20.488875
2022-12-23 19:33:12,755 WARNING [train.py:1060] (2/4) Exclude cut with ID 2929-85685-0079-62383-0 from training. Duration: 29.735
2022-12-23 19:33:41,582 INFO [zipformer.py:660] (2/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=78272.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 19:33:49,220 INFO [scaling.py:679] (2/4) Whitening: num_groups=8, num_channels=192, metric=1.52 vs. limit=2.0
2022-12-23 19:33:51,833 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.6757, 1.5824, 1.6552, 1.7469, 1.2423, 3.9425, 1.6266, 2.0444],
       device='cuda:2'), covar=tensor([0.3109, 0.2077, 0.2024, 0.1989, 0.1442, 0.0160, 0.1586, 0.0877],
       device='cuda:2'), in_proj_covar=tensor([0.0133, 0.0118, 0.0125, 0.0121, 0.0104, 0.0096, 0.0091, 0.0089],
       device='cuda:2'), out_proj_covar=tensor([0.0005, 0.0005, 0.0005, 0.0005, 0.0004, 0.0003, 0.0004, 0.0004],
       device='cuda:2')
2022-12-23 19:34:00,730 INFO [train.py:894] (2/4) Epoch 23, batch 1150, loss[loss=0.1722, simple_loss=0.2547, pruned_loss=0.04483, over 18369.00 frames. ], tot_loss[loss=0.1723, simple_loss=0.2615, pruned_loss=0.04159, over 3701369.23 frames. ], batch size: 46, lr: 5.00e-03, grad_scale: 8.0
2022-12-23 19:34:12,659 INFO [zipformer.py:660] (2/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=78293.0, num_to_drop=1, layers_to_drop={3}
2022-12-23 19:34:34,558 WARNING [train.py:1060] (2/4) Exclude cut with ID 7276-92427-0014-12983-0_sp0.9 from training. Duration: 21.3055625
2022-12-23 19:34:35,994 WARNING [train.py:1060] (2/4) Exclude cut with ID 1025-75365-0008-79168-0_sp0.9 from training. Duration: 22.0666875
2022-12-23 19:34:53,260 INFO [zipformer.py:660] (2/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=78320.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 19:35:12,644 INFO [zipformer.py:660] (2/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=78333.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 19:35:15,189 INFO [train.py:894] (2/4) Epoch 23, batch 1200, loss[loss=0.1951, simple_loss=0.282, pruned_loss=0.05414, over 18596.00 frames. ], tot_loss[loss=0.1722, simple_loss=0.2614, pruned_loss=0.04154, over 3704523.42 frames. ], batch size: 51, lr: 5.00e-03, grad_scale: 8.0
2022-12-23 19:35:15,921 INFO [scaling.py:679] (2/4) Whitening: num_groups=8, num_channels=96, metric=1.20 vs. limit=2.0
2022-12-23 19:35:27,165 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 2.472e+02 3.425e+02 4.075e+02 4.983e+02 9.037e+02, threshold=8.150e+02, percent-clipped=1.0
2022-12-23 19:35:52,935 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.5785, 1.3909, 1.4094, 1.7741, 1.5584, 3.2229, 1.3050, 1.4280],
       device='cuda:2'), covar=tensor([0.0858, 0.1772, 0.1127, 0.0974, 0.1548, 0.0252, 0.1405, 0.1512],
       device='cuda:2'), in_proj_covar=tensor([0.0073, 0.0082, 0.0073, 0.0076, 0.0091, 0.0076, 0.0085, 0.0078],
       device='cuda:2'), out_proj_covar=tensor([0.0004, 0.0004, 0.0004, 0.0004, 0.0004, 0.0004, 0.0004, 0.0004],
       device='cuda:2')
2022-12-23 19:36:24,124 INFO [zipformer.py:660] (2/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=78381.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 19:36:26,698 WARNING [train.py:1060] (2/4) Exclude cut with ID 2929-85685-0062-62366-0 from training. Duration: 20.26
2022-12-23 19:36:29,570 INFO [train.py:894] (2/4) Epoch 23, batch 1250, loss[loss=0.1991, simple_loss=0.2838, pruned_loss=0.05717, over 18534.00 frames. ], tot_loss[loss=0.1719, simple_loss=0.2611, pruned_loss=0.04137, over 3706525.37 frames. ], batch size: 55, lr: 5.00e-03, grad_scale: 8.0
2022-12-23 19:36:38,579 WARNING [train.py:1060] (2/4) Exclude cut with ID 5239-32139-0030-9324-0_sp0.9 from training. Duration: 21.3444375
2022-12-23 19:37:37,461 WARNING [train.py:1060] (2/4) Exclude cut with ID 497-129325-0061-62254-0_sp1.1 from training. Duration: 0.97725
2022-12-23 19:37:45,283 INFO [train.py:894] (2/4) Epoch 23, batch 1300, loss[loss=0.1434, simple_loss=0.2272, pruned_loss=0.02981, over 18693.00 frames. ], tot_loss[loss=0.1712, simple_loss=0.2606, pruned_loss=0.04087, over 3708556.94 frames. ], batch size: 46, lr: 5.00e-03, grad_scale: 8.0
2022-12-23 19:37:57,538 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 2.004e+02 3.125e+02 3.830e+02 4.659e+02 1.286e+03, threshold=7.660e+02, percent-clipped=4.0
2022-12-23 19:38:21,869 WARNING [train.py:1060] (2/4) Exclude cut with ID 7395-89880-0031-39906-0_sp0.9 from training. Duration: 22.97225
2022-12-23 19:38:23,343 INFO [zipformer.py:660] (2/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=78460.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 19:38:51,340 WARNING [train.py:1060] (2/4) Exclude cut with ID 7395-89880-0047-39922-0_sp0.9 from training. Duration: 21.97775
2022-12-23 19:38:53,165 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([2.0028, 1.0509, 1.6451, 1.6222, 1.9454, 2.0997, 1.6863, 1.7248],
       device='cuda:2'), covar=tensor([0.2281, 0.3561, 0.2917, 0.3033, 0.2586, 0.1173, 0.3618, 0.1518],
       device='cuda:2'), in_proj_covar=tensor([0.0267, 0.0298, 0.0279, 0.0318, 0.0308, 0.0251, 0.0342, 0.0240],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-23 19:38:59,646 INFO [train.py:894] (2/4) Epoch 23, batch 1350, loss[loss=0.1703, simple_loss=0.2717, pruned_loss=0.03442, over 18671.00 frames. ], tot_loss[loss=0.1707, simple_loss=0.2597, pruned_loss=0.04081, over 3708821.70 frames. ], batch size: 69, lr: 5.00e-03, grad_scale: 8.0
2022-12-23 19:39:02,903 WARNING [train.py:1060] (2/4) Exclude cut with ID 1112-1043-0006-89194-0_sp0.9 from training. Duration: 21.8333125
2022-12-23 19:39:12,292 WARNING [train.py:1060] (2/4) Exclude cut with ID 1914-133440-0031-94921-0 from training. Duration: 20.47
2022-12-23 19:39:55,057 INFO [zipformer.py:660] (2/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=78521.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 19:40:13,638 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([2.1820, 1.7433, 2.0674, 2.6655, 2.0966, 4.8063, 1.7419, 1.9000],
       device='cuda:2'), covar=tensor([0.0717, 0.1622, 0.0859, 0.0849, 0.1350, 0.0158, 0.1292, 0.1418],
       device='cuda:2'), in_proj_covar=tensor([0.0073, 0.0082, 0.0072, 0.0075, 0.0090, 0.0076, 0.0085, 0.0077],
       device='cuda:2'), out_proj_covar=tensor([0.0004, 0.0004, 0.0004, 0.0004, 0.0004, 0.0003, 0.0004, 0.0004],
       device='cuda:2')
2022-12-23 19:40:14,603 INFO [train.py:894] (2/4) Epoch 23, batch 1400, loss[loss=0.1513, simple_loss=0.2348, pruned_loss=0.03391, over 18540.00 frames. ], tot_loss[loss=0.1716, simple_loss=0.2604, pruned_loss=0.04143, over 3710558.40 frames. ], batch size: 44, lr: 5.00e-03, grad_scale: 8.0
2022-12-23 19:40:18,898 WARNING [train.py:1060] (2/4) Exclude cut with ID 7395-89880-0037-39912-0_sp0.9 from training. Duration: 20.67225
2022-12-23 19:40:26,386 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 2.565e+02 3.335e+02 3.901e+02 5.226e+02 1.409e+03, threshold=7.802e+02, percent-clipped=5.0
2022-12-23 19:40:30,279 INFO [scaling.py:679] (2/4) Whitening: num_groups=8, num_channels=96, metric=1.24 vs. limit=2.0
2022-12-23 19:40:34,405 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([2.2709, 1.5665, 1.9201, 1.8935, 2.2533, 2.3021, 2.0207, 1.8447],
       device='cuda:2'), covar=tensor([0.2312, 0.3549, 0.2659, 0.3019, 0.2112, 0.0964, 0.3605, 0.1362],
       device='cuda:2'), in_proj_covar=tensor([0.0269, 0.0299, 0.0281, 0.0319, 0.0310, 0.0253, 0.0345, 0.0242],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-23 19:40:36,743 WARNING [train.py:1060] (2/4) Exclude cut with ID 1914-133440-0024-94914-0_sp0.9 from training. Duration: 25.2444375
2022-12-23 19:40:46,170 INFO [zipformer.py:660] (2/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=78556.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 19:40:59,526 WARNING [train.py:1060] (2/4) Exclude cut with ID 3340-169293-0021-76797-0_sp0.9 from training. Duration: 21.1445
2022-12-23 19:41:08,922 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.5131, 1.8412, 1.4395, 2.1928, 2.4626, 1.5965, 1.4780, 1.2403],
       device='cuda:2'), covar=tensor([0.1981, 0.1816, 0.1708, 0.1046, 0.1252, 0.1138, 0.2248, 0.1609],
       device='cuda:2'), in_proj_covar=tensor([0.0247, 0.0224, 0.0215, 0.0198, 0.0257, 0.0196, 0.0223, 0.0199],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-23 19:41:29,578 INFO [train.py:894] (2/4) Epoch 23, batch 1450, loss[loss=0.1555, simple_loss=0.2395, pruned_loss=0.03574, over 18666.00 frames. ], tot_loss[loss=0.1716, simple_loss=0.2603, pruned_loss=0.04142, over 3711087.80 frames. ], batch size: 41, lr: 4.99e-03, grad_scale: 8.0
2022-12-23 19:41:33,946 INFO [zipformer.py:660] (2/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=78588.0, num_to_drop=1, layers_to_drop={3}
2022-12-23 19:41:41,845 INFO [scaling.py:679] (2/4) Whitening: num_groups=1, num_channels=384, metric=3.17 vs. limit=5.0
2022-12-23 19:42:15,429 WARNING [train.py:1060] (2/4) Exclude cut with ID 2929-85685-0079-62383-0_sp0.9 from training. Duration: 33.038875
2022-12-23 19:42:18,563 INFO [zipformer.py:660] (2/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=78617.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 19:42:40,115 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([2.7496, 2.5318, 2.0617, 0.9255, 1.9436, 2.2379, 1.9734, 2.2407],
       device='cuda:2'), covar=tensor([0.0657, 0.0524, 0.1233, 0.1745, 0.1337, 0.1393, 0.1509, 0.0779],
       device='cuda:2'), in_proj_covar=tensor([0.0172, 0.0185, 0.0205, 0.0188, 0.0209, 0.0199, 0.0213, 0.0198],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-23 19:42:45,869 INFO [train.py:894] (2/4) Epoch 23, batch 1500, loss[loss=0.166, simple_loss=0.2591, pruned_loss=0.03648, over 18706.00 frames. ], tot_loss[loss=0.1713, simple_loss=0.26, pruned_loss=0.0413, over 3712071.55 frames. ], batch size: 52, lr: 4.99e-03, grad_scale: 8.0
2022-12-23 19:42:53,079 WARNING [train.py:1060] (2/4) Exclude cut with ID 6426-64291-0000-16059-0_sp0.9 from training. Duration: 20.0944375
2022-12-23 19:42:57,188 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 2.151e+02 3.273e+02 3.888e+02 4.739e+02 1.159e+03, threshold=7.776e+02, percent-clipped=4.0
2022-12-23 19:42:58,156 INFO [scaling.py:679] (2/4) Whitening: num_groups=8, num_channels=192, metric=1.98 vs. limit=2.0
2022-12-23 19:43:08,934 WARNING [train.py:1060] (2/4) Exclude cut with ID 2929-85685-0044-62348-0_sp1.1 from training. Duration: 20.4
2022-12-23 19:43:15,399 WARNING [train.py:1060] (2/4) Exclude cut with ID 6330-62851-0022-91297-0 from training. Duration: 20.085
2022-12-23 19:43:27,869 WARNING [train.py:1060] (2/4) Exclude cut with ID 4860-13185-0032-76709-0_sp0.9 from training. Duration: 23.07775
2022-12-23 19:44:00,106 INFO [train.py:894] (2/4) Epoch 23, batch 1550, loss[loss=0.1776, simple_loss=0.2718, pruned_loss=0.04165, over 18463.00 frames. ], tot_loss[loss=0.1717, simple_loss=0.2604, pruned_loss=0.04148, over 3712888.72 frames. ], batch size: 64, lr: 4.99e-03, grad_scale: 8.0
2022-12-23 19:44:06,600 INFO [scaling.py:679] (2/4) Whitening: num_groups=8, num_channels=192, metric=1.51 vs. limit=2.0
2022-12-23 19:44:12,110 WARNING [train.py:1060] (2/4) Exclude cut with ID 2929-85685-0044-62348-0_sp0.9 from training. Duration: 24.9333125
2022-12-23 19:44:57,437 WARNING [train.py:1060] (2/4) Exclude cut with ID 5118-111612-0016-124680-0_sp0.9 from training. Duration: 20.388875
2022-12-23 19:45:03,196 WARNING [train.py:1060] (2/4) Exclude cut with ID 432-122774-0017-62487-0_sp1.1 from training. Duration: 20.3590625
2022-12-23 19:45:14,752 INFO [train.py:894] (2/4) Epoch 23, batch 1600, loss[loss=0.163, simple_loss=0.2486, pruned_loss=0.03865, over 18403.00 frames. ], tot_loss[loss=0.1723, simple_loss=0.2616, pruned_loss=0.04152, over 3713436.61 frames. ], batch size: 46, lr: 4.99e-03, grad_scale: 8.0
2022-12-23 19:45:27,223 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 2.292e+02 3.292e+02 3.867e+02 4.586e+02 1.296e+03, threshold=7.734e+02, percent-clipped=3.0
2022-12-23 19:45:38,758 INFO [scaling.py:679] (2/4) Whitening: num_groups=8, num_channels=96, metric=1.14 vs. limit=2.0
2022-12-23 19:46:12,253 WARNING [train.py:1060] (2/4) Exclude cut with ID 3557-8342-0013-54691-0_sp1.1 from training. Duration: 0.836375
2022-12-23 19:46:31,062 INFO [train.py:894] (2/4) Epoch 23, batch 1650, loss[loss=0.1728, simple_loss=0.2631, pruned_loss=0.04127, over 18579.00 frames. ], tot_loss[loss=0.1732, simple_loss=0.2618, pruned_loss=0.04228, over 3713134.07 frames. ], batch size: 51, lr: 4.99e-03, grad_scale: 8.0
2022-12-23 19:46:42,439 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([2.1158, 1.5821, 2.0409, 2.5711, 2.1643, 4.7183, 1.6577, 1.8146],
       device='cuda:2'), covar=tensor([0.0726, 0.1769, 0.0914, 0.0862, 0.1352, 0.0172, 0.1360, 0.1489],
       device='cuda:2'), in_proj_covar=tensor([0.0072, 0.0082, 0.0072, 0.0074, 0.0090, 0.0075, 0.0084, 0.0077],
       device='cuda:2'), out_proj_covar=tensor([0.0004, 0.0004, 0.0004, 0.0004, 0.0004, 0.0003, 0.0004, 0.0004],
       device='cuda:2')
2022-12-23 19:46:54,675 WARNING [train.py:1060] (2/4) Exclude cut with ID 8565-290391-0049-67394-0_sp0.9 from training. Duration: 21.3166875
2022-12-23 19:47:11,769 INFO [scaling.py:679] (2/4) Whitening: num_groups=8, num_channels=192, metric=1.82 vs. limit=2.0
2022-12-23 19:47:15,624 INFO [scaling.py:679] (2/4) Whitening: num_groups=1, num_channels=384, metric=3.37 vs. limit=5.0
2022-12-23 19:47:19,295 INFO [zipformer.py:660] (2/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=78816.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 19:47:26,390 WARNING [train.py:1060] (2/4) Exclude cut with ID 6533-399-0029-104863-0_sp0.9 from training. Duration: 22.1055625
2022-12-23 19:47:36,867 WARNING [train.py:1060] (2/4) Exclude cut with ID 7699-105389-0094-26379-0_sp1.1 from training. Duration: 21.77725
2022-12-23 19:47:46,976 INFO [train.py:894] (2/4) Epoch 23, batch 1700, loss[loss=0.1828, simple_loss=0.2696, pruned_loss=0.04797, over 18579.00 frames. ], tot_loss[loss=0.1745, simple_loss=0.2626, pruned_loss=0.04318, over 3713380.56 frames. ], batch size: 56, lr: 4.99e-03, grad_scale: 8.0
2022-12-23 19:47:55,484 WARNING [train.py:1060] (2/4) Exclude cut with ID 7255-291500-0005-134304-0_sp0.9 from training. Duration: 27.8166875
2022-12-23 19:48:00,051 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 2.331e+02 3.650e+02 4.431e+02 5.330e+02 1.177e+03, threshold=8.861e+02, percent-clipped=5.0
2022-12-23 19:48:02,508 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([2.6339, 2.3665, 1.9531, 1.0254, 2.0638, 2.0962, 1.7381, 1.9876],
       device='cuda:2'), covar=tensor([0.0687, 0.0594, 0.1288, 0.1717, 0.1171, 0.1518, 0.1765, 0.0918],
       device='cuda:2'), in_proj_covar=tensor([0.0175, 0.0188, 0.0208, 0.0191, 0.0213, 0.0202, 0.0217, 0.0201],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-23 19:48:08,468 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([2.0604, 2.0052, 2.3139, 1.3202, 2.3741, 2.2926, 1.6684, 2.6849],
       device='cuda:2'), covar=tensor([0.1280, 0.1847, 0.1387, 0.2145, 0.0756, 0.1318, 0.2318, 0.0608],
       device='cuda:2'), in_proj_covar=tensor([0.0196, 0.0212, 0.0207, 0.0194, 0.0174, 0.0215, 0.0211, 0.0193],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-23 19:48:14,652 INFO [scaling.py:679] (2/4) Whitening: num_groups=8, num_channels=96, metric=1.29 vs. limit=2.0
2022-12-23 19:48:21,291 WARNING [train.py:1060] (2/4) Exclude cut with ID 7357-94126-0021-15852-0_sp1.1 from training. Duration: 22.5090625
2022-12-23 19:48:28,037 WARNING [train.py:1060] (2/4) Exclude cut with ID 7255-291500-0005-134304-0 from training. Duration: 25.035
2022-12-23 19:48:44,666 WARNING [train.py:1060] (2/4) Exclude cut with ID 774-127930-0014-10412-0_sp1.1 from training. Duration: 0.95
2022-12-23 19:49:02,881 WARNING [train.py:1060] (2/4) Exclude cut with ID 3033-130750-0096-55598-0_sp0.9 from training. Duration: 0.92225
2022-12-23 19:49:04,286 INFO [train.py:894] (2/4) Epoch 23, batch 1750, loss[loss=0.2364, simple_loss=0.3195, pruned_loss=0.07666, over 18519.00 frames. ], tot_loss[loss=0.1762, simple_loss=0.2634, pruned_loss=0.04453, over 3712750.49 frames. ], batch size: 58, lr: 4.98e-03, grad_scale: 8.0
2022-12-23 19:49:08,919 INFO [zipformer.py:660] (2/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=78888.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 19:49:31,082 WARNING [train.py:1060] (2/4) Exclude cut with ID 4511-76322-0006-80011-0 from training. Duration: 21.97
2022-12-23 19:49:44,860 INFO [zipformer.py:660] (2/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=78912.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 19:49:49,577 WARNING [train.py:1060] (2/4) Exclude cut with ID 7492-105653-0055-62765-0_sp0.9 from training. Duration: 21.97225
2022-12-23 19:49:51,109 WARNING [train.py:1060] (2/4) Exclude cut with ID 453-131332-0000-47844-0_sp0.9 from training. Duration: 25.3333125
2022-12-23 19:50:01,663 WARNING [train.py:1060] (2/4) Exclude cut with ID 5172-29468-0015-19128-0_sp0.9 from training. Duration: 21.5055625
2022-12-23 19:50:10,211 WARNING [train.py:1060] (2/4) Exclude cut with ID 453-131332-0000-47844-0_sp1.1 from training. Duration: 20.72725
2022-12-23 19:50:19,182 INFO [train.py:894] (2/4) Epoch 23, batch 1800, loss[loss=0.1872, simple_loss=0.2749, pruned_loss=0.04973, over 18585.00 frames. ], tot_loss[loss=0.1779, simple_loss=0.2639, pruned_loss=0.04599, over 3713895.07 frames. ], batch size: 78, lr: 4.98e-03, grad_scale: 8.0
2022-12-23 19:50:20,893 INFO [zipformer.py:660] (2/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=78936.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 19:50:32,365 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 3.020e+02 4.798e+02 5.662e+02 6.863e+02 1.253e+03, threshold=1.132e+03, percent-clipped=4.0
2022-12-23 19:50:43,750 WARNING [train.py:1060] (2/4) Exclude cut with ID 8631-249866-0030-130156-0_sp0.9 from training. Duration: 26.32775
2022-12-23 19:50:56,531 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([2.1790, 1.3670, 1.9151, 1.7894, 2.1752, 2.2134, 1.9244, 1.8532],
       device='cuda:2'), covar=tensor([0.2168, 0.3212, 0.2542, 0.2794, 0.2021, 0.0947, 0.3022, 0.1302],
       device='cuda:2'), in_proj_covar=tensor([0.0268, 0.0298, 0.0280, 0.0318, 0.0308, 0.0252, 0.0344, 0.0241],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-23 19:51:04,758 INFO [zipformer.py:660] (2/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=78965.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 19:51:13,295 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.9702, 1.8797, 1.4947, 1.6580, 1.6599, 1.8098, 1.6660, 1.7530],
       device='cuda:2'), covar=tensor([0.2284, 0.2897, 0.2062, 0.2397, 0.3226, 0.1127, 0.2769, 0.1089],
       device='cuda:2'), in_proj_covar=tensor([0.0297, 0.0295, 0.0249, 0.0349, 0.0275, 0.0230, 0.0291, 0.0217],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0003, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-23 19:51:15,708 WARNING [train.py:1060] (2/4) Exclude cut with ID 3867-173237-0077-144769-0 from training. Duration: 20.025
2022-12-23 19:51:18,371 WARNING [train.py:1060] (2/4) Exclude cut with ID 6709-74022-0004-86860-0_sp1.1 from training. Duration: 0.9409375
2022-12-23 19:51:18,381 WARNING [train.py:1060] (2/4) Exclude cut with ID 4757-1811-0023-62229-0_sp0.9 from training. Duration: 21.37775
2022-12-23 19:51:34,064 INFO [train.py:894] (2/4) Epoch 23, batch 1850, loss[loss=0.1914, simple_loss=0.2771, pruned_loss=0.05282, over 18524.00 frames. ], tot_loss[loss=0.1801, simple_loss=0.2651, pruned_loss=0.04754, over 3713911.51 frames. ], batch size: 52, lr: 4.98e-03, grad_scale: 8.0
2022-12-23 19:51:41,085 WARNING [train.py:1060] (2/4) Exclude cut with ID 1250-135782-0004-25974-0_sp0.9 from training. Duration: 21.17225
2022-12-23 19:51:41,099 WARNING [train.py:1060] (2/4) Exclude cut with ID 7357-94126-0021-15852-0_sp0.9 from training. Duration: 27.511125
2022-12-23 19:51:53,670 INFO [scaling.py:679] (2/4) Whitening: num_groups=1, num_channels=384, metric=3.12 vs. limit=5.0
2022-12-23 19:52:12,936 WARNING [train.py:1060] (2/4) Exclude cut with ID 453-131332-0000-47844-0 from training. Duration: 22.8
2022-12-23 19:52:17,455 WARNING [train.py:1060] (2/4) Exclude cut with ID 4964-30587-0040-44509-0 from training. Duration: 22.585
2022-12-23 19:52:17,929 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.4765, 2.8523, 3.1008, 1.1713, 2.7475, 3.4065, 2.6906, 2.6186],
       device='cuda:2'), covar=tensor([0.0804, 0.0336, 0.0295, 0.0543, 0.0384, 0.0337, 0.0346, 0.0717],
       device='cuda:2'), in_proj_covar=tensor([0.0151, 0.0172, 0.0130, 0.0141, 0.0149, 0.0144, 0.0168, 0.0177],
       device='cuda:2'), out_proj_covar=tensor([1.1428e-04, 1.3062e-04, 9.6547e-05, 1.0458e-04, 1.1045e-04, 1.0946e-04,
        1.2822e-04, 1.3398e-04], device='cuda:2')
2022-12-23 19:52:38,020 INFO [zipformer.py:660] (2/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=79026.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 19:52:46,917 WARNING [train.py:1060] (2/4) Exclude cut with ID 6978-92210-0001-146967-0_sp0.9 from training. Duration: 22.0166875
2022-12-23 19:52:51,170 INFO [train.py:894] (2/4) Epoch 23, batch 1900, loss[loss=0.1587, simple_loss=0.2509, pruned_loss=0.03326, over 18577.00 frames. ], tot_loss[loss=0.1811, simple_loss=0.2651, pruned_loss=0.04851, over 3714672.01 frames. ], batch size: 56, lr: 4.98e-03, grad_scale: 8.0
2022-12-23 19:53:01,435 WARNING [train.py:1060] (2/4) Exclude cut with ID 7255-291500-0003-134302-0_sp1.1 from training. Duration: 24.395375
2022-12-23 19:53:04,451 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 2.829e+02 4.333e+02 5.435e+02 7.142e+02 1.314e+03, threshold=1.087e+03, percent-clipped=4.0
2022-12-23 19:53:07,780 WARNING [train.py:1060] (2/4) Exclude cut with ID 3488-85273-0017-41203-0_sp0.9 from training. Duration: 27.47775
2022-12-23 19:53:14,123 WARNING [train.py:1060] (2/4) Exclude cut with ID 432-122774-0017-62487-0_sp0.9 from training. Duration: 24.8833125
2022-12-23 19:53:16,982 WARNING [train.py:1060] (2/4) Exclude cut with ID 6758-72288-0033-108368-0 from training. Duration: 23.39
2022-12-23 19:53:21,403 WARNING [train.py:1060] (2/4) Exclude cut with ID 8291-282929-0007-12994-0_sp0.9 from training. Duration: 28.72225
2022-12-23 19:53:29,886 WARNING [train.py:1060] (2/4) Exclude cut with ID 585-294811-0110-133686-0_sp0.9 from training. Duration: 20.8944375
2022-12-23 19:53:44,473 WARNING [train.py:1060] (2/4) Exclude cut with ID 5796-66357-0007-116447-0_sp0.9 from training. Duration: 23.8444375
2022-12-23 19:54:05,744 INFO [scaling.py:679] (2/4) Whitening: num_groups=1, num_channels=384, metric=3.36 vs. limit=5.0
2022-12-23 19:54:07,792 INFO [train.py:894] (2/4) Epoch 23, batch 1950, loss[loss=0.1744, simple_loss=0.2634, pruned_loss=0.04267, over 18458.00 frames. ], tot_loss[loss=0.1821, simple_loss=0.2656, pruned_loss=0.04931, over 3715411.66 frames. ], batch size: 54, lr: 4.98e-03, grad_scale: 8.0
2022-12-23 19:54:10,790 WARNING [train.py:1060] (2/4) Exclude cut with ID 8291-282929-0007-12994-0 from training. Duration: 25.85
2022-12-23 19:54:10,801 WARNING [train.py:1060] (2/4) Exclude cut with ID 8291-282929-0023-13010-0 from training. Duration: 21.39
2022-12-23 19:54:22,540 WARNING [train.py:1060] (2/4) Exclude cut with ID 7357-94126-0014-15845-0 from training. Duration: 27.92
2022-12-23 19:54:23,006 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([2.1689, 2.0893, 1.7216, 1.7832, 2.3833, 2.8232, 2.6022, 2.0266],
       device='cuda:2'), covar=tensor([0.0381, 0.0343, 0.0489, 0.0283, 0.0245, 0.0368, 0.0358, 0.0341],
       device='cuda:2'), in_proj_covar=tensor([0.0097, 0.0129, 0.0155, 0.0126, 0.0119, 0.0124, 0.0101, 0.0130],
       device='cuda:2'), out_proj_covar=tensor([7.6864e-05, 1.0235e-04, 1.2749e-04, 1.0031e-04, 9.6186e-05, 9.5081e-05,
        7.8457e-05, 1.0217e-04], device='cuda:2')
2022-12-23 19:54:49,926 WARNING [train.py:1060] (2/4) Exclude cut with ID 8631-249866-0039-130165-0_sp0.9 from training. Duration: 20.661125
2022-12-23 19:54:54,501 INFO [zipformer.py:660] (2/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=79116.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 19:55:11,692 WARNING [train.py:1060] (2/4) Exclude cut with ID 7357-94126-0043-15874-0_sp0.9 from training. Duration: 20.07225
2022-12-23 19:55:17,909 WARNING [train.py:1060] (2/4) Exclude cut with ID 1085-156170-0017-128270-0 from training. Duration: 21.01
2022-12-23 19:55:23,804 INFO [train.py:894] (2/4) Epoch 23, batch 2000, loss[loss=0.1841, simple_loss=0.256, pruned_loss=0.05615, over 18417.00 frames. ], tot_loss[loss=0.1829, simple_loss=0.2656, pruned_loss=0.05012, over 3715375.68 frames. ], batch size: 42, lr: 4.98e-03, grad_scale: 8.0
2022-12-23 19:55:36,309 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 2.867e+02 4.288e+02 5.070e+02 5.717e+02 1.182e+03, threshold=1.014e+03, percent-clipped=2.0
2022-12-23 19:55:48,917 INFO [scaling.py:679] (2/4) Whitening: num_groups=8, num_channels=96, metric=1.58 vs. limit=2.0
2022-12-23 19:56:08,481 INFO [zipformer.py:660] (2/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=79164.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 19:56:25,590 WARNING [train.py:1060] (2/4) Exclude cut with ID 2195-150901-0045-59933-0 from training. Duration: 20.65
2022-12-23 19:56:34,268 WARNING [train.py:1060] (2/4) Exclude cut with ID 5796-66357-0007-116447-0 from training. Duration: 21.46
2022-12-23 19:56:37,528 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.8237, 1.3153, 1.6784, 1.9832, 1.7130, 3.5494, 1.3318, 1.4583],
       device='cuda:2'), covar=tensor([0.0825, 0.1971, 0.1134, 0.1004, 0.1538, 0.0284, 0.1589, 0.1639],
       device='cuda:2'), in_proj_covar=tensor([0.0073, 0.0082, 0.0072, 0.0075, 0.0091, 0.0076, 0.0084, 0.0077],
       device='cuda:2'), out_proj_covar=tensor([0.0004, 0.0004, 0.0004, 0.0004, 0.0004, 0.0003, 0.0004, 0.0004],
       device='cuda:2')
2022-12-23 19:56:39,815 INFO [train.py:894] (2/4) Epoch 23, batch 2050, loss[loss=0.2069, simple_loss=0.2864, pruned_loss=0.06375, over 18606.00 frames. ], tot_loss[loss=0.1835, simple_loss=0.2661, pruned_loss=0.05042, over 3714808.07 frames. ], batch size: 97, lr: 4.98e-03, grad_scale: 8.0
2022-12-23 19:57:02,161 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([2.2715, 2.3245, 1.6237, 2.6540, 2.5141, 2.1678, 3.0776, 2.3695],
       device='cuda:2'), covar=tensor([0.0799, 0.1710, 0.2717, 0.1701, 0.1592, 0.0821, 0.0910, 0.1146],
       device='cuda:2'), in_proj_covar=tensor([0.0178, 0.0210, 0.0250, 0.0289, 0.0238, 0.0190, 0.0205, 0.0204],
       device='cuda:2'), out_proj_covar=tensor([0.0001, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-23 19:57:11,091 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.5367, 1.4082, 1.3866, 0.8264, 1.7146, 1.5764, 1.4959, 1.2725],
       device='cuda:2'), covar=tensor([0.0412, 0.0576, 0.0511, 0.0798, 0.0419, 0.0417, 0.0467, 0.1055],
       device='cuda:2'), in_proj_covar=tensor([0.0126, 0.0130, 0.0131, 0.0120, 0.0102, 0.0124, 0.0134, 0.0161],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0001, 0.0001, 0.0001, 0.0002, 0.0002],
       device='cuda:2')
2022-12-23 19:57:19,560 WARNING [train.py:1060] (2/4) Exclude cut with ID 3557-8342-0013-54691-0 from training. Duration: 0.92
2022-12-23 19:57:19,680 INFO [zipformer.py:660] (2/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=79212.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 19:57:27,458 WARNING [train.py:1060] (2/4) Exclude cut with ID 8291-282929-0023-13010-0_sp0.9 from training. Duration: 23.7666875
2022-12-23 19:57:54,618 INFO [train.py:894] (2/4) Epoch 23, batch 2100, loss[loss=0.155, simple_loss=0.2256, pruned_loss=0.04224, over 18531.00 frames. ], tot_loss[loss=0.1829, simple_loss=0.2652, pruned_loss=0.05027, over 3713370.20 frames. ], batch size: 44, lr: 4.97e-03, grad_scale: 8.0
2022-12-23 19:58:02,618 WARNING [train.py:1060] (2/4) Exclude cut with ID 8544-281189-0060-101339-0_sp0.9 from training. Duration: 20.861125
2022-12-23 19:58:06,582 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 2.746e+02 4.115e+02 5.224e+02 6.588e+02 1.506e+03, threshold=1.045e+03, percent-clipped=3.0
2022-12-23 19:58:11,512 WARNING [train.py:1060] (2/4) Exclude cut with ID 3488-65654-0031-41259-0_sp0.9 from training. Duration: 22.711125
2022-12-23 19:58:26,749 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.5246, 1.9055, 1.6447, 2.2799, 2.4176, 1.6074, 1.4567, 1.2858],
       device='cuda:2'), covar=tensor([0.2112, 0.1826, 0.1656, 0.1066, 0.1308, 0.1176, 0.2280, 0.1716],
       device='cuda:2'), in_proj_covar=tensor([0.0251, 0.0228, 0.0218, 0.0201, 0.0262, 0.0198, 0.0227, 0.0202],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-23 19:58:32,780 INFO [zipformer.py:660] (2/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=79260.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 19:58:55,658 WARNING [train.py:1060] (2/4) Exclude cut with ID 6951-79737-0043-132310-0_sp1.1 from training. Duration: 22.986375
2022-12-23 19:59:10,425 INFO [train.py:894] (2/4) Epoch 23, batch 2150, loss[loss=0.1628, simple_loss=0.2396, pruned_loss=0.04298, over 18410.00 frames. ], tot_loss[loss=0.1834, simple_loss=0.2656, pruned_loss=0.05065, over 3713371.62 frames. ], batch size: 42, lr: 4.97e-03, grad_scale: 8.0
2022-12-23 19:59:10,498 WARNING [train.py:1060] (2/4) Exclude cut with ID 8040-260924-0003-80960-0_sp0.9 from training. Duration: 22.07225
2022-12-23 19:59:10,810 INFO [zipformer.py:660] (2/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=79285.0, num_to_drop=1, layers_to_drop={1}
2022-12-23 19:59:15,185 WARNING [train.py:1060] (2/4) Exclude cut with ID 7699-105389-0045-26330-0_sp0.9 from training. Duration: 20.3055625
2022-12-23 19:59:19,140 WARNING [train.py:1060] (2/4) Exclude cut with ID 6356-271890-0060-94317-0_sp0.9 from training. Duration: 20.72225
2022-12-23 19:59:36,723 WARNING [train.py:1060] (2/4) Exclude cut with ID 3488-85273-0017-41203-0_sp1.1 from training. Duration: 22.4818125
2022-12-23 19:59:50,517 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([2.6444, 3.8350, 3.7111, 1.4988, 3.9304, 3.0230, 1.0177, 2.4795],
       device='cuda:2'), covar=tensor([0.2063, 0.1223, 0.1432, 0.3626, 0.0914, 0.0928, 0.4547, 0.1644],
       device='cuda:2'), in_proj_covar=tensor([0.0149, 0.0144, 0.0160, 0.0125, 0.0147, 0.0116, 0.0146, 0.0117],
       device='cuda:2'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0002, 0.0003, 0.0002, 0.0003, 0.0002],
       device='cuda:2')
2022-12-23 20:00:03,908 WARNING [train.py:1060] (2/4) Exclude cut with ID 4964-30587-0040-44509-0_sp0.9 from training. Duration: 25.0944375
2022-12-23 20:00:04,058 INFO [zipformer.py:660] (2/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=79321.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 20:00:08,107 WARNING [train.py:1060] (2/4) Exclude cut with ID 6533-399-0047-104881-0 from training. Duration: 21.515
2022-12-23 20:00:14,012 WARNING [train.py:1060] (2/4) Exclude cut with ID 7357-94126-0009-15840-0_sp0.9 from training. Duration: 27.02225
2022-12-23 20:00:18,769 WARNING [train.py:1060] (2/4) Exclude cut with ID 432-122774-0010-62480-0_sp0.9 from training. Duration: 22.22225
2022-12-23 20:00:21,903 INFO [zipformer.py:660] (2/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=79333.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 20:00:24,469 INFO [train.py:894] (2/4) Epoch 23, batch 2200, loss[loss=0.1692, simple_loss=0.2458, pruned_loss=0.04627, over 18578.00 frames. ], tot_loss[loss=0.1835, simple_loss=0.2655, pruned_loss=0.05077, over 3714164.92 frames. ], batch size: 49, lr: 4.97e-03, grad_scale: 8.0
2022-12-23 20:00:24,536 WARNING [train.py:1060] (2/4) Exclude cut with ID 4964-30587-0085-44554-0_sp0.9 from training. Duration: 20.85
2022-12-23 20:00:36,433 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 2.368e+02 4.433e+02 5.321e+02 6.846e+02 1.189e+03, threshold=1.064e+03, percent-clipped=4.0
2022-12-23 20:00:41,255 INFO [zipformer.py:660] (2/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=79346.0, num_to_drop=1, layers_to_drop={0}
2022-12-23 20:01:00,221 WARNING [train.py:1060] (2/4) Exclude cut with ID 4295-39940-0007-92567-0 from training. Duration: 21.54
2022-12-23 20:01:03,826 WARNING [train.py:1060] (2/4) Exclude cut with ID 4964-30587-0040-44509-0_sp1.1 from training. Duration: 20.5318125
2022-12-23 20:01:08,206 INFO [zipformer.py:660] (2/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=79364.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 20:01:13,994 WARNING [train.py:1060] (2/4) Exclude cut with ID 7255-291500-0012-134311-0_sp0.9 from training. Duration: 21.9333125
2022-12-23 20:01:24,542 INFO [zipformer.py:660] (2/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=79375.0, num_to_drop=1, layers_to_drop={1}
2022-12-23 20:01:39,057 INFO [train.py:894] (2/4) Epoch 23, batch 2250, loss[loss=0.2111, simple_loss=0.2921, pruned_loss=0.0651, over 18678.00 frames. ], tot_loss[loss=0.1836, simple_loss=0.2653, pruned_loss=0.05099, over 3713369.18 frames. ], batch size: 99, lr: 4.97e-03, grad_scale: 8.0
2022-12-23 20:01:52,937 INFO [zipformer.py:660] (2/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=79394.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 20:02:01,481 WARNING [train.py:1060] (2/4) Exclude cut with ID 8631-249866-0025-130151-0_sp0.9 from training. Duration: 21.7944375
2022-12-23 20:02:15,893 WARNING [train.py:1060] (2/4) Exclude cut with ID 8291-282929-0002-12989-0_sp0.9 from training. Duration: 22.4666875
2022-12-23 20:02:21,762 WARNING [train.py:1060] (2/4) Exclude cut with ID 6121-9014-0076-24124-0 from training. Duration: 21.635
2022-12-23 20:02:28,826 WARNING [train.py:1060] (2/4) Exclude cut with ID 6121-9014-0076-24124-0_sp0.9 from training. Duration: 24.038875
2022-12-23 20:02:39,813 INFO [zipformer.py:660] (2/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=79425.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 20:02:53,955 INFO [train.py:894] (2/4) Epoch 23, batch 2300, loss[loss=0.1854, simple_loss=0.2711, pruned_loss=0.04982, over 18674.00 frames. ], tot_loss[loss=0.1836, simple_loss=0.2652, pruned_loss=0.05097, over 3713499.89 frames. ], batch size: 60, lr: 4.97e-03, grad_scale: 8.0
2022-12-23 20:02:55,761 INFO [zipformer.py:660] (2/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=79436.0, num_to_drop=1, layers_to_drop={1}
2022-12-23 20:03:05,585 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 2.680e+02 4.602e+02 5.590e+02 7.460e+02 2.406e+03, threshold=1.118e+03, percent-clipped=8.0
2022-12-23 20:03:09,863 WARNING [train.py:1060] (2/4) Exclude cut with ID 7255-291500-0009-134308-0_sp1.1 from training. Duration: 21.786375
2022-12-23 20:03:18,715 INFO [scaling.py:679] (2/4) Whitening: num_groups=1, num_channels=384, metric=4.13 vs. limit=5.0
2022-12-23 20:03:24,233 WARNING [train.py:1060] (2/4) Exclude cut with ID 8291-282929-0002-12989-0 from training. Duration: 20.22
2022-12-23 20:04:09,401 INFO [train.py:894] (2/4) Epoch 23, batch 2350, loss[loss=0.1865, simple_loss=0.2746, pruned_loss=0.04924, over 18631.00 frames. ], tot_loss[loss=0.1845, simple_loss=0.2663, pruned_loss=0.05137, over 3714362.10 frames. ], batch size: 78, lr: 4.97e-03, grad_scale: 8.0
2022-12-23 20:05:18,898 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([3.1529, 3.1323, 2.3059, 1.7220, 3.7008, 3.5539, 3.0895, 2.6436],
       device='cuda:2'), covar=tensor([0.0392, 0.0395, 0.0532, 0.0758, 0.0220, 0.0357, 0.0426, 0.0723],
       device='cuda:2'), in_proj_covar=tensor([0.0125, 0.0129, 0.0130, 0.0119, 0.0102, 0.0123, 0.0133, 0.0159],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0001, 0.0001, 0.0001, 0.0002, 0.0002],
       device='cuda:2')
2022-12-23 20:05:21,542 WARNING [train.py:1060] (2/4) Exclude cut with ID 6951-79737-0043-132310-0 from training. Duration: 25.285
2022-12-23 20:05:26,459 INFO [train.py:894] (2/4) Epoch 23, batch 2400, loss[loss=0.1674, simple_loss=0.2637, pruned_loss=0.03556, over 18641.00 frames. ], tot_loss[loss=0.1839, simple_loss=0.2657, pruned_loss=0.05111, over 3713864.78 frames. ], batch size: 53, lr: 4.96e-03, grad_scale: 8.0
2022-12-23 20:05:38,580 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 2.551e+02 4.138e+02 4.947e+02 5.790e+02 8.686e+02, threshold=9.894e+02, percent-clipped=0.0
2022-12-23 20:05:39,425 INFO [scaling.py:679] (2/4) Whitening: num_groups=8, num_channels=96, metric=1.50 vs. limit=2.0
2022-12-23 20:06:27,725 WARNING [train.py:1060] (2/4) Exclude cut with ID 811-130148-0001-63453-0_sp0.9 from training. Duration: 20.861125
2022-12-23 20:06:29,421 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([2.4703, 1.8249, 0.9546, 1.9766, 2.9454, 1.8817, 2.1707, 2.3563],
       device='cuda:2'), covar=tensor([0.1390, 0.1904, 0.2148, 0.1388, 0.1321, 0.1568, 0.1325, 0.1500],
       device='cuda:2'), in_proj_covar=tensor([0.0093, 0.0097, 0.0115, 0.0095, 0.0117, 0.0091, 0.0098, 0.0093],
       device='cuda:2'), out_proj_covar=tensor([0.0003, 0.0003, 0.0004, 0.0003, 0.0004, 0.0003, 0.0003, 0.0003],
       device='cuda:2')
2022-12-23 20:06:42,120 INFO [train.py:894] (2/4) Epoch 23, batch 2450, loss[loss=0.1808, simple_loss=0.2658, pruned_loss=0.04787, over 18577.00 frames. ], tot_loss[loss=0.1823, simple_loss=0.2641, pruned_loss=0.05024, over 3713644.84 frames. ], batch size: 69, lr: 4.96e-03, grad_scale: 8.0
2022-12-23 20:06:42,532 INFO [zipformer.py:660] (2/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=79585.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 20:06:51,635 WARNING [train.py:1060] (2/4) Exclude cut with ID 6010-56788-0055-90261-0 from training. Duration: 20.88
2022-12-23 20:07:22,969 WARNING [train.py:1060] (2/4) Exclude cut with ID 7357-94126-0045-15876-0_sp0.9 from training. Duration: 23.4166875
2022-12-23 20:07:37,223 INFO [zipformer.py:660] (2/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=79621.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 20:07:42,225 INFO [scaling.py:679] (2/4) Whitening: num_groups=8, num_channels=192, metric=1.54 vs. limit=2.0
2022-12-23 20:07:47,869 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([2.2472, 1.6231, 2.3369, 4.4784, 3.4348, 2.7720, 0.9423, 3.3748],
       device='cuda:2'), covar=tensor([0.1596, 0.1516, 0.1519, 0.0573, 0.0785, 0.1159, 0.2099, 0.0716],
       device='cuda:2'), in_proj_covar=tensor([0.0104, 0.0118, 0.0137, 0.0153, 0.0107, 0.0144, 0.0130, 0.0114],
       device='cuda:2'), out_proj_covar=tensor([0.0003, 0.0003, 0.0004, 0.0004, 0.0003, 0.0004, 0.0004, 0.0003],
       device='cuda:2')
2022-12-23 20:07:58,606 INFO [train.py:894] (2/4) Epoch 23, batch 2500, loss[loss=0.1687, simple_loss=0.2525, pruned_loss=0.04239, over 18618.00 frames. ], tot_loss[loss=0.1822, simple_loss=0.2643, pruned_loss=0.05006, over 3714724.91 frames. ], batch size: 69, lr: 4.96e-03, grad_scale: 8.0
2022-12-23 20:08:08,449 INFO [zipformer.py:660] (2/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=79641.0, num_to_drop=1, layers_to_drop={3}
2022-12-23 20:08:11,485 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 2.617e+02 4.368e+02 5.588e+02 6.657e+02 1.417e+03, threshold=1.118e+03, percent-clipped=5.0
2022-12-23 20:08:16,432 INFO [zipformer.py:660] (2/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=79646.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 20:08:32,498 INFO [zipformer.py:660] (2/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=79657.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 20:08:35,894 INFO [scaling.py:679] (2/4) Whitening: num_groups=1, num_channels=384, metric=2.89 vs. limit=5.0
2022-12-23 20:08:36,470 WARNING [train.py:1060] (2/4) Exclude cut with ID 7255-291500-0006-134305-0 from training. Duration: 21.24
2022-12-23 20:08:36,483 WARNING [train.py:1060] (2/4) Exclude cut with ID 6533-399-0047-104881-0_sp0.9 from training. Duration: 23.9055625
2022-12-23 20:08:49,542 INFO [zipformer.py:660] (2/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=79669.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 20:09:10,155 WARNING [train.py:1060] (2/4) Exclude cut with ID 6758-72288-0033-108368-0_sp0.9 from training. Duration: 25.988875
2022-12-23 20:09:12,911 INFO [train.py:894] (2/4) Epoch 23, batch 2550, loss[loss=0.2012, simple_loss=0.2748, pruned_loss=0.06383, over 18478.00 frames. ], tot_loss[loss=0.1832, simple_loss=0.2651, pruned_loss=0.05067, over 3713645.74 frames. ], batch size: 54, lr: 4.96e-03, grad_scale: 8.0
2022-12-23 20:09:19,558 WARNING [train.py:1060] (2/4) Exclude cut with ID 7255-291500-0001-134300-0_sp0.9 from training. Duration: 20.67225
2022-12-23 20:09:19,701 INFO [zipformer.py:660] (2/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=79689.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 20:09:46,137 INFO [scaling.py:679] (2/4) Whitening: num_groups=8, num_channels=192, metric=1.52 vs. limit=2.0
2022-12-23 20:10:04,293 INFO [zipformer.py:660] (2/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=79718.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 20:10:06,864 INFO [zipformer.py:660] (2/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=79720.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 20:10:08,135 WARNING [train.py:1060] (2/4) Exclude cut with ID 3488-85273-0038-41224-0 from training. Duration: 20.34
2022-12-23 20:10:23,151 INFO [zipformer.py:660] (2/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=79731.0, num_to_drop=1, layers_to_drop={2}
2022-12-23 20:10:29,805 INFO [train.py:894] (2/4) Epoch 23, batch 2600, loss[loss=0.1817, simple_loss=0.2688, pruned_loss=0.04727, over 18595.00 frames. ], tot_loss[loss=0.1821, simple_loss=0.2643, pruned_loss=0.05001, over 3713037.92 frames. ], batch size: 57, lr: 4.96e-03, grad_scale: 8.0
2022-12-23 20:10:35,325 INFO [scaling.py:679] (2/4) Whitening: num_groups=8, num_channels=96, metric=1.31 vs. limit=2.0
2022-12-23 20:10:41,467 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 2.544e+02 4.012e+02 4.948e+02 6.016e+02 1.385e+03, threshold=9.896e+02, percent-clipped=1.0
2022-12-23 20:11:04,977 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([2.4121, 1.5871, 2.0803, 1.9854, 2.3063, 2.3215, 2.1755, 1.9627],
       device='cuda:2'), covar=tensor([0.2095, 0.3349, 0.2456, 0.2648, 0.2023, 0.0947, 0.3194, 0.1339],
       device='cuda:2'), in_proj_covar=tensor([0.0272, 0.0302, 0.0284, 0.0321, 0.0313, 0.0256, 0.0348, 0.0245],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-23 20:11:10,438 INFO [zipformer.py:660] (2/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=79762.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 20:11:22,897 WARNING [train.py:1060] (2/4) Exclude cut with ID 7357-94126-0026-15857-0_sp0.9 from training. Duration: 25.061125
2022-12-23 20:11:33,651 WARNING [train.py:1060] (2/4) Exclude cut with ID 3033-130750-0096-55598-0 from training. Duration: 0.83
2022-12-23 20:11:35,932 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.7011, 2.1212, 1.6454, 2.3294, 2.5479, 1.7278, 1.7840, 1.4533],
       device='cuda:2'), covar=tensor([0.1987, 0.1636, 0.1642, 0.1044, 0.1327, 0.1104, 0.2141, 0.1576],
       device='cuda:2'), in_proj_covar=tensor([0.0251, 0.0230, 0.0218, 0.0203, 0.0264, 0.0198, 0.0228, 0.0202],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-23 20:11:44,508 INFO [train.py:894] (2/4) Epoch 23, batch 2650, loss[loss=0.2074, simple_loss=0.2947, pruned_loss=0.06006, over 18568.00 frames. ], tot_loss[loss=0.1817, simple_loss=0.2637, pruned_loss=0.04981, over 3714029.81 frames. ], batch size: 57, lr: 4.96e-03, grad_scale: 8.0
2022-12-23 20:11:58,507 WARNING [train.py:1060] (2/4) Exclude cut with ID 3488-85273-0017-41203-0 from training. Duration: 24.73
2022-12-23 20:12:12,108 WARNING [train.py:1060] (2/4) Exclude cut with ID 7255-291500-0009-134308-0 from training. Duration: 23.965
2022-12-23 20:12:20,576 WARNING [train.py:1060] (2/4) Exclude cut with ID 6978-92210-0030-146996-0_sp0.9 from training. Duration: 22.088875
2022-12-23 20:12:36,661 WARNING [train.py:1060] (2/4) Exclude cut with ID 7255-291500-0006-134305-0_sp0.9 from training. Duration: 23.6
2022-12-23 20:12:38,315 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.7917, 1.3332, 0.6263, 1.4558, 2.1653, 1.1753, 1.5017, 1.6534],
       device='cuda:2'), covar=tensor([0.1571, 0.2072, 0.2336, 0.1404, 0.1639, 0.1725, 0.1473, 0.1717],
       device='cuda:2'), in_proj_covar=tensor([0.0093, 0.0098, 0.0116, 0.0096, 0.0118, 0.0091, 0.0099, 0.0094],
       device='cuda:2'), out_proj_covar=tensor([0.0003, 0.0003, 0.0004, 0.0003, 0.0004, 0.0003, 0.0003, 0.0003],
       device='cuda:2')
2022-12-23 20:12:41,162 INFO [zipformer.py:660] (2/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=79823.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 20:13:00,066 INFO [train.py:894] (2/4) Epoch 23, batch 2700, loss[loss=0.1679, simple_loss=0.2555, pruned_loss=0.04012, over 18447.00 frames. ], tot_loss[loss=0.1816, simple_loss=0.2639, pruned_loss=0.04962, over 3713123.14 frames. ], batch size: 50, lr: 4.95e-03, grad_scale: 8.0
2022-12-23 20:13:12,606 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 2.225e+02 4.087e+02 5.258e+02 6.716e+02 1.165e+03, threshold=1.052e+03, percent-clipped=3.0
2022-12-23 20:13:55,411 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.7252, 1.5487, 1.5962, 1.8828, 1.8152, 3.5657, 1.7569, 1.6089],
       device='cuda:2'), covar=tensor([0.0854, 0.1832, 0.1165, 0.0973, 0.1416, 0.0269, 0.1351, 0.1621],
       device='cuda:2'), in_proj_covar=tensor([0.0072, 0.0081, 0.0072, 0.0074, 0.0091, 0.0076, 0.0084, 0.0077],
       device='cuda:2'), out_proj_covar=tensor([0.0004, 0.0004, 0.0004, 0.0004, 0.0004, 0.0003, 0.0004, 0.0004],
       device='cuda:2')
2022-12-23 20:14:16,439 INFO [train.py:894] (2/4) Epoch 23, batch 2750, loss[loss=0.2, simple_loss=0.2848, pruned_loss=0.0576, over 18551.00 frames. ], tot_loss[loss=0.1814, simple_loss=0.2639, pruned_loss=0.04941, over 3713661.07 frames. ], batch size: 55, lr: 4.95e-03, grad_scale: 8.0
2022-12-23 20:14:17,066 WARNING [train.py:1060] (2/4) Exclude cut with ID 8291-282929-0024-13011-0 from training. Duration: 23.795
2022-12-23 20:14:33,584 WARNING [train.py:1060] (2/4) Exclude cut with ID 8631-249866-0030-130156-0_sp1.1 from training. Duration: 21.5409375
2022-12-23 20:14:36,510 WARNING [train.py:1060] (2/4) Exclude cut with ID 6978-92210-0019-146985-0_sp0.9 from training. Duration: 24.97775
2022-12-23 20:14:46,780 WARNING [train.py:1060] (2/4) Exclude cut with ID 1085-156170-0017-128270-0_sp0.9 from training. Duration: 23.3444375
2022-12-23 20:15:01,998 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([3.8694, 3.3575, 3.3834, 3.7912, 3.4788, 3.3383, 3.9763, 1.3542],
       device='cuda:2'), covar=tensor([0.0878, 0.0864, 0.0753, 0.0877, 0.1642, 0.1405, 0.0775, 0.4888],
       device='cuda:2'), in_proj_covar=tensor([0.0345, 0.0229, 0.0238, 0.0276, 0.0327, 0.0269, 0.0293, 0.0286],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0001, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-23 20:15:02,183 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.4121, 1.2250, 1.3516, 1.2814, 1.7263, 1.5437, 1.5224, 1.1574],
       device='cuda:2'), covar=tensor([0.0291, 0.0240, 0.0468, 0.0214, 0.0174, 0.0352, 0.0261, 0.0305],
       device='cuda:2'), in_proj_covar=tensor([0.0096, 0.0127, 0.0153, 0.0125, 0.0118, 0.0122, 0.0099, 0.0128],
       device='cuda:2'), out_proj_covar=tensor([7.5936e-05, 1.0081e-04, 1.2600e-04, 9.9509e-05, 9.4988e-05, 9.3696e-05,
        7.7414e-05, 1.0103e-04], device='cuda:2')
2022-12-23 20:15:16,337 WARNING [train.py:1060] (2/4) Exclude cut with ID 6010-56788-0055-90261-0_sp0.9 from training. Duration: 23.2
2022-12-23 20:15:22,234 WARNING [train.py:1060] (2/4) Exclude cut with ID 5653-46179-0060-117930-0_sp0.9 from training. Duration: 21.17225
2022-12-23 20:15:24,157 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.9951, 1.8894, 1.5270, 1.5967, 1.6997, 1.7821, 1.6923, 1.8272],
       device='cuda:2'), covar=tensor([0.2155, 0.2763, 0.1954, 0.2466, 0.3110, 0.1139, 0.2621, 0.0998],
       device='cuda:2'), in_proj_covar=tensor([0.0298, 0.0294, 0.0247, 0.0349, 0.0274, 0.0229, 0.0291, 0.0217],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0003, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-23 20:15:24,540 INFO [scaling.py:679] (2/4) Whitening: num_groups=1, num_channels=384, metric=4.22 vs. limit=5.0
2022-12-23 20:15:32,913 INFO [train.py:894] (2/4) Epoch 23, batch 2800, loss[loss=0.1737, simple_loss=0.2607, pruned_loss=0.04337, over 18522.00 frames. ], tot_loss[loss=0.1816, simple_loss=0.2641, pruned_loss=0.04955, over 3713996.70 frames. ], batch size: 58, lr: 4.95e-03, grad_scale: 8.0
2022-12-23 20:15:40,672 WARNING [train.py:1060] (2/4) Exclude cut with ID 7255-291500-0008-134307-0_sp0.9 from training. Duration: 24.6555625
2022-12-23 20:15:42,339 INFO [zipformer.py:660] (2/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=79941.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 20:15:42,555 INFO [zipformer.py:660] (2/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=79941.0, num_to_drop=1, layers_to_drop={1}
2022-12-23 20:15:45,136 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 2.473e+02 3.775e+02 4.575e+02 5.641e+02 1.095e+03, threshold=9.150e+02, percent-clipped=1.0
2022-12-23 20:16:36,471 WARNING [train.py:1060] (2/4) Exclude cut with ID 3488-65654-0031-41259-0 from training. Duration: 20.44
2022-12-23 20:16:49,083 INFO [train.py:894] (2/4) Epoch 23, batch 2850, loss[loss=0.1453, simple_loss=0.2272, pruned_loss=0.03167, over 18459.00 frames. ], tot_loss[loss=0.1811, simple_loss=0.2636, pruned_loss=0.04926, over 3714384.36 frames. ], batch size: 43, lr: 4.95e-03, grad_scale: 8.0
2022-12-23 20:16:49,460 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.8541, 1.1869, 0.7124, 1.3847, 2.2272, 1.2580, 1.6073, 1.7626],
       device='cuda:2'), covar=tensor([0.1584, 0.2280, 0.2356, 0.1545, 0.1722, 0.1844, 0.1518, 0.1709],
       device='cuda:2'), in_proj_covar=tensor([0.0093, 0.0098, 0.0116, 0.0096, 0.0118, 0.0092, 0.0099, 0.0094],
       device='cuda:2'), out_proj_covar=tensor([0.0003, 0.0003, 0.0004, 0.0003, 0.0004, 0.0003, 0.0003, 0.0003],
       device='cuda:2')
2022-12-23 20:16:53,646 WARNING [train.py:1060] (2/4) Exclude cut with ID 6951-79737-0018-132285-0_sp0.9 from training. Duration: 23.45
2022-12-23 20:16:55,232 INFO [zipformer.py:660] (2/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=79989.0, num_to_drop=1, layers_to_drop={1}
2022-12-23 20:16:55,254 INFO [zipformer.py:660] (2/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=79989.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 20:17:24,438 WARNING [train.py:1060] (2/4) Exclude cut with ID 6945-60535-0076-12784-0_sp0.9 from training. Duration: 20.52225
2022-12-23 20:17:32,286 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([4.2296, 3.6676, 3.7821, 4.1086, 3.9089, 3.7483, 4.2553, 2.3405],
       device='cuda:2'), covar=tensor([0.0672, 0.0659, 0.0609, 0.0840, 0.1081, 0.1036, 0.0773, 0.3629],
       device='cuda:2'), in_proj_covar=tensor([0.0344, 0.0228, 0.0238, 0.0276, 0.0327, 0.0269, 0.0292, 0.0286],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0001, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-23 20:17:33,614 WARNING [train.py:1060] (2/4) Exclude cut with ID 7255-291500-0008-134307-0 from training. Duration: 22.19
2022-12-23 20:17:35,312 INFO [zipformer.py:660] (2/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=80013.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 20:17:43,621 WARNING [train.py:1060] (2/4) Exclude cut with ID 7357-94126-0014-15845-0_sp1.1 from training. Duration: 25.3818125
2022-12-23 20:17:45,323 INFO [zipformer.py:660] (2/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=80020.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 20:18:00,187 WARNING [train.py:1060] (2/4) Exclude cut with ID 6951-79737-0043-132310-0_sp0.9 from training. Duration: 28.0944375
2022-12-23 20:18:02,005 INFO [zipformer.py:660] (2/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=80031.0, num_to_drop=1, layers_to_drop={0}
2022-12-23 20:18:02,148 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([0.3679, 1.7397, 1.9067, 1.0723, 1.2267, 2.0533, 1.8764, 1.6776],
       device='cuda:2'), covar=tensor([0.0774, 0.0326, 0.0344, 0.0356, 0.0409, 0.0467, 0.0249, 0.0696],
       device='cuda:2'), in_proj_covar=tensor([0.0149, 0.0170, 0.0129, 0.0141, 0.0149, 0.0144, 0.0168, 0.0176],
       device='cuda:2'), out_proj_covar=tensor([1.1293e-04, 1.2896e-04, 9.6340e-05, 1.0426e-04, 1.0985e-04, 1.0976e-04,
        1.2798e-04, 1.3305e-04], device='cuda:2')
2022-12-23 20:18:05,972 WARNING [train.py:1060] (2/4) Exclude cut with ID 2195-150901-0045-59933-0_sp0.9 from training. Duration: 22.9444375
2022-12-23 20:18:07,431 INFO [train.py:894] (2/4) Epoch 23, batch 2900, loss[loss=0.1807, simple_loss=0.261, pruned_loss=0.05018, over 18532.00 frames. ], tot_loss[loss=0.1804, simple_loss=0.2626, pruned_loss=0.04916, over 3714193.14 frames. ], batch size: 47, lr: 4.95e-03, grad_scale: 8.0
2022-12-23 20:18:10,371 INFO [zipformer.py:660] (2/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=80037.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 20:18:13,146 WARNING [train.py:1060] (2/4) Exclude cut with ID 8291-282929-0024-13011-0_sp1.1 from training. Duration: 21.6318125
2022-12-23 20:18:18,881 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 2.810e+02 4.423e+02 5.143e+02 6.532e+02 1.388e+03, threshold=1.029e+03, percent-clipped=7.0
2022-12-23 20:18:29,724 WARNING [train.py:1060] (2/4) Exclude cut with ID 8631-249866-0030-130156-0 from training. Duration: 23.695
2022-12-23 20:18:56,098 WARNING [train.py:1060] (2/4) Exclude cut with ID 7699-105389-0094-26379-0 from training. Duration: 23.955
2022-12-23 20:18:57,666 INFO [zipformer.py:660] (2/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=80068.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 20:19:13,656 INFO [zipformer.py:660] (2/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=80079.0, num_to_drop=1, layers_to_drop={1}
2022-12-23 20:19:22,080 INFO [train.py:894] (2/4) Epoch 23, batch 2950, loss[loss=0.1691, simple_loss=0.2596, pruned_loss=0.03929, over 18622.00 frames. ], tot_loss[loss=0.181, simple_loss=0.2633, pruned_loss=0.04938, over 3714218.12 frames. ], batch size: 53, lr: 4.95e-03, grad_scale: 8.0
2022-12-23 20:19:26,449 WARNING [train.py:1060] (2/4) Exclude cut with ID 8291-282929-0024-13011-0_sp0.9 from training. Duration: 26.438875
2022-12-23 20:20:10,767 WARNING [train.py:1060] (2/4) Exclude cut with ID 7699-105389-0021-26306-0_sp0.9 from training. Duration: 21.2444375
2022-12-23 20:20:12,024 WARNING [train.py:1060] (2/4) Exclude cut with ID 7357-94126-0014-15845-0_sp0.9 from training. Duration: 31.02225
2022-12-23 20:20:12,197 INFO [zipformer.py:660] (2/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=80118.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 20:20:19,163 WARNING [train.py:1060] (2/4) Exclude cut with ID 432-122774-0017-62487-0 from training. Duration: 22.395
2022-12-23 20:20:36,798 INFO [train.py:894] (2/4) Epoch 23, batch 3000, loss[loss=0.1912, simple_loss=0.2757, pruned_loss=0.05338, over 18468.00 frames. ], tot_loss[loss=0.1818, simple_loss=0.2642, pruned_loss=0.04968, over 3713942.50 frames. ], batch size: 54, lr: 4.95e-03, grad_scale: 8.0
2022-12-23 20:20:36,799 INFO [train.py:919] (2/4) Computing validation loss
2022-12-23 20:20:47,841 INFO [train.py:928] (2/4) Epoch 23, validation: loss=0.1641, simple_loss=0.261, pruned_loss=0.03364, over 944034.00 frames. 
2022-12-23 20:20:47,841 INFO [train.py:929] (2/4) Maximum memory allocated so far is 24680MB
2022-12-23 20:20:47,888 WARNING [train.py:1060] (2/4) Exclude cut with ID 7357-94126-0045-15876-0 from training. Duration: 21.075
2022-12-23 20:20:52,147 WARNING [train.py:1060] (2/4) Exclude cut with ID 6482-98857-0025-147532-0_sp0.9 from training. Duration: 20.0055625
2022-12-23 20:20:52,154 WARNING [train.py:1060] (2/4) Exclude cut with ID 6951-79737-0037-132304-0_sp0.9 from training. Duration: 22.05
2022-12-23 20:20:52,166 WARNING [train.py:1060] (2/4) Exclude cut with ID 7255-291500-0003-134302-0 from training. Duration: 26.8349375
2022-12-23 20:20:55,180 WARNING [train.py:1060] (2/4) Exclude cut with ID 7357-94126-0009-15840-0_sp1.1 from training. Duration: 22.1090625
2022-12-23 20:21:00,209 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 3.015e+02 4.784e+02 5.536e+02 6.733e+02 1.386e+03, threshold=1.107e+03, percent-clipped=4.0
2022-12-23 20:21:03,074 WARNING [train.py:1060] (2/4) Exclude cut with ID 7699-105389-0094-26379-0_sp0.9 from training. Duration: 26.6166875
2022-12-23 20:21:03,480 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.0680, 0.9967, 0.9696, 1.1340, 1.2932, 1.1836, 1.1189, 0.9970],
       device='cuda:2'), covar=tensor([0.0299, 0.0242, 0.0519, 0.0210, 0.0222, 0.0384, 0.0241, 0.0306],
       device='cuda:2'), in_proj_covar=tensor([0.0096, 0.0128, 0.0154, 0.0126, 0.0118, 0.0122, 0.0100, 0.0128],
       device='cuda:2'), out_proj_covar=tensor([7.6303e-05, 1.0122e-04, 1.2613e-04, 1.0032e-04, 9.5421e-05, 9.3896e-05,
        7.7714e-05, 1.0089e-04], device='cuda:2')
2022-12-23 20:21:21,868 WARNING [train.py:1060] (2/4) Exclude cut with ID 2046-178027-0000-53705-0_sp0.9 from training. Duration: 20.3055625
2022-12-23 20:21:42,185 WARNING [train.py:1060] (2/4) Exclude cut with ID 7205-50138-0008-5373-0_sp0.9 from training. Duration: 20.7
2022-12-23 20:22:02,876 INFO [train.py:894] (2/4) Epoch 23, batch 3050, loss[loss=0.201, simple_loss=0.278, pruned_loss=0.062, over 18505.00 frames. ], tot_loss[loss=0.1822, simple_loss=0.2647, pruned_loss=0.04988, over 3714415.89 frames. ], batch size: 64, lr: 4.94e-03, grad_scale: 16.0
2022-12-23 20:22:28,917 WARNING [train.py:1060] (2/4) Exclude cut with ID 6978-92210-0019-146985-0 from training. Duration: 22.48
2022-12-23 20:22:37,803 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.5763, 1.4780, 1.5324, 1.8418, 1.7464, 3.4968, 1.4568, 1.6905],
       device='cuda:2'), covar=tensor([0.0886, 0.1754, 0.1075, 0.0961, 0.1433, 0.0258, 0.1440, 0.1454],
       device='cuda:2'), in_proj_covar=tensor([0.0072, 0.0081, 0.0072, 0.0074, 0.0091, 0.0076, 0.0084, 0.0076],
       device='cuda:2'), out_proj_covar=tensor([0.0004, 0.0004, 0.0004, 0.0004, 0.0004, 0.0003, 0.0004, 0.0004],
       device='cuda:2')
2022-12-23 20:22:43,475 WARNING [train.py:1060] (2/4) Exclude cut with ID 7255-291500-0003-134302-0_sp0.9 from training. Duration: 29.816625
2022-12-23 20:22:51,235 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.7181, 1.6863, 1.7766, 1.6332, 1.2180, 3.4220, 1.5956, 2.0230],
       device='cuda:2'), covar=tensor([0.3174, 0.1974, 0.1866, 0.2073, 0.1474, 0.0219, 0.1500, 0.0787],
       device='cuda:2'), in_proj_covar=tensor([0.0132, 0.0118, 0.0125, 0.0121, 0.0105, 0.0097, 0.0091, 0.0090],
       device='cuda:2'), out_proj_covar=tensor([0.0005, 0.0005, 0.0005, 0.0005, 0.0004, 0.0003, 0.0004, 0.0004],
       device='cuda:2')
2022-12-23 20:23:04,917 WARNING [train.py:1060] (2/4) Exclude cut with ID 7255-291500-0005-134304-0_sp1.1 from training. Duration: 22.7590625
2022-12-23 20:23:06,772 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.6030, 1.5152, 1.4608, 1.4001, 1.8566, 1.7257, 1.7529, 1.2762],
       device='cuda:2'), covar=tensor([0.0315, 0.0254, 0.0484, 0.0230, 0.0198, 0.0398, 0.0270, 0.0332],
       device='cuda:2'), in_proj_covar=tensor([0.0096, 0.0128, 0.0154, 0.0126, 0.0118, 0.0123, 0.0100, 0.0128],
       device='cuda:2'), out_proj_covar=tensor([7.6430e-05, 1.0132e-04, 1.2609e-04, 9.9994e-05, 9.5214e-05, 9.4102e-05,
        7.8012e-05, 1.0116e-04], device='cuda:2')
2022-12-23 20:23:10,821 WARNING [train.py:1060] (2/4) Exclude cut with ID 7357-94126-0026-15857-0 from training. Duration: 22.555
2022-12-23 20:23:19,538 INFO [train.py:894] (2/4) Epoch 23, batch 3100, loss[loss=0.1925, simple_loss=0.2823, pruned_loss=0.05129, over 18568.00 frames. ], tot_loss[loss=0.1819, simple_loss=0.2646, pruned_loss=0.04964, over 3714006.66 frames. ], batch size: 69, lr: 4.94e-03, grad_scale: 16.0
2022-12-23 20:23:29,229 INFO [zipformer.py:660] (2/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=80241.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 20:23:32,360 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 2.802e+02 4.142e+02 4.904e+02 7.040e+02 2.138e+03, threshold=9.809e+02, percent-clipped=3.0
2022-12-23 20:23:32,378 WARNING [train.py:1060] (2/4) Exclude cut with ID 1250-135782-0005-25975-0_sp0.9 from training. Duration: 21.688875
2022-12-23 20:23:59,481 INFO [scaling.py:679] (2/4) Whitening: num_groups=1, num_channels=384, metric=2.99 vs. limit=5.0
2022-12-23 20:24:04,783 WARNING [train.py:1060] (2/4) Exclude cut with ID 3488-85273-0038-41224-0_sp0.9 from training. Duration: 22.6
2022-12-23 20:24:05,547 INFO [scaling.py:679] (2/4) Whitening: num_groups=1, num_channels=384, metric=2.77 vs. limit=5.0
2022-12-23 20:24:32,739 INFO [zipformer.py:660] (2/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=80283.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 20:24:35,764 INFO [train.py:894] (2/4) Epoch 23, batch 3150, loss[loss=0.2139, simple_loss=0.2901, pruned_loss=0.06883, over 18626.00 frames. ], tot_loss[loss=0.1833, simple_loss=0.2657, pruned_loss=0.05041, over 3713470.40 frames. ], batch size: 53, lr: 4.94e-03, grad_scale: 16.0
2022-12-23 20:24:41,664 INFO [zipformer.py:660] (2/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=80289.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 20:24:43,339 WARNING [train.py:1060] (2/4) Exclude cut with ID 7357-94126-0009-15840-0 from training. Duration: 24.32
2022-12-23 20:25:06,417 INFO [scaling.py:679] (2/4) Whitening: num_groups=8, num_channels=96, metric=1.37 vs. limit=2.0
2022-12-23 20:25:17,180 INFO [zipformer.py:660] (2/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=80313.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 20:25:40,389 WARNING [train.py:1060] (2/4) Exclude cut with ID 8291-276745-0093-13116-0_sp0.9 from training. Duration: 21.061125
2022-12-23 20:25:51,249 INFO [train.py:894] (2/4) Epoch 23, batch 3200, loss[loss=0.1801, simple_loss=0.2636, pruned_loss=0.04836, over 18509.00 frames. ], tot_loss[loss=0.1826, simple_loss=0.2652, pruned_loss=0.05001, over 3713114.04 frames. ], batch size: 64, lr: 4.94e-03, grad_scale: 16.0
2022-12-23 20:25:54,289 WARNING [train.py:1060] (2/4) Exclude cut with ID 7357-94126-0024-15855-0_sp0.9 from training. Duration: 20.32225
2022-12-23 20:26:03,086 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 2.560e+02 4.270e+02 5.295e+02 6.467e+02 1.258e+03, threshold=1.059e+03, percent-clipped=3.0
2022-12-23 20:26:04,942 INFO [zipformer.py:660] (2/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=80344.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 20:26:06,013 WARNING [train.py:1060] (2/4) Exclude cut with ID 3033-130750-0096-55598-0_sp1.1 from training. Duration: 0.7545625
2022-12-23 20:26:20,736 WARNING [train.py:1060] (2/4) Exclude cut with ID 4295-39940-0007-92567-0_sp0.9 from training. Duration: 23.9333125
2022-12-23 20:26:29,508 INFO [zipformer.py:660] (2/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=80361.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 20:26:57,704 WARNING [train.py:1060] (2/4) Exclude cut with ID 7255-291500-0008-134307-0_sp1.1 from training. Duration: 20.17275
2022-12-23 20:27:01,629 INFO [scaling.py:679] (2/4) Whitening: num_groups=8, num_channels=96, metric=1.19 vs. limit=2.0
2022-12-23 20:27:02,423 WARNING [train.py:1060] (2/4) Exclude cut with ID 6978-92210-0019-146985-0_sp1.1 from training. Duration: 20.436375
2022-12-23 20:27:06,852 INFO [train.py:894] (2/4) Epoch 23, batch 3250, loss[loss=0.2067, simple_loss=0.2849, pruned_loss=0.06418, over 18672.00 frames. ], tot_loss[loss=0.1815, simple_loss=0.2642, pruned_loss=0.04936, over 3714347.88 frames. ], batch size: 179, lr: 4.94e-03, grad_scale: 16.0
2022-12-23 20:27:19,175 INFO [zipformer.py:660] (2/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=80393.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 20:27:57,466 INFO [zipformer.py:660] (2/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=80418.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 20:28:21,782 WARNING [train.py:1060] (2/4) Exclude cut with ID 4234-40345-0022-142709-0_sp0.9 from training. Duration: 23.1055625
2022-12-23 20:28:23,432 INFO [train.py:894] (2/4) Epoch 23, batch 3300, loss[loss=0.1733, simple_loss=0.2683, pruned_loss=0.03913, over 18724.00 frames. ], tot_loss[loss=0.1803, simple_loss=0.2629, pruned_loss=0.04888, over 3713819.99 frames. ], batch size: 54, lr: 4.94e-03, grad_scale: 16.0
2022-12-23 20:28:24,870 WARNING [train.py:1060] (2/4) Exclude cut with ID 8291-282929-0007-12994-0_sp1.1 from training. Duration: 23.5
2022-12-23 20:28:35,117 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 3.089e+02 4.153e+02 5.096e+02 6.144e+02 1.740e+03, threshold=1.019e+03, percent-clipped=6.0
2022-12-23 20:28:36,531 WARNING [train.py:1060] (2/4) Exclude cut with ID 7255-291500-0009-134308-0_sp0.9 from training. Duration: 26.62775
2022-12-23 20:28:49,867 WARNING [train.py:1060] (2/4) Exclude cut with ID 6951-79737-0018-132285-0 from training. Duration: 21.105
2022-12-23 20:28:51,437 INFO [zipformer.py:660] (2/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=80454.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 20:28:52,481 WARNING [train.py:1060] (2/4) Exclude cut with ID 4511-76322-0006-80011-0_sp0.9 from training. Duration: 24.411125
2022-12-23 20:29:04,936 INFO [zipformer.py:660] (2/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=80463.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 20:29:08,675 INFO [zipformer.py:660] (2/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=80466.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 20:29:20,825 WARNING [train.py:1060] (2/4) Exclude cut with ID 6758-72288-0033-108368-0_sp1.1 from training. Duration: 21.263625
2022-12-23 20:29:37,277 INFO [train.py:894] (2/4) Epoch 23, batch 3350, loss[loss=0.2139, simple_loss=0.289, pruned_loss=0.06944, over 18585.00 frames. ], tot_loss[loss=0.1804, simple_loss=0.2631, pruned_loss=0.04887, over 3714792.19 frames. ], batch size: 51, lr: 4.94e-03, grad_scale: 16.0
2022-12-23 20:29:50,872 WARNING [train.py:1060] (2/4) Exclude cut with ID 4234-40345-0022-142709-0 from training. Duration: 20.795
2022-12-23 20:29:51,197 INFO [zipformer.py:660] (2/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=80494.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 20:30:01,110 WARNING [train.py:1060] (2/4) Exclude cut with ID 7357-94126-0021-15852-0 from training. Duration: 24.76
2022-12-23 20:30:01,125 WARNING [train.py:1060] (2/4) Exclude cut with ID 3867-173237-0077-144769-0_sp0.9 from training. Duration: 22.25
2022-12-23 20:30:28,399 WARNING [train.py:1060] (2/4) Exclude cut with ID 7357-94126-0026-15857-0_sp1.1 from training. Duration: 20.5045625
2022-12-23 20:30:36,498 INFO [zipformer.py:660] (2/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=80524.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 20:30:53,194 INFO [train.py:894] (2/4) Epoch 23, batch 3400, loss[loss=0.1599, simple_loss=0.2485, pruned_loss=0.03563, over 18466.00 frames. ], tot_loss[loss=0.1802, simple_loss=0.263, pruned_loss=0.04872, over 3715420.63 frames. ], batch size: 50, lr: 4.93e-03, grad_scale: 8.0
2022-12-23 20:31:06,191 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 2.766e+02 4.348e+02 5.300e+02 6.386e+02 1.214e+03, threshold=1.060e+03, percent-clipped=1.0
2022-12-23 20:31:23,070 INFO [zipformer.py:660] (2/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=80555.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 20:32:04,416 INFO [train.py:894] (2/4) Epoch 23, batch 3450, loss[loss=0.1463, simple_loss=0.222, pruned_loss=0.03529, over 18490.00 frames. ], tot_loss[loss=0.1801, simple_loss=0.2631, pruned_loss=0.04857, over 3716175.01 frames. ], batch size: 43, lr: 4.93e-03, grad_scale: 8.0
2022-12-23 20:33:19,549 INFO [train.py:894] (2/4) Epoch 23, batch 3500, loss[loss=0.2016, simple_loss=0.2821, pruned_loss=0.06055, over 18590.00 frames. ], tot_loss[loss=0.1808, simple_loss=0.2636, pruned_loss=0.04899, over 3717034.74 frames. ], batch size: 173, lr: 4.93e-03, grad_scale: 8.0
2022-12-23 20:33:26,372 INFO [zipformer.py:660] (2/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=80639.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 20:33:40,546 WARNING [train.py:1060] (2/4) Exclude cut with ID 7859-102521-0017-7548-0_sp1.1 from training. Duration: 22.2954375
2022-12-23 20:33:51,323 INFO [train.py:894] (2/4) Epoch 24, batch 0, loss[loss=0.1799, simple_loss=0.2674, pruned_loss=0.04624, over 18601.00 frames. ], tot_loss[loss=0.1799, simple_loss=0.2674, pruned_loss=0.04624, over 18601.00 frames. ], batch size: 56, lr: 4.82e-03, grad_scale: 8.0
2022-12-23 20:33:51,323 INFO [train.py:919] (2/4) Computing validation loss
2022-12-23 20:34:02,130 INFO [train.py:928] (2/4) Epoch 24, validation: loss=0.1635, simple_loss=0.2609, pruned_loss=0.03309, over 944034.00 frames. 
2022-12-23 20:34:02,131 INFO [train.py:929] (2/4) Maximum memory allocated so far is 24680MB
2022-12-23 20:34:06,644 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 2.712e+02 4.449e+02 5.372e+02 6.832e+02 1.472e+03, threshold=1.074e+03, percent-clipped=3.0
2022-12-23 20:34:30,439 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.3942, 1.2870, 1.2712, 1.2589, 1.6738, 1.4750, 1.4909, 1.1103],
       device='cuda:2'), covar=tensor([0.0356, 0.0266, 0.0585, 0.0250, 0.0217, 0.0464, 0.0298, 0.0378],
       device='cuda:2'), in_proj_covar=tensor([0.0097, 0.0130, 0.0156, 0.0128, 0.0119, 0.0124, 0.0101, 0.0130],
       device='cuda:2'), out_proj_covar=tensor([7.7245e-05, 1.0308e-04, 1.2772e-04, 1.0175e-04, 9.5951e-05, 9.5445e-05,
        7.8557e-05, 1.0269e-04], device='cuda:2')
2022-12-23 20:34:51,977 WARNING [train.py:1060] (2/4) Exclude cut with ID 298-126791-0067-24026-0_sp0.9 from training. Duration: 21.438875
2022-12-23 20:34:58,277 WARNING [train.py:1060] (2/4) Exclude cut with ID 5652-39938-0025-23684-0_sp0.9 from training. Duration: 22.2055625
2022-12-23 20:35:18,395 INFO [train.py:894] (2/4) Epoch 24, batch 50, loss[loss=0.1591, simple_loss=0.2366, pruned_loss=0.04081, over 18476.00 frames. ], tot_loss[loss=0.1689, simple_loss=0.257, pruned_loss=0.04038, over 837316.98 frames. ], batch size: 43, lr: 4.82e-03, grad_scale: 8.0
2022-12-23 20:36:14,485 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([2.0018, 1.9241, 1.4883, 1.9771, 2.1557, 1.8806, 2.5874, 2.0641],
       device='cuda:2'), covar=tensor([0.0888, 0.1783, 0.2823, 0.1851, 0.1831, 0.0923, 0.0947, 0.1273],
       device='cuda:2'), in_proj_covar=tensor([0.0179, 0.0214, 0.0253, 0.0293, 0.0240, 0.0194, 0.0210, 0.0207],
       device='cuda:2'), out_proj_covar=tensor([0.0001, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-23 20:36:27,818 INFO [zipformer.py:660] (2/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=80736.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 20:36:30,721 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([5.8729, 4.9639, 5.0935, 5.8730, 5.4157, 5.1896, 5.9514, 1.6849],
       device='cuda:2'), covar=tensor([0.0534, 0.0673, 0.0582, 0.0614, 0.1127, 0.1079, 0.0345, 0.5286],
       device='cuda:2'), in_proj_covar=tensor([0.0351, 0.0233, 0.0244, 0.0280, 0.0335, 0.0274, 0.0297, 0.0291],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0001, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-23 20:36:34,714 INFO [train.py:894] (2/4) Epoch 24, batch 100, loss[loss=0.154, simple_loss=0.2485, pruned_loss=0.02973, over 18508.00 frames. ], tot_loss[loss=0.1705, simple_loss=0.2599, pruned_loss=0.04054, over 1474943.85 frames. ], batch size: 52, lr: 4.82e-03, grad_scale: 8.0
2022-12-23 20:36:39,362 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 2.348e+02 3.139e+02 3.771e+02 4.682e+02 1.046e+03, threshold=7.543e+02, percent-clipped=0.0
2022-12-23 20:36:47,515 INFO [zipformer.py:660] (2/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=80749.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 20:37:22,381 INFO [zipformer.py:660] (2/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=80773.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 20:37:49,412 INFO [train.py:894] (2/4) Epoch 24, batch 150, loss[loss=0.2088, simple_loss=0.2885, pruned_loss=0.06452, over 18571.00 frames. ], tot_loss[loss=0.1707, simple_loss=0.2598, pruned_loss=0.04082, over 1972721.09 frames. ], batch size: 188, lr: 4.82e-03, grad_scale: 8.0
2022-12-23 20:37:56,481 WARNING [train.py:1060] (2/4) Exclude cut with ID 7859-102521-0017-7548-0 from training. Duration: 24.525
2022-12-23 20:37:59,638 INFO [zipformer.py:660] (2/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=80797.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 20:38:31,055 WARNING [train.py:1060] (2/4) Exclude cut with ID 3699-47246-0007-3408-0_sp0.9 from training. Duration: 20.26675
2022-12-23 20:38:31,143 INFO [zipformer.py:660] (2/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=80819.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 20:38:37,457 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([2.6296, 2.2141, 1.7174, 2.3094, 2.0146, 2.0961, 2.0517, 2.5093],
       device='cuda:2'), covar=tensor([0.2051, 0.3130, 0.1971, 0.2950, 0.3697, 0.1102, 0.3096, 0.0946],
       device='cuda:2'), in_proj_covar=tensor([0.0303, 0.0299, 0.0251, 0.0356, 0.0279, 0.0234, 0.0295, 0.0222],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0003, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-23 20:38:44,160 WARNING [train.py:1060] (2/4) Exclude cut with ID 7859-102521-0017-7548-0_sp0.9 from training. Duration: 27.25
2022-12-23 20:38:53,827 INFO [zipformer.py:660] (2/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=80834.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 20:39:03,940 INFO [train.py:894] (2/4) Epoch 24, batch 200, loss[loss=0.1672, simple_loss=0.2591, pruned_loss=0.03765, over 18414.00 frames. ], tot_loss[loss=0.169, simple_loss=0.2573, pruned_loss=0.04031, over 2357641.04 frames. ], batch size: 53, lr: 4.82e-03, grad_scale: 8.0
2022-12-23 20:39:09,444 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 1.872e+02 2.949e+02 3.615e+02 4.514e+02 7.125e+02, threshold=7.230e+02, percent-clipped=0.0
2022-12-23 20:39:18,430 INFO [zipformer.py:660] (2/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=80850.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 20:40:01,187 WARNING [train.py:1060] (2/4) Exclude cut with ID 6426-64292-0017-15984-0 from training. Duration: 21.68
2022-12-23 20:40:11,485 WARNING [train.py:1060] (2/4) Exclude cut with ID 4278-13270-0007-59342-0 from training. Duration: 21.6300625
2022-12-23 20:40:19,787 INFO [train.py:894] (2/4) Epoch 24, batch 250, loss[loss=0.1518, simple_loss=0.2339, pruned_loss=0.03483, over 18662.00 frames. ], tot_loss[loss=0.1693, simple_loss=0.2576, pruned_loss=0.04047, over 2659012.18 frames. ], batch size: 46, lr: 4.82e-03, grad_scale: 8.0
2022-12-23 20:40:34,952 WARNING [train.py:1060] (2/4) Exclude cut with ID 4278-13270-0007-59342-0_sp0.9 from training. Duration: 24.033375
2022-12-23 20:41:16,514 INFO [scaling.py:679] (2/4) Whitening: num_groups=8, num_channels=96, metric=1.21 vs. limit=2.0
2022-12-23 20:41:30,822 INFO [zipformer.py:660] (2/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=80939.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 20:41:32,071 WARNING [train.py:1060] (2/4) Exclude cut with ID 4278-13270-0009-59344-0 from training. Duration: 22.905
2022-12-23 20:41:34,073 INFO [train.py:894] (2/4) Epoch 24, batch 300, loss[loss=0.1793, simple_loss=0.2672, pruned_loss=0.04568, over 18617.00 frames. ], tot_loss[loss=0.1692, simple_loss=0.2575, pruned_loss=0.04048, over 2892224.66 frames. ], batch size: 53, lr: 4.82e-03, grad_scale: 8.0
2022-12-23 20:41:34,078 WARNING [train.py:1060] (2/4) Exclude cut with ID 5622-44585-0006-90525-0_sp1.1 from training. Duration: 23.4318125
2022-12-23 20:41:38,664 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 2.257e+02 3.429e+02 4.110e+02 5.616e+02 1.556e+03, threshold=8.221e+02, percent-clipped=7.0
2022-12-23 20:42:13,879 INFO [scaling.py:679] (2/4) Whitening: num_groups=8, num_channels=96, metric=1.28 vs. limit=2.0
2022-12-23 20:42:43,288 INFO [zipformer.py:660] (2/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=80987.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 20:42:49,389 INFO [train.py:894] (2/4) Epoch 24, batch 350, loss[loss=0.146, simple_loss=0.229, pruned_loss=0.0315, over 18574.00 frames. ], tot_loss[loss=0.1692, simple_loss=0.2581, pruned_loss=0.04014, over 3074263.05 frames. ], batch size: 44, lr: 4.81e-03, grad_scale: 8.0
2022-12-23 20:42:50,088 INFO [scaling.py:679] (2/4) Whitening: num_groups=1, num_channels=384, metric=3.28 vs. limit=5.0
2022-12-23 20:42:51,953 INFO [scaling.py:679] (2/4) Whitening: num_groups=1, num_channels=384, metric=3.78 vs. limit=5.0
2022-12-23 20:43:30,957 WARNING [train.py:1060] (2/4) Exclude cut with ID 4278-13270-0009-59344-0_sp1.1 from training. Duration: 20.82275
2022-12-23 20:43:31,519 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([2.0478, 1.9173, 1.5386, 1.5806, 1.7617, 1.8669, 1.7609, 1.8109],
       device='cuda:2'), covar=tensor([0.2261, 0.3107, 0.2073, 0.2581, 0.3523, 0.1096, 0.2839, 0.1087],
       device='cuda:2'), in_proj_covar=tensor([0.0302, 0.0299, 0.0252, 0.0353, 0.0280, 0.0235, 0.0296, 0.0222],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0003, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-23 20:43:32,515 WARNING [train.py:1060] (2/4) Exclude cut with ID 4278-13270-0009-59344-0_sp0.9 from training. Duration: 25.45
2022-12-23 20:44:07,086 INFO [train.py:894] (2/4) Epoch 24, batch 400, loss[loss=0.1496, simple_loss=0.2311, pruned_loss=0.03401, over 18425.00 frames. ], tot_loss[loss=0.1701, simple_loss=0.2593, pruned_loss=0.04044, over 3215964.45 frames. ], batch size: 42, lr: 4.81e-03, grad_scale: 8.0
2022-12-23 20:44:11,335 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 2.000e+02 3.338e+02 3.884e+02 4.805e+02 8.135e+02, threshold=7.768e+02, percent-clipped=0.0
2022-12-23 20:44:19,467 INFO [zipformer.py:660] (2/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=81049.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 20:44:32,727 WARNING [train.py:1060] (2/4) Exclude cut with ID 5622-44585-0006-90525-0 from training. Duration: 25.775
2022-12-23 20:44:54,612 WARNING [train.py:1060] (2/4) Exclude cut with ID 2929-85685-0071-62375-0_sp0.9 from training. Duration: 22.25
2022-12-23 20:45:06,918 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([2.0012, 2.0179, 2.2482, 1.3224, 2.3123, 2.3989, 1.7077, 2.6502],
       device='cuda:2'), covar=tensor([0.1312, 0.1987, 0.1374, 0.2227, 0.0742, 0.1188, 0.2327, 0.0584],
       device='cuda:2'), in_proj_covar=tensor([0.0197, 0.0215, 0.0207, 0.0194, 0.0173, 0.0216, 0.0214, 0.0194],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-23 20:45:21,083 WARNING [train.py:1060] (2/4) Exclude cut with ID 3972-170212-0014-23379-0 from training. Duration: 26.205
2022-12-23 20:45:22,646 INFO [train.py:894] (2/4) Epoch 24, batch 450, loss[loss=0.172, simple_loss=0.27, pruned_loss=0.03704, over 18527.00 frames. ], tot_loss[loss=0.1711, simple_loss=0.2607, pruned_loss=0.04077, over 3327182.01 frames. ], batch size: 77, lr: 4.81e-03, grad_scale: 8.0
2022-12-23 20:45:24,953 INFO [zipformer.py:660] (2/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=81092.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 20:45:32,217 INFO [zipformer.py:660] (2/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=81097.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 20:45:37,959 WARNING [train.py:1060] (2/4) Exclude cut with ID 5239-32139-0047-9341-0_sp0.9 from training. Duration: 30.1555625
2022-12-23 20:45:43,632 WARNING [train.py:1060] (2/4) Exclude cut with ID 1265-135635-0050-6781-0_sp0.9 from training. Duration: 21.8333125
2022-12-23 20:45:51,957 WARNING [train.py:1060] (2/4) Exclude cut with ID 1914-133440-0024-94914-0_sp1.1 from training. Duration: 20.6545625
2022-12-23 20:45:59,992 INFO [zipformer.py:660] (2/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=81116.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 20:46:04,062 INFO [zipformer.py:660] (2/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=81119.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 20:46:18,925 INFO [zipformer.py:660] (2/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=81129.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 20:46:32,828 WARNING [train.py:1060] (2/4) Exclude cut with ID 7395-89880-0045-39920-0_sp0.9 from training. Duration: 20.52225
2022-12-23 20:46:37,962 INFO [train.py:894] (2/4) Epoch 24, batch 500, loss[loss=0.1534, simple_loss=0.2371, pruned_loss=0.03492, over 18412.00 frames. ], tot_loss[loss=0.1717, simple_loss=0.2608, pruned_loss=0.04127, over 3412687.85 frames. ], batch size: 46, lr: 4.81e-03, grad_scale: 8.0
2022-12-23 20:46:42,459 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 2.626e+02 3.730e+02 4.462e+02 5.472e+02 2.665e+03, threshold=8.924e+02, percent-clipped=9.0
2022-12-23 20:46:51,242 INFO [zipformer.py:660] (2/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=81150.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 20:46:54,093 WARNING [train.py:1060] (2/4) Exclude cut with ID 3972-170212-0014-23379-0_sp0.9 from training. Duration: 29.1166875
2022-12-23 20:47:16,090 INFO [zipformer.py:660] (2/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=81167.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 20:47:31,548 INFO [zipformer.py:660] (2/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=81177.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 20:47:53,252 INFO [train.py:894] (2/4) Epoch 24, batch 550, loss[loss=0.1908, simple_loss=0.2848, pruned_loss=0.04845, over 18728.00 frames. ], tot_loss[loss=0.1718, simple_loss=0.2608, pruned_loss=0.04143, over 3479157.91 frames. ], batch size: 54, lr: 4.81e-03, grad_scale: 8.0
2022-12-23 20:47:53,295 WARNING [train.py:1060] (2/4) Exclude cut with ID 543-133211-0007-59831-0_sp0.9 from training. Duration: 21.388875
2022-12-23 20:48:04,026 INFO [zipformer.py:660] (2/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=81198.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 20:48:11,397 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([2.4145, 3.7193, 3.5492, 1.4205, 3.8906, 2.8928, 0.9028, 2.3272],
       device='cuda:2'), covar=tensor([0.2262, 0.1057, 0.1426, 0.3722, 0.0775, 0.0874, 0.4696, 0.1632],
       device='cuda:2'), in_proj_covar=tensor([0.0149, 0.0144, 0.0159, 0.0125, 0.0147, 0.0115, 0.0144, 0.0116],
       device='cuda:2'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0002, 0.0003, 0.0002, 0.0003, 0.0002],
       device='cuda:2')
2022-12-23 20:48:28,950 WARNING [train.py:1060] (2/4) Exclude cut with ID 1914-133440-0024-94914-0 from training. Duration: 22.72
2022-12-23 20:48:30,575 WARNING [train.py:1060] (2/4) Exclude cut with ID 1914-133440-0031-94921-0_sp0.9 from training. Duration: 22.7444375
2022-12-23 20:49:09,399 INFO [train.py:894] (2/4) Epoch 24, batch 600, loss[loss=0.1582, simple_loss=0.24, pruned_loss=0.0382, over 18539.00 frames. ], tot_loss[loss=0.1717, simple_loss=0.261, pruned_loss=0.04119, over 3530598.92 frames. ], batch size: 44, lr: 4.81e-03, grad_scale: 8.0
2022-12-23 20:49:14,227 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 2.016e+02 3.241e+02 3.841e+02 4.795e+02 1.272e+03, threshold=7.682e+02, percent-clipped=3.0
2022-12-23 20:49:15,650 WARNING [train.py:1060] (2/4) Exclude cut with ID 4133-6541-0027-40495-0_sp1.1 from training. Duration: 0.9681875
2022-12-23 20:49:18,600 WARNING [train.py:1060] (2/4) Exclude cut with ID 6330-62851-0022-91297-0_sp0.9 from training. Duration: 22.3166875
2022-12-23 20:49:24,180 WARNING [train.py:1060] (2/4) Exclude cut with ID 543-133212-0015-59917-0_sp0.9 from training. Duration: 21.8166875
2022-12-23 20:50:04,710 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([5.9918, 5.0383, 5.1720, 5.9446, 5.5526, 5.2178, 6.0635, 1.7402],
       device='cuda:2'), covar=tensor([0.0565, 0.0647, 0.0620, 0.0715, 0.1119, 0.1159, 0.0413, 0.5468],
       device='cuda:2'), in_proj_covar=tensor([0.0347, 0.0230, 0.0242, 0.0277, 0.0330, 0.0270, 0.0296, 0.0289],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0001, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-23 20:50:24,740 INFO [train.py:894] (2/4) Epoch 24, batch 650, loss[loss=0.1859, simple_loss=0.2787, pruned_loss=0.04652, over 18548.00 frames. ], tot_loss[loss=0.1718, simple_loss=0.2613, pruned_loss=0.04117, over 3571504.56 frames. ], batch size: 69, lr: 4.81e-03, grad_scale: 8.0
2022-12-23 20:50:36,984 INFO [zipformer.py:660] (2/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=81299.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 20:50:39,786 INFO [zipformer.py:660] (2/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=81301.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 20:51:05,740 WARNING [train.py:1060] (2/4) Exclude cut with ID 4957-30119-0041-23990-0_sp0.9 from training. Duration: 20.22775
2022-12-23 20:51:39,380 INFO [train.py:894] (2/4) Epoch 24, batch 700, loss[loss=0.1674, simple_loss=0.2619, pruned_loss=0.03645, over 18716.00 frames. ], tot_loss[loss=0.1729, simple_loss=0.2625, pruned_loss=0.04162, over 3602940.95 frames. ], batch size: 52, lr: 4.80e-03, grad_scale: 8.0
2022-12-23 20:51:43,775 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 2.135e+02 3.349e+02 3.927e+02 5.051e+02 1.281e+03, threshold=7.854e+02, percent-clipped=5.0
2022-12-23 20:51:44,468 INFO [scaling.py:679] (2/4) Whitening: num_groups=1, num_channels=384, metric=3.06 vs. limit=5.0
2022-12-23 20:51:48,199 WARNING [train.py:1060] (2/4) Exclude cut with ID 5239-32139-0047-9341-0_sp1.1 from training. Duration: 24.67275
2022-12-23 20:52:07,658 INFO [zipformer.py:660] (2/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=81360.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 20:52:10,605 INFO [zipformer.py:660] (2/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=81362.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 20:52:17,849 WARNING [train.py:1060] (2/4) Exclude cut with ID 3082-165428-0081-50734-0_sp0.9 from training. Duration: 21.8055625
2022-12-23 20:52:55,133 INFO [train.py:894] (2/4) Epoch 24, batch 750, loss[loss=0.2069, simple_loss=0.2896, pruned_loss=0.06204, over 18600.00 frames. ], tot_loss[loss=0.1723, simple_loss=0.2617, pruned_loss=0.04147, over 3628122.01 frames. ], batch size: 180, lr: 4.80e-03, grad_scale: 8.0
2022-12-23 20:52:57,145 INFO [zipformer.py:660] (2/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=81392.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 20:52:58,194 WARNING [train.py:1060] (2/4) Exclude cut with ID 3340-169293-0054-76830-0_sp0.9 from training. Duration: 22.6666875
2022-12-23 20:53:52,768 INFO [zipformer.py:660] (2/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=81429.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 20:54:00,245 WARNING [train.py:1060] (2/4) Exclude cut with ID 2411-132532-0017-82279-0_sp1.1 from training. Duration: 0.9681875
2022-12-23 20:54:04,073 INFO [scaling.py:679] (2/4) Whitening: num_groups=8, num_channels=96, metric=1.22 vs. limit=2.0
2022-12-23 20:54:09,056 INFO [zipformer.py:660] (2/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=81440.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 20:54:09,308 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([2.4903, 2.2163, 1.7407, 0.7656, 1.6750, 2.0492, 1.7281, 1.9406],
       device='cuda:2'), covar=tensor([0.0696, 0.0536, 0.1310, 0.1801, 0.1339, 0.1509, 0.1707, 0.0826],
       device='cuda:2'), in_proj_covar=tensor([0.0173, 0.0189, 0.0208, 0.0192, 0.0213, 0.0203, 0.0219, 0.0204],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-23 20:54:10,204 INFO [train.py:894] (2/4) Epoch 24, batch 800, loss[loss=0.172, simple_loss=0.2713, pruned_loss=0.03633, over 18619.00 frames. ], tot_loss[loss=0.1719, simple_loss=0.2612, pruned_loss=0.04131, over 3647049.14 frames. ], batch size: 53, lr: 4.80e-03, grad_scale: 8.0
2022-12-23 20:54:14,887 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 1.956e+02 3.275e+02 3.803e+02 4.513e+02 1.046e+03, threshold=7.605e+02, percent-clipped=6.0
2022-12-23 20:54:25,549 WARNING [train.py:1060] (2/4) Exclude cut with ID 6330-62850-0007-91323-0 from training. Duration: 22.485
2022-12-23 20:54:57,625 INFO [zipformer.py:660] (2/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=81472.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 20:55:04,571 WARNING [train.py:1060] (2/4) Exclude cut with ID 3972-170212-0014-23379-0_sp1.1 from training. Duration: 23.82275
2022-12-23 20:55:05,341 INFO [zipformer.py:660] (2/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=81477.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 20:55:19,134 WARNING [train.py:1060] (2/4) Exclude cut with ID 4860-13185-0032-76709-0 from training. Duration: 20.77
2022-12-23 20:55:22,842 INFO [scaling.py:679] (2/4) Whitening: num_groups=1, num_channels=384, metric=2.96 vs. limit=5.0
2022-12-23 20:55:25,340 WARNING [train.py:1060] (2/4) Exclude cut with ID 6426-64292-0017-15984-0_sp0.9 from training. Duration: 24.088875
2022-12-23 20:55:26,779 INFO [train.py:894] (2/4) Epoch 24, batch 850, loss[loss=0.1771, simple_loss=0.2696, pruned_loss=0.04233, over 18721.00 frames. ], tot_loss[loss=0.1715, simple_loss=0.2611, pruned_loss=0.0409, over 3662024.51 frames. ], batch size: 52, lr: 4.80e-03, grad_scale: 8.0
2022-12-23 20:55:54,285 WARNING [train.py:1060] (2/4) Exclude cut with ID 6330-62850-0007-91323-0_sp1.1 from training. Duration: 20.4409375
2022-12-23 20:55:59,007 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([3.7348, 3.1641, 2.9025, 1.2904, 2.7159, 2.6151, 2.3590, 2.7680],
       device='cuda:2'), covar=tensor([0.0575, 0.0590, 0.1363, 0.1931, 0.1625, 0.1375, 0.1461, 0.0941],
       device='cuda:2'), in_proj_covar=tensor([0.0171, 0.0186, 0.0205, 0.0189, 0.0209, 0.0201, 0.0215, 0.0201],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-23 20:56:41,813 INFO [train.py:894] (2/4) Epoch 24, batch 900, loss[loss=0.2346, simple_loss=0.3092, pruned_loss=0.07997, over 18663.00 frames. ], tot_loss[loss=0.172, simple_loss=0.2613, pruned_loss=0.04133, over 3673226.21 frames. ], batch size: 182, lr: 4.80e-03, grad_scale: 8.0
2022-12-23 20:56:46,272 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 2.385e+02 3.273e+02 3.910e+02 4.576e+02 1.457e+03, threshold=7.821e+02, percent-clipped=4.0
2022-12-23 20:56:46,859 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([2.1886, 1.6325, 1.8591, 1.8719, 2.1908, 2.1775, 2.0754, 1.8276],
       device='cuda:2'), covar=tensor([0.2332, 0.3320, 0.2593, 0.2933, 0.2082, 0.0996, 0.3353, 0.1334],
       device='cuda:2'), in_proj_covar=tensor([0.0269, 0.0298, 0.0281, 0.0319, 0.0311, 0.0254, 0.0346, 0.0242],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-23 20:57:06,964 WARNING [train.py:1060] (2/4) Exclude cut with ID 2929-85685-0062-62366-0_sp0.9 from training. Duration: 22.511125
2022-12-23 20:57:08,341 WARNING [train.py:1060] (2/4) Exclude cut with ID 7395-89880-0031-39906-0 from training. Duration: 20.675
2022-12-23 20:57:57,312 INFO [train.py:894] (2/4) Epoch 24, batch 950, loss[loss=0.1633, simple_loss=0.256, pruned_loss=0.03529, over 18661.00 frames. ], tot_loss[loss=0.1713, simple_loss=0.2609, pruned_loss=0.04088, over 3681870.40 frames. ], batch size: 48, lr: 4.80e-03, grad_scale: 8.0
2022-12-23 20:58:05,123 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.4912, 1.3459, 1.3313, 1.2679, 1.7591, 1.5551, 1.5286, 1.1377],
       device='cuda:2'), covar=tensor([0.0300, 0.0243, 0.0495, 0.0233, 0.0196, 0.0397, 0.0277, 0.0320],
       device='cuda:2'), in_proj_covar=tensor([0.0097, 0.0130, 0.0156, 0.0127, 0.0117, 0.0124, 0.0102, 0.0129],
       device='cuda:2'), out_proj_covar=tensor([7.6845e-05, 1.0281e-04, 1.2779e-04, 1.0132e-04, 9.4686e-05, 9.5267e-05,
        7.9125e-05, 1.0165e-04], device='cuda:2')
2022-12-23 20:58:28,519 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([0.5271, 1.9641, 2.1981, 1.1448, 1.4708, 2.3094, 2.1846, 1.7577],
       device='cuda:2'), covar=tensor([0.0811, 0.0341, 0.0279, 0.0419, 0.0411, 0.0463, 0.0230, 0.0759],
       device='cuda:2'), in_proj_covar=tensor([0.0148, 0.0171, 0.0128, 0.0141, 0.0149, 0.0143, 0.0165, 0.0176],
       device='cuda:2'), out_proj_covar=tensor([1.1244e-04, 1.2944e-04, 9.4806e-05, 1.0393e-04, 1.0987e-04, 1.0854e-04,
        1.2632e-04, 1.3299e-04], device='cuda:2')
2022-12-23 20:58:51,279 WARNING [train.py:1060] (2/4) Exclude cut with ID 6330-62850-0007-91323-0_sp0.9 from training. Duration: 24.9833125
2022-12-23 20:59:07,223 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([4.1959, 3.5986, 3.5821, 4.1207, 3.8464, 3.6338, 4.3208, 1.2696],
       device='cuda:2'), covar=tensor([0.0697, 0.0783, 0.0801, 0.0801, 0.1368, 0.1246, 0.0648, 0.5369],
       device='cuda:2'), in_proj_covar=tensor([0.0345, 0.0229, 0.0240, 0.0275, 0.0327, 0.0268, 0.0293, 0.0287],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0001, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-23 20:59:13,228 INFO [train.py:894] (2/4) Epoch 24, batch 1000, loss[loss=0.1584, simple_loss=0.2366, pruned_loss=0.04008, over 18655.00 frames. ], tot_loss[loss=0.172, simple_loss=0.2614, pruned_loss=0.04127, over 3688668.54 frames. ], batch size: 41, lr: 4.79e-03, grad_scale: 8.0
2022-12-23 20:59:17,609 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 2.354e+02 3.373e+02 4.012e+02 4.702e+02 7.910e+02, threshold=8.023e+02, percent-clipped=1.0
2022-12-23 20:59:21,937 WARNING [train.py:1060] (2/4) Exclude cut with ID 5239-32139-0047-9341-0 from training. Duration: 27.14
2022-12-23 20:59:22,411 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.6375, 1.9139, 1.5735, 2.2017, 2.4720, 1.6420, 1.5115, 1.3639],
       device='cuda:2'), covar=tensor([0.1922, 0.1821, 0.1675, 0.1110, 0.1271, 0.1154, 0.2341, 0.1602],
       device='cuda:2'), in_proj_covar=tensor([0.0248, 0.0227, 0.0217, 0.0201, 0.0261, 0.0196, 0.0225, 0.0201],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-23 20:59:33,986 INFO [zipformer.py:660] (2/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=81655.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 20:59:35,221 WARNING [train.py:1060] (2/4) Exclude cut with ID 2929-85685-0044-62348-0 from training. Duration: 22.44
2022-12-23 20:59:36,762 INFO [zipformer.py:660] (2/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=81657.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 20:59:56,840 INFO [scaling.py:679] (2/4) Whitening: num_groups=8, num_channels=192, metric=1.87 vs. limit=2.0
2022-12-23 20:59:58,354 INFO [zipformer.py:660] (2/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=81670.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 21:00:24,062 INFO [scaling.py:679] (2/4) Whitening: num_groups=1, num_channels=384, metric=4.07 vs. limit=5.0
2022-12-23 21:00:29,154 INFO [train.py:894] (2/4) Epoch 24, batch 1050, loss[loss=0.1889, simple_loss=0.2807, pruned_loss=0.04855, over 18619.00 frames. ], tot_loss[loss=0.1725, simple_loss=0.2622, pruned_loss=0.04138, over 3695284.30 frames. ], batch size: 62, lr: 4.79e-03, grad_scale: 8.0
2022-12-23 21:00:54,290 WARNING [train.py:1060] (2/4) Exclude cut with ID 2929-85685-0060-62364-0_sp0.9 from training. Duration: 21.361125
2022-12-23 21:01:01,506 WARNING [train.py:1060] (2/4) Exclude cut with ID 2929-85685-0079-62383-0_sp1.1 from training. Duration: 27.0318125
2022-12-23 21:01:11,209 WARNING [train.py:1060] (2/4) Exclude cut with ID 5622-44585-0006-90525-0_sp0.9 from training. Duration: 28.638875
2022-12-23 21:01:26,356 WARNING [train.py:1060] (2/4) Exclude cut with ID 3340-169293-0054-76830-0 from training. Duration: 20.4
2022-12-23 21:01:29,931 INFO [zipformer.py:660] (2/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=81731.0, num_to_drop=1, layers_to_drop={0}
2022-12-23 21:01:43,891 INFO [train.py:894] (2/4) Epoch 24, batch 1100, loss[loss=0.177, simple_loss=0.2722, pruned_loss=0.04091, over 18659.00 frames. ], tot_loss[loss=0.172, simple_loss=0.2615, pruned_loss=0.04122, over 3700110.67 frames. ], batch size: 69, lr: 4.79e-03, grad_scale: 8.0
2022-12-23 21:01:47,097 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([2.7154, 4.2018, 3.9969, 1.7712, 4.3043, 3.1661, 0.7384, 2.6454],
       device='cuda:2'), covar=tensor([0.1836, 0.0876, 0.1179, 0.3296, 0.0661, 0.0786, 0.4813, 0.1362],
       device='cuda:2'), in_proj_covar=tensor([0.0148, 0.0144, 0.0158, 0.0124, 0.0145, 0.0114, 0.0144, 0.0115],
       device='cuda:2'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0002, 0.0003, 0.0002, 0.0003, 0.0002],
       device='cuda:2')
2022-12-23 21:01:48,277 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 2.333e+02 3.535e+02 4.177e+02 5.058e+02 9.147e+02, threshold=8.355e+02, percent-clipped=1.0
2022-12-23 21:01:58,851 WARNING [train.py:1060] (2/4) Exclude cut with ID 2929-85685-0071-62375-0 from training. Duration: 20.025
2022-12-23 21:01:58,860 WARNING [train.py:1060] (2/4) Exclude cut with ID 2364-131735-0112-64612-0_sp0.9 from training. Duration: 20.488875
2022-12-23 21:02:04,947 WARNING [train.py:1060] (2/4) Exclude cut with ID 2929-85685-0079-62383-0 from training. Duration: 29.735
2022-12-23 21:02:31,813 INFO [zipformer.py:660] (2/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=81772.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 21:02:59,301 INFO [train.py:894] (2/4) Epoch 24, batch 1150, loss[loss=0.1699, simple_loss=0.2635, pruned_loss=0.03815, over 18618.00 frames. ], tot_loss[loss=0.1709, simple_loss=0.2606, pruned_loss=0.04064, over 3703030.96 frames. ], batch size: 69, lr: 4.79e-03, grad_scale: 8.0
2022-12-23 21:03:15,286 INFO [scaling.py:679] (2/4) Whitening: num_groups=8, num_channels=96, metric=1.20 vs. limit=2.0
2022-12-23 21:03:25,693 WARNING [train.py:1060] (2/4) Exclude cut with ID 7276-92427-0014-12983-0_sp0.9 from training. Duration: 21.3055625
2022-12-23 21:03:27,219 WARNING [train.py:1060] (2/4) Exclude cut with ID 1025-75365-0008-79168-0_sp0.9 from training. Duration: 22.0666875
2022-12-23 21:03:45,330 INFO [zipformer.py:660] (2/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=81820.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 21:04:11,697 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([4.6146, 3.9394, 3.9270, 4.4912, 4.3025, 4.0323, 4.7414, 1.3376],
       device='cuda:2'), covar=tensor([0.0749, 0.0809, 0.0751, 0.0899, 0.1307, 0.1276, 0.0617, 0.5743],
       device='cuda:2'), in_proj_covar=tensor([0.0344, 0.0228, 0.0239, 0.0274, 0.0326, 0.0267, 0.0293, 0.0287],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0001, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-23 21:04:15,786 INFO [train.py:894] (2/4) Epoch 24, batch 1200, loss[loss=0.1593, simple_loss=0.2475, pruned_loss=0.03561, over 18432.00 frames. ], tot_loss[loss=0.1701, simple_loss=0.26, pruned_loss=0.04011, over 3704729.34 frames. ], batch size: 48, lr: 4.79e-03, grad_scale: 8.0
2022-12-23 21:04:19,890 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 1.558e+02 3.274e+02 3.968e+02 4.873e+02 1.280e+03, threshold=7.935e+02, percent-clipped=3.0
2022-12-23 21:04:23,847 INFO [scaling.py:679] (2/4) Whitening: num_groups=1, num_channels=384, metric=4.05 vs. limit=5.0
2022-12-23 21:05:14,405 INFO [scaling.py:679] (2/4) Whitening: num_groups=1, num_channels=384, metric=2.98 vs. limit=5.0
2022-12-23 21:05:19,325 WARNING [train.py:1060] (2/4) Exclude cut with ID 2929-85685-0062-62366-0 from training. Duration: 20.26
2022-12-23 21:05:30,577 INFO [train.py:894] (2/4) Epoch 24, batch 1250, loss[loss=0.1776, simple_loss=0.2623, pruned_loss=0.04647, over 18412.00 frames. ], tot_loss[loss=0.1701, simple_loss=0.2604, pruned_loss=0.03991, over 3706525.69 frames. ], batch size: 48, lr: 4.79e-03, grad_scale: 8.0
2022-12-23 21:05:32,102 WARNING [train.py:1060] (2/4) Exclude cut with ID 5239-32139-0030-9324-0_sp0.9 from training. Duration: 21.3444375
2022-12-23 21:05:39,754 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([2.0714, 1.9315, 2.1399, 1.2041, 2.1778, 2.2787, 1.6673, 2.6727],
       device='cuda:2'), covar=tensor([0.1115, 0.1817, 0.1243, 0.1966, 0.0800, 0.1131, 0.2309, 0.0474],
       device='cuda:2'), in_proj_covar=tensor([0.0195, 0.0212, 0.0206, 0.0192, 0.0173, 0.0215, 0.0212, 0.0192],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-23 21:06:21,947 INFO [scaling.py:679] (2/4) Whitening: num_groups=8, num_channels=192, metric=1.95 vs. limit=2.0
2022-12-23 21:06:28,804 WARNING [train.py:1060] (2/4) Exclude cut with ID 497-129325-0061-62254-0_sp1.1 from training. Duration: 0.97725
2022-12-23 21:06:46,112 INFO [train.py:894] (2/4) Epoch 24, batch 1300, loss[loss=0.1685, simple_loss=0.2607, pruned_loss=0.03812, over 18627.00 frames. ], tot_loss[loss=0.1697, simple_loss=0.2599, pruned_loss=0.03973, over 3708119.44 frames. ], batch size: 53, lr: 4.79e-03, grad_scale: 8.0
2022-12-23 21:06:50,343 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 2.062e+02 3.110e+02 3.685e+02 4.502e+02 1.281e+03, threshold=7.371e+02, percent-clipped=3.0
2022-12-23 21:07:07,391 INFO [zipformer.py:660] (2/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=81955.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 21:07:10,259 WARNING [train.py:1060] (2/4) Exclude cut with ID 7395-89880-0031-39906-0_sp0.9 from training. Duration: 22.97225
2022-12-23 21:07:10,506 INFO [zipformer.py:660] (2/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=81957.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 21:07:31,818 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([2.1162, 1.0779, 1.8451, 1.7024, 2.0826, 2.1266, 1.7588, 1.8547],
       device='cuda:2'), covar=tensor([0.2197, 0.3468, 0.2658, 0.2720, 0.2227, 0.1037, 0.3459, 0.1378],
       device='cuda:2'), in_proj_covar=tensor([0.0269, 0.0299, 0.0282, 0.0320, 0.0312, 0.0255, 0.0347, 0.0243],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-23 21:07:40,770 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.5636, 1.5186, 1.5160, 1.4551, 0.9258, 3.0383, 1.1508, 1.7048],
       device='cuda:2'), covar=tensor([0.3173, 0.2179, 0.1993, 0.2145, 0.1636, 0.0219, 0.1805, 0.0918],
       device='cuda:2'), in_proj_covar=tensor([0.0131, 0.0117, 0.0123, 0.0120, 0.0104, 0.0096, 0.0090, 0.0089],
       device='cuda:2'), out_proj_covar=tensor([0.0005, 0.0005, 0.0005, 0.0005, 0.0004, 0.0003, 0.0004, 0.0004],
       device='cuda:2')
2022-12-23 21:07:41,935 WARNING [train.py:1060] (2/4) Exclude cut with ID 7395-89880-0047-39922-0_sp0.9 from training. Duration: 21.97775
2022-12-23 21:07:55,393 WARNING [train.py:1060] (2/4) Exclude cut with ID 1112-1043-0006-89194-0_sp0.9 from training. Duration: 21.8333125
2022-12-23 21:07:57,293 INFO [zipformer.py:660] (2/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=81987.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 21:08:02,962 INFO [train.py:894] (2/4) Epoch 24, batch 1350, loss[loss=0.1846, simple_loss=0.278, pruned_loss=0.04563, over 18700.00 frames. ], tot_loss[loss=0.1701, simple_loss=0.26, pruned_loss=0.04003, over 3710108.81 frames. ], batch size: 60, lr: 4.78e-03, grad_scale: 8.0
2022-12-23 21:08:05,982 WARNING [train.py:1060] (2/4) Exclude cut with ID 1914-133440-0031-94921-0 from training. Duration: 20.47
2022-12-23 21:08:25,308 INFO [zipformer.py:660] (2/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=82003.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 21:08:28,681 INFO [zipformer.py:660] (2/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=82005.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 21:09:01,122 INFO [zipformer.py:660] (2/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=82026.0, num_to_drop=1, layers_to_drop={3}
2022-12-23 21:09:16,382 WARNING [train.py:1060] (2/4) Exclude cut with ID 7395-89880-0037-39912-0_sp0.9 from training. Duration: 20.67225
2022-12-23 21:09:19,836 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([2.8828, 2.3415, 1.8123, 2.5114, 2.1916, 2.2878, 2.1670, 2.7360],
       device='cuda:2'), covar=tensor([0.1875, 0.3242, 0.1985, 0.2787, 0.3573, 0.1041, 0.3081, 0.0869],
       device='cuda:2'), in_proj_covar=tensor([0.0296, 0.0294, 0.0247, 0.0343, 0.0274, 0.0230, 0.0289, 0.0218],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-23 21:09:22,125 INFO [train.py:894] (2/4) Epoch 24, batch 1400, loss[loss=0.171, simple_loss=0.2582, pruned_loss=0.04192, over 18573.00 frames. ], tot_loss[loss=0.1698, simple_loss=0.2596, pruned_loss=0.03997, over 3710799.52 frames. ], batch size: 57, lr: 4.78e-03, grad_scale: 8.0
2022-12-23 21:09:26,691 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 2.199e+02 3.133e+02 3.786e+02 4.539e+02 8.263e+02, threshold=7.572e+02, percent-clipped=1.0
2022-12-23 21:09:33,585 INFO [zipformer.py:660] (2/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=82048.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 21:09:34,732 WARNING [train.py:1060] (2/4) Exclude cut with ID 1914-133440-0024-94914-0_sp0.9 from training. Duration: 25.2444375
2022-12-23 21:09:45,229 INFO [zipformer.py:660] (2/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=82055.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 21:09:48,974 INFO [scaling.py:679] (2/4) Whitening: num_groups=8, num_channels=192, metric=2.00 vs. limit=2.0
2022-12-23 21:10:00,180 WARNING [train.py:1060] (2/4) Exclude cut with ID 3340-169293-0021-76797-0_sp0.9 from training. Duration: 21.1445
2022-12-23 21:10:37,962 INFO [train.py:894] (2/4) Epoch 24, batch 1450, loss[loss=0.1547, simple_loss=0.243, pruned_loss=0.03321, over 18690.00 frames. ], tot_loss[loss=0.1698, simple_loss=0.2595, pruned_loss=0.04005, over 3712196.74 frames. ], batch size: 50, lr: 4.78e-03, grad_scale: 8.0
2022-12-23 21:11:15,443 WARNING [train.py:1060] (2/4) Exclude cut with ID 2929-85685-0079-62383-0_sp0.9 from training. Duration: 33.038875
2022-12-23 21:11:17,134 INFO [zipformer.py:660] (2/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=82116.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 21:11:52,242 WARNING [train.py:1060] (2/4) Exclude cut with ID 6426-64291-0000-16059-0_sp0.9 from training. Duration: 20.0944375
2022-12-23 21:11:53,452 INFO [train.py:894] (2/4) Epoch 24, batch 1500, loss[loss=0.1595, simple_loss=0.2469, pruned_loss=0.03609, over 18422.00 frames. ], tot_loss[loss=0.1693, simple_loss=0.2591, pruned_loss=0.03977, over 3711903.63 frames. ], batch size: 48, lr: 4.78e-03, grad_scale: 8.0
2022-12-23 21:11:57,690 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 2.080e+02 3.225e+02 3.712e+02 4.583e+02 1.375e+03, threshold=7.423e+02, percent-clipped=2.0
2022-12-23 21:12:07,777 WARNING [train.py:1060] (2/4) Exclude cut with ID 2929-85685-0044-62348-0_sp1.1 from training. Duration: 20.4
2022-12-23 21:12:15,657 WARNING [train.py:1060] (2/4) Exclude cut with ID 6330-62851-0022-91297-0 from training. Duration: 20.085
2022-12-23 21:12:26,463 WARNING [train.py:1060] (2/4) Exclude cut with ID 4860-13185-0032-76709-0_sp0.9 from training. Duration: 23.07775
2022-12-23 21:13:07,890 INFO [train.py:894] (2/4) Epoch 24, batch 1550, loss[loss=0.1688, simple_loss=0.268, pruned_loss=0.0348, over 18464.00 frames. ], tot_loss[loss=0.1694, simple_loss=0.259, pruned_loss=0.03994, over 3711723.37 frames. ], batch size: 54, lr: 4.78e-03, grad_scale: 8.0
2022-12-23 21:13:12,588 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([2.4895, 2.5456, 1.8273, 3.1984, 2.8350, 2.4125, 3.7593, 2.5992],
       device='cuda:2'), covar=tensor([0.0805, 0.1834, 0.2730, 0.1691, 0.1679, 0.0854, 0.0798, 0.1154],
       device='cuda:2'), in_proj_covar=tensor([0.0177, 0.0213, 0.0252, 0.0288, 0.0239, 0.0192, 0.0206, 0.0206],
       device='cuda:2'), out_proj_covar=tensor([0.0001, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-23 21:13:13,544 WARNING [train.py:1060] (2/4) Exclude cut with ID 2929-85685-0044-62348-0_sp0.9 from training. Duration: 24.9333125
2022-12-23 21:13:23,390 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.2411, 1.2413, 0.9836, 1.3349, 1.3208, 1.1714, 1.5498, 1.3369],
       device='cuda:2'), covar=tensor([0.0792, 0.1320, 0.2115, 0.1285, 0.1461, 0.0808, 0.0788, 0.1032],
       device='cuda:2'), in_proj_covar=tensor([0.0177, 0.0214, 0.0253, 0.0289, 0.0239, 0.0192, 0.0206, 0.0206],
       device='cuda:2'), out_proj_covar=tensor([0.0001, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-23 21:13:59,151 WARNING [train.py:1060] (2/4) Exclude cut with ID 5118-111612-0016-124680-0_sp0.9 from training. Duration: 20.388875
2022-12-23 21:14:05,173 WARNING [train.py:1060] (2/4) Exclude cut with ID 432-122774-0017-62487-0_sp1.1 from training. Duration: 20.3590625
2022-12-23 21:14:23,210 INFO [train.py:894] (2/4) Epoch 24, batch 1600, loss[loss=0.183, simple_loss=0.2823, pruned_loss=0.04185, over 18590.00 frames. ], tot_loss[loss=0.1697, simple_loss=0.2592, pruned_loss=0.04009, over 3712253.69 frames. ], batch size: 56, lr: 4.78e-03, grad_scale: 8.0
2022-12-23 21:14:25,125 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([4.0337, 3.2608, 2.9896, 1.4299, 2.8520, 2.8506, 2.3238, 2.9344],
       device='cuda:2'), covar=tensor([0.0485, 0.0638, 0.1420, 0.1832, 0.1573, 0.1194, 0.1554, 0.1018],
       device='cuda:2'), in_proj_covar=tensor([0.0172, 0.0187, 0.0207, 0.0190, 0.0211, 0.0203, 0.0218, 0.0204],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-23 21:14:27,464 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 2.380e+02 3.366e+02 4.050e+02 5.021e+02 8.278e+02, threshold=8.100e+02, percent-clipped=2.0
2022-12-23 21:14:30,915 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.6079, 1.3812, 1.5260, 1.9585, 1.7083, 3.4079, 1.3225, 1.5307],
       device='cuda:2'), covar=tensor([0.0840, 0.1843, 0.1076, 0.0884, 0.1429, 0.0245, 0.1458, 0.1559],
       device='cuda:2'), in_proj_covar=tensor([0.0073, 0.0082, 0.0072, 0.0075, 0.0091, 0.0076, 0.0084, 0.0077],
       device='cuda:2'), out_proj_covar=tensor([0.0004, 0.0004, 0.0004, 0.0004, 0.0004, 0.0004, 0.0004, 0.0004],
       device='cuda:2')
2022-12-23 21:14:49,321 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([5.9137, 4.9787, 5.4846, 5.6438, 5.1210, 4.7892, 5.9142, 1.5489],
       device='cuda:2'), covar=tensor([0.0733, 0.1039, 0.0771, 0.1429, 0.1765, 0.1749, 0.0714, 0.6702],
       device='cuda:2'), in_proj_covar=tensor([0.0343, 0.0228, 0.0239, 0.0273, 0.0324, 0.0266, 0.0292, 0.0285],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0001, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-23 21:15:15,716 WARNING [train.py:1060] (2/4) Exclude cut with ID 3557-8342-0013-54691-0_sp1.1 from training. Duration: 0.836375
2022-12-23 21:15:20,733 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([2.9595, 5.5002, 5.0098, 2.5617, 5.5783, 4.2324, 1.1670, 3.7998],
       device='cuda:2'), covar=tensor([0.2023, 0.1035, 0.1161, 0.3130, 0.0546, 0.0751, 0.4813, 0.1177],
       device='cuda:2'), in_proj_covar=tensor([0.0148, 0.0144, 0.0158, 0.0125, 0.0145, 0.0115, 0.0145, 0.0114],
       device='cuda:2'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0002, 0.0003, 0.0002, 0.0003, 0.0002],
       device='cuda:2')
2022-12-23 21:15:38,368 INFO [train.py:894] (2/4) Epoch 24, batch 1650, loss[loss=0.1734, simple_loss=0.2594, pruned_loss=0.0437, over 18624.00 frames. ], tot_loss[loss=0.1705, simple_loss=0.2599, pruned_loss=0.04062, over 3712491.10 frames. ], batch size: 53, lr: 4.78e-03, grad_scale: 8.0
2022-12-23 21:16:01,390 WARNING [train.py:1060] (2/4) Exclude cut with ID 8565-290391-0049-67394-0_sp0.9 from training. Duration: 21.3166875
2022-12-23 21:16:05,737 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([2.7836, 4.1021, 3.9183, 1.8284, 4.2104, 3.1823, 0.9919, 2.8268],
       device='cuda:2'), covar=tensor([0.2091, 0.1253, 0.1307, 0.3390, 0.0780, 0.0852, 0.4498, 0.1368],
       device='cuda:2'), in_proj_covar=tensor([0.0148, 0.0144, 0.0158, 0.0124, 0.0144, 0.0115, 0.0144, 0.0114],
       device='cuda:2'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0002, 0.0003, 0.0002, 0.0003, 0.0002],
       device='cuda:2')
2022-12-23 21:16:31,565 INFO [zipformer.py:660] (2/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=82326.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 21:16:32,766 WARNING [train.py:1060] (2/4) Exclude cut with ID 6533-399-0029-104863-0_sp0.9 from training. Duration: 22.1055625
2022-12-23 21:16:43,029 WARNING [train.py:1060] (2/4) Exclude cut with ID 7699-105389-0094-26379-0_sp1.1 from training. Duration: 21.77725
2022-12-23 21:16:53,960 INFO [train.py:894] (2/4) Epoch 24, batch 1700, loss[loss=0.1635, simple_loss=0.2572, pruned_loss=0.03486, over 18622.00 frames. ], tot_loss[loss=0.1731, simple_loss=0.2616, pruned_loss=0.0423, over 3713236.50 frames. ], batch size: 53, lr: 4.77e-03, grad_scale: 8.0
2022-12-23 21:16:57,084 INFO [zipformer.py:660] (2/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=82343.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 21:16:58,410 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 2.394e+02 3.814e+02 4.522e+02 5.754e+02 2.798e+03, threshold=9.045e+02, percent-clipped=8.0
2022-12-23 21:17:01,695 INFO [zipformer.py:660] (2/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=82346.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 21:17:04,249 WARNING [train.py:1060] (2/4) Exclude cut with ID 7255-291500-0005-134304-0_sp0.9 from training. Duration: 27.8166875
2022-12-23 21:17:32,341 WARNING [train.py:1060] (2/4) Exclude cut with ID 7357-94126-0021-15852-0_sp1.1 from training. Duration: 22.5090625
2022-12-23 21:17:38,434 WARNING [train.py:1060] (2/4) Exclude cut with ID 7255-291500-0005-134304-0 from training. Duration: 25.035
2022-12-23 21:17:44,253 INFO [zipformer.py:660] (2/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=82374.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 21:17:57,029 WARNING [train.py:1060] (2/4) Exclude cut with ID 774-127930-0014-10412-0_sp1.1 from training. Duration: 0.95
2022-12-23 21:18:09,470 INFO [train.py:894] (2/4) Epoch 24, batch 1750, loss[loss=0.2153, simple_loss=0.2893, pruned_loss=0.0706, over 18672.00 frames. ], tot_loss[loss=0.1739, simple_loss=0.2615, pruned_loss=0.04318, over 3713565.33 frames. ], batch size: 180, lr: 4.77e-03, grad_scale: 8.0
2022-12-23 21:18:15,917 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([2.9228, 3.0197, 1.9390, 1.5324, 3.5274, 3.5138, 3.0786, 2.2836],
       device='cuda:2'), covar=tensor([0.0438, 0.0399, 0.0652, 0.0803, 0.0243, 0.0366, 0.0417, 0.0845],
       device='cuda:2'), in_proj_covar=tensor([0.0122, 0.0126, 0.0127, 0.0119, 0.0101, 0.0123, 0.0132, 0.0157],
       device='cuda:2'), out_proj_covar=tensor([0.0001, 0.0002, 0.0002, 0.0001, 0.0001, 0.0001, 0.0002, 0.0002],
       device='cuda:2')
2022-12-23 21:18:17,024 WARNING [train.py:1060] (2/4) Exclude cut with ID 3033-130750-0096-55598-0_sp0.9 from training. Duration: 0.92225
2022-12-23 21:18:25,326 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([3.0067, 4.8120, 4.5458, 2.4352, 4.8817, 3.7843, 1.1716, 3.2465],
       device='cuda:2'), covar=tensor([0.2005, 0.0967, 0.1236, 0.3025, 0.0692, 0.0754, 0.4632, 0.1291],
       device='cuda:2'), in_proj_covar=tensor([0.0148, 0.0145, 0.0158, 0.0125, 0.0145, 0.0116, 0.0144, 0.0114],
       device='cuda:2'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0002, 0.0003, 0.0002, 0.0003, 0.0002],
       device='cuda:2')
2022-12-23 21:18:34,184 INFO [zipformer.py:660] (2/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=82407.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 21:18:39,721 INFO [zipformer.py:660] (2/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=82411.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 21:18:44,521 WARNING [train.py:1060] (2/4) Exclude cut with ID 4511-76322-0006-80011-0 from training. Duration: 21.97
2022-12-23 21:19:01,039 WARNING [train.py:1060] (2/4) Exclude cut with ID 7492-105653-0055-62765-0_sp0.9 from training. Duration: 21.97225
2022-12-23 21:19:01,079 WARNING [train.py:1060] (2/4) Exclude cut with ID 453-131332-0000-47844-0_sp0.9 from training. Duration: 25.3333125
2022-12-23 21:19:12,080 WARNING [train.py:1060] (2/4) Exclude cut with ID 5172-29468-0015-19128-0_sp0.9 from training. Duration: 21.5055625
2022-12-23 21:19:22,291 WARNING [train.py:1060] (2/4) Exclude cut with ID 453-131332-0000-47844-0_sp1.1 from training. Duration: 20.72725
2022-12-23 21:19:24,922 INFO [train.py:894] (2/4) Epoch 24, batch 1800, loss[loss=0.1792, simple_loss=0.2589, pruned_loss=0.04974, over 18644.00 frames. ], tot_loss[loss=0.1766, simple_loss=0.2632, pruned_loss=0.04497, over 3714730.49 frames. ], batch size: 78, lr: 4.77e-03, grad_scale: 8.0
2022-12-23 21:19:29,071 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 2.173e+02 4.124e+02 5.123e+02 6.216e+02 1.031e+03, threshold=1.025e+03, percent-clipped=5.0
2022-12-23 21:19:52,617 WARNING [train.py:1060] (2/4) Exclude cut with ID 8631-249866-0030-130156-0_sp0.9 from training. Duration: 26.32775
2022-12-23 21:20:23,946 WARNING [train.py:1060] (2/4) Exclude cut with ID 3867-173237-0077-144769-0 from training. Duration: 20.025
2022-12-23 21:20:28,418 WARNING [train.py:1060] (2/4) Exclude cut with ID 6709-74022-0004-86860-0_sp1.1 from training. Duration: 0.9409375
2022-12-23 21:20:28,424 WARNING [train.py:1060] (2/4) Exclude cut with ID 4757-1811-0023-62229-0_sp0.9 from training. Duration: 21.37775
2022-12-23 21:20:40,168 INFO [train.py:894] (2/4) Epoch 24, batch 1850, loss[loss=0.1578, simple_loss=0.2361, pruned_loss=0.03975, over 18701.00 frames. ], tot_loss[loss=0.1778, simple_loss=0.2632, pruned_loss=0.04618, over 3714291.45 frames. ], batch size: 46, lr: 4.77e-03, grad_scale: 8.0
2022-12-23 21:20:43,281 INFO [zipformer.py:660] (2/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=82493.0, num_to_drop=1, layers_to_drop={0}
2022-12-23 21:20:49,434 WARNING [train.py:1060] (2/4) Exclude cut with ID 1250-135782-0004-25974-0_sp0.9 from training. Duration: 21.17225
2022-12-23 21:20:49,447 WARNING [train.py:1060] (2/4) Exclude cut with ID 7357-94126-0021-15852-0_sp0.9 from training. Duration: 27.511125
2022-12-23 21:21:22,580 WARNING [train.py:1060] (2/4) Exclude cut with ID 453-131332-0000-47844-0 from training. Duration: 22.8
2022-12-23 21:21:26,795 WARNING [train.py:1060] (2/4) Exclude cut with ID 4964-30587-0040-44509-0 from training. Duration: 22.585
2022-12-23 21:21:53,437 INFO [train.py:894] (2/4) Epoch 24, batch 1900, loss[loss=0.1704, simple_loss=0.2557, pruned_loss=0.04257, over 18579.00 frames. ], tot_loss[loss=0.1791, simple_loss=0.2636, pruned_loss=0.04731, over 3715104.63 frames. ], batch size: 51, lr: 4.77e-03, grad_scale: 16.0
2022-12-23 21:21:58,950 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 2.610e+02 4.545e+02 5.793e+02 7.153e+02 1.257e+03, threshold=1.159e+03, percent-clipped=6.0
2022-12-23 21:21:59,011 WARNING [train.py:1060] (2/4) Exclude cut with ID 6978-92210-0001-146967-0_sp0.9 from training. Duration: 22.0166875
2022-12-23 21:22:14,563 INFO [zipformer.py:660] (2/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=82554.0, num_to_drop=1, layers_to_drop={3}
2022-12-23 21:22:17,087 WARNING [train.py:1060] (2/4) Exclude cut with ID 7255-291500-0003-134302-0_sp1.1 from training. Duration: 24.395375
2022-12-23 21:22:23,026 WARNING [train.py:1060] (2/4) Exclude cut with ID 3488-85273-0017-41203-0_sp0.9 from training. Duration: 27.47775
2022-12-23 21:22:28,795 WARNING [train.py:1060] (2/4) Exclude cut with ID 432-122774-0017-62487-0_sp0.9 from training. Duration: 24.8833125
2022-12-23 21:22:30,253 WARNING [train.py:1060] (2/4) Exclude cut with ID 6758-72288-0033-108368-0 from training. Duration: 23.39
2022-12-23 21:22:36,506 INFO [zipformer.py:660] (2/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=82569.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 21:22:37,621 WARNING [train.py:1060] (2/4) Exclude cut with ID 8291-282929-0007-12994-0_sp0.9 from training. Duration: 28.72225
2022-12-23 21:22:48,519 WARNING [train.py:1060] (2/4) Exclude cut with ID 585-294811-0110-133686-0_sp0.9 from training. Duration: 20.8944375
2022-12-23 21:23:04,039 WARNING [train.py:1060] (2/4) Exclude cut with ID 5796-66357-0007-116447-0_sp0.9 from training. Duration: 23.8444375
2022-12-23 21:23:05,922 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.8395, 1.7182, 1.8007, 1.7198, 1.3366, 3.8413, 1.6112, 2.2011],
       device='cuda:2'), covar=tensor([0.3106, 0.2013, 0.1881, 0.2130, 0.1498, 0.0182, 0.1576, 0.0804],
       device='cuda:2'), in_proj_covar=tensor([0.0133, 0.0118, 0.0125, 0.0122, 0.0105, 0.0097, 0.0091, 0.0090],
       device='cuda:2'), out_proj_covar=tensor([0.0005, 0.0005, 0.0005, 0.0005, 0.0004, 0.0003, 0.0004, 0.0004],
       device='cuda:2')
2022-12-23 21:23:08,471 INFO [train.py:894] (2/4) Epoch 24, batch 1950, loss[loss=0.1959, simple_loss=0.28, pruned_loss=0.05591, over 18580.00 frames. ], tot_loss[loss=0.179, simple_loss=0.2631, pruned_loss=0.04744, over 3714126.08 frames. ], batch size: 69, lr: 4.77e-03, grad_scale: 16.0
2022-12-23 21:23:27,284 WARNING [train.py:1060] (2/4) Exclude cut with ID 8291-282929-0007-12994-0 from training. Duration: 25.85
2022-12-23 21:23:27,294 WARNING [train.py:1060] (2/4) Exclude cut with ID 8291-282929-0023-13010-0 from training. Duration: 21.39
2022-12-23 21:23:37,536 WARNING [train.py:1060] (2/4) Exclude cut with ID 7357-94126-0014-15845-0 from training. Duration: 27.92
2022-12-23 21:24:06,465 WARNING [train.py:1060] (2/4) Exclude cut with ID 8631-249866-0039-130165-0_sp0.9 from training. Duration: 20.661125
2022-12-23 21:24:09,580 INFO [zipformer.py:660] (2/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=82630.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 21:24:26,713 INFO [train.py:894] (2/4) Epoch 24, batch 2000, loss[loss=0.1636, simple_loss=0.238, pruned_loss=0.04464, over 18594.00 frames. ], tot_loss[loss=0.179, simple_loss=0.2628, pruned_loss=0.04764, over 3714845.61 frames. ], batch size: 45, lr: 4.77e-03, grad_scale: 16.0
2022-12-23 21:24:29,836 INFO [zipformer.py:660] (2/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=82643.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 21:24:30,848 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 2.140e+02 4.292e+02 5.088e+02 6.391e+02 1.331e+03, threshold=1.018e+03, percent-clipped=1.0
2022-12-23 21:24:30,872 WARNING [train.py:1060] (2/4) Exclude cut with ID 7357-94126-0043-15874-0_sp0.9 from training. Duration: 20.07225
2022-12-23 21:24:37,220 WARNING [train.py:1060] (2/4) Exclude cut with ID 1085-156170-0017-128270-0 from training. Duration: 21.01
2022-12-23 21:24:48,095 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([2.4792, 2.0808, 1.6374, 2.2608, 1.9318, 1.9393, 1.8916, 2.4269],
       device='cuda:2'), covar=tensor([0.2352, 0.3279, 0.2263, 0.2560, 0.3772, 0.1307, 0.3548, 0.0953],
       device='cuda:2'), in_proj_covar=tensor([0.0299, 0.0296, 0.0248, 0.0348, 0.0277, 0.0231, 0.0292, 0.0219],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0003, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-23 21:24:49,796 INFO [scaling.py:679] (2/4) Whitening: num_groups=8, num_channels=96, metric=1.16 vs. limit=2.0
2022-12-23 21:25:42,843 INFO [train.py:894] (2/4) Epoch 24, batch 2050, loss[loss=0.1979, simple_loss=0.2826, pruned_loss=0.05665, over 18532.00 frames. ], tot_loss[loss=0.18, simple_loss=0.2631, pruned_loss=0.0485, over 3714755.57 frames. ], batch size: 58, lr: 4.76e-03, grad_scale: 16.0
2022-12-23 21:25:43,038 INFO [zipformer.py:660] (2/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=82691.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 21:25:47,425 WARNING [train.py:1060] (2/4) Exclude cut with ID 2195-150901-0045-59933-0 from training. Duration: 20.65
2022-12-23 21:25:53,916 WARNING [train.py:1060] (2/4) Exclude cut with ID 5796-66357-0007-116447-0 from training. Duration: 21.46
2022-12-23 21:25:59,845 INFO [zipformer.py:660] (2/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=82702.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 21:26:12,809 INFO [zipformer.py:660] (2/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=82711.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 21:26:40,204 WARNING [train.py:1060] (2/4) Exclude cut with ID 3557-8342-0013-54691-0 from training. Duration: 0.92
2022-12-23 21:26:47,999 WARNING [train.py:1060] (2/4) Exclude cut with ID 8291-282929-0023-13010-0_sp0.9 from training. Duration: 23.7666875
2022-12-23 21:26:57,935 INFO [train.py:894] (2/4) Epoch 24, batch 2100, loss[loss=0.1528, simple_loss=0.238, pruned_loss=0.03381, over 18649.00 frames. ], tot_loss[loss=0.1806, simple_loss=0.2636, pruned_loss=0.04884, over 3715506.87 frames. ], batch size: 48, lr: 4.76e-03, grad_scale: 16.0
2022-12-23 21:27:03,146 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 3.401e+02 4.641e+02 5.481e+02 6.606e+02 1.267e+03, threshold=1.096e+03, percent-clipped=3.0
2022-12-23 21:27:22,192 WARNING [train.py:1060] (2/4) Exclude cut with ID 8544-281189-0060-101339-0_sp0.9 from training. Duration: 20.861125
2022-12-23 21:27:25,209 INFO [zipformer.py:660] (2/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=82759.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 21:27:33,500 WARNING [train.py:1060] (2/4) Exclude cut with ID 3488-65654-0031-41259-0_sp0.9 from training. Duration: 22.711125
2022-12-23 21:28:14,011 INFO [train.py:894] (2/4) Epoch 24, batch 2150, loss[loss=0.1551, simple_loss=0.231, pruned_loss=0.03957, over 18392.00 frames. ], tot_loss[loss=0.1807, simple_loss=0.2635, pruned_loss=0.04898, over 3714045.98 frames. ], batch size: 42, lr: 4.76e-03, grad_scale: 16.0
2022-12-23 21:28:16,082 WARNING [train.py:1060] (2/4) Exclude cut with ID 6951-79737-0043-132310-0_sp1.1 from training. Duration: 22.986375
2022-12-23 21:28:32,492 WARNING [train.py:1060] (2/4) Exclude cut with ID 8040-260924-0003-80960-0_sp0.9 from training. Duration: 22.07225
2022-12-23 21:28:36,508 WARNING [train.py:1060] (2/4) Exclude cut with ID 7699-105389-0045-26330-0_sp0.9 from training. Duration: 20.3055625
2022-12-23 21:28:39,751 WARNING [train.py:1060] (2/4) Exclude cut with ID 6356-271890-0060-94317-0_sp0.9 from training. Duration: 20.72225
2022-12-23 21:28:56,009 WARNING [train.py:1060] (2/4) Exclude cut with ID 3488-85273-0017-41203-0_sp1.1 from training. Duration: 22.4818125
2022-12-23 21:29:22,941 WARNING [train.py:1060] (2/4) Exclude cut with ID 4964-30587-0040-44509-0_sp0.9 from training. Duration: 25.0944375
2022-12-23 21:29:26,272 WARNING [train.py:1060] (2/4) Exclude cut with ID 6533-399-0047-104881-0 from training. Duration: 21.515
2022-12-23 21:29:30,882 INFO [train.py:894] (2/4) Epoch 24, batch 2200, loss[loss=0.1809, simple_loss=0.2739, pruned_loss=0.04388, over 18721.00 frames. ], tot_loss[loss=0.1807, simple_loss=0.2635, pruned_loss=0.04898, over 3713077.05 frames. ], batch size: 65, lr: 4.76e-03, grad_scale: 16.0
2022-12-23 21:29:33,896 WARNING [train.py:1060] (2/4) Exclude cut with ID 7357-94126-0009-15840-0_sp0.9 from training. Duration: 27.02225
2022-12-23 21:29:35,206 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 2.795e+02 4.049e+02 5.149e+02 6.445e+02 1.343e+03, threshold=1.030e+03, percent-clipped=3.0
2022-12-23 21:29:38,102 WARNING [train.py:1060] (2/4) Exclude cut with ID 432-122774-0010-62480-0_sp0.9 from training. Duration: 22.22225
2022-12-23 21:29:42,646 INFO [zipformer.py:660] (2/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=82849.0, num_to_drop=1, layers_to_drop={3}
2022-12-23 21:29:45,310 WARNING [train.py:1060] (2/4) Exclude cut with ID 4964-30587-0085-44554-0_sp0.9 from training. Duration: 20.85
2022-12-23 21:30:00,165 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([2.0424, 1.2435, 0.7640, 1.4878, 2.3229, 1.4090, 1.8716, 1.9130],
       device='cuda:2'), covar=tensor([0.1446, 0.2180, 0.2481, 0.1462, 0.1704, 0.1798, 0.1349, 0.1545],
       device='cuda:2'), in_proj_covar=tensor([0.0094, 0.0099, 0.0119, 0.0098, 0.0120, 0.0093, 0.0099, 0.0094],
       device='cuda:2'), out_proj_covar=tensor([0.0003, 0.0003, 0.0004, 0.0003, 0.0004, 0.0003, 0.0003, 0.0003],
       device='cuda:2')
2022-12-23 21:30:17,818 WARNING [train.py:1060] (2/4) Exclude cut with ID 4295-39940-0007-92567-0 from training. Duration: 21.54
2022-12-23 21:30:21,274 WARNING [train.py:1060] (2/4) Exclude cut with ID 4964-30587-0040-44509-0_sp1.1 from training. Duration: 20.5318125
2022-12-23 21:30:30,155 WARNING [train.py:1060] (2/4) Exclude cut with ID 7255-291500-0012-134311-0_sp0.9 from training. Duration: 21.9333125
2022-12-23 21:30:36,997 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([3.4955, 2.6072, 2.1163, 3.2474, 2.4208, 2.5292, 2.6108, 3.5653],
       device='cuda:2'), covar=tensor([0.1725, 0.3145, 0.1889, 0.2815, 0.3794, 0.0972, 0.3179, 0.0737],
       device='cuda:2'), in_proj_covar=tensor([0.0299, 0.0295, 0.0249, 0.0348, 0.0277, 0.0232, 0.0292, 0.0219],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0003, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-23 21:30:48,589 INFO [train.py:894] (2/4) Epoch 24, batch 2250, loss[loss=0.2369, simple_loss=0.3038, pruned_loss=0.08498, over 18616.00 frames. ], tot_loss[loss=0.1812, simple_loss=0.2637, pruned_loss=0.04936, over 3712829.96 frames. ], batch size: 186, lr: 4.76e-03, grad_scale: 16.0
2022-12-23 21:31:18,820 WARNING [train.py:1060] (2/4) Exclude cut with ID 8631-249866-0025-130151-0_sp0.9 from training. Duration: 21.7944375
2022-12-23 21:31:27,289 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([2.3424, 2.2029, 1.9349, 1.3448, 1.9196, 2.0107, 1.8421, 2.0675],
       device='cuda:2'), covar=tensor([0.0613, 0.0431, 0.1067, 0.1234, 0.0938, 0.0972, 0.1128, 0.0651],
       device='cuda:2'), in_proj_covar=tensor([0.0171, 0.0186, 0.0204, 0.0189, 0.0207, 0.0200, 0.0215, 0.0201],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-23 21:31:29,986 WARNING [train.py:1060] (2/4) Exclude cut with ID 8291-282929-0002-12989-0_sp0.9 from training. Duration: 22.4666875
2022-12-23 21:31:35,867 WARNING [train.py:1060] (2/4) Exclude cut with ID 6121-9014-0076-24124-0 from training. Duration: 21.635
2022-12-23 21:31:40,573 WARNING [train.py:1060] (2/4) Exclude cut with ID 6121-9014-0076-24124-0_sp0.9 from training. Duration: 24.038875
2022-12-23 21:31:40,678 INFO [zipformer.py:660] (2/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=82925.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 21:32:06,056 INFO [train.py:894] (2/4) Epoch 24, batch 2300, loss[loss=0.1632, simple_loss=0.2363, pruned_loss=0.04511, over 18545.00 frames. ], tot_loss[loss=0.1805, simple_loss=0.2626, pruned_loss=0.04917, over 3713373.18 frames. ], batch size: 44, lr: 4.76e-03, grad_scale: 16.0
2022-12-23 21:32:10,496 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 2.645e+02 4.227e+02 4.886e+02 6.108e+02 1.315e+03, threshold=9.771e+02, percent-clipped=2.0
2022-12-23 21:32:24,019 WARNING [train.py:1060] (2/4) Exclude cut with ID 7255-291500-0009-134308-0_sp1.1 from training. Duration: 21.786375
2022-12-23 21:32:37,046 WARNING [train.py:1060] (2/4) Exclude cut with ID 8291-282929-0002-12989-0 from training. Duration: 20.22
2022-12-23 21:32:40,220 INFO [zipformer.py:660] (2/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=82963.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 21:33:17,122 INFO [scaling.py:679] (2/4) Whitening: num_groups=1, num_channels=384, metric=2.76 vs. limit=5.0
2022-12-23 21:33:21,970 INFO [train.py:894] (2/4) Epoch 24, batch 2350, loss[loss=0.1875, simple_loss=0.2781, pruned_loss=0.04844, over 18477.00 frames. ], tot_loss[loss=0.1808, simple_loss=0.2631, pruned_loss=0.04922, over 3713393.06 frames. ], batch size: 58, lr: 4.76e-03, grad_scale: 16.0
2022-12-23 21:33:39,531 INFO [zipformer.py:660] (2/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=83002.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 21:34:12,435 INFO [zipformer.py:660] (2/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=83024.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 21:34:37,750 INFO [train.py:894] (2/4) Epoch 24, batch 2400, loss[loss=0.1718, simple_loss=0.2587, pruned_loss=0.04246, over 18502.00 frames. ], tot_loss[loss=0.1809, simple_loss=0.2631, pruned_loss=0.04936, over 3713800.08 frames. ], batch size: 52, lr: 4.75e-03, grad_scale: 16.0
2022-12-23 21:34:39,338 WARNING [train.py:1060] (2/4) Exclude cut with ID 6951-79737-0043-132310-0 from training. Duration: 25.285
2022-12-23 21:34:41,926 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 2.252e+02 3.943e+02 4.892e+02 6.290e+02 1.556e+03, threshold=9.783e+02, percent-clipped=4.0
2022-12-23 21:34:51,782 INFO [zipformer.py:660] (2/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=83050.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 21:35:10,997 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([3.4660, 2.4569, 2.1630, 3.2477, 2.3858, 2.5625, 2.5408, 3.6460],
       device='cuda:2'), covar=tensor([0.1820, 0.3340, 0.1876, 0.2798, 0.3938, 0.0962, 0.3534, 0.0756],
       device='cuda:2'), in_proj_covar=tensor([0.0297, 0.0294, 0.0248, 0.0346, 0.0275, 0.0230, 0.0290, 0.0218],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0003, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-23 21:35:32,485 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.7613, 1.7030, 1.5460, 1.6713, 2.0110, 2.0041, 2.1243, 1.3884],
       device='cuda:2'), covar=tensor([0.0385, 0.0299, 0.0483, 0.0228, 0.0206, 0.0378, 0.0245, 0.0337],
       device='cuda:2'), in_proj_covar=tensor([0.0096, 0.0126, 0.0152, 0.0124, 0.0116, 0.0121, 0.0100, 0.0127],
       device='cuda:2'), out_proj_covar=tensor([7.5827e-05, 9.9853e-05, 1.2481e-04, 9.8321e-05, 9.3605e-05, 9.2562e-05,
        7.7560e-05, 9.9752e-05], device='cuda:2')
2022-12-23 21:35:45,711 WARNING [train.py:1060] (2/4) Exclude cut with ID 811-130148-0001-63453-0_sp0.9 from training. Duration: 20.861125
2022-12-23 21:35:54,366 INFO [train.py:894] (2/4) Epoch 24, batch 2450, loss[loss=0.212, simple_loss=0.2983, pruned_loss=0.06284, over 18528.00 frames. ], tot_loss[loss=0.1801, simple_loss=0.2623, pruned_loss=0.04892, over 3713158.83 frames. ], batch size: 58, lr: 4.75e-03, grad_scale: 16.0
2022-12-23 21:36:04,356 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([2.2735, 1.4845, 2.0119, 1.9059, 2.2989, 2.2877, 2.0599, 1.9629],
       device='cuda:2'), covar=tensor([0.2182, 0.3254, 0.2559, 0.2792, 0.1987, 0.0985, 0.3035, 0.1294],
       device='cuda:2'), in_proj_covar=tensor([0.0270, 0.0300, 0.0284, 0.0321, 0.0313, 0.0257, 0.0349, 0.0245],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-23 21:36:06,717 WARNING [train.py:1060] (2/4) Exclude cut with ID 6010-56788-0055-90261-0 from training. Duration: 20.88
2022-12-23 21:36:34,736 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([2.2052, 2.1635, 1.8999, 1.8371, 2.3456, 2.7160, 2.5527, 2.1063],
       device='cuda:2'), covar=tensor([0.0349, 0.0312, 0.0400, 0.0258, 0.0257, 0.0353, 0.0310, 0.0306],
       device='cuda:2'), in_proj_covar=tensor([0.0096, 0.0127, 0.0152, 0.0124, 0.0117, 0.0121, 0.0100, 0.0127],
       device='cuda:2'), out_proj_covar=tensor([7.6082e-05, 1.0015e-04, 1.2482e-04, 9.8705e-05, 9.4057e-05, 9.3130e-05,
        7.7912e-05, 1.0005e-04], device='cuda:2')
2022-12-23 21:36:40,280 WARNING [train.py:1060] (2/4) Exclude cut with ID 7357-94126-0045-15876-0_sp0.9 from training. Duration: 23.4166875
2022-12-23 21:37:10,374 INFO [train.py:894] (2/4) Epoch 24, batch 2500, loss[loss=0.1672, simple_loss=0.2606, pruned_loss=0.03693, over 18724.00 frames. ], tot_loss[loss=0.1798, simple_loss=0.2622, pruned_loss=0.04869, over 3712110.87 frames. ], batch size: 52, lr: 4.75e-03, grad_scale: 16.0
2022-12-23 21:37:15,185 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 2.926e+02 4.415e+02 5.249e+02 6.665e+02 1.083e+03, threshold=1.050e+03, percent-clipped=2.0
2022-12-23 21:37:22,612 INFO [zipformer.py:660] (2/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=83149.0, num_to_drop=1, layers_to_drop={0}
2022-12-23 21:37:35,925 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([2.5514, 2.1855, 1.7394, 2.1922, 1.9021, 2.0842, 2.0020, 2.3697],
       device='cuda:2'), covar=tensor([0.2021, 0.2995, 0.1966, 0.2715, 0.3521, 0.1074, 0.2839, 0.0999],
       device='cuda:2'), in_proj_covar=tensor([0.0300, 0.0297, 0.0250, 0.0350, 0.0278, 0.0233, 0.0293, 0.0220],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0003, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-23 21:37:57,652 WARNING [train.py:1060] (2/4) Exclude cut with ID 7255-291500-0006-134305-0 from training. Duration: 21.24
2022-12-23 21:37:57,667 WARNING [train.py:1060] (2/4) Exclude cut with ID 6533-399-0047-104881-0_sp0.9 from training. Duration: 23.9055625
2022-12-23 21:38:15,770 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([2.2197, 2.1497, 1.6573, 2.2148, 2.3912, 2.0331, 2.8628, 2.2525],
       device='cuda:2'), covar=tensor([0.0812, 0.1565, 0.2787, 0.1811, 0.1696, 0.0870, 0.0941, 0.1152],
       device='cuda:2'), in_proj_covar=tensor([0.0179, 0.0214, 0.0255, 0.0291, 0.0240, 0.0193, 0.0207, 0.0207],
       device='cuda:2'), out_proj_covar=tensor([0.0001, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-23 21:38:24,786 INFO [train.py:894] (2/4) Epoch 24, batch 2550, loss[loss=0.1785, simple_loss=0.2573, pruned_loss=0.04986, over 18647.00 frames. ], tot_loss[loss=0.1802, simple_loss=0.2628, pruned_loss=0.04884, over 3711827.11 frames. ], batch size: 48, lr: 4.75e-03, grad_scale: 16.0
2022-12-23 21:38:26,095 INFO [scaling.py:679] (2/4) Whitening: num_groups=8, num_channels=96, metric=1.11 vs. limit=2.0
2022-12-23 21:38:31,600 WARNING [train.py:1060] (2/4) Exclude cut with ID 6758-72288-0033-108368-0_sp0.9 from training. Duration: 25.988875
2022-12-23 21:38:33,442 INFO [zipformer.py:660] (2/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=83196.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 21:38:34,622 INFO [zipformer.py:660] (2/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=83197.0, num_to_drop=1, layers_to_drop={1}
2022-12-23 21:38:40,974 WARNING [train.py:1060] (2/4) Exclude cut with ID 7255-291500-0001-134300-0_sp0.9 from training. Duration: 20.67225
2022-12-23 21:39:08,296 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([2.6421, 2.7751, 2.9332, 1.7100, 3.1358, 3.0280, 2.2324, 3.1520],
       device='cuda:2'), covar=tensor([0.1082, 0.1559, 0.1331, 0.2101, 0.0658, 0.1103, 0.1986, 0.0569],
       device='cuda:2'), in_proj_covar=tensor([0.0196, 0.0214, 0.0206, 0.0192, 0.0174, 0.0214, 0.0214, 0.0192],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-23 21:39:17,898 INFO [zipformer.py:660] (2/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=83225.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 21:39:26,423 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([2.1621, 2.0215, 2.2473, 1.4007, 2.3646, 2.4325, 1.6265, 2.7209],
       device='cuda:2'), covar=tensor([0.1108, 0.1902, 0.1413, 0.1892, 0.0740, 0.1107, 0.2372, 0.0529],
       device='cuda:2'), in_proj_covar=tensor([0.0196, 0.0215, 0.0207, 0.0193, 0.0174, 0.0215, 0.0215, 0.0193],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-23 21:39:27,527 WARNING [train.py:1060] (2/4) Exclude cut with ID 3488-85273-0038-41224-0 from training. Duration: 20.34
2022-12-23 21:39:41,996 INFO [train.py:894] (2/4) Epoch 24, batch 2600, loss[loss=0.1717, simple_loss=0.2546, pruned_loss=0.04439, over 18456.00 frames. ], tot_loss[loss=0.1792, simple_loss=0.262, pruned_loss=0.04818, over 3712213.90 frames. ], batch size: 50, lr: 4.75e-03, grad_scale: 8.0
2022-12-23 21:39:47,859 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 2.557e+02 4.021e+02 4.692e+02 5.692e+02 9.573e+02, threshold=9.385e+02, percent-clipped=0.0
2022-12-23 21:39:54,172 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.8515, 1.3866, 0.9770, 1.5508, 2.3061, 1.3673, 1.7392, 1.7837],
       device='cuda:2'), covar=tensor([0.1514, 0.1946, 0.2036, 0.1363, 0.1554, 0.1682, 0.1334, 0.1587],
       device='cuda:2'), in_proj_covar=tensor([0.0094, 0.0098, 0.0117, 0.0096, 0.0118, 0.0092, 0.0098, 0.0094],
       device='cuda:2'), out_proj_covar=tensor([0.0003, 0.0003, 0.0004, 0.0003, 0.0004, 0.0003, 0.0003, 0.0003],
       device='cuda:2')
2022-12-23 21:40:06,971 INFO [zipformer.py:660] (2/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=83257.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 21:40:34,320 INFO [zipformer.py:660] (2/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=83273.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 21:40:42,918 WARNING [train.py:1060] (2/4) Exclude cut with ID 7357-94126-0026-15857-0_sp0.9 from training. Duration: 25.061125
2022-12-23 21:40:46,438 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.6706, 1.2598, 2.0909, 3.4224, 2.4530, 2.5027, 0.9256, 2.5177],
       device='cuda:2'), covar=tensor([0.1742, 0.1833, 0.1445, 0.0716, 0.1014, 0.1191, 0.2154, 0.0894],
       device='cuda:2'), in_proj_covar=tensor([0.0103, 0.0118, 0.0136, 0.0154, 0.0107, 0.0145, 0.0130, 0.0114],
       device='cuda:2'), out_proj_covar=tensor([0.0003, 0.0003, 0.0004, 0.0004, 0.0003, 0.0004, 0.0004, 0.0003],
       device='cuda:2')
2022-12-23 21:40:51,140 INFO [zipformer.py:660] (2/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=83284.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 21:40:55,195 WARNING [train.py:1060] (2/4) Exclude cut with ID 3033-130750-0096-55598-0 from training. Duration: 0.83
2022-12-23 21:41:01,171 INFO [train.py:894] (2/4) Epoch 24, batch 2650, loss[loss=0.1753, simple_loss=0.259, pruned_loss=0.04582, over 18549.00 frames. ], tot_loss[loss=0.1789, simple_loss=0.2619, pruned_loss=0.04796, over 3712657.83 frames. ], batch size: 47, lr: 4.75e-03, grad_scale: 8.0
2022-12-23 21:41:19,945 WARNING [train.py:1060] (2/4) Exclude cut with ID 3488-85273-0017-41203-0 from training. Duration: 24.73
2022-12-23 21:41:31,879 WARNING [train.py:1060] (2/4) Exclude cut with ID 7255-291500-0009-134308-0 from training. Duration: 23.965
2022-12-23 21:41:40,114 WARNING [train.py:1060] (2/4) Exclude cut with ID 6978-92210-0030-146996-0_sp0.9 from training. Duration: 22.088875
2022-12-23 21:41:44,525 INFO [zipformer.py:660] (2/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=83319.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 21:41:45,023 INFO [scaling.py:679] (2/4) Whitening: num_groups=8, num_channels=192, metric=1.53 vs. limit=2.0
2022-12-23 21:41:57,049 WARNING [train.py:1060] (2/4) Exclude cut with ID 7255-291500-0006-134305-0_sp0.9 from training. Duration: 23.6
2022-12-23 21:42:18,620 INFO [train.py:894] (2/4) Epoch 24, batch 2700, loss[loss=0.1892, simple_loss=0.271, pruned_loss=0.05372, over 18518.00 frames. ], tot_loss[loss=0.1795, simple_loss=0.262, pruned_loss=0.04852, over 3714007.37 frames. ], batch size: 77, lr: 4.75e-03, grad_scale: 8.0
2022-12-23 21:42:24,629 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 2.744e+02 4.043e+02 4.781e+02 6.098e+02 9.863e+02, threshold=9.563e+02, percent-clipped=3.0
2022-12-23 21:42:25,132 INFO [zipformer.py:660] (2/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=83345.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 21:43:35,056 INFO [train.py:894] (2/4) Epoch 24, batch 2750, loss[loss=0.1594, simple_loss=0.245, pruned_loss=0.03696, over 18380.00 frames. ], tot_loss[loss=0.1804, simple_loss=0.2625, pruned_loss=0.04917, over 3714525.19 frames. ], batch size: 46, lr: 4.74e-03, grad_scale: 8.0
2022-12-23 21:43:35,149 WARNING [train.py:1060] (2/4) Exclude cut with ID 8291-282929-0024-13011-0 from training. Duration: 23.795
2022-12-23 21:43:38,564 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([2.1357, 1.4619, 1.8528, 1.7724, 2.1343, 2.1281, 1.9788, 1.7850],
       device='cuda:2'), covar=tensor([0.2341, 0.3372, 0.2716, 0.3089, 0.2194, 0.1119, 0.3218, 0.1435],
       device='cuda:2'), in_proj_covar=tensor([0.0272, 0.0301, 0.0284, 0.0322, 0.0314, 0.0258, 0.0350, 0.0245],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-23 21:43:50,846 WARNING [train.py:1060] (2/4) Exclude cut with ID 8631-249866-0030-130156-0_sp1.1 from training. Duration: 21.5409375
2022-12-23 21:43:54,060 WARNING [train.py:1060] (2/4) Exclude cut with ID 6978-92210-0019-146985-0_sp0.9 from training. Duration: 24.97775
2022-12-23 21:44:04,714 WARNING [train.py:1060] (2/4) Exclude cut with ID 1085-156170-0017-128270-0_sp0.9 from training. Duration: 23.3444375
2022-12-23 21:44:12,890 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.3435, 1.4252, 1.1966, 1.6154, 1.6170, 3.0197, 1.4355, 1.5676],
       device='cuda:2'), covar=tensor([0.0970, 0.1824, 0.1105, 0.0954, 0.1499, 0.0276, 0.1422, 0.1553],
       device='cuda:2'), in_proj_covar=tensor([0.0073, 0.0082, 0.0072, 0.0074, 0.0092, 0.0077, 0.0084, 0.0076],
       device='cuda:2'), out_proj_covar=tensor([0.0004, 0.0004, 0.0004, 0.0004, 0.0004, 0.0004, 0.0004, 0.0004],
       device='cuda:2')
2022-12-23 21:44:33,550 WARNING [train.py:1060] (2/4) Exclude cut with ID 6010-56788-0055-90261-0_sp0.9 from training. Duration: 23.2
2022-12-23 21:44:38,952 WARNING [train.py:1060] (2/4) Exclude cut with ID 5653-46179-0060-117930-0_sp0.9 from training. Duration: 21.17225
2022-12-23 21:44:51,824 INFO [train.py:894] (2/4) Epoch 24, batch 2800, loss[loss=0.1464, simple_loss=0.2293, pruned_loss=0.03169, over 18692.00 frames. ], tot_loss[loss=0.1805, simple_loss=0.2629, pruned_loss=0.04905, over 3714036.80 frames. ], batch size: 46, lr: 4.74e-03, grad_scale: 8.0
2022-12-23 21:44:57,540 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 3.010e+02 4.230e+02 5.439e+02 6.718e+02 1.478e+03, threshold=1.088e+03, percent-clipped=3.0
2022-12-23 21:44:58,849 WARNING [train.py:1060] (2/4) Exclude cut with ID 7255-291500-0008-134307-0_sp0.9 from training. Duration: 24.6555625
2022-12-23 21:45:53,871 WARNING [train.py:1060] (2/4) Exclude cut with ID 3488-65654-0031-41259-0 from training. Duration: 20.44
2022-12-23 21:46:08,183 INFO [train.py:894] (2/4) Epoch 24, batch 2850, loss[loss=0.1766, simple_loss=0.2552, pruned_loss=0.04895, over 18383.00 frames. ], tot_loss[loss=0.1805, simple_loss=0.2628, pruned_loss=0.04915, over 3713812.67 frames. ], batch size: 46, lr: 4.74e-03, grad_scale: 8.0
2022-12-23 21:46:09,832 WARNING [train.py:1060] (2/4) Exclude cut with ID 6951-79737-0018-132285-0_sp0.9 from training. Duration: 23.45
2022-12-23 21:46:35,551 WARNING [train.py:1060] (2/4) Exclude cut with ID 6945-60535-0076-12784-0_sp0.9 from training. Duration: 20.52225
2022-12-23 21:46:43,265 WARNING [train.py:1060] (2/4) Exclude cut with ID 7255-291500-0008-134307-0 from training. Duration: 22.19
2022-12-23 21:46:52,182 WARNING [train.py:1060] (2/4) Exclude cut with ID 7357-94126-0014-15845-0_sp1.1 from training. Duration: 25.3818125
2022-12-23 21:47:07,807 WARNING [train.py:1060] (2/4) Exclude cut with ID 6951-79737-0043-132310-0_sp0.9 from training. Duration: 28.0944375
2022-12-23 21:47:14,477 WARNING [train.py:1060] (2/4) Exclude cut with ID 2195-150901-0045-59933-0_sp0.9 from training. Duration: 22.9444375
2022-12-23 21:47:23,283 WARNING [train.py:1060] (2/4) Exclude cut with ID 8291-282929-0024-13011-0_sp1.1 from training. Duration: 21.6318125
2022-12-23 21:47:26,370 INFO [train.py:894] (2/4) Epoch 24, batch 2900, loss[loss=0.2189, simple_loss=0.2933, pruned_loss=0.07229, over 18654.00 frames. ], tot_loss[loss=0.181, simple_loss=0.2632, pruned_loss=0.04945, over 3714243.18 frames. ], batch size: 175, lr: 4.74e-03, grad_scale: 8.0
2022-12-23 21:47:31,833 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 2.246e+02 4.184e+02 5.440e+02 6.811e+02 1.934e+03, threshold=1.088e+03, percent-clipped=6.0
2022-12-23 21:47:39,894 WARNING [train.py:1060] (2/4) Exclude cut with ID 8631-249866-0030-130156-0 from training. Duration: 23.695
2022-12-23 21:47:43,545 INFO [zipformer.py:660] (2/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=83552.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 21:48:05,363 WARNING [train.py:1060] (2/4) Exclude cut with ID 7699-105389-0094-26379-0 from training. Duration: 23.955
2022-12-23 21:48:30,100 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.8417, 1.2682, 0.7713, 1.4570, 2.1876, 1.2882, 1.5792, 1.7853],
       device='cuda:2'), covar=tensor([0.1642, 0.2196, 0.2316, 0.1560, 0.1798, 0.1893, 0.1509, 0.1675],
       device='cuda:2'), in_proj_covar=tensor([0.0094, 0.0098, 0.0117, 0.0097, 0.0119, 0.0092, 0.0099, 0.0094],
       device='cuda:2'), out_proj_covar=tensor([0.0003, 0.0003, 0.0004, 0.0003, 0.0004, 0.0003, 0.0003, 0.0003],
       device='cuda:2')
2022-12-23 21:48:40,664 WARNING [train.py:1060] (2/4) Exclude cut with ID 8291-282929-0024-13011-0_sp0.9 from training. Duration: 26.438875
2022-12-23 21:48:42,068 INFO [train.py:894] (2/4) Epoch 24, batch 2950, loss[loss=0.1991, simple_loss=0.2796, pruned_loss=0.05933, over 18721.00 frames. ], tot_loss[loss=0.1814, simple_loss=0.2637, pruned_loss=0.04954, over 3714443.17 frames. ], batch size: 52, lr: 4.74e-03, grad_scale: 8.0
2022-12-23 21:49:25,343 INFO [zipformer.py:660] (2/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=83619.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 21:49:26,605 WARNING [train.py:1060] (2/4) Exclude cut with ID 7699-105389-0021-26306-0_sp0.9 from training. Duration: 21.2444375
2022-12-23 21:49:27,912 WARNING [train.py:1060] (2/4) Exclude cut with ID 7357-94126-0014-15845-0_sp0.9 from training. Duration: 31.02225
2022-12-23 21:49:39,533 WARNING [train.py:1060] (2/4) Exclude cut with ID 432-122774-0017-62487-0 from training. Duration: 22.395
2022-12-23 21:49:57,050 WARNING [train.py:1060] (2/4) Exclude cut with ID 7357-94126-0045-15876-0 from training. Duration: 21.075
2022-12-23 21:49:57,164 INFO [zipformer.py:660] (2/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=83640.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 21:49:58,481 INFO [train.py:894] (2/4) Epoch 24, batch 3000, loss[loss=0.1612, simple_loss=0.238, pruned_loss=0.04221, over 18431.00 frames. ], tot_loss[loss=0.1805, simple_loss=0.2632, pruned_loss=0.04896, over 3713018.84 frames. ], batch size: 42, lr: 4.74e-03, grad_scale: 8.0
2022-12-23 21:49:58,481 INFO [train.py:919] (2/4) Computing validation loss
2022-12-23 21:50:09,481 INFO [train.py:928] (2/4) Epoch 24, validation: loss=0.1636, simple_loss=0.2603, pruned_loss=0.03339, over 944034.00 frames. 
2022-12-23 21:50:09,482 INFO [train.py:929] (2/4) Maximum memory allocated so far is 24680MB
2022-12-23 21:50:13,807 WARNING [train.py:1060] (2/4) Exclude cut with ID 6482-98857-0025-147532-0_sp0.9 from training. Duration: 20.0055625
2022-12-23 21:50:13,815 WARNING [train.py:1060] (2/4) Exclude cut with ID 6951-79737-0037-132304-0_sp0.9 from training. Duration: 22.05
2022-12-23 21:50:13,834 WARNING [train.py:1060] (2/4) Exclude cut with ID 7255-291500-0003-134302-0 from training. Duration: 26.8349375
2022-12-23 21:50:15,121 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 1.874e+02 4.185e+02 5.061e+02 6.325e+02 1.554e+03, threshold=1.012e+03, percent-clipped=3.0
2022-12-23 21:50:18,056 WARNING [train.py:1060] (2/4) Exclude cut with ID 7357-94126-0009-15840-0_sp1.1 from training. Duration: 22.1090625
2022-12-23 21:50:23,813 WARNING [train.py:1060] (2/4) Exclude cut with ID 7699-105389-0094-26379-0_sp0.9 from training. Duration: 26.6166875
2022-12-23 21:50:43,403 WARNING [train.py:1060] (2/4) Exclude cut with ID 2046-178027-0000-53705-0_sp0.9 from training. Duration: 20.3055625
2022-12-23 21:50:48,961 INFO [zipformer.py:660] (2/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=83667.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 21:51:03,950 WARNING [train.py:1060] (2/4) Exclude cut with ID 7205-50138-0008-5373-0_sp0.9 from training. Duration: 20.7
2022-12-23 21:51:24,968 INFO [train.py:894] (2/4) Epoch 24, batch 3050, loss[loss=0.2039, simple_loss=0.2898, pruned_loss=0.059, over 18515.00 frames. ], tot_loss[loss=0.1802, simple_loss=0.2629, pruned_loss=0.04868, over 3712853.88 frames. ], batch size: 52, lr: 4.74e-03, grad_scale: 8.0
2022-12-23 21:51:48,732 WARNING [train.py:1060] (2/4) Exclude cut with ID 6978-92210-0019-146985-0 from training. Duration: 22.48
2022-12-23 21:52:05,710 WARNING [train.py:1060] (2/4) Exclude cut with ID 7255-291500-0003-134302-0_sp0.9 from training. Duration: 29.816625
2022-12-23 21:52:24,368 WARNING [train.py:1060] (2/4) Exclude cut with ID 7255-291500-0005-134304-0_sp1.1 from training. Duration: 22.7590625
2022-12-23 21:52:28,500 WARNING [train.py:1060] (2/4) Exclude cut with ID 7357-94126-0026-15857-0 from training. Duration: 22.555
2022-12-23 21:52:40,211 INFO [train.py:894] (2/4) Epoch 24, batch 3100, loss[loss=0.1872, simple_loss=0.2763, pruned_loss=0.04902, over 18594.00 frames. ], tot_loss[loss=0.1805, simple_loss=0.2634, pruned_loss=0.04879, over 3713391.60 frames. ], batch size: 57, lr: 4.73e-03, grad_scale: 8.0
2022-12-23 21:52:46,094 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 2.916e+02 4.123e+02 4.986e+02 6.362e+02 1.111e+03, threshold=9.972e+02, percent-clipped=1.0
2022-12-23 21:52:47,262 INFO [scaling.py:679] (2/4) Whitening: num_groups=8, num_channels=192, metric=1.95 vs. limit=2.0
2022-12-23 21:52:50,262 WARNING [train.py:1060] (2/4) Exclude cut with ID 1250-135782-0005-25975-0_sp0.9 from training. Duration: 21.688875
2022-12-23 21:53:09,627 INFO [zipformer.py:660] (2/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=83760.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 21:53:27,972 WARNING [train.py:1060] (2/4) Exclude cut with ID 3488-85273-0038-41224-0_sp0.9 from training. Duration: 22.6
2022-12-23 21:53:56,001 INFO [train.py:894] (2/4) Epoch 24, batch 3150, loss[loss=0.1724, simple_loss=0.2688, pruned_loss=0.03794, over 18475.00 frames. ], tot_loss[loss=0.1795, simple_loss=0.2622, pruned_loss=0.04844, over 3714485.13 frames. ], batch size: 54, lr: 4.73e-03, grad_scale: 8.0
2022-12-23 21:54:04,933 WARNING [train.py:1060] (2/4) Exclude cut with ID 7357-94126-0009-15840-0 from training. Duration: 24.32
2022-12-23 21:54:43,409 INFO [zipformer.py:660] (2/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=83821.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 21:55:01,071 WARNING [train.py:1060] (2/4) Exclude cut with ID 8291-276745-0093-13116-0_sp0.9 from training. Duration: 21.061125
2022-12-23 21:55:13,155 INFO [train.py:894] (2/4) Epoch 24, batch 3200, loss[loss=0.1909, simple_loss=0.2671, pruned_loss=0.05735, over 18695.00 frames. ], tot_loss[loss=0.1799, simple_loss=0.2626, pruned_loss=0.04864, over 3714098.30 frames. ], batch size: 50, lr: 4.73e-03, grad_scale: 8.0
2022-12-23 21:55:13,228 WARNING [train.py:1060] (2/4) Exclude cut with ID 7357-94126-0024-15855-0_sp0.9 from training. Duration: 20.32225
2022-12-23 21:55:19,613 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 2.688e+02 4.275e+02 5.159e+02 6.611e+02 1.406e+03, threshold=1.032e+03, percent-clipped=3.0
2022-12-23 21:55:25,873 WARNING [train.py:1060] (2/4) Exclude cut with ID 3033-130750-0096-55598-0_sp1.1 from training. Duration: 0.7545625
2022-12-23 21:55:30,755 INFO [zipformer.py:660] (2/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=83852.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 21:55:38,361 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.5319, 1.4786, 1.5233, 1.5508, 1.1589, 3.4170, 1.4178, 1.9885],
       device='cuda:2'), covar=tensor([0.4484, 0.2804, 0.2534, 0.2836, 0.1715, 0.0299, 0.1800, 0.0935],
       device='cuda:2'), in_proj_covar=tensor([0.0133, 0.0117, 0.0125, 0.0121, 0.0105, 0.0097, 0.0091, 0.0090],
       device='cuda:2'), out_proj_covar=tensor([0.0005, 0.0005, 0.0005, 0.0005, 0.0004, 0.0003, 0.0004, 0.0004],
       device='cuda:2')
2022-12-23 21:55:39,416 WARNING [train.py:1060] (2/4) Exclude cut with ID 4295-39940-0007-92567-0_sp0.9 from training. Duration: 23.9333125
2022-12-23 21:56:12,206 WARNING [train.py:1060] (2/4) Exclude cut with ID 7255-291500-0008-134307-0_sp1.1 from training. Duration: 20.17275
2022-12-23 21:56:15,097 WARNING [train.py:1060] (2/4) Exclude cut with ID 6978-92210-0019-146985-0_sp1.1 from training. Duration: 20.436375
2022-12-23 21:56:30,602 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([2.0965, 1.9444, 2.1233, 2.1479, 1.6766, 4.5756, 2.1030, 2.6276],
       device='cuda:2'), covar=tensor([0.2853, 0.1909, 0.1716, 0.1928, 0.1331, 0.0145, 0.1548, 0.0792],
       device='cuda:2'), in_proj_covar=tensor([0.0133, 0.0117, 0.0125, 0.0121, 0.0105, 0.0097, 0.0091, 0.0090],
       device='cuda:2'), out_proj_covar=tensor([0.0005, 0.0005, 0.0005, 0.0005, 0.0004, 0.0003, 0.0004, 0.0004],
       device='cuda:2')
2022-12-23 21:56:31,573 INFO [train.py:894] (2/4) Epoch 24, batch 3250, loss[loss=0.1927, simple_loss=0.2847, pruned_loss=0.05034, over 18555.00 frames. ], tot_loss[loss=0.1801, simple_loss=0.263, pruned_loss=0.04862, over 3713459.20 frames. ], batch size: 97, lr: 4.73e-03, grad_scale: 8.0
2022-12-23 21:56:44,749 INFO [zipformer.py:660] (2/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=83900.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 21:56:55,720 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.2932, 1.2678, 0.8887, 1.4483, 1.4305, 2.4642, 1.2584, 1.4352],
       device='cuda:2'), covar=tensor([0.0852, 0.1783, 0.1113, 0.0836, 0.1549, 0.0341, 0.1412, 0.1492],
       device='cuda:2'), in_proj_covar=tensor([0.0073, 0.0082, 0.0072, 0.0074, 0.0091, 0.0076, 0.0084, 0.0076],
       device='cuda:2'), out_proj_covar=tensor([0.0004, 0.0004, 0.0004, 0.0004, 0.0004, 0.0004, 0.0004, 0.0004],
       device='cuda:2')
2022-12-23 21:57:39,336 WARNING [train.py:1060] (2/4) Exclude cut with ID 4234-40345-0022-142709-0_sp0.9 from training. Duration: 23.1055625
2022-12-23 21:57:39,573 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([4.2031, 3.5888, 3.6295, 4.1095, 3.8306, 3.6092, 4.2682, 1.3126],
       device='cuda:2'), covar=tensor([0.0742, 0.0823, 0.0761, 0.0833, 0.1366, 0.1313, 0.0742, 0.5138],
       device='cuda:2'), in_proj_covar=tensor([0.0355, 0.0235, 0.0246, 0.0283, 0.0336, 0.0277, 0.0302, 0.0292],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0001, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-23 21:57:40,778 WARNING [train.py:1060] (2/4) Exclude cut with ID 8291-282929-0007-12994-0_sp1.1 from training. Duration: 23.5
2022-12-23 21:57:44,707 INFO [zipformer.py:660] (2/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=83940.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 21:57:45,945 INFO [train.py:894] (2/4) Epoch 24, batch 3300, loss[loss=0.1947, simple_loss=0.2724, pruned_loss=0.05855, over 18670.00 frames. ], tot_loss[loss=0.1792, simple_loss=0.2622, pruned_loss=0.04813, over 3713397.12 frames. ], batch size: 184, lr: 4.73e-03, grad_scale: 8.0
2022-12-23 21:57:51,364 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 2.326e+02 4.315e+02 5.085e+02 6.619e+02 1.531e+03, threshold=1.017e+03, percent-clipped=2.0
2022-12-23 21:57:52,709 WARNING [train.py:1060] (2/4) Exclude cut with ID 7255-291500-0009-134308-0_sp0.9 from training. Duration: 26.62775
2022-12-23 21:57:58,145 INFO [zipformer.py:660] (2/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=83949.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 21:58:07,490 WARNING [train.py:1060] (2/4) Exclude cut with ID 6951-79737-0018-132285-0 from training. Duration: 21.105
2022-12-23 21:58:10,518 WARNING [train.py:1060] (2/4) Exclude cut with ID 4511-76322-0006-80011-0_sp0.9 from training. Duration: 24.411125
2022-12-23 21:58:38,842 WARNING [train.py:1060] (2/4) Exclude cut with ID 6758-72288-0033-108368-0_sp1.1 from training. Duration: 21.263625
2022-12-23 21:58:48,783 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.7367, 1.4879, 1.0933, 0.2617, 1.1480, 1.6036, 1.4363, 1.4768],
       device='cuda:2'), covar=tensor([0.0681, 0.0569, 0.1061, 0.1654, 0.1096, 0.1590, 0.1657, 0.0712],
       device='cuda:2'), in_proj_covar=tensor([0.0176, 0.0190, 0.0210, 0.0194, 0.0212, 0.0206, 0.0220, 0.0205],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-23 21:58:54,227 INFO [zipformer.py:660] (2/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=83985.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 21:58:58,258 INFO [zipformer.py:660] (2/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=83988.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 21:59:02,661 INFO [train.py:894] (2/4) Epoch 24, batch 3350, loss[loss=0.1789, simple_loss=0.248, pruned_loss=0.05486, over 18438.00 frames. ], tot_loss[loss=0.1797, simple_loss=0.2622, pruned_loss=0.0486, over 3713534.42 frames. ], batch size: 42, lr: 4.73e-03, grad_scale: 8.0
2022-12-23 21:59:10,393 WARNING [train.py:1060] (2/4) Exclude cut with ID 4234-40345-0022-142709-0 from training. Duration: 20.795
2022-12-23 21:59:25,186 WARNING [train.py:1060] (2/4) Exclude cut with ID 7357-94126-0021-15852-0 from training. Duration: 24.76
2022-12-23 21:59:25,205 WARNING [train.py:1060] (2/4) Exclude cut with ID 3867-173237-0077-144769-0_sp0.9 from training. Duration: 22.25
2022-12-23 21:59:32,246 INFO [scaling.py:679] (2/4) Whitening: num_groups=8, num_channels=96, metric=1.53 vs. limit=2.0
2022-12-23 21:59:35,940 INFO [zipformer.py:660] (2/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=84010.0, num_to_drop=1, layers_to_drop={2}
2022-12-23 21:59:49,859 WARNING [train.py:1060] (2/4) Exclude cut with ID 7357-94126-0026-15857-0_sp1.1 from training. Duration: 20.5045625
2022-12-23 22:00:21,132 INFO [train.py:894] (2/4) Epoch 24, batch 3400, loss[loss=0.1743, simple_loss=0.2612, pruned_loss=0.04368, over 18639.00 frames. ], tot_loss[loss=0.1788, simple_loss=0.261, pruned_loss=0.04825, over 3712963.03 frames. ], batch size: 53, lr: 4.73e-03, grad_scale: 8.0
2022-12-23 22:00:23,825 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.3355, 1.3429, 1.4857, 0.9061, 1.3800, 1.3716, 1.1468, 1.6267],
       device='cuda:2'), covar=tensor([0.1016, 0.1896, 0.1216, 0.1442, 0.0756, 0.1014, 0.2624, 0.0591],
       device='cuda:2'), in_proj_covar=tensor([0.0195, 0.0213, 0.0206, 0.0193, 0.0172, 0.0213, 0.0214, 0.0192],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-23 22:00:27,540 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 3.020e+02 4.214e+02 5.258e+02 6.086e+02 1.654e+03, threshold=1.052e+03, percent-clipped=1.0
2022-12-23 22:00:29,294 INFO [zipformer.py:660] (2/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=84046.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 22:01:18,142 INFO [scaling.py:679] (2/4) Whitening: num_groups=8, num_channels=96, metric=1.54 vs. limit=2.0
2022-12-23 22:01:20,748 INFO [scaling.py:679] (2/4) Whitening: num_groups=8, num_channels=96, metric=1.14 vs. limit=2.0
2022-12-23 22:01:33,006 INFO [train.py:894] (2/4) Epoch 24, batch 3450, loss[loss=0.1566, simple_loss=0.2335, pruned_loss=0.03988, over 18530.00 frames. ], tot_loss[loss=0.1787, simple_loss=0.2609, pruned_loss=0.04821, over 3713036.36 frames. ], batch size: 44, lr: 4.72e-03, grad_scale: 8.0
2022-12-23 22:01:41,737 INFO [zipformer.py:660] (2/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=84097.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 22:01:49,283 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([2.3413, 2.8229, 2.6971, 1.2127, 2.9897, 2.1566, 0.5402, 1.7773],
       device='cuda:2'), covar=tensor([0.2026, 0.1464, 0.1850, 0.3869, 0.1145, 0.1166, 0.4884, 0.1824],
       device='cuda:2'), in_proj_covar=tensor([0.0151, 0.0149, 0.0164, 0.0126, 0.0151, 0.0117, 0.0147, 0.0117],
       device='cuda:2'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0002, 0.0003, 0.0002, 0.0003, 0.0002],
       device='cuda:2')
2022-12-23 22:02:09,328 INFO [zipformer.py:660] (2/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=84116.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 22:02:15,620 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([2.3035, 2.3152, 1.6481, 2.5493, 2.5524, 2.1793, 3.0802, 2.3640],
       device='cuda:2'), covar=tensor([0.0929, 0.1907, 0.3020, 0.1981, 0.1801, 0.0968, 0.0991, 0.1317],
       device='cuda:2'), in_proj_covar=tensor([0.0181, 0.0215, 0.0257, 0.0295, 0.0243, 0.0195, 0.0210, 0.0209],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-23 22:02:30,310 INFO [scaling.py:679] (2/4) Whitening: num_groups=1, num_channels=384, metric=2.79 vs. limit=5.0
2022-12-23 22:02:41,183 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([2.1982, 1.6361, 2.3170, 3.9820, 2.9916, 2.8905, 0.9914, 2.8334],
       device='cuda:2'), covar=tensor([0.1617, 0.1642, 0.1502, 0.0517, 0.0890, 0.1525, 0.2228, 0.0980],
       device='cuda:2'), in_proj_covar=tensor([0.0103, 0.0118, 0.0137, 0.0154, 0.0107, 0.0145, 0.0130, 0.0115],
       device='cuda:2'), out_proj_covar=tensor([0.0003, 0.0003, 0.0004, 0.0004, 0.0003, 0.0004, 0.0004, 0.0003],
       device='cuda:2')
2022-12-23 22:02:45,416 INFO [train.py:894] (2/4) Epoch 24, batch 3500, loss[loss=0.2256, simple_loss=0.3057, pruned_loss=0.07274, over 18509.00 frames. ], tot_loss[loss=0.1791, simple_loss=0.2613, pruned_loss=0.0485, over 3713590.56 frames. ], batch size: 64, lr: 4.72e-03, grad_scale: 8.0
2022-12-23 22:02:50,006 INFO [scaling.py:679] (2/4) Whitening: num_groups=1, num_channels=384, metric=4.68 vs. limit=5.0
2022-12-23 22:02:52,081 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 3.085e+02 4.483e+02 5.516e+02 7.129e+02 1.524e+03, threshold=1.103e+03, percent-clipped=5.0
2022-12-23 22:03:06,731 WARNING [train.py:1060] (2/4) Exclude cut with ID 7859-102521-0017-7548-0_sp1.1 from training. Duration: 22.2954375
2022-12-23 22:03:16,195 INFO [train.py:894] (2/4) Epoch 25, batch 0, loss[loss=0.1676, simple_loss=0.2516, pruned_loss=0.04181, over 18544.00 frames. ], tot_loss[loss=0.1676, simple_loss=0.2516, pruned_loss=0.04181, over 18544.00 frames. ], batch size: 44, lr: 4.63e-03, grad_scale: 8.0
2022-12-23 22:03:16,195 INFO [train.py:919] (2/4) Computing validation loss
2022-12-23 22:03:27,094 INFO [train.py:928] (2/4) Epoch 25, validation: loss=0.1626, simple_loss=0.2599, pruned_loss=0.03263, over 944034.00 frames. 
2022-12-23 22:03:27,095 INFO [train.py:929] (2/4) Maximum memory allocated so far is 24680MB
2022-12-23 22:03:44,439 INFO [zipformer.py:660] (2/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=84158.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 22:03:50,758 INFO [scaling.py:679] (2/4) Whitening: num_groups=1, num_channels=384, metric=3.14 vs. limit=5.0
2022-12-23 22:04:20,817 WARNING [train.py:1060] (2/4) Exclude cut with ID 298-126791-0067-24026-0_sp0.9 from training. Duration: 21.438875
2022-12-23 22:04:25,120 WARNING [train.py:1060] (2/4) Exclude cut with ID 5652-39938-0025-23684-0_sp0.9 from training. Duration: 22.2055625
2022-12-23 22:04:41,761 INFO [train.py:894] (2/4) Epoch 25, batch 50, loss[loss=0.1644, simple_loss=0.2456, pruned_loss=0.04157, over 18607.00 frames. ], tot_loss[loss=0.1714, simple_loss=0.2598, pruned_loss=0.04151, over 837209.90 frames. ], batch size: 45, lr: 4.62e-03, grad_scale: 8.0
2022-12-23 22:04:45,780 INFO [zipformer.py:660] (2/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=84199.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 22:05:40,585 INFO [zipformer.py:660] (2/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=84236.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 22:05:54,057 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 2.027e+02 3.187e+02 3.567e+02 4.493e+02 9.384e+02, threshold=7.134e+02, percent-clipped=0.0
2022-12-23 22:05:56,768 INFO [train.py:894] (2/4) Epoch 25, batch 100, loss[loss=0.1793, simple_loss=0.2679, pruned_loss=0.0454, over 18610.00 frames. ], tot_loss[loss=0.1694, simple_loss=0.2584, pruned_loss=0.04021, over 1474875.18 frames. ], batch size: 53, lr: 4.62e-03, grad_scale: 8.0
2022-12-23 22:06:10,235 INFO [scaling.py:679] (2/4) Whitening: num_groups=8, num_channels=96, metric=1.44 vs. limit=2.0
2022-12-23 22:06:16,902 INFO [zipformer.py:660] (2/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=84260.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 22:07:10,955 INFO [train.py:894] (2/4) Epoch 25, batch 150, loss[loss=0.1484, simple_loss=0.2366, pruned_loss=0.03014, over 18600.00 frames. ], tot_loss[loss=0.1683, simple_loss=0.2568, pruned_loss=0.03992, over 1971257.76 frames. ], batch size: 45, lr: 4.62e-03, grad_scale: 8.0
2022-12-23 22:07:11,454 INFO [zipformer.py:660] (2/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=84297.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 22:07:23,443 INFO [zipformer.py:660] (2/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=84305.0, num_to_drop=1, layers_to_drop={3}
2022-12-23 22:07:27,696 WARNING [train.py:1060] (2/4) Exclude cut with ID 7859-102521-0017-7548-0 from training. Duration: 24.525
2022-12-23 22:07:30,981 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.7742, 1.2817, 0.9317, 1.4003, 2.1310, 1.4414, 1.5318, 1.7617],
       device='cuda:2'), covar=tensor([0.1681, 0.2181, 0.2161, 0.1534, 0.1784, 0.1738, 0.1458, 0.1725],
       device='cuda:2'), in_proj_covar=tensor([0.0094, 0.0098, 0.0117, 0.0096, 0.0119, 0.0092, 0.0098, 0.0094],
       device='cuda:2'), out_proj_covar=tensor([0.0003, 0.0003, 0.0004, 0.0003, 0.0004, 0.0003, 0.0003, 0.0003],
       device='cuda:2')
2022-12-23 22:08:02,229 WARNING [train.py:1060] (2/4) Exclude cut with ID 3699-47246-0007-3408-0_sp0.9 from training. Duration: 20.26675
2022-12-23 22:08:09,895 INFO [zipformer.py:660] (2/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=84336.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 22:08:16,145 WARNING [train.py:1060] (2/4) Exclude cut with ID 7859-102521-0017-7548-0_sp0.9 from training. Duration: 27.25
2022-12-23 22:08:17,450 INFO [zipformer.py:660] (2/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=84341.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 22:08:23,126 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 1.898e+02 3.025e+02 3.482e+02 4.265e+02 1.190e+03, threshold=6.964e+02, percent-clipped=5.0
2022-12-23 22:08:26,658 INFO [train.py:894] (2/4) Epoch 25, batch 200, loss[loss=0.1818, simple_loss=0.2704, pruned_loss=0.0466, over 18684.00 frames. ], tot_loss[loss=0.1675, simple_loss=0.2562, pruned_loss=0.03943, over 2356378.53 frames. ], batch size: 60, lr: 4.62e-03, grad_scale: 8.0
2022-12-23 22:08:53,208 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([2.1778, 1.3238, 1.9358, 1.7731, 2.1701, 2.1672, 1.9449, 1.8825],
       device='cuda:2'), covar=tensor([0.2282, 0.3654, 0.2743, 0.3044, 0.2270, 0.1082, 0.3324, 0.1389],
       device='cuda:2'), in_proj_covar=tensor([0.0271, 0.0301, 0.0285, 0.0323, 0.0315, 0.0258, 0.0350, 0.0246],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-23 22:09:26,933 WARNING [train.py:1060] (2/4) Exclude cut with ID 6426-64292-0017-15984-0 from training. Duration: 21.68
2022-12-23 22:09:38,738 WARNING [train.py:1060] (2/4) Exclude cut with ID 4278-13270-0007-59342-0 from training. Duration: 21.6300625
2022-12-23 22:09:41,364 INFO [train.py:894] (2/4) Epoch 25, batch 250, loss[loss=0.1753, simple_loss=0.2653, pruned_loss=0.0427, over 18392.00 frames. ], tot_loss[loss=0.169, simple_loss=0.2577, pruned_loss=0.04009, over 2658213.61 frames. ], batch size: 53, lr: 4.62e-03, grad_scale: 8.0
2022-12-23 22:09:41,822 INFO [zipformer.py:660] (2/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=84397.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 22:10:02,564 WARNING [train.py:1060] (2/4) Exclude cut with ID 4278-13270-0007-59342-0_sp0.9 from training. Duration: 24.033375
2022-12-23 22:10:09,745 INFO [zipformer.py:660] (2/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=84416.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 22:10:53,305 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 2.137e+02 3.339e+02 3.977e+02 4.779e+02 8.289e+02, threshold=7.955e+02, percent-clipped=6.0
2022-12-23 22:10:56,495 INFO [train.py:894] (2/4) Epoch 25, batch 300, loss[loss=0.1797, simple_loss=0.2696, pruned_loss=0.04489, over 18595.00 frames. ], tot_loss[loss=0.1685, simple_loss=0.2572, pruned_loss=0.03988, over 2892106.20 frames. ], batch size: 51, lr: 4.62e-03, grad_scale: 8.0
2022-12-23 22:10:58,045 WARNING [train.py:1060] (2/4) Exclude cut with ID 4278-13270-0009-59344-0 from training. Duration: 22.905
2022-12-23 22:10:59,413 WARNING [train.py:1060] (2/4) Exclude cut with ID 5622-44585-0006-90525-0_sp1.1 from training. Duration: 23.4318125
2022-12-23 22:11:01,397 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.6839, 3.1075, 3.3099, 1.2709, 2.9866, 3.8063, 2.7932, 2.9454],
       device='cuda:2'), covar=tensor([0.0720, 0.0356, 0.0276, 0.0534, 0.0350, 0.0343, 0.0381, 0.0602],
       device='cuda:2'), in_proj_covar=tensor([0.0151, 0.0174, 0.0131, 0.0143, 0.0151, 0.0145, 0.0167, 0.0178],
       device='cuda:2'), out_proj_covar=tensor([1.1433e-04, 1.3150e-04, 9.6979e-05, 1.0534e-04, 1.1164e-04, 1.0984e-04,
        1.2689e-04, 1.3404e-04], device='cuda:2')
2022-12-23 22:11:06,070 INFO [zipformer.py:660] (2/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=84453.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 22:11:22,660 INFO [zipformer.py:660] (2/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=84464.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 22:11:32,138 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([3.2103, 5.7169, 5.0561, 3.2380, 5.8102, 4.5924, 1.3367, 3.8901],
       device='cuda:2'), covar=tensor([0.1803, 0.0665, 0.1232, 0.2502, 0.0495, 0.0562, 0.4649, 0.1214],
       device='cuda:2'), in_proj_covar=tensor([0.0148, 0.0145, 0.0160, 0.0123, 0.0147, 0.0114, 0.0143, 0.0115],
       device='cuda:2'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0002, 0.0003, 0.0002, 0.0003, 0.0002],
       device='cuda:2')
2022-12-23 22:12:12,042 INFO [train.py:894] (2/4) Epoch 25, batch 350, loss[loss=0.1692, simple_loss=0.2631, pruned_loss=0.03763, over 18585.00 frames. ], tot_loss[loss=0.1679, simple_loss=0.2569, pruned_loss=0.03949, over 3074719.56 frames. ], batch size: 57, lr: 4.62e-03, grad_scale: 8.0
2022-12-23 22:12:18,113 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([2.0062, 1.8599, 1.6772, 1.0238, 2.2419, 2.0478, 1.8314, 1.5943],
       device='cuda:2'), covar=tensor([0.0405, 0.0491, 0.0549, 0.0849, 0.0340, 0.0437, 0.0495, 0.0951],
       device='cuda:2'), in_proj_covar=tensor([0.0123, 0.0129, 0.0128, 0.0119, 0.0101, 0.0125, 0.0133, 0.0159],
       device='cuda:2'), out_proj_covar=tensor([0.0001, 0.0002, 0.0002, 0.0001, 0.0001, 0.0001, 0.0002, 0.0002],
       device='cuda:2')
2022-12-23 22:12:23,788 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.5181, 1.0807, 1.7751, 2.9141, 2.2145, 2.4012, 0.6899, 2.0926],
       device='cuda:2'), covar=tensor([0.1857, 0.1789, 0.1544, 0.0626, 0.0988, 0.1179, 0.2422, 0.1076],
       device='cuda:2'), in_proj_covar=tensor([0.0102, 0.0117, 0.0135, 0.0153, 0.0106, 0.0144, 0.0129, 0.0113],
       device='cuda:2'), out_proj_covar=tensor([0.0003, 0.0003, 0.0004, 0.0004, 0.0003, 0.0004, 0.0003, 0.0003],
       device='cuda:2')
2022-12-23 22:12:54,683 INFO [zipformer.py:660] (2/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=84525.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 22:12:57,209 WARNING [train.py:1060] (2/4) Exclude cut with ID 4278-13270-0009-59344-0_sp1.1 from training. Duration: 20.82275
2022-12-23 22:12:58,709 WARNING [train.py:1060] (2/4) Exclude cut with ID 4278-13270-0009-59344-0_sp0.9 from training. Duration: 25.45
2022-12-23 22:13:24,702 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 2.178e+02 3.360e+02 4.039e+02 5.320e+02 1.136e+03, threshold=8.077e+02, percent-clipped=4.0
2022-12-23 22:13:27,741 INFO [train.py:894] (2/4) Epoch 25, batch 400, loss[loss=0.1732, simple_loss=0.2655, pruned_loss=0.04044, over 18591.00 frames. ], tot_loss[loss=0.1679, simple_loss=0.2574, pruned_loss=0.03919, over 3215209.67 frames. ], batch size: 51, lr: 4.61e-03, grad_scale: 8.0
2022-12-23 22:13:35,417 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([2.2850, 1.7736, 2.1812, 2.6072, 2.2596, 4.8854, 1.9214, 1.9415],
       device='cuda:2'), covar=tensor([0.0728, 0.1643, 0.0903, 0.0867, 0.1274, 0.0148, 0.1286, 0.1418],
       device='cuda:2'), in_proj_covar=tensor([0.0072, 0.0081, 0.0071, 0.0073, 0.0091, 0.0076, 0.0084, 0.0076],
       device='cuda:2'), out_proj_covar=tensor([0.0004, 0.0004, 0.0004, 0.0004, 0.0004, 0.0004, 0.0004, 0.0004],
       device='cuda:2')
2022-12-23 22:13:40,359 INFO [zipformer.py:660] (2/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=84555.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 22:13:58,004 WARNING [train.py:1060] (2/4) Exclude cut with ID 5622-44585-0006-90525-0 from training. Duration: 25.775
2022-12-23 22:14:20,063 WARNING [train.py:1060] (2/4) Exclude cut with ID 2929-85685-0071-62375-0_sp0.9 from training. Duration: 22.25
2022-12-23 22:14:21,162 INFO [scaling.py:679] (2/4) Whitening: num_groups=8, num_channels=96, metric=1.35 vs. limit=2.0
2022-12-23 22:14:26,199 INFO [zipformer.py:660] (2/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=84586.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 22:14:35,575 INFO [zipformer.py:660] (2/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=84592.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 22:14:42,391 INFO [train.py:894] (2/4) Epoch 25, batch 450, loss[loss=0.1507, simple_loss=0.241, pruned_loss=0.03023, over 18386.00 frames. ], tot_loss[loss=0.1678, simple_loss=0.2574, pruned_loss=0.03907, over 3325395.25 frames. ], batch size: 46, lr: 4.61e-03, grad_scale: 8.0
2022-12-23 22:14:46,833 WARNING [train.py:1060] (2/4) Exclude cut with ID 3972-170212-0014-23379-0 from training. Duration: 26.205
2022-12-23 22:14:55,040 INFO [zipformer.py:660] (2/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=84605.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 22:15:02,639 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([0.3057, 1.7321, 1.8898, 1.0922, 1.2203, 2.0709, 1.8565, 1.6248],
       device='cuda:2'), covar=tensor([0.0807, 0.0303, 0.0313, 0.0375, 0.0395, 0.0427, 0.0240, 0.0621],
       device='cuda:2'), in_proj_covar=tensor([0.0150, 0.0173, 0.0131, 0.0143, 0.0151, 0.0145, 0.0167, 0.0177],
       device='cuda:2'), out_proj_covar=tensor([1.1378e-04, 1.3133e-04, 9.6872e-05, 1.0550e-04, 1.1111e-04, 1.0995e-04,
        1.2680e-04, 1.3355e-04], device='cuda:2')
2022-12-23 22:15:05,364 WARNING [train.py:1060] (2/4) Exclude cut with ID 5239-32139-0047-9341-0_sp0.9 from training. Duration: 30.1555625
2022-12-23 22:15:11,959 WARNING [train.py:1060] (2/4) Exclude cut with ID 1265-135635-0050-6781-0_sp0.9 from training. Duration: 21.8333125
2022-12-23 22:15:20,687 WARNING [train.py:1060] (2/4) Exclude cut with ID 1914-133440-0024-94914-0_sp1.1 from training. Duration: 20.6545625
2022-12-23 22:15:37,902 INFO [zipformer.py:660] (2/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=84633.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 22:15:43,998 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([2.6974, 2.2888, 1.7217, 2.5691, 2.0443, 2.2463, 2.0901, 2.6435],
       device='cuda:2'), covar=tensor([0.2062, 0.3111, 0.2028, 0.2581, 0.3779, 0.1058, 0.3144, 0.0954],
       device='cuda:2'), in_proj_covar=tensor([0.0297, 0.0292, 0.0248, 0.0346, 0.0275, 0.0230, 0.0292, 0.0218],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-23 22:15:50,221 INFO [zipformer.py:660] (2/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=84641.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 22:15:55,472 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 2.349e+02 3.119e+02 3.751e+02 4.585e+02 8.162e+02, threshold=7.502e+02, percent-clipped=1.0
2022-12-23 22:15:58,513 INFO [train.py:894] (2/4) Epoch 25, batch 500, loss[loss=0.1643, simple_loss=0.2644, pruned_loss=0.03213, over 18517.00 frames. ], tot_loss[loss=0.1682, simple_loss=0.2575, pruned_loss=0.03941, over 3412238.10 frames. ], batch size: 52, lr: 4.61e-03, grad_scale: 8.0
2022-12-23 22:16:01,587 WARNING [train.py:1060] (2/4) Exclude cut with ID 7395-89880-0045-39920-0_sp0.9 from training. Duration: 20.52225
2022-12-23 22:16:07,908 INFO [zipformer.py:660] (2/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=84653.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 22:16:21,227 WARNING [train.py:1060] (2/4) Exclude cut with ID 3972-170212-0014-23379-0_sp0.9 from training. Duration: 29.1166875
2022-12-23 22:16:40,118 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.6865, 1.6017, 1.6639, 1.7028, 1.2235, 3.7342, 1.5588, 2.0916],
       device='cuda:2'), covar=tensor([0.3022, 0.2027, 0.1955, 0.2013, 0.1471, 0.0168, 0.1584, 0.0836],
       device='cuda:2'), in_proj_covar=tensor([0.0131, 0.0117, 0.0124, 0.0120, 0.0104, 0.0097, 0.0090, 0.0089],
       device='cuda:2'), out_proj_covar=tensor([0.0005, 0.0005, 0.0005, 0.0005, 0.0004, 0.0003, 0.0004, 0.0004],
       device='cuda:2')
2022-12-23 22:16:42,230 INFO [scaling.py:679] (2/4) Whitening: num_groups=8, num_channels=96, metric=1.24 vs. limit=2.0
2022-12-23 22:17:02,969 INFO [zipformer.py:660] (2/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=84689.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 22:17:07,077 INFO [zipformer.py:660] (2/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=84692.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 22:17:10,233 INFO [zipformer.py:660] (2/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=84694.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 22:17:14,082 INFO [train.py:894] (2/4) Epoch 25, batch 550, loss[loss=0.196, simple_loss=0.2848, pruned_loss=0.05363, over 18564.00 frames. ], tot_loss[loss=0.1696, simple_loss=0.2593, pruned_loss=0.03998, over 3479470.25 frames. ], batch size: 98, lr: 4.61e-03, grad_scale: 8.0
2022-12-23 22:17:19,743 WARNING [train.py:1060] (2/4) Exclude cut with ID 543-133211-0007-59831-0_sp0.9 from training. Duration: 21.388875
2022-12-23 22:17:33,102 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.2848, 1.3493, 1.4325, 0.8925, 1.3667, 1.3732, 1.1686, 1.6316],
       device='cuda:2'), covar=tensor([0.1144, 0.2039, 0.1206, 0.1509, 0.0832, 0.1094, 0.2588, 0.0612],
       device='cuda:2'), in_proj_covar=tensor([0.0199, 0.0216, 0.0209, 0.0196, 0.0176, 0.0217, 0.0216, 0.0194],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-23 22:17:57,265 WARNING [train.py:1060] (2/4) Exclude cut with ID 1914-133440-0024-94914-0 from training. Duration: 22.72
2022-12-23 22:17:57,304 WARNING [train.py:1060] (2/4) Exclude cut with ID 1914-133440-0031-94921-0_sp0.9 from training. Duration: 22.7444375
2022-12-23 22:18:25,919 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 2.064e+02 3.578e+02 4.188e+02 5.086e+02 7.637e+02, threshold=8.377e+02, percent-clipped=1.0
2022-12-23 22:18:29,240 INFO [train.py:894] (2/4) Epoch 25, batch 600, loss[loss=0.1947, simple_loss=0.2815, pruned_loss=0.05389, over 18607.00 frames. ], tot_loss[loss=0.1707, simple_loss=0.2599, pruned_loss=0.04072, over 3531195.58 frames. ], batch size: 69, lr: 4.61e-03, grad_scale: 8.0
2022-12-23 22:18:38,627 INFO [zipformer.py:660] (2/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=84753.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 22:18:39,842 WARNING [train.py:1060] (2/4) Exclude cut with ID 4133-6541-0027-40495-0_sp1.1 from training. Duration: 0.9681875
2022-12-23 22:18:44,134 WARNING [train.py:1060] (2/4) Exclude cut with ID 6330-62851-0022-91297-0_sp0.9 from training. Duration: 22.3166875
2022-12-23 22:18:50,398 WARNING [train.py:1060] (2/4) Exclude cut with ID 543-133212-0015-59917-0_sp0.9 from training. Duration: 21.8166875
2022-12-23 22:19:44,490 INFO [train.py:894] (2/4) Epoch 25, batch 650, loss[loss=0.1728, simple_loss=0.2684, pruned_loss=0.03858, over 18574.00 frames. ], tot_loss[loss=0.1712, simple_loss=0.2606, pruned_loss=0.04089, over 3572584.54 frames. ], batch size: 51, lr: 4.61e-03, grad_scale: 8.0
2022-12-23 22:19:49,765 INFO [zipformer.py:660] (2/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=84800.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 22:19:51,040 INFO [zipformer.py:660] (2/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=84801.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 22:20:20,461 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([0.1949, 1.4334, 1.6234, 0.8749, 1.0501, 1.7437, 1.6418, 1.4982],
       device='cuda:2'), covar=tensor([0.0859, 0.0357, 0.0344, 0.0384, 0.0419, 0.0485, 0.0286, 0.0736],
       device='cuda:2'), in_proj_covar=tensor([0.0150, 0.0173, 0.0130, 0.0143, 0.0150, 0.0145, 0.0166, 0.0177],
       device='cuda:2'), out_proj_covar=tensor([1.1365e-04, 1.3070e-04, 9.6418e-05, 1.0515e-04, 1.1073e-04, 1.0984e-04,
        1.2604e-04, 1.3332e-04], device='cuda:2')
2022-12-23 22:20:31,414 WARNING [train.py:1060] (2/4) Exclude cut with ID 4957-30119-0041-23990-0_sp0.9 from training. Duration: 20.22775
2022-12-23 22:20:54,923 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 2.139e+02 3.497e+02 4.132e+02 4.951e+02 1.041e+03, threshold=8.263e+02, percent-clipped=1.0
2022-12-23 22:20:57,873 INFO [train.py:894] (2/4) Epoch 25, batch 700, loss[loss=0.1471, simple_loss=0.2296, pruned_loss=0.03225, over 18415.00 frames. ], tot_loss[loss=0.1714, simple_loss=0.2605, pruned_loss=0.04113, over 3603303.36 frames. ], batch size: 42, lr: 4.61e-03, grad_scale: 8.0
2022-12-23 22:21:02,328 INFO [scaling.py:679] (2/4) Whitening: num_groups=8, num_channels=96, metric=1.24 vs. limit=2.0
2022-12-23 22:21:10,316 INFO [zipformer.py:660] (2/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=84855.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 22:21:17,964 WARNING [train.py:1060] (2/4) Exclude cut with ID 5239-32139-0047-9341-0_sp1.1 from training. Duration: 24.67275
2022-12-23 22:21:19,734 INFO [zipformer.py:660] (2/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=84861.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 22:21:22,799 INFO [scaling.py:679] (2/4) Whitening: num_groups=8, num_channels=96, metric=1.31 vs. limit=2.0
2022-12-23 22:21:43,251 WARNING [train.py:1060] (2/4) Exclude cut with ID 3082-165428-0081-50734-0_sp0.9 from training. Duration: 21.8055625
2022-12-23 22:21:49,926 INFO [zipformer.py:660] (2/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=84881.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 22:22:05,823 INFO [zipformer.py:660] (2/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=84892.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 22:22:13,730 INFO [train.py:894] (2/4) Epoch 25, batch 750, loss[loss=0.1911, simple_loss=0.2815, pruned_loss=0.05031, over 18675.00 frames. ], tot_loss[loss=0.1706, simple_loss=0.2597, pruned_loss=0.04072, over 3627419.87 frames. ], batch size: 60, lr: 4.61e-03, grad_scale: 8.0
2022-12-23 22:22:21,298 WARNING [train.py:1060] (2/4) Exclude cut with ID 3340-169293-0054-76830-0_sp0.9 from training. Duration: 22.6666875
2022-12-23 22:22:22,851 INFO [zipformer.py:660] (2/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=84903.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 22:23:06,206 INFO [scaling.py:679] (2/4) Whitening: num_groups=8, num_channels=96, metric=1.23 vs. limit=2.0
2022-12-23 22:23:20,275 INFO [zipformer.py:660] (2/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=84940.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 22:23:23,204 WARNING [train.py:1060] (2/4) Exclude cut with ID 2411-132532-0017-82279-0_sp1.1 from training. Duration: 0.9681875
2022-12-23 22:23:27,847 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 2.164e+02 3.238e+02 4.165e+02 4.876e+02 9.107e+02, threshold=8.330e+02, percent-clipped=1.0
2022-12-23 22:23:30,809 INFO [train.py:894] (2/4) Epoch 25, batch 800, loss[loss=0.1906, simple_loss=0.2859, pruned_loss=0.0476, over 18498.00 frames. ], tot_loss[loss=0.1707, simple_loss=0.2601, pruned_loss=0.04063, over 3645704.88 frames. ], batch size: 52, lr: 4.60e-03, grad_scale: 8.0
2022-12-23 22:23:37,255 INFO [zipformer.py:660] (2/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=84951.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 22:23:49,153 WARNING [train.py:1060] (2/4) Exclude cut with ID 6330-62850-0007-91323-0 from training. Duration: 22.485
2022-12-23 22:24:04,661 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.9628, 1.2406, 0.6844, 1.5556, 2.1823, 1.4927, 1.8071, 1.8907],
       device='cuda:2'), covar=tensor([0.1613, 0.2077, 0.2418, 0.1502, 0.1728, 0.1701, 0.1317, 0.1639],
       device='cuda:2'), in_proj_covar=tensor([0.0094, 0.0098, 0.0118, 0.0097, 0.0120, 0.0092, 0.0099, 0.0095],
       device='cuda:2'), out_proj_covar=tensor([0.0003, 0.0003, 0.0004, 0.0003, 0.0004, 0.0003, 0.0003, 0.0003],
       device='cuda:2')
2022-12-23 22:24:27,100 WARNING [train.py:1060] (2/4) Exclude cut with ID 3972-170212-0014-23379-0_sp1.1 from training. Duration: 23.82275
2022-12-23 22:24:34,831 INFO [zipformer.py:660] (2/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=84989.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 22:24:39,282 INFO [zipformer.py:660] (2/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=84992.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 22:24:40,992 WARNING [train.py:1060] (2/4) Exclude cut with ID 4860-13185-0032-76709-0 from training. Duration: 20.77
2022-12-23 22:24:46,811 INFO [train.py:894] (2/4) Epoch 25, batch 850, loss[loss=0.1564, simple_loss=0.2373, pruned_loss=0.03774, over 18400.00 frames. ], tot_loss[loss=0.1702, simple_loss=0.2594, pruned_loss=0.04046, over 3660671.32 frames. ], batch size: 46, lr: 4.60e-03, grad_scale: 8.0
2022-12-23 22:24:48,322 WARNING [train.py:1060] (2/4) Exclude cut with ID 6426-64292-0017-15984-0_sp0.9 from training. Duration: 24.088875
2022-12-23 22:25:10,786 INFO [zipformer.py:660] (2/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=85012.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 22:25:18,942 WARNING [train.py:1060] (2/4) Exclude cut with ID 6330-62850-0007-91323-0_sp1.1 from training. Duration: 20.4409375
2022-12-23 22:25:51,497 INFO [zipformer.py:660] (2/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=85040.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 22:25:59,254 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 2.350e+02 3.402e+02 3.934e+02 4.822e+02 1.128e+03, threshold=7.867e+02, percent-clipped=2.0
2022-12-23 22:26:02,196 INFO [train.py:894] (2/4) Epoch 25, batch 900, loss[loss=0.1763, simple_loss=0.2673, pruned_loss=0.0427, over 18469.00 frames. ], tot_loss[loss=0.1704, simple_loss=0.2596, pruned_loss=0.04063, over 3672670.46 frames. ], batch size: 54, lr: 4.60e-03, grad_scale: 8.0
2022-12-23 22:26:33,728 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([2.8044, 4.2439, 3.9756, 2.0609, 4.3837, 3.3357, 0.6930, 2.9440],
       device='cuda:2'), covar=tensor([0.2074, 0.0872, 0.1418, 0.3333, 0.0653, 0.0818, 0.5232, 0.1455],
       device='cuda:2'), in_proj_covar=tensor([0.0148, 0.0144, 0.0159, 0.0123, 0.0147, 0.0113, 0.0143, 0.0114],
       device='cuda:2'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0002, 0.0003, 0.0002, 0.0003, 0.0002],
       device='cuda:2')
2022-12-23 22:26:34,064 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([2.3440, 1.6224, 1.9883, 1.9802, 2.2922, 2.3243, 2.1265, 1.8923],
       device='cuda:2'), covar=tensor([0.2300, 0.3410, 0.2623, 0.3058, 0.2132, 0.1005, 0.3382, 0.1371],
       device='cuda:2'), in_proj_covar=tensor([0.0270, 0.0299, 0.0283, 0.0321, 0.0314, 0.0257, 0.0349, 0.0245],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-23 22:26:35,014 WARNING [train.py:1060] (2/4) Exclude cut with ID 2929-85685-0062-62366-0_sp0.9 from training. Duration: 22.511125
2022-12-23 22:26:35,036 WARNING [train.py:1060] (2/4) Exclude cut with ID 7395-89880-0031-39906-0 from training. Duration: 20.675
2022-12-23 22:26:42,872 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([4.7237, 4.0610, 4.0718, 4.6665, 4.3121, 4.2803, 4.8783, 1.4790],
       device='cuda:2'), covar=tensor([0.0705, 0.0727, 0.0659, 0.0751, 0.1462, 0.1080, 0.0580, 0.5268],
       device='cuda:2'), in_proj_covar=tensor([0.0350, 0.0232, 0.0243, 0.0275, 0.0329, 0.0271, 0.0296, 0.0288],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0001, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-23 22:27:16,898 INFO [train.py:894] (2/4) Epoch 25, batch 950, loss[loss=0.175, simple_loss=0.2724, pruned_loss=0.03877, over 18678.00 frames. ], tot_loss[loss=0.1714, simple_loss=0.2609, pruned_loss=0.04098, over 3682141.48 frames. ], batch size: 62, lr: 4.60e-03, grad_scale: 8.0
2022-12-23 22:27:32,968 INFO [zipformer.py:660] (2/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=85107.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 22:27:44,448 INFO [zipformer.py:660] (2/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=85115.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 22:27:58,286 INFO [scaling.py:679] (2/4) Whitening: num_groups=8, num_channels=96, metric=1.24 vs. limit=2.0
2022-12-23 22:28:09,988 INFO [scaling.py:679] (2/4) Whitening: num_groups=8, num_channels=96, metric=1.55 vs. limit=2.0
2022-12-23 22:28:13,671 WARNING [train.py:1060] (2/4) Exclude cut with ID 6330-62850-0007-91323-0_sp0.9 from training. Duration: 24.9833125
2022-12-23 22:28:20,433 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([2.3486, 3.4164, 3.2238, 1.2921, 3.4508, 2.6366, 0.5140, 2.0469],
       device='cuda:2'), covar=tensor([0.2465, 0.1441, 0.1525, 0.3922, 0.0977, 0.1022, 0.5250, 0.1831],
       device='cuda:2'), in_proj_covar=tensor([0.0148, 0.0144, 0.0159, 0.0122, 0.0147, 0.0113, 0.0143, 0.0114],
       device='cuda:2'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0002, 0.0003, 0.0002, 0.0003, 0.0002],
       device='cuda:2')
2022-12-23 22:28:29,076 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 1.943e+02 3.246e+02 3.932e+02 4.617e+02 8.441e+02, threshold=7.864e+02, percent-clipped=1.0
2022-12-23 22:28:32,093 INFO [train.py:894] (2/4) Epoch 25, batch 1000, loss[loss=0.1797, simple_loss=0.2797, pruned_loss=0.0398, over 18629.00 frames. ], tot_loss[loss=0.1708, simple_loss=0.2605, pruned_loss=0.04054, over 3688857.13 frames. ], batch size: 53, lr: 4.60e-03, grad_scale: 8.0
2022-12-23 22:28:46,631 WARNING [train.py:1060] (2/4) Exclude cut with ID 5239-32139-0047-9341-0 from training. Duration: 27.14
2022-12-23 22:28:46,881 INFO [zipformer.py:660] (2/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=85156.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 22:29:00,954 WARNING [train.py:1060] (2/4) Exclude cut with ID 2929-85685-0044-62348-0 from training. Duration: 22.44
2022-12-23 22:29:04,425 INFO [zipformer.py:660] (2/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=85168.0, num_to_drop=1, layers_to_drop={1}
2022-12-23 22:29:16,467 INFO [zipformer.py:660] (2/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=85176.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 22:29:23,913 INFO [zipformer.py:660] (2/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=85181.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 22:29:37,760 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.8839, 1.5113, 1.7742, 2.0723, 1.6953, 3.5644, 1.6366, 1.6351],
       device='cuda:2'), covar=tensor([0.0818, 0.1807, 0.1056, 0.0941, 0.1539, 0.0230, 0.1339, 0.1484],
       device='cuda:2'), in_proj_covar=tensor([0.0073, 0.0082, 0.0072, 0.0074, 0.0092, 0.0077, 0.0085, 0.0077],
       device='cuda:2'), out_proj_covar=tensor([0.0004, 0.0004, 0.0004, 0.0004, 0.0004, 0.0004, 0.0004, 0.0004],
       device='cuda:2')
2022-12-23 22:29:48,015 INFO [train.py:894] (2/4) Epoch 25, batch 1050, loss[loss=0.1841, simple_loss=0.2774, pruned_loss=0.04542, over 18573.00 frames. ], tot_loss[loss=0.1703, simple_loss=0.2601, pruned_loss=0.04024, over 3694445.03 frames. ], batch size: 98, lr: 4.60e-03, grad_scale: 8.0
2022-12-23 22:30:15,691 INFO [scaling.py:679] (2/4) Whitening: num_groups=8, num_channels=192, metric=1.96 vs. limit=2.0
2022-12-23 22:30:20,625 INFO [scaling.py:679] (2/4) Whitening: num_groups=8, num_channels=96, metric=1.16 vs. limit=2.0
2022-12-23 22:30:21,437 WARNING [train.py:1060] (2/4) Exclude cut with ID 2929-85685-0060-62364-0_sp0.9 from training. Duration: 21.361125
2022-12-23 22:30:28,617 WARNING [train.py:1060] (2/4) Exclude cut with ID 2929-85685-0079-62383-0_sp1.1 from training. Duration: 27.0318125
2022-12-23 22:30:37,426 INFO [zipformer.py:660] (2/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=85229.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 22:30:38,791 WARNING [train.py:1060] (2/4) Exclude cut with ID 5622-44585-0006-90525-0_sp0.9 from training. Duration: 28.638875
2022-12-23 22:30:52,416 WARNING [train.py:1060] (2/4) Exclude cut with ID 3340-169293-0054-76830-0 from training. Duration: 20.4
2022-12-23 22:31:01,075 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 2.159e+02 3.176e+02 3.831e+02 4.678e+02 1.344e+03, threshold=7.662e+02, percent-clipped=2.0
2022-12-23 22:31:04,281 INFO [train.py:894] (2/4) Epoch 25, batch 1100, loss[loss=0.1793, simple_loss=0.2705, pruned_loss=0.04405, over 18531.00 frames. ], tot_loss[loss=0.17, simple_loss=0.2597, pruned_loss=0.04013, over 3698026.09 frames. ], batch size: 55, lr: 4.60e-03, grad_scale: 16.0
2022-12-23 22:31:25,864 WARNING [train.py:1060] (2/4) Exclude cut with ID 2929-85685-0071-62375-0 from training. Duration: 20.025
2022-12-23 22:31:25,878 WARNING [train.py:1060] (2/4) Exclude cut with ID 2364-131735-0112-64612-0_sp0.9 from training. Duration: 20.488875
2022-12-23 22:31:30,297 WARNING [train.py:1060] (2/4) Exclude cut with ID 2929-85685-0079-62383-0 from training. Duration: 29.735
2022-12-23 22:32:07,493 INFO [scaling.py:679] (2/4) Whitening: num_groups=8, num_channels=96, metric=1.15 vs. limit=2.0
2022-12-23 22:32:08,339 INFO [zipformer.py:660] (2/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=85289.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 22:32:19,892 INFO [train.py:894] (2/4) Epoch 25, batch 1150, loss[loss=0.1559, simple_loss=0.2432, pruned_loss=0.03426, over 18686.00 frames. ], tot_loss[loss=0.1695, simple_loss=0.2593, pruned_loss=0.03988, over 3700949.01 frames. ], batch size: 46, lr: 4.59e-03, grad_scale: 16.0
2022-12-23 22:32:35,403 INFO [zipformer.py:660] (2/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=85307.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 22:32:54,204 WARNING [train.py:1060] (2/4) Exclude cut with ID 7276-92427-0014-12983-0_sp0.9 from training. Duration: 21.3055625
2022-12-23 22:32:55,664 WARNING [train.py:1060] (2/4) Exclude cut with ID 1025-75365-0008-79168-0_sp0.9 from training. Duration: 22.0666875
2022-12-23 22:33:19,336 INFO [zipformer.py:660] (2/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=85337.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 22:33:31,170 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 2.188e+02 3.174e+02 3.958e+02 5.147e+02 9.508e+02, threshold=7.916e+02, percent-clipped=3.0
2022-12-23 22:33:34,603 INFO [train.py:894] (2/4) Epoch 25, batch 1200, loss[loss=0.1694, simple_loss=0.2674, pruned_loss=0.03575, over 18534.00 frames. ], tot_loss[loss=0.1694, simple_loss=0.259, pruned_loss=0.03989, over 3702931.56 frames. ], batch size: 55, lr: 4.59e-03, grad_scale: 16.0
2022-12-23 22:34:14,184 INFO [scaling.py:679] (2/4) Whitening: num_groups=1, num_channels=384, metric=4.75 vs. limit=5.0
2022-12-23 22:34:18,115 INFO [scaling.py:679] (2/4) Whitening: num_groups=8, num_channels=96, metric=1.18 vs. limit=2.0
2022-12-23 22:34:42,449 WARNING [train.py:1060] (2/4) Exclude cut with ID 2929-85685-0062-62366-0 from training. Duration: 20.26
2022-12-23 22:34:47,913 INFO [train.py:894] (2/4) Epoch 25, batch 1250, loss[loss=0.1597, simple_loss=0.2622, pruned_loss=0.02862, over 18622.00 frames. ], tot_loss[loss=0.1696, simple_loss=0.2596, pruned_loss=0.0398, over 3705300.61 frames. ], batch size: 53, lr: 4.59e-03, grad_scale: 16.0
2022-12-23 22:34:58,010 WARNING [train.py:1060] (2/4) Exclude cut with ID 5239-32139-0030-9324-0_sp0.9 from training. Duration: 21.3444375
2022-12-23 22:35:03,305 INFO [scaling.py:679] (2/4) Whitening: num_groups=8, num_channels=192, metric=1.77 vs. limit=2.0
2022-12-23 22:35:34,911 INFO [zipformer.py:660] (2/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=85428.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 22:35:53,817 WARNING [train.py:1060] (2/4) Exclude cut with ID 497-129325-0061-62254-0_sp1.1 from training. Duration: 0.97725
2022-12-23 22:36:00,830 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 1.956e+02 3.502e+02 4.076e+02 5.122e+02 9.671e+02, threshold=8.152e+02, percent-clipped=4.0
2022-12-23 22:36:03,833 INFO [train.py:894] (2/4) Epoch 25, batch 1300, loss[loss=0.1522, simple_loss=0.2386, pruned_loss=0.03295, over 18392.00 frames. ], tot_loss[loss=0.1697, simple_loss=0.2598, pruned_loss=0.03976, over 3707920.90 frames. ], batch size: 46, lr: 4.59e-03, grad_scale: 16.0
2022-12-23 22:36:11,205 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([2.3647, 2.0575, 1.6131, 2.0361, 1.8146, 2.1155, 1.8719, 2.1252],
       device='cuda:2'), covar=tensor([0.2205, 0.3353, 0.2139, 0.2544, 0.3693, 0.1093, 0.2995, 0.1082],
       device='cuda:2'), in_proj_covar=tensor([0.0299, 0.0295, 0.0250, 0.0348, 0.0277, 0.0232, 0.0293, 0.0219],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0003, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-23 22:36:16,912 INFO [zipformer.py:660] (2/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=85456.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 22:36:27,062 INFO [zipformer.py:660] (2/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=85463.0, num_to_drop=1, layers_to_drop={2}
2022-12-23 22:36:34,055 WARNING [train.py:1060] (2/4) Exclude cut with ID 7395-89880-0031-39906-0_sp0.9 from training. Duration: 22.97225
2022-12-23 22:36:38,385 INFO [zipformer.py:660] (2/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=85471.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 22:36:56,343 INFO [zipformer.py:660] (2/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=85483.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 22:37:04,642 WARNING [train.py:1060] (2/4) Exclude cut with ID 7395-89880-0047-39922-0_sp0.9 from training. Duration: 21.97775
2022-12-23 22:37:05,020 INFO [zipformer.py:660] (2/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=85489.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 22:37:17,333 INFO [train.py:894] (2/4) Epoch 25, batch 1350, loss[loss=0.1558, simple_loss=0.2404, pruned_loss=0.03559, over 18523.00 frames. ], tot_loss[loss=0.1699, simple_loss=0.2598, pruned_loss=0.04, over 3709483.02 frames. ], batch size: 47, lr: 4.59e-03, grad_scale: 16.0
2022-12-23 22:37:20,178 WARNING [train.py:1060] (2/4) Exclude cut with ID 1112-1043-0006-89194-0_sp0.9 from training. Duration: 21.8333125
2022-12-23 22:37:27,332 INFO [zipformer.py:660] (2/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=85504.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 22:37:28,674 WARNING [train.py:1060] (2/4) Exclude cut with ID 1914-133440-0031-94921-0 from training. Duration: 20.47
2022-12-23 22:37:35,155 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.9479, 1.5541, 1.8074, 1.7131, 1.9414, 1.9178, 1.8127, 1.7735],
       device='cuda:2'), covar=tensor([0.1826, 0.2537, 0.1958, 0.2335, 0.1826, 0.0849, 0.2709, 0.1043],
       device='cuda:2'), in_proj_covar=tensor([0.0272, 0.0300, 0.0284, 0.0322, 0.0316, 0.0258, 0.0350, 0.0246],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-23 22:38:06,266 INFO [zipformer.py:660] (2/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=85530.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 22:38:27,647 INFO [zipformer.py:660] (2/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=85544.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 22:38:28,668 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 1.968e+02 3.372e+02 4.467e+02 5.937e+02 1.365e+03, threshold=8.935e+02, percent-clipped=9.0
2022-12-23 22:38:31,866 INFO [train.py:894] (2/4) Epoch 25, batch 1400, loss[loss=0.1741, simple_loss=0.261, pruned_loss=0.04359, over 18387.00 frames. ], tot_loss[loss=0.1699, simple_loss=0.2597, pruned_loss=0.04004, over 3709956.42 frames. ], batch size: 46, lr: 4.59e-03, grad_scale: 16.0
2022-12-23 22:38:31,878 WARNING [train.py:1060] (2/4) Exclude cut with ID 7395-89880-0037-39912-0_sp0.9 from training. Duration: 20.67225
2022-12-23 22:38:36,673 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.0643, 1.0018, 0.9720, 1.1860, 1.2893, 1.2259, 1.1165, 1.0123],
       device='cuda:2'), covar=tensor([0.0320, 0.0239, 0.0579, 0.0215, 0.0239, 0.0406, 0.0322, 0.0327],
       device='cuda:2'), in_proj_covar=tensor([0.0094, 0.0125, 0.0151, 0.0122, 0.0115, 0.0119, 0.0098, 0.0126],
       device='cuda:2'), out_proj_covar=tensor([7.4852e-05, 9.8839e-05, 1.2407e-04, 9.7115e-05, 9.2375e-05, 9.1257e-05,
        7.6465e-05, 9.9480e-05], device='cuda:2')
2022-12-23 22:38:44,800 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([0.4816, 1.6503, 1.9401, 0.9821, 1.3239, 2.0901, 1.8713, 1.6432],
       device='cuda:2'), covar=tensor([0.0828, 0.0446, 0.0340, 0.0489, 0.0441, 0.0478, 0.0280, 0.0788],
       device='cuda:2'), in_proj_covar=tensor([0.0149, 0.0171, 0.0130, 0.0141, 0.0149, 0.0145, 0.0165, 0.0176],
       device='cuda:2'), out_proj_covar=tensor([1.1300e-04, 1.2966e-04, 9.6256e-05, 1.0380e-04, 1.0950e-04, 1.0949e-04,
        1.2545e-04, 1.3281e-04], device='cuda:2')
2022-12-23 22:38:46,061 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.8723, 2.2590, 1.7731, 2.5539, 2.6901, 1.8546, 1.7238, 1.5483],
       device='cuda:2'), covar=tensor([0.1794, 0.1581, 0.1515, 0.0918, 0.1176, 0.1036, 0.1976, 0.1474],
       device='cuda:2'), in_proj_covar=tensor([0.0248, 0.0230, 0.0220, 0.0202, 0.0262, 0.0197, 0.0226, 0.0203],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-23 22:38:50,193 WARNING [train.py:1060] (2/4) Exclude cut with ID 1914-133440-0024-94914-0_sp0.9 from training. Duration: 25.2444375
2022-12-23 22:39:13,352 WARNING [train.py:1060] (2/4) Exclude cut with ID 3340-169293-0021-76797-0_sp0.9 from training. Duration: 21.1445
2022-12-23 22:39:24,520 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([4.4261, 3.8518, 3.8069, 4.3490, 4.0609, 3.9175, 4.6005, 1.3590],
       device='cuda:2'), covar=tensor([0.0712, 0.0684, 0.0704, 0.0761, 0.1392, 0.1194, 0.0556, 0.5270],
       device='cuda:2'), in_proj_covar=tensor([0.0347, 0.0230, 0.0242, 0.0275, 0.0328, 0.0272, 0.0296, 0.0288],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0001, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-23 22:39:38,453 INFO [zipformer.py:660] (2/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=85591.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 22:39:47,005 INFO [train.py:894] (2/4) Epoch 25, batch 1450, loss[loss=0.176, simple_loss=0.278, pruned_loss=0.03704, over 18594.00 frames. ], tot_loss[loss=0.1704, simple_loss=0.26, pruned_loss=0.04041, over 3710734.75 frames. ], batch size: 69, lr: 4.59e-03, grad_scale: 16.0
2022-12-23 22:40:02,279 INFO [zipformer.py:660] (2/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=85607.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 22:40:28,413 WARNING [train.py:1060] (2/4) Exclude cut with ID 2929-85685-0079-62383-0_sp0.9 from training. Duration: 33.038875
2022-12-23 22:40:28,795 INFO [zipformer.py:660] (2/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=85625.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 22:40:59,332 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 2.452e+02 3.397e+02 4.023e+02 4.690e+02 9.642e+02, threshold=8.046e+02, percent-clipped=1.0
2022-12-23 22:41:02,423 INFO [train.py:894] (2/4) Epoch 25, batch 1500, loss[loss=0.1826, simple_loss=0.2699, pruned_loss=0.04763, over 18657.00 frames. ], tot_loss[loss=0.1706, simple_loss=0.2599, pruned_loss=0.04067, over 3711744.21 frames. ], batch size: 178, lr: 4.59e-03, grad_scale: 16.0
2022-12-23 22:41:06,813 WARNING [train.py:1060] (2/4) Exclude cut with ID 6426-64291-0000-16059-0_sp0.9 from training. Duration: 20.0944375
2022-12-23 22:41:14,978 INFO [zipformer.py:660] (2/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=85655.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 22:41:21,933 WARNING [train.py:1060] (2/4) Exclude cut with ID 2929-85685-0044-62348-0_sp1.1 from training. Duration: 20.4
2022-12-23 22:41:30,402 WARNING [train.py:1060] (2/4) Exclude cut with ID 6330-62851-0022-91297-0 from training. Duration: 20.085
2022-12-23 22:41:42,394 WARNING [train.py:1060] (2/4) Exclude cut with ID 4860-13185-0032-76709-0_sp0.9 from training. Duration: 23.07775
2022-12-23 22:42:00,703 INFO [zipformer.py:660] (2/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=85686.0, num_to_drop=1, layers_to_drop={0}
2022-12-23 22:42:16,526 INFO [train.py:894] (2/4) Epoch 25, batch 1550, loss[loss=0.1468, simple_loss=0.2261, pruned_loss=0.03376, over 18414.00 frames. ], tot_loss[loss=0.1701, simple_loss=0.2591, pruned_loss=0.04052, over 3711925.83 frames. ], batch size: 42, lr: 4.58e-03, grad_scale: 16.0
2022-12-23 22:42:27,771 WARNING [train.py:1060] (2/4) Exclude cut with ID 2929-85685-0044-62348-0_sp0.9 from training. Duration: 24.9333125
2022-12-23 22:43:12,642 WARNING [train.py:1060] (2/4) Exclude cut with ID 5118-111612-0016-124680-0_sp0.9 from training. Duration: 20.388875
2022-12-23 22:43:19,169 WARNING [train.py:1060] (2/4) Exclude cut with ID 432-122774-0017-62487-0_sp1.1 from training. Duration: 20.3590625
2022-12-23 22:43:29,017 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 1.994e+02 3.114e+02 3.818e+02 4.657e+02 1.147e+03, threshold=7.635e+02, percent-clipped=4.0
2022-12-23 22:43:31,801 INFO [train.py:894] (2/4) Epoch 25, batch 1600, loss[loss=0.1709, simple_loss=0.2629, pruned_loss=0.03947, over 18588.00 frames. ], tot_loss[loss=0.1705, simple_loss=0.2596, pruned_loss=0.04069, over 3711729.50 frames. ], batch size: 96, lr: 4.58e-03, grad_scale: 16.0
2022-12-23 22:43:55,496 INFO [zipformer.py:660] (2/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=85763.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 22:44:07,577 INFO [zipformer.py:660] (2/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=85771.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 22:44:26,654 WARNING [train.py:1060] (2/4) Exclude cut with ID 3557-8342-0013-54691-0_sp1.1 from training. Duration: 0.836375
2022-12-23 22:44:26,800 INFO [zipformer.py:660] (2/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=85784.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 22:44:45,846 INFO [train.py:894] (2/4) Epoch 25, batch 1650, loss[loss=0.1737, simple_loss=0.2684, pruned_loss=0.03951, over 18676.00 frames. ], tot_loss[loss=0.17, simple_loss=0.2591, pruned_loss=0.04047, over 3712477.77 frames. ], batch size: 62, lr: 4.58e-03, grad_scale: 16.0
2022-12-23 22:44:54,521 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.5253, 1.5108, 1.5336, 1.4964, 1.0753, 2.9661, 1.3321, 1.7794],
       device='cuda:2'), covar=tensor([0.3246, 0.2217, 0.2125, 0.2224, 0.1694, 0.0319, 0.1831, 0.0952],
       device='cuda:2'), in_proj_covar=tensor([0.0132, 0.0118, 0.0125, 0.0121, 0.0105, 0.0096, 0.0090, 0.0090],
       device='cuda:2'), out_proj_covar=tensor([0.0005, 0.0005, 0.0005, 0.0005, 0.0004, 0.0003, 0.0004, 0.0004],
       device='cuda:2')
2022-12-23 22:45:06,475 INFO [zipformer.py:660] (2/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=85811.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 22:45:10,585 WARNING [train.py:1060] (2/4) Exclude cut with ID 8565-290391-0049-67394-0_sp0.9 from training. Duration: 21.3166875
2022-12-23 22:45:18,182 INFO [zipformer.py:660] (2/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=85819.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 22:45:41,568 WARNING [train.py:1060] (2/4) Exclude cut with ID 6533-399-0029-104863-0_sp0.9 from training. Duration: 22.1055625
2022-12-23 22:45:47,988 INFO [zipformer.py:660] (2/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=85839.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 22:45:53,773 WARNING [train.py:1060] (2/4) Exclude cut with ID 7699-105389-0094-26379-0_sp1.1 from training. Duration: 21.77725
2022-12-23 22:45:56,759 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 2.405e+02 3.603e+02 4.272e+02 4.915e+02 8.898e+02, threshold=8.543e+02, percent-clipped=2.0
2022-12-23 22:45:59,552 INFO [train.py:894] (2/4) Epoch 25, batch 1700, loss[loss=0.2134, simple_loss=0.2908, pruned_loss=0.06798, over 18489.00 frames. ], tot_loss[loss=0.1711, simple_loss=0.259, pruned_loss=0.04153, over 3712628.52 frames. ], batch size: 64, lr: 4.58e-03, grad_scale: 16.0
2022-12-23 22:46:11,867 WARNING [train.py:1060] (2/4) Exclude cut with ID 7255-291500-0005-134304-0_sp0.9 from training. Duration: 27.8166875
2022-12-23 22:46:38,727 WARNING [train.py:1060] (2/4) Exclude cut with ID 7357-94126-0021-15852-0_sp1.1 from training. Duration: 22.5090625
2022-12-23 22:46:42,149 INFO [zipformer.py:660] (2/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=85875.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 22:46:44,943 WARNING [train.py:1060] (2/4) Exclude cut with ID 7255-291500-0005-134304-0 from training. Duration: 25.035
2022-12-23 22:46:58,475 INFO [zipformer.py:660] (2/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=85886.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 22:47:02,071 WARNING [train.py:1060] (2/4) Exclude cut with ID 774-127930-0014-10412-0_sp1.1 from training. Duration: 0.95
2022-12-23 22:47:15,474 INFO [train.py:894] (2/4) Epoch 25, batch 1750, loss[loss=0.1732, simple_loss=0.252, pruned_loss=0.04722, over 18605.00 frames. ], tot_loss[loss=0.173, simple_loss=0.2601, pruned_loss=0.04298, over 3713106.36 frames. ], batch size: 45, lr: 4.58e-03, grad_scale: 16.0
2022-12-23 22:47:20,380 WARNING [train.py:1060] (2/4) Exclude cut with ID 3033-130750-0096-55598-0_sp0.9 from training. Duration: 0.92225
2022-12-23 22:47:41,425 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.5304, 1.2312, 1.8883, 2.6843, 1.9186, 2.2991, 0.9792, 2.0085],
       device='cuda:2'), covar=tensor([0.1758, 0.1599, 0.1218, 0.0662, 0.1050, 0.1086, 0.1828, 0.1038],
       device='cuda:2'), in_proj_covar=tensor([0.0103, 0.0118, 0.0136, 0.0154, 0.0106, 0.0144, 0.0129, 0.0114],
       device='cuda:2'), out_proj_covar=tensor([0.0003, 0.0003, 0.0004, 0.0004, 0.0003, 0.0004, 0.0004, 0.0003],
       device='cuda:2')
2022-12-23 22:47:44,333 WARNING [train.py:1060] (2/4) Exclude cut with ID 4511-76322-0006-80011-0 from training. Duration: 21.97
2022-12-23 22:48:02,329 WARNING [train.py:1060] (2/4) Exclude cut with ID 7492-105653-0055-62765-0_sp0.9 from training. Duration: 21.97225
2022-12-23 22:48:03,775 WARNING [train.py:1060] (2/4) Exclude cut with ID 453-131332-0000-47844-0_sp0.9 from training. Duration: 25.3333125
2022-12-23 22:48:05,738 INFO [zipformer.py:660] (2/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=85930.0, num_to_drop=1, layers_to_drop={0}
2022-12-23 22:48:14,210 WARNING [train.py:1060] (2/4) Exclude cut with ID 5172-29468-0015-19128-0_sp0.9 from training. Duration: 21.5055625
2022-12-23 22:48:14,578 INFO [zipformer.py:660] (2/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=85936.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 22:48:25,188 WARNING [train.py:1060] (2/4) Exclude cut with ID 453-131332-0000-47844-0_sp1.1 from training. Duration: 20.72725
2022-12-23 22:48:27,933 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 2.947e+02 4.240e+02 5.276e+02 6.362e+02 1.395e+03, threshold=1.055e+03, percent-clipped=9.0
2022-12-23 22:48:30,845 INFO [train.py:894] (2/4) Epoch 25, batch 1800, loss[loss=0.1827, simple_loss=0.2721, pruned_loss=0.0466, over 18586.00 frames. ], tot_loss[loss=0.1752, simple_loss=0.2617, pruned_loss=0.04436, over 3713201.46 frames. ], batch size: 69, lr: 4.58e-03, grad_scale: 16.0
2022-12-23 22:48:45,864 INFO [zipformer.py:660] (2/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=85957.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 22:48:57,724 WARNING [train.py:1060] (2/4) Exclude cut with ID 8631-249866-0030-130156-0_sp0.9 from training. Duration: 26.32775
2022-12-23 22:49:02,889 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([2.5117, 2.6338, 2.9153, 1.6714, 3.1050, 3.1401, 2.0415, 3.2755],
       device='cuda:2'), covar=tensor([0.1226, 0.1517, 0.1305, 0.2029, 0.0706, 0.1005, 0.2118, 0.0583],
       device='cuda:2'), in_proj_covar=tensor([0.0195, 0.0212, 0.0206, 0.0191, 0.0172, 0.0213, 0.0213, 0.0191],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-23 22:49:05,889 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([3.1852, 2.5747, 1.9671, 2.9765, 2.3804, 2.4668, 2.5253, 3.2420],
       device='cuda:2'), covar=tensor([0.1915, 0.3152, 0.1967, 0.2950, 0.3728, 0.1028, 0.3171, 0.0859],
       device='cuda:2'), in_proj_covar=tensor([0.0299, 0.0297, 0.0251, 0.0350, 0.0278, 0.0233, 0.0295, 0.0219],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0003, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-23 22:49:21,009 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.5919, 1.2893, 1.9954, 2.8728, 2.0625, 2.4439, 0.9543, 2.0892],
       device='cuda:2'), covar=tensor([0.1797, 0.1648, 0.1339, 0.0719, 0.1099, 0.1172, 0.2074, 0.1142],
       device='cuda:2'), in_proj_covar=tensor([0.0103, 0.0117, 0.0136, 0.0153, 0.0106, 0.0143, 0.0129, 0.0114],
       device='cuda:2'), out_proj_covar=tensor([0.0003, 0.0003, 0.0004, 0.0004, 0.0003, 0.0004, 0.0003, 0.0003],
       device='cuda:2')
2022-12-23 22:49:22,888 INFO [zipformer.py:660] (2/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=85981.0, num_to_drop=1, layers_to_drop={3}
2022-12-23 22:49:24,561 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.9233, 1.8114, 2.1025, 2.4539, 2.1065, 4.1882, 1.9541, 1.9094],
       device='cuda:2'), covar=tensor([0.0854, 0.1680, 0.1043, 0.0853, 0.1318, 0.0298, 0.1264, 0.1467],
       device='cuda:2'), in_proj_covar=tensor([0.0072, 0.0081, 0.0071, 0.0073, 0.0090, 0.0075, 0.0084, 0.0076],
       device='cuda:2'), out_proj_covar=tensor([0.0004, 0.0004, 0.0004, 0.0004, 0.0004, 0.0003, 0.0004, 0.0004],
       device='cuda:2')
2022-12-23 22:49:27,379 WARNING [train.py:1060] (2/4) Exclude cut with ID 3867-173237-0077-144769-0 from training. Duration: 20.025
2022-12-23 22:49:33,325 WARNING [train.py:1060] (2/4) Exclude cut with ID 6709-74022-0004-86860-0_sp1.1 from training. Duration: 0.9409375
2022-12-23 22:49:33,332 WARNING [train.py:1060] (2/4) Exclude cut with ID 4757-1811-0023-62229-0_sp0.9 from training. Duration: 21.37775
2022-12-23 22:49:34,297 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.5630, 1.6034, 1.8156, 1.0723, 1.7270, 1.8306, 1.3705, 2.1684],
       device='cuda:2'), covar=tensor([0.1155, 0.1857, 0.1128, 0.1674, 0.0789, 0.1052, 0.2475, 0.0553],
       device='cuda:2'), in_proj_covar=tensor([0.0196, 0.0212, 0.0206, 0.0191, 0.0172, 0.0214, 0.0214, 0.0192],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-23 22:49:38,493 INFO [zipformer.py:660] (2/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=85991.0, num_to_drop=1, layers_to_drop={1}
2022-12-23 22:49:46,964 INFO [train.py:894] (2/4) Epoch 25, batch 1850, loss[loss=0.1782, simple_loss=0.2576, pruned_loss=0.04939, over 18534.00 frames. ], tot_loss[loss=0.1767, simple_loss=0.2625, pruned_loss=0.04541, over 3713243.77 frames. ], batch size: 47, lr: 4.58e-03, grad_scale: 16.0
2022-12-23 22:49:57,568 WARNING [train.py:1060] (2/4) Exclude cut with ID 1250-135782-0004-25974-0_sp0.9 from training. Duration: 21.17225
2022-12-23 22:49:57,580 WARNING [train.py:1060] (2/4) Exclude cut with ID 7357-94126-0021-15852-0_sp0.9 from training. Duration: 27.511125
2022-12-23 22:49:57,980 INFO [zipformer.py:660] (2/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=86002.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 22:50:20,929 INFO [scaling.py:679] (2/4) Whitening: num_groups=8, num_channels=192, metric=1.94 vs. limit=2.0
2022-12-23 22:50:23,226 INFO [zipformer.py:660] (2/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=86018.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 22:50:31,414 WARNING [train.py:1060] (2/4) Exclude cut with ID 453-131332-0000-47844-0 from training. Duration: 22.8
2022-12-23 22:50:36,346 WARNING [train.py:1060] (2/4) Exclude cut with ID 4964-30587-0040-44509-0 from training. Duration: 22.585
2022-12-23 22:51:02,098 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 3.089e+02 4.084e+02 4.781e+02 5.802e+02 1.070e+03, threshold=9.561e+02, percent-clipped=1.0
2022-12-23 22:51:04,971 INFO [train.py:894] (2/4) Epoch 25, batch 1900, loss[loss=0.1909, simple_loss=0.2796, pruned_loss=0.05109, over 18573.00 frames. ], tot_loss[loss=0.1771, simple_loss=0.2623, pruned_loss=0.04594, over 3713072.54 frames. ], batch size: 56, lr: 4.57e-03, grad_scale: 16.0
2022-12-23 22:51:08,094 WARNING [train.py:1060] (2/4) Exclude cut with ID 6978-92210-0001-146967-0_sp0.9 from training. Duration: 22.0166875
2022-12-23 22:51:12,632 INFO [zipformer.py:660] (2/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=86052.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 22:51:24,371 WARNING [train.py:1060] (2/4) Exclude cut with ID 7255-291500-0003-134302-0_sp1.1 from training. Duration: 24.395375
2022-12-23 22:51:29,281 INFO [zipformer.py:660] (2/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=86063.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 22:51:31,072 WARNING [train.py:1060] (2/4) Exclude cut with ID 3488-85273-0017-41203-0_sp0.9 from training. Duration: 27.47775
2022-12-23 22:51:36,790 WARNING [train.py:1060] (2/4) Exclude cut with ID 432-122774-0017-62487-0_sp0.9 from training. Duration: 24.8833125
2022-12-23 22:51:38,427 INFO [zipformer.py:660] (2/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=86069.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 22:51:39,490 WARNING [train.py:1060] (2/4) Exclude cut with ID 6758-72288-0033-108368-0 from training. Duration: 23.39
2022-12-23 22:51:45,425 WARNING [train.py:1060] (2/4) Exclude cut with ID 8291-282929-0007-12994-0_sp0.9 from training. Duration: 28.72225
2022-12-23 22:51:55,024 WARNING [train.py:1060] (2/4) Exclude cut with ID 585-294811-0110-133686-0_sp0.9 from training. Duration: 20.8944375
2022-12-23 22:52:00,930 INFO [zipformer.py:660] (2/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=86084.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 22:52:07,210 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([2.4745, 2.2186, 1.9434, 1.2230, 2.7622, 2.5654, 2.2310, 1.8257],
       device='cuda:2'), covar=tensor([0.0383, 0.0509, 0.0554, 0.0833, 0.0311, 0.0397, 0.0486, 0.0957],
       device='cuda:2'), in_proj_covar=tensor([0.0124, 0.0130, 0.0129, 0.0119, 0.0103, 0.0124, 0.0133, 0.0159],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0001, 0.0001, 0.0001, 0.0002, 0.0002],
       device='cuda:2')
2022-12-23 22:52:09,474 WARNING [train.py:1060] (2/4) Exclude cut with ID 5796-66357-0007-116447-0_sp0.9 from training. Duration: 23.8444375
2022-12-23 22:52:20,026 INFO [train.py:894] (2/4) Epoch 25, batch 1950, loss[loss=0.1639, simple_loss=0.2488, pruned_loss=0.03947, over 18704.00 frames. ], tot_loss[loss=0.1778, simple_loss=0.2622, pruned_loss=0.04672, over 3714608.11 frames. ], batch size: 50, lr: 4.57e-03, grad_scale: 16.0
2022-12-23 22:52:32,175 WARNING [train.py:1060] (2/4) Exclude cut with ID 8291-282929-0007-12994-0 from training. Duration: 25.85
2022-12-23 22:52:32,185 WARNING [train.py:1060] (2/4) Exclude cut with ID 8291-282929-0023-13010-0 from training. Duration: 21.39
2022-12-23 22:52:44,124 WARNING [train.py:1060] (2/4) Exclude cut with ID 7357-94126-0014-15845-0 from training. Duration: 27.92
2022-12-23 22:52:44,526 INFO [zipformer.py:660] (2/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=86113.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 22:53:09,631 INFO [zipformer.py:660] (2/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=86130.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 22:53:11,927 WARNING [train.py:1060] (2/4) Exclude cut with ID 8631-249866-0039-130165-0_sp0.9 from training. Duration: 20.661125
2022-12-23 22:53:12,037 INFO [zipformer.py:660] (2/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=86132.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 22:53:22,993 INFO [zipformer.py:660] (2/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=86139.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 22:53:31,647 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 2.321e+02 4.247e+02 5.001e+02 6.020e+02 1.322e+03, threshold=1.000e+03, percent-clipped=4.0
2022-12-23 22:53:34,768 INFO [train.py:894] (2/4) Epoch 25, batch 2000, loss[loss=0.2121, simple_loss=0.2955, pruned_loss=0.06441, over 18663.00 frames. ], tot_loss[loss=0.1794, simple_loss=0.2633, pruned_loss=0.04771, over 3714469.58 frames. ], batch size: 60, lr: 4.57e-03, grad_scale: 16.0
2022-12-23 22:53:36,277 WARNING [train.py:1060] (2/4) Exclude cut with ID 7357-94126-0043-15874-0_sp0.9 from training. Duration: 20.07225
2022-12-23 22:53:44,155 WARNING [train.py:1060] (2/4) Exclude cut with ID 1085-156170-0017-128270-0 from training. Duration: 21.01
2022-12-23 22:54:13,233 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([3.6024, 3.0029, 2.6944, 1.2544, 2.7436, 2.5619, 2.0228, 2.4829],
       device='cuda:2'), covar=tensor([0.0653, 0.0723, 0.1483, 0.1877, 0.1499, 0.1350, 0.1749, 0.1105],
       device='cuda:2'), in_proj_covar=tensor([0.0174, 0.0189, 0.0209, 0.0191, 0.0211, 0.0204, 0.0219, 0.0203],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-23 22:54:34,855 INFO [zipformer.py:660] (2/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=86186.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 22:54:36,160 INFO [zipformer.py:660] (2/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=86187.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 22:54:49,890 WARNING [train.py:1060] (2/4) Exclude cut with ID 2195-150901-0045-59933-0 from training. Duration: 20.65
2022-12-23 22:54:51,337 INFO [train.py:894] (2/4) Epoch 25, batch 2050, loss[loss=0.183, simple_loss=0.2703, pruned_loss=0.04782, over 18537.00 frames. ], tot_loss[loss=0.1809, simple_loss=0.2643, pruned_loss=0.04875, over 3714998.22 frames. ], batch size: 55, lr: 4.57e-03, grad_scale: 16.0
2022-12-23 22:54:56,315 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([2.3695, 2.0747, 1.6685, 1.9334, 1.7774, 1.9842, 1.8901, 2.1749],
       device='cuda:2'), covar=tensor([0.2110, 0.2976, 0.1987, 0.2724, 0.3621, 0.1133, 0.2965, 0.1062],
       device='cuda:2'), in_proj_covar=tensor([0.0298, 0.0296, 0.0251, 0.0349, 0.0278, 0.0233, 0.0294, 0.0219],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0003, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-23 22:54:57,558 WARNING [train.py:1060] (2/4) Exclude cut with ID 5796-66357-0007-116447-0 from training. Duration: 21.46
2022-12-23 22:55:18,334 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.8011, 1.8450, 1.5632, 1.5431, 2.0246, 2.0135, 2.0009, 1.3655],
       device='cuda:2'), covar=tensor([0.0327, 0.0271, 0.0468, 0.0236, 0.0185, 0.0355, 0.0294, 0.0336],
       device='cuda:2'), in_proj_covar=tensor([0.0095, 0.0126, 0.0151, 0.0123, 0.0115, 0.0120, 0.0099, 0.0127],
       device='cuda:2'), out_proj_covar=tensor([7.5385e-05, 9.9207e-05, 1.2390e-04, 9.7457e-05, 9.2652e-05, 9.2137e-05,
        7.7309e-05, 9.9821e-05], device='cuda:2')
2022-12-23 22:55:31,665 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([2.0341, 1.1464, 1.8461, 1.6621, 2.0938, 2.1163, 1.7882, 1.8241],
       device='cuda:2'), covar=tensor([0.2337, 0.3384, 0.2751, 0.2851, 0.2177, 0.1042, 0.3268, 0.1402],
       device='cuda:2'), in_proj_covar=tensor([0.0272, 0.0302, 0.0285, 0.0324, 0.0316, 0.0258, 0.0351, 0.0247],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-23 22:55:37,768 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([2.2541, 2.7976, 2.6797, 1.1977, 2.9798, 2.1291, 0.4646, 1.8011],
       device='cuda:2'), covar=tensor([0.2361, 0.1848, 0.1929, 0.3949, 0.1224, 0.1176, 0.4869, 0.1818],
       device='cuda:2'), in_proj_covar=tensor([0.0149, 0.0147, 0.0161, 0.0124, 0.0149, 0.0114, 0.0144, 0.0115],
       device='cuda:2'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0002, 0.0003, 0.0002, 0.0003, 0.0002],
       device='cuda:2')
2022-12-23 22:55:42,231 WARNING [train.py:1060] (2/4) Exclude cut with ID 3557-8342-0013-54691-0 from training. Duration: 0.92
2022-12-23 22:55:43,878 INFO [zipformer.py:660] (2/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=86231.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 22:55:48,115 WARNING [train.py:1060] (2/4) Exclude cut with ID 8291-282929-0023-13010-0_sp0.9 from training. Duration: 23.7666875
2022-12-23 22:55:48,226 INFO [zipformer.py:660] (2/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=86234.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 22:55:48,608 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([0.5580, 1.4269, 1.4708, 0.9698, 1.1132, 1.4926, 1.5589, 1.2845],
       device='cuda:2'), covar=tensor([0.0686, 0.0345, 0.0302, 0.0383, 0.0381, 0.0521, 0.0233, 0.0630],
       device='cuda:2'), in_proj_covar=tensor([0.0151, 0.0174, 0.0130, 0.0142, 0.0150, 0.0146, 0.0167, 0.0178],
       device='cuda:2'), out_proj_covar=tensor([1.1448e-04, 1.3173e-04, 9.6554e-05, 1.0444e-04, 1.1020e-04, 1.1001e-04,
        1.2669e-04, 1.3422e-04], device='cuda:2')
2022-12-23 22:56:04,974 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 3.056e+02 4.151e+02 5.152e+02 6.542e+02 1.596e+03, threshold=1.030e+03, percent-clipped=4.0
2022-12-23 22:56:07,278 INFO [zipformer.py:660] (2/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=86246.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 22:56:08,503 INFO [train.py:894] (2/4) Epoch 25, batch 2100, loss[loss=0.2272, simple_loss=0.2981, pruned_loss=0.07817, over 18612.00 frames. ], tot_loss[loss=0.1825, simple_loss=0.2655, pruned_loss=0.04974, over 3714378.56 frames. ], batch size: 179, lr: 4.57e-03, grad_scale: 16.0
2022-12-23 22:56:22,373 WARNING [train.py:1060] (2/4) Exclude cut with ID 8544-281189-0060-101339-0_sp0.9 from training. Duration: 20.861125
2022-12-23 22:56:32,123 WARNING [train.py:1060] (2/4) Exclude cut with ID 3488-65654-0031-41259-0_sp0.9 from training. Duration: 22.711125
2022-12-23 22:57:02,493 INFO [zipformer.py:660] (2/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=86281.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 22:57:09,263 INFO [zipformer.py:660] (2/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=86286.0, num_to_drop=1, layers_to_drop={3}
2022-12-23 22:57:15,371 WARNING [train.py:1060] (2/4) Exclude cut with ID 6951-79737-0043-132310-0_sp1.1 from training. Duration: 22.986375
2022-12-23 22:57:26,188 INFO [train.py:894] (2/4) Epoch 25, batch 2150, loss[loss=0.1867, simple_loss=0.2692, pruned_loss=0.05215, over 18591.00 frames. ], tot_loss[loss=0.1816, simple_loss=0.2646, pruned_loss=0.04927, over 3714518.61 frames. ], batch size: 51, lr: 4.57e-03, grad_scale: 16.0
2022-12-23 22:57:30,726 WARNING [train.py:1060] (2/4) Exclude cut with ID 8040-260924-0003-80960-0_sp0.9 from training. Duration: 22.07225
2022-12-23 22:57:35,156 WARNING [train.py:1060] (2/4) Exclude cut with ID 7699-105389-0045-26330-0_sp0.9 from training. Duration: 20.3055625
2022-12-23 22:57:38,049 WARNING [train.py:1060] (2/4) Exclude cut with ID 6356-271890-0060-94317-0_sp0.9 from training. Duration: 20.72225
2022-12-23 22:57:40,591 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([2.3219, 2.0240, 1.5938, 1.9219, 1.7894, 1.9947, 1.9077, 2.1863],
       device='cuda:2'), covar=tensor([0.2101, 0.3101, 0.2099, 0.2763, 0.3519, 0.1139, 0.2916, 0.1018],
       device='cuda:2'), in_proj_covar=tensor([0.0298, 0.0294, 0.0251, 0.0348, 0.0277, 0.0232, 0.0294, 0.0219],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0003, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-23 22:57:41,941 INFO [zipformer.py:660] (2/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=86307.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 22:57:50,575 INFO [zipformer.py:660] (2/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=86313.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 22:57:53,462 INFO [scaling.py:679] (2/4) Whitening: num_groups=8, num_channels=192, metric=1.90 vs. limit=2.0
2022-12-23 22:57:58,375 WARNING [train.py:1060] (2/4) Exclude cut with ID 3488-85273-0017-41203-0_sp1.1 from training. Duration: 22.4818125
2022-12-23 22:58:15,621 INFO [zipformer.py:660] (2/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=86329.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 22:58:24,032 WARNING [train.py:1060] (2/4) Exclude cut with ID 4964-30587-0040-44509-0_sp0.9 from training. Duration: 25.0944375
2022-12-23 22:58:28,538 WARNING [train.py:1060] (2/4) Exclude cut with ID 6533-399-0047-104881-0 from training. Duration: 21.515
2022-12-23 22:58:34,267 WARNING [train.py:1060] (2/4) Exclude cut with ID 7357-94126-0009-15840-0_sp0.9 from training. Duration: 27.02225
2022-12-23 22:58:39,249 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 2.878e+02 4.086e+02 4.973e+02 6.168e+02 9.928e+02, threshold=9.946e+02, percent-clipped=0.0
2022-12-23 22:58:39,334 WARNING [train.py:1060] (2/4) Exclude cut with ID 432-122774-0010-62480-0_sp0.9 from training. Duration: 22.22225
2022-12-23 22:58:42,237 INFO [train.py:894] (2/4) Epoch 25, batch 2200, loss[loss=0.1989, simple_loss=0.2821, pruned_loss=0.05788, over 18579.00 frames. ], tot_loss[loss=0.181, simple_loss=0.264, pruned_loss=0.04896, over 3713245.20 frames. ], batch size: 57, lr: 4.57e-03, grad_scale: 16.0
2022-12-23 22:58:46,760 WARNING [train.py:1060] (2/4) Exclude cut with ID 4964-30587-0085-44554-0_sp0.9 from training. Duration: 20.85
2022-12-23 22:58:59,829 INFO [zipformer.py:660] (2/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=86358.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 22:59:20,015 WARNING [train.py:1060] (2/4) Exclude cut with ID 4295-39940-0007-92567-0 from training. Duration: 21.54
2022-12-23 22:59:21,469 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.4107, 1.3187, 1.3732, 1.2631, 0.7939, 2.2965, 0.7874, 1.4072],
       device='cuda:2'), covar=tensor([0.3399, 0.2379, 0.2265, 0.2441, 0.1878, 0.0393, 0.1883, 0.1007],
       device='cuda:2'), in_proj_covar=tensor([0.0133, 0.0118, 0.0126, 0.0122, 0.0106, 0.0097, 0.0090, 0.0090],
       device='cuda:2'), out_proj_covar=tensor([0.0005, 0.0005, 0.0005, 0.0005, 0.0004, 0.0003, 0.0004, 0.0004],
       device='cuda:2')
2022-12-23 22:59:24,511 WARNING [train.py:1060] (2/4) Exclude cut with ID 4964-30587-0040-44509-0_sp1.1 from training. Duration: 20.5318125
2022-12-23 22:59:34,839 WARNING [train.py:1060] (2/4) Exclude cut with ID 7255-291500-0012-134311-0_sp0.9 from training. Duration: 21.9333125
2022-12-23 22:59:56,997 INFO [train.py:894] (2/4) Epoch 25, batch 2250, loss[loss=0.1788, simple_loss=0.2627, pruned_loss=0.0474, over 18477.00 frames. ], tot_loss[loss=0.1807, simple_loss=0.2635, pruned_loss=0.0489, over 3714654.65 frames. ], batch size: 54, lr: 4.57e-03, grad_scale: 16.0
2022-12-23 23:00:14,596 INFO [zipformer.py:660] (2/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=86408.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 23:00:22,247 WARNING [train.py:1060] (2/4) Exclude cut with ID 8631-249866-0025-130151-0_sp0.9 from training. Duration: 21.7944375
2022-12-23 23:00:30,220 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.6985, 1.7036, 1.2567, 1.5451, 1.7453, 1.5073, 2.2200, 1.7919],
       device='cuda:2'), covar=tensor([0.1076, 0.1647, 0.2950, 0.1735, 0.1957, 0.1161, 0.0990, 0.1410],
       device='cuda:2'), in_proj_covar=tensor([0.0181, 0.0214, 0.0255, 0.0291, 0.0241, 0.0194, 0.0208, 0.0208],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-23 23:00:34,950 WARNING [train.py:1060] (2/4) Exclude cut with ID 8291-282929-0002-12989-0_sp0.9 from training. Duration: 22.4666875
2022-12-23 23:00:40,803 INFO [zipformer.py:660] (2/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=86425.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 23:00:42,141 WARNING [train.py:1060] (2/4) Exclude cut with ID 6121-9014-0076-24124-0 from training. Duration: 21.635
2022-12-23 23:00:47,855 WARNING [train.py:1060] (2/4) Exclude cut with ID 6121-9014-0076-24124-0_sp0.9 from training. Duration: 24.038875
2022-12-23 23:01:03,555 INFO [zipformer.py:660] (2/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=86440.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 23:01:10,456 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 2.162e+02 3.881e+02 4.977e+02 5.869e+02 1.458e+03, threshold=9.954e+02, percent-clipped=4.0
2022-12-23 23:01:10,811 INFO [zipformer.py:660] (2/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=86445.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 23:01:13,391 INFO [train.py:894] (2/4) Epoch 25, batch 2300, loss[loss=0.1583, simple_loss=0.2407, pruned_loss=0.03791, over 18690.00 frames. ], tot_loss[loss=0.181, simple_loss=0.264, pruned_loss=0.04897, over 3714175.17 frames. ], batch size: 46, lr: 4.56e-03, grad_scale: 16.0
2022-12-23 23:01:34,083 WARNING [train.py:1060] (2/4) Exclude cut with ID 7255-291500-0009-134308-0_sp1.1 from training. Duration: 21.786375
2022-12-23 23:01:44,922 WARNING [train.py:1060] (2/4) Exclude cut with ID 8291-282929-0002-12989-0 from training. Duration: 20.22
2022-12-23 23:02:30,732 INFO [train.py:894] (2/4) Epoch 25, batch 2350, loss[loss=0.1774, simple_loss=0.2703, pruned_loss=0.04227, over 18482.00 frames. ], tot_loss[loss=0.1805, simple_loss=0.2632, pruned_loss=0.04883, over 3714623.47 frames. ], batch size: 54, lr: 4.56e-03, grad_scale: 16.0
2022-12-23 23:02:33,863 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.5212, 1.5286, 1.3996, 1.3646, 1.8029, 1.6934, 1.6728, 1.2047],
       device='cuda:2'), covar=tensor([0.0340, 0.0222, 0.0464, 0.0219, 0.0183, 0.0388, 0.0283, 0.0305],
       device='cuda:2'), in_proj_covar=tensor([0.0094, 0.0124, 0.0149, 0.0122, 0.0114, 0.0119, 0.0098, 0.0125],
       device='cuda:2'), out_proj_covar=tensor([7.4268e-05, 9.7870e-05, 1.2227e-04, 9.6611e-05, 9.1969e-05, 9.1192e-05,
        7.6162e-05, 9.8325e-05], device='cuda:2')
2022-12-23 23:02:37,256 INFO [zipformer.py:660] (2/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=86501.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 23:02:42,369 INFO [scaling.py:679] (2/4) Whitening: num_groups=8, num_channels=96, metric=1.18 vs. limit=2.0
2022-12-23 23:02:44,740 INFO [zipformer.py:660] (2/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=86506.0, num_to_drop=1, layers_to_drop={1}
2022-12-23 23:02:46,146 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.1927, 1.3229, 1.5128, 0.8966, 1.2794, 1.3571, 1.1990, 1.6368],
       device='cuda:2'), covar=tensor([0.1260, 0.2100, 0.1171, 0.1541, 0.0895, 0.1113, 0.2629, 0.0643],
       device='cuda:2'), in_proj_covar=tensor([0.0198, 0.0215, 0.0209, 0.0193, 0.0174, 0.0217, 0.0217, 0.0194],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-23 23:02:47,761 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.4448, 3.0030, 3.3840, 1.1018, 2.7986, 3.6059, 2.6998, 2.8003],
       device='cuda:2'), covar=tensor([0.0782, 0.0356, 0.0254, 0.0548, 0.0406, 0.0371, 0.0417, 0.0637],
       device='cuda:2'), in_proj_covar=tensor([0.0151, 0.0173, 0.0130, 0.0142, 0.0150, 0.0145, 0.0167, 0.0178],
       device='cuda:2'), out_proj_covar=tensor([1.1459e-04, 1.3099e-04, 9.6450e-05, 1.0433e-04, 1.1009e-04, 1.0979e-04,
        1.2685e-04, 1.3403e-04], device='cuda:2')
2022-12-23 23:03:22,411 INFO [zipformer.py:660] (2/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=86531.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 23:03:43,757 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 2.777e+02 3.901e+02 5.093e+02 6.285e+02 1.254e+03, threshold=1.019e+03, percent-clipped=1.0
2022-12-23 23:03:43,832 WARNING [train.py:1060] (2/4) Exclude cut with ID 6951-79737-0043-132310-0 from training. Duration: 25.285
2022-12-23 23:03:46,470 INFO [train.py:894] (2/4) Epoch 25, batch 2400, loss[loss=0.176, simple_loss=0.262, pruned_loss=0.04502, over 18707.00 frames. ], tot_loss[loss=0.1803, simple_loss=0.263, pruned_loss=0.04882, over 3713472.47 frames. ], batch size: 50, lr: 4.56e-03, grad_scale: 16.0
2022-12-23 23:04:00,005 INFO [scaling.py:679] (2/4) Whitening: num_groups=8, num_channels=96, metric=1.20 vs. limit=2.0
2022-12-23 23:04:36,259 INFO [zipformer.py:660] (2/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=86579.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 23:04:38,018 INFO [zipformer.py:660] (2/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=86580.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 23:04:47,240 WARNING [train.py:1060] (2/4) Exclude cut with ID 811-130148-0001-63453-0_sp0.9 from training. Duration: 20.861125
2022-12-23 23:04:47,388 INFO [zipformer.py:660] (2/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=86586.0, num_to_drop=1, layers_to_drop={1}
2022-12-23 23:05:03,676 INFO [train.py:894] (2/4) Epoch 25, batch 2450, loss[loss=0.1521, simple_loss=0.2327, pruned_loss=0.03577, over 18693.00 frames. ], tot_loss[loss=0.1799, simple_loss=0.2626, pruned_loss=0.04861, over 3714556.32 frames. ], batch size: 46, lr: 4.56e-03, grad_scale: 16.0
2022-12-23 23:05:08,234 WARNING [train.py:1060] (2/4) Exclude cut with ID 6010-56788-0055-90261-0 from training. Duration: 20.88
2022-12-23 23:05:12,213 INFO [zipformer.py:660] (2/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=86602.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 23:05:28,315 INFO [zipformer.py:660] (2/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=86613.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 23:05:41,687 WARNING [train.py:1060] (2/4) Exclude cut with ID 7357-94126-0045-15876-0_sp0.9 from training. Duration: 23.4166875
2022-12-23 23:06:00,847 INFO [zipformer.py:660] (2/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=86634.0, num_to_drop=1, layers_to_drop={0}
2022-12-23 23:06:11,081 INFO [zipformer.py:660] (2/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=86641.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 23:06:17,293 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 2.818e+02 4.531e+02 5.452e+02 6.835e+02 1.192e+03, threshold=1.090e+03, percent-clipped=5.0
2022-12-23 23:06:20,414 INFO [train.py:894] (2/4) Epoch 25, batch 2500, loss[loss=0.1742, simple_loss=0.2593, pruned_loss=0.04459, over 18588.00 frames. ], tot_loss[loss=0.1799, simple_loss=0.2627, pruned_loss=0.04859, over 3714390.78 frames. ], batch size: 100, lr: 4.56e-03, grad_scale: 16.0
2022-12-23 23:06:37,458 INFO [zipformer.py:660] (2/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=86658.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 23:06:41,800 INFO [zipformer.py:660] (2/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=86661.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 23:06:58,241 WARNING [train.py:1060] (2/4) Exclude cut with ID 7255-291500-0006-134305-0 from training. Duration: 21.24
2022-12-23 23:06:58,251 WARNING [train.py:1060] (2/4) Exclude cut with ID 6533-399-0047-104881-0_sp0.9 from training. Duration: 23.9055625
2022-12-23 23:07:31,338 WARNING [train.py:1060] (2/4) Exclude cut with ID 6758-72288-0033-108368-0_sp0.9 from training. Duration: 25.988875
2022-12-23 23:07:35,993 INFO [train.py:894] (2/4) Epoch 25, batch 2550, loss[loss=0.1679, simple_loss=0.2556, pruned_loss=0.04005, over 18597.00 frames. ], tot_loss[loss=0.1804, simple_loss=0.2628, pruned_loss=0.04893, over 3713306.00 frames. ], batch size: 69, lr: 4.56e-03, grad_scale: 16.0
2022-12-23 23:07:40,353 WARNING [train.py:1060] (2/4) Exclude cut with ID 7255-291500-0001-134300-0_sp0.9 from training. Duration: 20.67225
2022-12-23 23:07:49,245 INFO [zipformer.py:660] (2/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=86706.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 23:07:52,324 INFO [zipformer.py:660] (2/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=86708.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 23:07:52,404 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([2.2674, 2.2674, 1.8144, 1.7125, 2.3224, 2.9514, 2.6339, 2.0675],
       device='cuda:2'), covar=tensor([0.0349, 0.0270, 0.0427, 0.0265, 0.0229, 0.0279, 0.0318, 0.0294],
       device='cuda:2'), in_proj_covar=tensor([0.0094, 0.0125, 0.0151, 0.0123, 0.0115, 0.0120, 0.0099, 0.0126],
       device='cuda:2'), out_proj_covar=tensor([7.4634e-05, 9.8932e-05, 1.2361e-04, 9.7111e-05, 9.2712e-05, 9.2310e-05,
        7.7415e-05, 9.9445e-05], device='cuda:2')
2022-12-23 23:08:17,798 INFO [zipformer.py:660] (2/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=86725.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 23:08:26,497 WARNING [train.py:1060] (2/4) Exclude cut with ID 3488-85273-0038-41224-0 from training. Duration: 20.34
2022-12-23 23:08:48,780 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 2.379e+02 3.888e+02 5.016e+02 6.192e+02 1.481e+03, threshold=1.003e+03, percent-clipped=3.0
2022-12-23 23:08:51,482 INFO [train.py:894] (2/4) Epoch 25, batch 2600, loss[loss=0.2271, simple_loss=0.3002, pruned_loss=0.07698, over 18541.00 frames. ], tot_loss[loss=0.1802, simple_loss=0.2629, pruned_loss=0.04873, over 3713803.10 frames. ], batch size: 69, lr: 4.56e-03, grad_scale: 16.0
2022-12-23 23:09:04,706 INFO [zipformer.py:660] (2/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=86756.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 23:09:30,651 INFO [zipformer.py:660] (2/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=86773.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 23:09:40,540 WARNING [train.py:1060] (2/4) Exclude cut with ID 7357-94126-0026-15857-0_sp0.9 from training. Duration: 25.061125
2022-12-23 23:09:51,612 WARNING [train.py:1060] (2/4) Exclude cut with ID 3033-130750-0096-55598-0 from training. Duration: 0.83
2022-12-23 23:10:05,348 INFO [zipformer.py:660] (2/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=86796.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 23:10:05,805 INFO [scaling.py:679] (2/4) Whitening: num_groups=8, num_channels=192, metric=1.58 vs. limit=2.0
2022-12-23 23:10:06,495 INFO [train.py:894] (2/4) Epoch 25, batch 2650, loss[loss=0.1989, simple_loss=0.2792, pruned_loss=0.05931, over 18533.00 frames. ], tot_loss[loss=0.1798, simple_loss=0.2628, pruned_loss=0.04842, over 3714905.11 frames. ], batch size: 58, lr: 4.55e-03, grad_scale: 16.0
2022-12-23 23:10:12,628 INFO [zipformer.py:660] (2/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=86801.0, num_to_drop=1, layers_to_drop={0}
2022-12-23 23:10:18,214 WARNING [train.py:1060] (2/4) Exclude cut with ID 3488-85273-0017-41203-0 from training. Duration: 24.73
2022-12-23 23:10:18,726 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.8850, 0.7284, 1.6821, 1.4699, 1.8720, 1.9140, 1.5309, 1.6801],
       device='cuda:2'), covar=tensor([0.2274, 0.3522, 0.2783, 0.2859, 0.2213, 0.1042, 0.3343, 0.1422],
       device='cuda:2'), in_proj_covar=tensor([0.0272, 0.0301, 0.0286, 0.0323, 0.0316, 0.0258, 0.0351, 0.0247],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-23 23:10:30,273 WARNING [train.py:1060] (2/4) Exclude cut with ID 7255-291500-0009-134308-0 from training. Duration: 23.965
2022-12-23 23:10:32,001 INFO [zipformer.py:660] (2/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=86814.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 23:10:39,962 WARNING [train.py:1060] (2/4) Exclude cut with ID 6978-92210-0030-146996-0_sp0.9 from training. Duration: 22.088875
2022-12-23 23:10:57,392 WARNING [train.py:1060] (2/4) Exclude cut with ID 7255-291500-0006-134305-0_sp0.9 from training. Duration: 23.6
2022-12-23 23:11:18,491 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 2.776e+02 4.332e+02 4.963e+02 5.753e+02 1.372e+03, threshold=9.925e+02, percent-clipped=1.0
2022-12-23 23:11:21,279 INFO [train.py:894] (2/4) Epoch 25, batch 2700, loss[loss=0.1501, simple_loss=0.2326, pruned_loss=0.03378, over 18529.00 frames. ], tot_loss[loss=0.1792, simple_loss=0.2616, pruned_loss=0.04835, over 3715398.29 frames. ], batch size: 44, lr: 4.55e-03, grad_scale: 16.0
2022-12-23 23:12:05,294 INFO [zipformer.py:660] (2/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=86875.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 23:12:09,838 INFO [zipformer.py:660] (2/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=86878.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 23:12:37,879 INFO [train.py:894] (2/4) Epoch 25, batch 2750, loss[loss=0.1664, simple_loss=0.2468, pruned_loss=0.04302, over 18587.00 frames. ], tot_loss[loss=0.1788, simple_loss=0.2615, pruned_loss=0.04806, over 3714722.44 frames. ], batch size: 49, lr: 4.55e-03, grad_scale: 8.0
2022-12-23 23:12:41,009 WARNING [train.py:1060] (2/4) Exclude cut with ID 8291-282929-0024-13011-0 from training. Duration: 23.795
2022-12-23 23:12:45,523 INFO [zipformer.py:660] (2/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=86902.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 23:12:58,816 WARNING [train.py:1060] (2/4) Exclude cut with ID 8631-249866-0030-130156-0_sp1.1 from training. Duration: 21.5409375
2022-12-23 23:13:01,633 WARNING [train.py:1060] (2/4) Exclude cut with ID 6978-92210-0019-146985-0_sp0.9 from training. Duration: 24.97775
2022-12-23 23:13:13,042 WARNING [train.py:1060] (2/4) Exclude cut with ID 1085-156170-0017-128270-0_sp0.9 from training. Duration: 23.3444375
2022-12-23 23:13:38,159 INFO [zipformer.py:660] (2/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=86936.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 23:13:42,710 WARNING [train.py:1060] (2/4) Exclude cut with ID 6010-56788-0055-90261-0_sp0.9 from training. Duration: 23.2
2022-12-23 23:13:43,164 INFO [zipformer.py:660] (2/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=86939.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 23:13:47,214 WARNING [train.py:1060] (2/4) Exclude cut with ID 5653-46179-0060-117930-0_sp0.9 from training. Duration: 21.17225
2022-12-23 23:13:50,734 INFO [scaling.py:679] (2/4) Whitening: num_groups=1, num_channels=384, metric=3.02 vs. limit=5.0
2022-12-23 23:13:52,950 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 2.402e+02 3.994e+02 4.847e+02 5.810e+02 1.077e+03, threshold=9.693e+02, percent-clipped=1.0
2022-12-23 23:13:54,454 INFO [train.py:894] (2/4) Epoch 25, batch 2800, loss[loss=0.1872, simple_loss=0.2722, pruned_loss=0.05104, over 18669.00 frames. ], tot_loss[loss=0.1771, simple_loss=0.2597, pruned_loss=0.04724, over 3714583.32 frames. ], batch size: 98, lr: 4.55e-03, grad_scale: 8.0
2022-12-23 23:13:58,944 INFO [zipformer.py:660] (2/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=86950.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 23:14:05,511 WARNING [train.py:1060] (2/4) Exclude cut with ID 7255-291500-0008-134307-0_sp0.9 from training. Duration: 24.6555625
2022-12-23 23:14:58,557 WARNING [train.py:1060] (2/4) Exclude cut with ID 3488-65654-0031-41259-0 from training. Duration: 20.44
2022-12-23 23:15:12,520 INFO [train.py:894] (2/4) Epoch 25, batch 2850, loss[loss=0.1836, simple_loss=0.2685, pruned_loss=0.0493, over 18540.00 frames. ], tot_loss[loss=0.1769, simple_loss=0.2596, pruned_loss=0.04712, over 3714835.02 frames. ], batch size: 57, lr: 4.55e-03, grad_scale: 8.0
2022-12-23 23:15:15,466 WARNING [train.py:1060] (2/4) Exclude cut with ID 6951-79737-0018-132285-0_sp0.9 from training. Duration: 23.45
2022-12-23 23:15:20,973 INFO [zipformer.py:660] (2/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=87002.0, num_to_drop=1, layers_to_drop={1}
2022-12-23 23:15:28,849 INFO [scaling.py:679] (2/4) Whitening: num_groups=8, num_channels=192, metric=1.63 vs. limit=2.0
2022-12-23 23:15:46,048 WARNING [train.py:1060] (2/4) Exclude cut with ID 6945-60535-0076-12784-0_sp0.9 from training. Duration: 20.52225
2022-12-23 23:15:53,770 WARNING [train.py:1060] (2/4) Exclude cut with ID 7255-291500-0008-134307-0 from training. Duration: 22.19
2022-12-23 23:16:01,809 INFO [scaling.py:679] (2/4) Whitening: num_groups=8, num_channels=96, metric=1.20 vs. limit=2.0
2022-12-23 23:16:04,103 WARNING [train.py:1060] (2/4) Exclude cut with ID 7357-94126-0014-15845-0_sp1.1 from training. Duration: 25.3818125
2022-12-23 23:16:21,538 WARNING [train.py:1060] (2/4) Exclude cut with ID 6951-79737-0043-132310-0_sp0.9 from training. Duration: 28.0944375
2022-12-23 23:16:26,182 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 2.838e+02 4.269e+02 5.078e+02 6.518e+02 1.318e+03, threshold=1.016e+03, percent-clipped=2.0
2022-12-23 23:16:27,878 INFO [train.py:894] (2/4) Epoch 25, batch 2900, loss[loss=0.1663, simple_loss=0.2515, pruned_loss=0.04054, over 18595.00 frames. ], tot_loss[loss=0.1768, simple_loss=0.2594, pruned_loss=0.04716, over 3714030.28 frames. ], batch size: 56, lr: 4.55e-03, grad_scale: 8.0
2022-12-23 23:16:27,927 WARNING [train.py:1060] (2/4) Exclude cut with ID 2195-150901-0045-59933-0_sp0.9 from training. Duration: 22.9444375
2022-12-23 23:16:37,095 WARNING [train.py:1060] (2/4) Exclude cut with ID 8291-282929-0024-13011-0_sp1.1 from training. Duration: 21.6318125
2022-12-23 23:16:53,065 INFO [zipformer.py:660] (2/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=87063.0, num_to_drop=1, layers_to_drop={3}
2022-12-23 23:16:54,082 WARNING [train.py:1060] (2/4) Exclude cut with ID 8631-249866-0030-130156-0 from training. Duration: 23.695
2022-12-23 23:17:20,426 WARNING [train.py:1060] (2/4) Exclude cut with ID 7699-105389-0094-26379-0 from training. Duration: 23.955
2022-12-23 23:17:25,334 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([2.1028, 1.2543, 1.8481, 1.6967, 2.0896, 2.1070, 1.8326, 1.8291],
       device='cuda:2'), covar=tensor([0.2342, 0.3646, 0.2814, 0.3081, 0.2285, 0.1099, 0.3507, 0.1434],
       device='cuda:2'), in_proj_covar=tensor([0.0273, 0.0302, 0.0287, 0.0324, 0.0317, 0.0260, 0.0353, 0.0248],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-23 23:17:41,843 INFO [zipformer.py:660] (2/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=87096.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 23:17:43,051 INFO [train.py:894] (2/4) Epoch 25, batch 2950, loss[loss=0.1633, simple_loss=0.2366, pruned_loss=0.04505, over 18676.00 frames. ], tot_loss[loss=0.1778, simple_loss=0.2603, pruned_loss=0.04762, over 3714709.66 frames. ], batch size: 46, lr: 4.55e-03, grad_scale: 8.0
2022-12-23 23:17:49,623 INFO [zipformer.py:660] (2/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=87101.0, num_to_drop=1, layers_to_drop={1}
2022-12-23 23:17:50,022 INFO [scaling.py:679] (2/4) Whitening: num_groups=1, num_channels=384, metric=2.96 vs. limit=5.0
2022-12-23 23:17:53,827 WARNING [train.py:1060] (2/4) Exclude cut with ID 8291-282929-0024-13011-0_sp0.9 from training. Duration: 26.438875
2022-12-23 23:18:35,996 WARNING [train.py:1060] (2/4) Exclude cut with ID 7699-105389-0021-26306-0_sp0.9 from training. Duration: 21.2444375
2022-12-23 23:18:36,025 WARNING [train.py:1060] (2/4) Exclude cut with ID 7357-94126-0014-15845-0_sp0.9 from training. Duration: 31.02225
2022-12-23 23:18:46,838 WARNING [train.py:1060] (2/4) Exclude cut with ID 432-122774-0017-62487-0 from training. Duration: 22.395
2022-12-23 23:18:50,734 INFO [scaling.py:679] (2/4) Whitening: num_groups=1, num_channels=384, metric=4.73 vs. limit=5.0
2022-12-23 23:18:55,894 INFO [zipformer.py:660] (2/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=87144.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 23:18:59,332 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 2.752e+02 4.188e+02 4.950e+02 6.486e+02 1.228e+03, threshold=9.900e+02, percent-clipped=4.0
2022-12-23 23:19:00,617 INFO [train.py:894] (2/4) Epoch 25, batch 3000, loss[loss=0.1772, simple_loss=0.2618, pruned_loss=0.0463, over 18479.00 frames. ], tot_loss[loss=0.178, simple_loss=0.2606, pruned_loss=0.04775, over 3714511.56 frames. ], batch size: 54, lr: 4.55e-03, grad_scale: 8.0
2022-12-23 23:19:00,618 INFO [train.py:919] (2/4) Computing validation loss
2022-12-23 23:19:04,402 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([4.5727, 4.0451, 3.9312, 4.5670, 4.1871, 4.0963, 4.6766, 1.6561],
       device='cuda:2'), covar=tensor([0.0557, 0.0608, 0.0656, 0.0611, 0.1215, 0.1053, 0.0434, 0.5278],
       device='cuda:2'), in_proj_covar=tensor([0.0351, 0.0234, 0.0243, 0.0278, 0.0333, 0.0274, 0.0296, 0.0289],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0001, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-23 23:19:11,394 INFO [train.py:928] (2/4) Epoch 25, validation: loss=0.1616, simple_loss=0.2586, pruned_loss=0.0323, over 944034.00 frames. 
2022-12-23 23:19:11,394 INFO [train.py:929] (2/4) Maximum memory allocated so far is 24680MB
2022-12-23 23:19:14,469 INFO [zipformer.py:660] (2/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=87149.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 23:19:17,280 WARNING [train.py:1060] (2/4) Exclude cut with ID 7357-94126-0045-15876-0 from training. Duration: 21.075
2022-12-23 23:19:21,447 WARNING [train.py:1060] (2/4) Exclude cut with ID 6482-98857-0025-147532-0_sp0.9 from training. Duration: 20.0055625
2022-12-23 23:19:21,459 WARNING [train.py:1060] (2/4) Exclude cut with ID 6951-79737-0037-132304-0_sp0.9 from training. Duration: 22.05
2022-12-23 23:19:21,471 WARNING [train.py:1060] (2/4) Exclude cut with ID 7255-291500-0003-134302-0 from training. Duration: 26.8349375
2022-12-23 23:19:21,838 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.8142, 1.7651, 1.8936, 1.8005, 1.1683, 3.6393, 1.5697, 2.0618],
       device='cuda:2'), covar=tensor([0.2975, 0.2054, 0.1812, 0.1964, 0.1564, 0.0195, 0.1591, 0.0844],
       device='cuda:2'), in_proj_covar=tensor([0.0132, 0.0118, 0.0125, 0.0121, 0.0105, 0.0097, 0.0090, 0.0090],
       device='cuda:2'), out_proj_covar=tensor([0.0005, 0.0005, 0.0005, 0.0005, 0.0004, 0.0003, 0.0004, 0.0004],
       device='cuda:2')
2022-12-23 23:19:26,252 WARNING [train.py:1060] (2/4) Exclude cut with ID 7357-94126-0009-15840-0_sp1.1 from training. Duration: 22.1090625
2022-12-23 23:19:32,420 WARNING [train.py:1060] (2/4) Exclude cut with ID 7699-105389-0094-26379-0_sp0.9 from training. Duration: 26.6166875
2022-12-23 23:19:40,558 INFO [scaling.py:679] (2/4) Whitening: num_groups=1, num_channels=384, metric=4.03 vs. limit=5.0
2022-12-23 23:19:46,480 INFO [zipformer.py:660] (2/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=87170.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 23:19:49,262 WARNING [train.py:1060] (2/4) Exclude cut with ID 2046-178027-0000-53705-0_sp0.9 from training. Duration: 20.3055625
2022-12-23 23:20:11,189 WARNING [train.py:1060] (2/4) Exclude cut with ID 7205-50138-0008-5373-0_sp0.9 from training. Duration: 20.7
2022-12-23 23:20:28,136 INFO [train.py:894] (2/4) Epoch 25, batch 3050, loss[loss=0.1662, simple_loss=0.2498, pruned_loss=0.0413, over 18564.00 frames. ], tot_loss[loss=0.1783, simple_loss=0.2609, pruned_loss=0.04788, over 3714814.83 frames. ], batch size: 49, lr: 4.54e-03, grad_scale: 8.0
2022-12-23 23:20:57,916 WARNING [train.py:1060] (2/4) Exclude cut with ID 6978-92210-0019-146985-0 from training. Duration: 22.48
2022-12-23 23:21:15,557 WARNING [train.py:1060] (2/4) Exclude cut with ID 7255-291500-0003-134302-0_sp0.9 from training. Duration: 29.816625
2022-12-23 23:21:17,430 INFO [zipformer.py:660] (2/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=87230.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 23:21:23,041 INFO [zipformer.py:660] (2/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=87234.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 23:21:23,209 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.7593, 1.3217, 0.8010, 1.4337, 2.0446, 1.2216, 1.5345, 1.6405],
       device='cuda:2'), covar=tensor([0.1659, 0.2201, 0.2374, 0.1548, 0.2042, 0.1835, 0.1522, 0.1739],
       device='cuda:2'), in_proj_covar=tensor([0.0095, 0.0099, 0.0118, 0.0097, 0.0121, 0.0093, 0.0099, 0.0095],
       device='cuda:2'), out_proj_covar=tensor([0.0003, 0.0003, 0.0004, 0.0003, 0.0004, 0.0003, 0.0003, 0.0003],
       device='cuda:2')
2022-12-23 23:21:25,985 INFO [zipformer.py:660] (2/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=87236.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 23:21:35,661 WARNING [train.py:1060] (2/4) Exclude cut with ID 7255-291500-0005-134304-0_sp1.1 from training. Duration: 22.7590625
2022-12-23 23:21:40,365 WARNING [train.py:1060] (2/4) Exclude cut with ID 7357-94126-0026-15857-0 from training. Duration: 22.555
2022-12-23 23:21:41,729 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 2.992e+02 4.307e+02 5.014e+02 6.364e+02 1.166e+03, threshold=1.003e+03, percent-clipped=4.0
2022-12-23 23:21:43,190 INFO [train.py:894] (2/4) Epoch 25, batch 3100, loss[loss=0.1681, simple_loss=0.244, pruned_loss=0.04612, over 18526.00 frames. ], tot_loss[loss=0.1774, simple_loss=0.2602, pruned_loss=0.04729, over 3714031.20 frames. ], batch size: 44, lr: 4.54e-03, grad_scale: 8.0
2022-12-23 23:21:58,935 WARNING [train.py:1060] (2/4) Exclude cut with ID 1250-135782-0005-25975-0_sp0.9 from training. Duration: 21.688875
2022-12-23 23:22:35,263 WARNING [train.py:1060] (2/4) Exclude cut with ID 3488-85273-0038-41224-0_sp0.9 from training. Duration: 22.6
2022-12-23 23:22:40,629 INFO [zipformer.py:660] (2/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=87284.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 23:22:52,537 INFO [zipformer.py:660] (2/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=87291.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 23:23:01,657 INFO [train.py:894] (2/4) Epoch 25, batch 3150, loss[loss=0.1521, simple_loss=0.2367, pruned_loss=0.03378, over 18666.00 frames. ], tot_loss[loss=0.177, simple_loss=0.2602, pruned_loss=0.04695, over 3714256.13 frames. ], batch size: 48, lr: 4.54e-03, grad_scale: 8.0
2022-12-23 23:23:09,176 WARNING [train.py:1060] (2/4) Exclude cut with ID 7357-94126-0009-15840-0 from training. Duration: 24.32
2022-12-23 23:24:09,102 WARNING [train.py:1060] (2/4) Exclude cut with ID 8291-276745-0093-13116-0_sp0.9 from training. Duration: 21.061125
2022-12-23 23:24:16,162 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 2.705e+02 4.200e+02 4.949e+02 5.939e+02 1.054e+03, threshold=9.899e+02, percent-clipped=1.0
2022-12-23 23:24:17,836 INFO [train.py:894] (2/4) Epoch 25, batch 3200, loss[loss=0.1769, simple_loss=0.2627, pruned_loss=0.04554, over 18594.00 frames. ], tot_loss[loss=0.1772, simple_loss=0.2605, pruned_loss=0.04696, over 3714270.14 frames. ], batch size: 51, lr: 4.54e-03, grad_scale: 8.0
2022-12-23 23:24:24,039 WARNING [train.py:1060] (2/4) Exclude cut with ID 7357-94126-0024-15855-0_sp0.9 from training. Duration: 20.32225
2022-12-23 23:24:34,882 INFO [zipformer.py:660] (2/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=87358.0, num_to_drop=1, layers_to_drop={1}
2022-12-23 23:24:36,367 WARNING [train.py:1060] (2/4) Exclude cut with ID 3033-130750-0096-55598-0_sp1.1 from training. Duration: 0.7545625
2022-12-23 23:24:41,584 INFO [zipformer.py:660] (2/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=87362.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 23:24:50,020 WARNING [train.py:1060] (2/4) Exclude cut with ID 4295-39940-0007-92567-0_sp0.9 from training. Duration: 23.9333125
2022-12-23 23:24:52,884 INFO [scaling.py:679] (2/4) Whitening: num_groups=1, num_channels=384, metric=4.69 vs. limit=5.0
2022-12-23 23:25:21,693 WARNING [train.py:1060] (2/4) Exclude cut with ID 7255-291500-0008-134307-0_sp1.1 from training. Duration: 20.17275
2022-12-23 23:25:27,094 WARNING [train.py:1060] (2/4) Exclude cut with ID 6978-92210-0019-146985-0_sp1.1 from training. Duration: 20.436375
2022-12-23 23:25:33,618 INFO [train.py:894] (2/4) Epoch 25, batch 3250, loss[loss=0.1866, simple_loss=0.2741, pruned_loss=0.04957, over 18674.00 frames. ], tot_loss[loss=0.1773, simple_loss=0.2606, pruned_loss=0.04704, over 3714428.74 frames. ], batch size: 97, lr: 4.54e-03, grad_scale: 8.0
2022-12-23 23:26:14,326 INFO [zipformer.py:660] (2/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=87423.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 23:26:47,658 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 3.020e+02 4.157e+02 5.173e+02 6.629e+02 1.386e+03, threshold=1.035e+03, percent-clipped=3.0
2022-12-23 23:26:49,119 INFO [train.py:894] (2/4) Epoch 25, batch 3300, loss[loss=0.1988, simple_loss=0.283, pruned_loss=0.05734, over 18528.00 frames. ], tot_loss[loss=0.1769, simple_loss=0.2603, pruned_loss=0.04678, over 3713935.67 frames. ], batch size: 55, lr: 4.54e-03, grad_scale: 8.0
2022-12-23 23:26:49,173 WARNING [train.py:1060] (2/4) Exclude cut with ID 4234-40345-0022-142709-0_sp0.9 from training. Duration: 23.1055625
2022-12-23 23:26:50,599 WARNING [train.py:1060] (2/4) Exclude cut with ID 8291-282929-0007-12994-0_sp1.1 from training. Duration: 23.5
2022-12-23 23:26:59,430 WARNING [train.py:1060] (2/4) Exclude cut with ID 7255-291500-0009-134308-0_sp0.9 from training. Duration: 26.62775
2022-12-23 23:27:10,767 INFO [zipformer.py:660] (2/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=87461.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 23:27:13,924 WARNING [train.py:1060] (2/4) Exclude cut with ID 6951-79737-0018-132285-0 from training. Duration: 21.105
2022-12-23 23:27:19,900 WARNING [train.py:1060] (2/4) Exclude cut with ID 4511-76322-0006-80011-0_sp0.9 from training. Duration: 24.411125
2022-12-23 23:27:25,108 INFO [zipformer.py:660] (2/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=87470.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 23:27:45,470 WARNING [train.py:1060] (2/4) Exclude cut with ID 6758-72288-0033-108368-0_sp1.1 from training. Duration: 21.263625
2022-12-23 23:28:05,684 INFO [train.py:894] (2/4) Epoch 25, batch 3350, loss[loss=0.2037, simple_loss=0.2833, pruned_loss=0.06211, over 18517.00 frames. ], tot_loss[loss=0.1788, simple_loss=0.2623, pruned_loss=0.04763, over 3715549.16 frames. ], batch size: 58, lr: 4.54e-03, grad_scale: 8.0
2022-12-23 23:28:18,133 WARNING [train.py:1060] (2/4) Exclude cut with ID 4234-40345-0022-142709-0 from training. Duration: 20.795
2022-12-23 23:28:19,202 INFO [zipformer.py:660] (2/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=87505.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 23:28:25,338 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([3.3632, 3.2097, 2.3074, 1.9092, 3.7523, 3.7667, 3.0505, 2.7667],
       device='cuda:2'), covar=tensor([0.0362, 0.0390, 0.0573, 0.0728, 0.0214, 0.0316, 0.0460, 0.0745],
       device='cuda:2'), in_proj_covar=tensor([0.0125, 0.0131, 0.0131, 0.0122, 0.0104, 0.0127, 0.0135, 0.0163],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0001, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-23 23:28:29,891 WARNING [train.py:1060] (2/4) Exclude cut with ID 7357-94126-0021-15852-0 from training. Duration: 24.76
2022-12-23 23:28:29,909 WARNING [train.py:1060] (2/4) Exclude cut with ID 3867-173237-0077-144769-0_sp0.9 from training. Duration: 22.25
2022-12-23 23:28:39,115 INFO [zipformer.py:660] (2/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=87518.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 23:28:45,030 INFO [zipformer.py:660] (2/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=87522.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 23:28:54,800 WARNING [train.py:1060] (2/4) Exclude cut with ID 7357-94126-0026-15857-0_sp1.1 from training. Duration: 20.5045625
2022-12-23 23:29:02,871 INFO [zipformer.py:660] (2/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=87534.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 23:29:20,008 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 2.807e+02 4.240e+02 5.058e+02 6.182e+02 1.379e+03, threshold=1.012e+03, percent-clipped=2.0
2022-12-23 23:29:21,514 INFO [train.py:894] (2/4) Epoch 25, batch 3400, loss[loss=0.1627, simple_loss=0.2454, pruned_loss=0.03995, over 18672.00 frames. ], tot_loss[loss=0.1784, simple_loss=0.2618, pruned_loss=0.04744, over 3715299.96 frames. ], batch size: 48, lr: 4.54e-03, grad_scale: 8.0
2022-12-23 23:29:50,320 INFO [zipformer.py:660] (2/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=87566.0, num_to_drop=1, layers_to_drop={1}
2022-12-23 23:30:13,760 INFO [zipformer.py:660] (2/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=87582.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 23:30:17,089 INFO [zipformer.py:660] (2/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=87584.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 23:30:20,122 INFO [zipformer.py:660] (2/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=87586.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 23:30:24,870 INFO [zipformer.py:660] (2/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=87589.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 23:30:36,103 INFO [train.py:894] (2/4) Epoch 25, batch 3450, loss[loss=0.1838, simple_loss=0.2679, pruned_loss=0.04981, over 18698.00 frames. ], tot_loss[loss=0.178, simple_loss=0.2614, pruned_loss=0.04734, over 3714982.01 frames. ], batch size: 62, lr: 4.53e-03, grad_scale: 8.0
2022-12-23 23:30:48,027 INFO [zipformer.py:660] (2/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=87605.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 23:31:46,420 INFO [zipformer.py:660] (2/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=87645.0, num_to_drop=1, layers_to_drop={2}
2022-12-23 23:31:47,682 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 2.307e+02 4.042e+02 4.989e+02 6.137e+02 1.168e+03, threshold=9.977e+02, percent-clipped=1.0
2022-12-23 23:31:49,097 INFO [train.py:894] (2/4) Epoch 25, batch 3500, loss[loss=0.1837, simple_loss=0.2734, pruned_loss=0.04697, over 18633.00 frames. ], tot_loss[loss=0.1785, simple_loss=0.2621, pruned_loss=0.04746, over 3716300.45 frames. ], batch size: 98, lr: 4.53e-03, grad_scale: 8.0
2022-12-23 23:31:54,167 INFO [zipformer.py:660] (2/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=87650.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 23:32:09,988 WARNING [train.py:1060] (2/4) Exclude cut with ID 7859-102521-0017-7548-0_sp1.1 from training. Duration: 22.2954375
2022-12-23 23:32:20,026 INFO [train.py:894] (2/4) Epoch 26, batch 0, loss[loss=0.1623, simple_loss=0.2416, pruned_loss=0.04147, over 18596.00 frames. ], tot_loss[loss=0.1623, simple_loss=0.2416, pruned_loss=0.04147, over 18596.00 frames. ], batch size: 41, lr: 4.44e-03, grad_scale: 8.0
2022-12-23 23:32:20,026 INFO [train.py:919] (2/4) Computing validation loss
2022-12-23 23:32:28,080 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.2409, 1.2332, 1.3619, 0.8827, 1.2762, 1.3156, 1.1246, 1.5038],
       device='cuda:2'), covar=tensor([0.1129, 0.2150, 0.1279, 0.1601, 0.0872, 0.1130, 0.2770, 0.0699],
       device='cuda:2'), in_proj_covar=tensor([0.0199, 0.0218, 0.0210, 0.0195, 0.0175, 0.0219, 0.0219, 0.0195],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-23 23:32:31,399 INFO [train.py:928] (2/4) Epoch 26, validation: loss=0.163, simple_loss=0.2598, pruned_loss=0.03308, over 944034.00 frames. 
2022-12-23 23:32:31,400 INFO [train.py:929] (2/4) Maximum memory allocated so far is 24680MB
2022-12-23 23:32:39,241 INFO [zipformer.py:660] (2/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=87658.0, num_to_drop=1, layers_to_drop={1}
2022-12-23 23:32:51,397 INFO [zipformer.py:660] (2/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=87666.0, num_to_drop=1, layers_to_drop={0}
2022-12-23 23:33:04,137 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.3584, 1.3837, 1.4752, 0.9229, 1.3025, 1.4375, 1.2599, 1.7056],
       device='cuda:2'), covar=tensor([0.1025, 0.2068, 0.1196, 0.1524, 0.0897, 0.1061, 0.2676, 0.0641],
       device='cuda:2'), in_proj_covar=tensor([0.0200, 0.0218, 0.0211, 0.0195, 0.0175, 0.0219, 0.0219, 0.0196],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-23 23:33:24,318 WARNING [train.py:1060] (2/4) Exclude cut with ID 298-126791-0067-24026-0_sp0.9 from training. Duration: 21.438875
2022-12-23 23:33:28,998 WARNING [train.py:1060] (2/4) Exclude cut with ID 5652-39938-0025-23684-0_sp0.9 from training. Duration: 22.2055625
2022-12-23 23:33:46,305 INFO [train.py:894] (2/4) Epoch 26, batch 50, loss[loss=0.1472, simple_loss=0.2412, pruned_loss=0.02662, over 18718.00 frames. ], tot_loss[loss=0.1712, simple_loss=0.2603, pruned_loss=0.04106, over 837821.73 frames. ], batch size: 50, lr: 4.44e-03, grad_scale: 8.0
2022-12-23 23:33:51,528 INFO [zipformer.py:660] (2/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=87706.0, num_to_drop=1, layers_to_drop={1}
2022-12-23 23:33:54,444 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([4.6721, 3.9980, 4.0508, 4.5895, 4.2880, 4.1348, 4.8020, 1.6578],
       device='cuda:2'), covar=tensor([0.0605, 0.0580, 0.0615, 0.0703, 0.1236, 0.1015, 0.0494, 0.4900],
       device='cuda:2'), in_proj_covar=tensor([0.0352, 0.0234, 0.0243, 0.0278, 0.0334, 0.0273, 0.0298, 0.0290],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0001, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-23 23:34:08,971 INFO [zipformer.py:660] (2/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=87718.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 23:34:49,306 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 1.982e+02 3.226e+02 3.859e+02 4.611e+02 8.239e+02, threshold=7.718e+02, percent-clipped=0.0
2022-12-23 23:34:59,999 INFO [train.py:894] (2/4) Epoch 26, batch 100, loss[loss=0.1589, simple_loss=0.2449, pruned_loss=0.0364, over 18501.00 frames. ], tot_loss[loss=0.1709, simple_loss=0.2603, pruned_loss=0.04071, over 1476561.67 frames. ], batch size: 43, lr: 4.44e-03, grad_scale: 8.0
2022-12-23 23:35:54,547 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([3.9065, 3.3500, 3.3641, 3.8693, 3.5463, 3.4236, 4.0180, 1.2365],
       device='cuda:2'), covar=tensor([0.0725, 0.0726, 0.0720, 0.0730, 0.1364, 0.1132, 0.0622, 0.5001],
       device='cuda:2'), in_proj_covar=tensor([0.0348, 0.0232, 0.0241, 0.0275, 0.0330, 0.0270, 0.0294, 0.0287],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0001, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-23 23:36:16,419 INFO [train.py:894] (2/4) Epoch 26, batch 150, loss[loss=0.149, simple_loss=0.2422, pruned_loss=0.02793, over 18451.00 frames. ], tot_loss[loss=0.1682, simple_loss=0.2572, pruned_loss=0.03967, over 1972184.36 frames. ], batch size: 50, lr: 4.44e-03, grad_scale: 8.0
2022-12-23 23:36:19,794 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([3.9388, 3.4366, 3.5015, 3.8459, 3.6411, 3.5320, 4.0261, 1.9914],
       device='cuda:2'), covar=tensor([0.0725, 0.0656, 0.0630, 0.0838, 0.1190, 0.1032, 0.0808, 0.4060],
       device='cuda:2'), in_proj_covar=tensor([0.0348, 0.0232, 0.0241, 0.0276, 0.0330, 0.0270, 0.0294, 0.0288],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0001, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-23 23:36:25,612 WARNING [train.py:1060] (2/4) Exclude cut with ID 7859-102521-0017-7548-0 from training. Duration: 24.525
2022-12-23 23:36:38,325 INFO [zipformer.py:660] (2/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=87817.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 23:36:59,986 WARNING [train.py:1060] (2/4) Exclude cut with ID 3699-47246-0007-3408-0_sp0.9 from training. Duration: 20.26675
2022-12-23 23:37:12,728 WARNING [train.py:1060] (2/4) Exclude cut with ID 7859-102521-0017-7548-0_sp0.9 from training. Duration: 27.25
2022-12-23 23:37:19,749 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 1.944e+02 3.316e+02 3.922e+02 4.543e+02 1.035e+03, threshold=7.845e+02, percent-clipped=6.0
2022-12-23 23:37:31,450 INFO [train.py:894] (2/4) Epoch 26, batch 200, loss[loss=0.1854, simple_loss=0.2791, pruned_loss=0.04585, over 18726.00 frames. ], tot_loss[loss=0.1681, simple_loss=0.2573, pruned_loss=0.03947, over 2359015.17 frames. ], batch size: 54, lr: 4.44e-03, grad_scale: 8.0
2022-12-23 23:37:44,455 INFO [zipformer.py:660] (2/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=87861.0, num_to_drop=1, layers_to_drop={3}
2022-12-23 23:38:20,775 INFO [zipformer.py:660] (2/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=87886.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 23:38:24,720 WARNING [train.py:1060] (2/4) Exclude cut with ID 6426-64292-0017-15984-0 from training. Duration: 21.68
2022-12-23 23:38:37,694 WARNING [train.py:1060] (2/4) Exclude cut with ID 4278-13270-0007-59342-0 from training. Duration: 21.6300625
2022-12-23 23:38:46,984 INFO [train.py:894] (2/4) Epoch 26, batch 250, loss[loss=0.181, simple_loss=0.2656, pruned_loss=0.04825, over 18645.00 frames. ], tot_loss[loss=0.167, simple_loss=0.2562, pruned_loss=0.03895, over 2659154.22 frames. ], batch size: 175, lr: 4.44e-03, grad_scale: 8.0
2022-12-23 23:39:01,945 WARNING [train.py:1060] (2/4) Exclude cut with ID 4278-13270-0007-59342-0_sp0.9 from training. Duration: 24.033375
2022-12-23 23:39:32,893 INFO [zipformer.py:660] (2/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=87934.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 23:39:42,092 INFO [zipformer.py:660] (2/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=87940.0, num_to_drop=1, layers_to_drop={3}
2022-12-23 23:39:49,836 INFO [zipformer.py:660] (2/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=87945.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 23:39:50,848 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 2.129e+02 3.146e+02 3.956e+02 4.545e+02 7.882e+02, threshold=7.912e+02, percent-clipped=1.0
2022-12-23 23:39:59,755 WARNING [train.py:1060] (2/4) Exclude cut with ID 4278-13270-0009-59344-0 from training. Duration: 22.905
2022-12-23 23:40:01,208 INFO [train.py:894] (2/4) Epoch 26, batch 300, loss[loss=0.1668, simple_loss=0.2603, pruned_loss=0.03668, over 18592.00 frames. ], tot_loss[loss=0.1674, simple_loss=0.2566, pruned_loss=0.03908, over 2892552.77 frames. ], batch size: 51, lr: 4.44e-03, grad_scale: 8.0
2022-12-23 23:40:01,275 WARNING [train.py:1060] (2/4) Exclude cut with ID 5622-44585-0006-90525-0_sp1.1 from training. Duration: 23.4318125
2022-12-23 23:40:13,557 INFO [zipformer.py:660] (2/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=87961.0, num_to_drop=1, layers_to_drop={2}
2022-12-23 23:41:19,133 INFO [train.py:894] (2/4) Epoch 26, batch 350, loss[loss=0.1413, simple_loss=0.2294, pruned_loss=0.02657, over 18538.00 frames. ], tot_loss[loss=0.1673, simple_loss=0.2565, pruned_loss=0.03906, over 3074555.72 frames. ], batch size: 44, lr: 4.43e-03, grad_scale: 8.0
2022-12-23 23:41:41,120 INFO [zipformer.py:660] (2/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=88018.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 23:41:59,600 WARNING [train.py:1060] (2/4) Exclude cut with ID 4278-13270-0009-59344-0_sp1.1 from training. Duration: 20.82275
2022-12-23 23:42:01,038 WARNING [train.py:1060] (2/4) Exclude cut with ID 4278-13270-0009-59344-0_sp0.9 from training. Duration: 25.45
2022-12-23 23:42:13,263 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.8174, 1.3439, 0.8434, 1.3910, 2.1180, 1.4606, 1.4980, 1.8862],
       device='cuda:2'), covar=tensor([0.1661, 0.2238, 0.2312, 0.1519, 0.1817, 0.1859, 0.1523, 0.1722],
       device='cuda:2'), in_proj_covar=tensor([0.0095, 0.0099, 0.0118, 0.0098, 0.0121, 0.0093, 0.0100, 0.0095],
       device='cuda:2'), out_proj_covar=tensor([0.0003, 0.0003, 0.0004, 0.0003, 0.0004, 0.0003, 0.0003, 0.0003],
       device='cuda:2')
2022-12-23 23:42:23,175 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 2.110e+02 3.234e+02 3.875e+02 5.002e+02 9.913e+02, threshold=7.750e+02, percent-clipped=2.0
2022-12-23 23:42:33,148 INFO [train.py:894] (2/4) Epoch 26, batch 400, loss[loss=0.1646, simple_loss=0.2471, pruned_loss=0.04109, over 18696.00 frames. ], tot_loss[loss=0.1683, simple_loss=0.2577, pruned_loss=0.03941, over 3216592.69 frames. ], batch size: 46, lr: 4.43e-03, grad_scale: 8.0
2022-12-23 23:42:52,323 INFO [zipformer.py:660] (2/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=88066.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 23:42:54,042 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([2.5465, 3.6901, 3.5174, 1.3685, 3.8381, 2.9478, 0.7247, 2.3158],
       device='cuda:2'), covar=tensor([0.2081, 0.1159, 0.1444, 0.3699, 0.0765, 0.0828, 0.4840, 0.1551],
       device='cuda:2'), in_proj_covar=tensor([0.0151, 0.0147, 0.0164, 0.0126, 0.0151, 0.0115, 0.0146, 0.0116],
       device='cuda:2'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0002, 0.0003, 0.0002, 0.0003, 0.0002],
       device='cuda:2')
2022-12-23 23:42:59,568 WARNING [train.py:1060] (2/4) Exclude cut with ID 5622-44585-0006-90525-0 from training. Duration: 25.775
2022-12-23 23:43:22,841 WARNING [train.py:1060] (2/4) Exclude cut with ID 2929-85685-0071-62375-0_sp0.9 from training. Duration: 22.25
2022-12-23 23:43:43,287 INFO [zipformer.py:660] (2/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=88100.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 23:43:47,790 INFO [train.py:894] (2/4) Epoch 26, batch 450, loss[loss=0.1553, simple_loss=0.2406, pruned_loss=0.03496, over 18537.00 frames. ], tot_loss[loss=0.1689, simple_loss=0.2585, pruned_loss=0.03965, over 3326965.25 frames. ], batch size: 47, lr: 4.43e-03, grad_scale: 8.0
2022-12-23 23:43:49,152 WARNING [train.py:1060] (2/4) Exclude cut with ID 3972-170212-0014-23379-0 from training. Duration: 26.205
2022-12-23 23:44:06,041 WARNING [train.py:1060] (2/4) Exclude cut with ID 5239-32139-0047-9341-0_sp0.9 from training. Duration: 30.1555625
2022-12-23 23:44:09,377 INFO [zipformer.py:660] (2/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=88117.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 23:44:10,480 WARNING [train.py:1060] (2/4) Exclude cut with ID 1265-135635-0050-6781-0_sp0.9 from training. Duration: 21.8333125
2022-12-23 23:44:19,473 WARNING [train.py:1060] (2/4) Exclude cut with ID 1914-133440-0024-94914-0_sp1.1 from training. Duration: 20.6545625
2022-12-23 23:44:52,909 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 2.473e+02 3.354e+02 3.991e+02 4.865e+02 8.439e+02, threshold=7.982e+02, percent-clipped=3.0
2022-12-23 23:45:01,877 WARNING [train.py:1060] (2/4) Exclude cut with ID 7395-89880-0045-39920-0_sp0.9 from training. Duration: 20.52225
2022-12-23 23:45:03,299 INFO [train.py:894] (2/4) Epoch 26, batch 500, loss[loss=0.1509, simple_loss=0.2397, pruned_loss=0.03108, over 18694.00 frames. ], tot_loss[loss=0.1697, simple_loss=0.2593, pruned_loss=0.04007, over 3412964.84 frames. ], batch size: 46, lr: 4.43e-03, grad_scale: 8.0
2022-12-23 23:45:15,265 INFO [zipformer.py:660] (2/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=88161.0, num_to_drop=1, layers_to_drop={2}
2022-12-23 23:45:15,338 INFO [zipformer.py:660] (2/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=88161.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 23:45:21,078 WARNING [train.py:1060] (2/4) Exclude cut with ID 3972-170212-0014-23379-0_sp0.9 from training. Duration: 29.1166875
2022-12-23 23:45:21,159 INFO [zipformer.py:660] (2/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=88165.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 23:46:17,802 INFO [train.py:894] (2/4) Epoch 26, batch 550, loss[loss=0.1609, simple_loss=0.2426, pruned_loss=0.0396, over 18524.00 frames. ], tot_loss[loss=0.1701, simple_loss=0.2598, pruned_loss=0.04019, over 3480249.77 frames. ], batch size: 44, lr: 4.43e-03, grad_scale: 8.0
2022-12-23 23:46:22,111 WARNING [train.py:1060] (2/4) Exclude cut with ID 543-133211-0007-59831-0_sp0.9 from training. Duration: 21.388875
2022-12-23 23:46:27,124 INFO [zipformer.py:660] (2/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=88209.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 23:46:56,614 WARNING [train.py:1060] (2/4) Exclude cut with ID 1914-133440-0024-94914-0 from training. Duration: 22.72
2022-12-23 23:46:58,069 WARNING [train.py:1060] (2/4) Exclude cut with ID 1914-133440-0031-94921-0_sp0.9 from training. Duration: 22.7444375
2022-12-23 23:47:13,858 INFO [zipformer.py:660] (2/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=88240.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 23:47:18,381 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.3101, 1.5264, 1.2304, 1.7466, 1.7473, 1.3799, 1.0677, 1.1943],
       device='cuda:2'), covar=tensor([0.2155, 0.2067, 0.1917, 0.1273, 0.1298, 0.1286, 0.2460, 0.1754],
       device='cuda:2'), in_proj_covar=tensor([0.0248, 0.0228, 0.0219, 0.0202, 0.0261, 0.0198, 0.0226, 0.0202],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-23 23:47:21,461 INFO [zipformer.py:660] (2/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=88245.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 23:47:22,437 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 2.456e+02 3.440e+02 4.053e+02 4.929e+02 1.052e+03, threshold=8.106e+02, percent-clipped=3.0
2022-12-23 23:47:33,654 INFO [train.py:894] (2/4) Epoch 26, batch 600, loss[loss=0.1626, simple_loss=0.2411, pruned_loss=0.04202, over 18623.00 frames. ], tot_loss[loss=0.1704, simple_loss=0.26, pruned_loss=0.04041, over 3531544.96 frames. ], batch size: 41, lr: 4.43e-03, grad_scale: 8.0
2022-12-23 23:47:40,131 INFO [scaling.py:679] (2/4) Whitening: num_groups=1, num_channels=384, metric=3.53 vs. limit=5.0
2022-12-23 23:47:42,310 WARNING [train.py:1060] (2/4) Exclude cut with ID 4133-6541-0027-40495-0_sp1.1 from training. Duration: 0.9681875
2022-12-23 23:47:45,219 WARNING [train.py:1060] (2/4) Exclude cut with ID 6330-62851-0022-91297-0_sp0.9 from training. Duration: 22.3166875
2022-12-23 23:47:45,438 INFO [zipformer.py:660] (2/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=88261.0, num_to_drop=1, layers_to_drop={1}
2022-12-23 23:47:49,208 WARNING [train.py:1060] (2/4) Exclude cut with ID 543-133212-0015-59917-0_sp0.9 from training. Duration: 21.8166875
2022-12-23 23:48:26,428 INFO [zipformer.py:660] (2/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=88288.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 23:48:31,001 INFO [zipformer.py:660] (2/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=88291.0, num_to_drop=1, layers_to_drop={1}
2022-12-23 23:48:34,246 INFO [zipformer.py:660] (2/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=88293.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 23:48:48,665 INFO [train.py:894] (2/4) Epoch 26, batch 650, loss[loss=0.1663, simple_loss=0.2686, pruned_loss=0.03194, over 18486.00 frames. ], tot_loss[loss=0.1705, simple_loss=0.2602, pruned_loss=0.04039, over 3572756.89 frames. ], batch size: 54, lr: 4.43e-03, grad_scale: 8.0
2022-12-23 23:48:57,889 INFO [zipformer.py:660] (2/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=88309.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 23:49:29,178 WARNING [train.py:1060] (2/4) Exclude cut with ID 4957-30119-0041-23990-0_sp0.9 from training. Duration: 20.22775
2022-12-23 23:49:53,361 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 2.132e+02 3.390e+02 4.009e+02 4.862e+02 9.909e+02, threshold=8.017e+02, percent-clipped=1.0
2022-12-23 23:50:02,386 INFO [zipformer.py:660] (2/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=88352.0, num_to_drop=1, layers_to_drop={3}
2022-12-23 23:50:03,575 INFO [train.py:894] (2/4) Epoch 26, batch 700, loss[loss=0.1779, simple_loss=0.2803, pruned_loss=0.03778, over 18681.00 frames. ], tot_loss[loss=0.1702, simple_loss=0.2599, pruned_loss=0.04028, over 3603194.05 frames. ], batch size: 60, lr: 4.43e-03, grad_scale: 8.0
2022-12-23 23:50:14,928 WARNING [train.py:1060] (2/4) Exclude cut with ID 5239-32139-0047-9341-0_sp1.1 from training. Duration: 24.67275
2022-12-23 23:50:16,803 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.5335, 1.8617, 1.5006, 2.1136, 2.3263, 1.5079, 1.4264, 1.2574],
       device='cuda:2'), covar=tensor([0.1914, 0.1778, 0.1636, 0.1022, 0.1269, 0.1154, 0.2162, 0.1605],
       device='cuda:2'), in_proj_covar=tensor([0.0248, 0.0228, 0.0220, 0.0202, 0.0262, 0.0198, 0.0227, 0.0202],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-23 23:50:40,220 WARNING [train.py:1060] (2/4) Exclude cut with ID 3082-165428-0081-50734-0_sp0.9 from training. Duration: 21.8055625
2022-12-23 23:50:49,229 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.6203, 1.5049, 1.4740, 1.7539, 1.7326, 3.2081, 1.5129, 1.6679],
       device='cuda:2'), covar=tensor([0.0873, 0.1855, 0.1161, 0.0979, 0.1471, 0.0263, 0.1422, 0.1540],
       device='cuda:2'), in_proj_covar=tensor([0.0073, 0.0083, 0.0072, 0.0074, 0.0092, 0.0077, 0.0085, 0.0077],
       device='cuda:2'), out_proj_covar=tensor([0.0004, 0.0004, 0.0004, 0.0004, 0.0004, 0.0004, 0.0004, 0.0004],
       device='cuda:2')
2022-12-23 23:51:18,935 INFO [train.py:894] (2/4) Epoch 26, batch 750, loss[loss=0.1697, simple_loss=0.2709, pruned_loss=0.03423, over 18522.00 frames. ], tot_loss[loss=0.1698, simple_loss=0.2593, pruned_loss=0.04012, over 3628787.01 frames. ], batch size: 52, lr: 4.42e-03, grad_scale: 8.0
2022-12-23 23:51:20,421 WARNING [train.py:1060] (2/4) Exclude cut with ID 3340-169293-0054-76830-0_sp0.9 from training. Duration: 22.6666875
2022-12-23 23:52:22,598 WARNING [train.py:1060] (2/4) Exclude cut with ID 2411-132532-0017-82279-0_sp1.1 from training. Duration: 0.9681875
2022-12-23 23:52:23,982 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 2.008e+02 3.060e+02 3.630e+02 4.640e+02 8.422e+02, threshold=7.260e+02, percent-clipped=1.0
2022-12-23 23:52:34,203 INFO [train.py:894] (2/4) Epoch 26, batch 800, loss[loss=0.1792, simple_loss=0.271, pruned_loss=0.04372, over 18725.00 frames. ], tot_loss[loss=0.17, simple_loss=0.2595, pruned_loss=0.0403, over 3647349.49 frames. ], batch size: 52, lr: 4.42e-03, grad_scale: 8.0
2022-12-23 23:52:38,803 INFO [zipformer.py:660] (2/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=88456.0, num_to_drop=0, layers_to_drop=set()
2022-12-23 23:52:45,594 WARNING [train.py:1060] (2/4) Exclude cut with ID 6330-62850-0007-91323-0 from training. Duration: 22.485
2022-12-23 23:53:11,425 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.0260, 1.0350, 0.9336, 1.1991, 1.2894, 1.1740, 1.1150, 0.9891],
       device='cuda:2'), covar=tensor([0.0330, 0.0247, 0.0598, 0.0233, 0.0247, 0.0442, 0.0332, 0.0330],
       device='cuda:2'), in_proj_covar=tensor([0.0095, 0.0127, 0.0153, 0.0122, 0.0117, 0.0121, 0.0100, 0.0127],
       device='cuda:2'), out_proj_covar=tensor([7.5071e-05, 1.0013e-04, 1.2512e-04, 9.6460e-05, 9.3536e-05, 9.2986e-05,
        7.8119e-05, 9.9969e-05], device='cuda:2')
2022-12-23 23:53:26,498 WARNING [train.py:1060] (2/4) Exclude cut with ID 3972-170212-0014-23379-0_sp1.1 from training. Duration: 23.82275
2022-12-23 23:53:41,070 WARNING [train.py:1060] (2/4) Exclude cut with ID 4860-13185-0032-76709-0 from training. Duration: 20.77
2022-12-23 23:53:48,536 INFO [train.py:894] (2/4) Epoch 26, batch 850, loss[loss=0.1748, simple_loss=0.2596, pruned_loss=0.04497, over 18435.00 frames. ], tot_loss[loss=0.1708, simple_loss=0.2601, pruned_loss=0.04078, over 3663151.07 frames. ], batch size: 48, lr: 4.42e-03, grad_scale: 8.0
2022-12-23 23:53:48,604 WARNING [train.py:1060] (2/4) Exclude cut with ID 6426-64292-0017-15984-0_sp0.9 from training. Duration: 24.088875
2022-12-23 23:54:17,186 WARNING [train.py:1060] (2/4) Exclude cut with ID 6330-62850-0007-91323-0_sp1.1 from training. Duration: 20.4409375
2022-12-23 23:54:53,870 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 1.940e+02 3.168e+02 3.813e+02 4.877e+02 1.035e+03, threshold=7.626e+02, percent-clipped=7.0
2022-12-23 23:55:03,865 INFO [train.py:894] (2/4) Epoch 26, batch 900, loss[loss=0.1671, simple_loss=0.2584, pruned_loss=0.03793, over 18672.00 frames. ], tot_loss[loss=0.1704, simple_loss=0.2596, pruned_loss=0.04064, over 3674447.20 frames. ], batch size: 60, lr: 4.42e-03, grad_scale: 8.0
2022-12-23 23:55:31,858 WARNING [train.py:1060] (2/4) Exclude cut with ID 2929-85685-0062-62366-0_sp0.9 from training. Duration: 22.511125
2022-12-23 23:55:33,349 WARNING [train.py:1060] (2/4) Exclude cut with ID 7395-89880-0031-39906-0 from training. Duration: 20.675
2022-12-23 23:56:21,084 INFO [train.py:894] (2/4) Epoch 26, batch 950, loss[loss=0.1338, simple_loss=0.2191, pruned_loss=0.02425, over 18557.00 frames. ], tot_loss[loss=0.1696, simple_loss=0.2589, pruned_loss=0.0402, over 3682370.97 frames. ], batch size: 41, lr: 4.42e-03, grad_scale: 8.0
2022-12-23 23:56:50,663 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.8162, 1.8055, 1.9451, 1.8142, 1.6034, 3.1170, 1.8898, 2.1649],
       device='cuda:2'), covar=tensor([0.2647, 0.1869, 0.1530, 0.1822, 0.1200, 0.0230, 0.1809, 0.0800],
       device='cuda:2'), in_proj_covar=tensor([0.0134, 0.0119, 0.0126, 0.0122, 0.0106, 0.0097, 0.0091, 0.0091],
       device='cuda:2'), out_proj_covar=tensor([0.0005, 0.0005, 0.0005, 0.0005, 0.0004, 0.0003, 0.0004, 0.0004],
       device='cuda:2')
2022-12-23 23:57:13,732 WARNING [train.py:1060] (2/4) Exclude cut with ID 6330-62850-0007-91323-0_sp0.9 from training. Duration: 24.9833125
2022-12-23 23:57:16,816 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.8472, 1.3004, 0.9632, 1.3789, 2.1065, 1.3668, 1.5281, 1.7795],
       device='cuda:2'), covar=tensor([0.1663, 0.2148, 0.2154, 0.1555, 0.1864, 0.1831, 0.1481, 0.1707],
       device='cuda:2'), in_proj_covar=tensor([0.0094, 0.0098, 0.0116, 0.0096, 0.0119, 0.0092, 0.0098, 0.0094],
       device='cuda:2'), out_proj_covar=tensor([0.0003, 0.0003, 0.0004, 0.0003, 0.0004, 0.0003, 0.0003, 0.0003],
       device='cuda:2')
2022-12-23 23:57:22,984 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([2.0868, 2.1029, 1.6357, 2.4062, 2.2201, 2.0040, 2.8995, 2.1536],
       device='cuda:2'), covar=tensor([0.0999, 0.1784, 0.2850, 0.1723, 0.1904, 0.0978, 0.0862, 0.1421],
       device='cuda:2'), in_proj_covar=tensor([0.0182, 0.0216, 0.0258, 0.0295, 0.0243, 0.0196, 0.0209, 0.0210],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-23 23:57:25,437 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 2.452e+02 3.318e+02 3.992e+02 4.999e+02 2.515e+03, threshold=7.985e+02, percent-clipped=6.0
2022-12-23 23:57:27,323 INFO [zipformer.py:660] (2/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=88647.0, num_to_drop=1, layers_to_drop={3}
2022-12-23 23:57:35,822 INFO [train.py:894] (2/4) Epoch 26, batch 1000, loss[loss=0.1499, simple_loss=0.2464, pruned_loss=0.0267, over 18626.00 frames. ], tot_loss[loss=0.169, simple_loss=0.258, pruned_loss=0.04001, over 3689490.25 frames. ], batch size: 53, lr: 4.42e-03, grad_scale: 8.0
2022-12-23 23:57:45,889 WARNING [train.py:1060] (2/4) Exclude cut with ID 5239-32139-0047-9341-0 from training. Duration: 27.14
2022-12-23 23:58:00,310 WARNING [train.py:1060] (2/4) Exclude cut with ID 2929-85685-0044-62348-0 from training. Duration: 22.44
2022-12-23 23:58:49,453 INFO [train.py:894] (2/4) Epoch 26, batch 1050, loss[loss=0.192, simple_loss=0.2862, pruned_loss=0.04891, over 18676.00 frames. ], tot_loss[loss=0.1694, simple_loss=0.2587, pruned_loss=0.04006, over 3694646.40 frames. ], batch size: 174, lr: 4.42e-03, grad_scale: 8.0
2022-12-23 23:59:16,372 WARNING [train.py:1060] (2/4) Exclude cut with ID 2929-85685-0060-62364-0_sp0.9 from training. Duration: 21.361125
2022-12-23 23:59:21,304 WARNING [train.py:1060] (2/4) Exclude cut with ID 2929-85685-0079-62383-0_sp1.1 from training. Duration: 27.0318125
2022-12-23 23:59:26,627 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([2.5488, 4.0982, 3.8227, 1.8548, 4.2384, 3.0714, 0.6076, 2.5684],
       device='cuda:2'), covar=tensor([0.2173, 0.0999, 0.1355, 0.3253, 0.0641, 0.0876, 0.5089, 0.1478],
       device='cuda:2'), in_proj_covar=tensor([0.0149, 0.0145, 0.0161, 0.0124, 0.0148, 0.0114, 0.0144, 0.0114],
       device='cuda:2'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0002, 0.0003, 0.0002, 0.0003, 0.0002],
       device='cuda:2')
2022-12-23 23:59:32,463 WARNING [train.py:1060] (2/4) Exclude cut with ID 5622-44585-0006-90525-0_sp0.9 from training. Duration: 28.638875
2022-12-23 23:59:47,034 WARNING [train.py:1060] (2/4) Exclude cut with ID 3340-169293-0054-76830-0 from training. Duration: 20.4
2022-12-23 23:59:53,977 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 1.848e+02 3.398e+02 4.016e+02 4.877e+02 1.233e+03, threshold=8.032e+02, percent-clipped=4.0
2022-12-24 00:00:04,371 INFO [train.py:894] (2/4) Epoch 26, batch 1100, loss[loss=0.1416, simple_loss=0.2258, pruned_loss=0.02868, over 18609.00 frames. ], tot_loss[loss=0.1692, simple_loss=0.2585, pruned_loss=0.03996, over 3699695.60 frames. ], batch size: 45, lr: 4.42e-03, grad_scale: 8.0
2022-12-24 00:00:09,267 INFO [zipformer.py:660] (2/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=88756.0, num_to_drop=0, layers_to_drop=set()
2022-12-24 00:00:20,098 WARNING [train.py:1060] (2/4) Exclude cut with ID 2929-85685-0071-62375-0 from training. Duration: 20.025
2022-12-24 00:00:20,107 WARNING [train.py:1060] (2/4) Exclude cut with ID 2364-131735-0112-64612-0_sp0.9 from training. Duration: 20.488875
2022-12-24 00:00:24,818 WARNING [train.py:1060] (2/4) Exclude cut with ID 2929-85685-0079-62383-0 from training. Duration: 29.735
2022-12-24 00:01:20,081 INFO [train.py:894] (2/4) Epoch 26, batch 1150, loss[loss=0.1788, simple_loss=0.2612, pruned_loss=0.04822, over 18565.00 frames. ], tot_loss[loss=0.17, simple_loss=0.2594, pruned_loss=0.04029, over 3703692.86 frames. ], batch size: 49, lr: 4.41e-03, grad_scale: 8.0
2022-12-24 00:01:21,761 INFO [zipformer.py:660] (2/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=88804.0, num_to_drop=0, layers_to_drop=set()
2022-12-24 00:01:48,967 WARNING [train.py:1060] (2/4) Exclude cut with ID 7276-92427-0014-12983-0_sp0.9 from training. Duration: 21.3055625
2022-12-24 00:01:50,515 WARNING [train.py:1060] (2/4) Exclude cut with ID 1025-75365-0008-79168-0_sp0.9 from training. Duration: 22.0666875
2022-12-24 00:02:24,918 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 2.168e+02 3.282e+02 3.913e+02 4.660e+02 1.073e+03, threshold=7.826e+02, percent-clipped=1.0
2022-12-24 00:02:35,229 INFO [train.py:894] (2/4) Epoch 26, batch 1200, loss[loss=0.1569, simple_loss=0.2423, pruned_loss=0.03572, over 18407.00 frames. ], tot_loss[loss=0.1695, simple_loss=0.2592, pruned_loss=0.03994, over 3707125.37 frames. ], batch size: 46, lr: 4.41e-03, grad_scale: 8.0
2022-12-24 00:02:45,718 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([2.7129, 2.4979, 2.0648, 0.9336, 1.9728, 2.0615, 1.8737, 2.1783],
       device='cuda:2'), covar=tensor([0.0731, 0.0562, 0.1338, 0.1978, 0.1398, 0.1724, 0.1761, 0.0878],
       device='cuda:2'), in_proj_covar=tensor([0.0174, 0.0188, 0.0206, 0.0191, 0.0209, 0.0202, 0.0216, 0.0202],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-24 00:03:09,904 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([2.8813, 5.3620, 4.9133, 2.7740, 5.5094, 4.0921, 0.6890, 3.6350],
       device='cuda:2'), covar=tensor([0.2223, 0.0916, 0.1419, 0.3009, 0.0547, 0.0737, 0.5278, 0.1182],
       device='cuda:2'), in_proj_covar=tensor([0.0148, 0.0143, 0.0159, 0.0123, 0.0147, 0.0114, 0.0143, 0.0113],
       device='cuda:2'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0002, 0.0003, 0.0002, 0.0003, 0.0002],
       device='cuda:2')
2022-12-24 00:03:15,871 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.5505, 1.3318, 1.4849, 1.4401, 0.9455, 3.0660, 1.1764, 1.6606],
       device='cuda:2'), covar=tensor([0.3219, 0.2256, 0.2100, 0.2212, 0.1687, 0.0222, 0.1696, 0.0907],
       device='cuda:2'), in_proj_covar=tensor([0.0134, 0.0119, 0.0126, 0.0123, 0.0106, 0.0097, 0.0091, 0.0091],
       device='cuda:2'), out_proj_covar=tensor([0.0005, 0.0005, 0.0005, 0.0005, 0.0004, 0.0003, 0.0004, 0.0004],
       device='cuda:2')
2022-12-24 00:03:37,299 WARNING [train.py:1060] (2/4) Exclude cut with ID 2929-85685-0062-62366-0 from training. Duration: 20.26
2022-12-24 00:03:50,599 INFO [train.py:894] (2/4) Epoch 26, batch 1250, loss[loss=0.1843, simple_loss=0.284, pruned_loss=0.04229, over 18581.00 frames. ], tot_loss[loss=0.1691, simple_loss=0.2588, pruned_loss=0.03971, over 3708296.25 frames. ], batch size: 56, lr: 4.41e-03, grad_scale: 16.0
2022-12-24 00:03:50,651 WARNING [train.py:1060] (2/4) Exclude cut with ID 5239-32139-0030-9324-0_sp0.9 from training. Duration: 21.3444375
2022-12-24 00:04:00,801 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([0.6948, 2.0395, 2.3371, 1.1431, 1.5955, 2.5048, 2.2293, 1.9016],
       device='cuda:2'), covar=tensor([0.0814, 0.0337, 0.0304, 0.0434, 0.0360, 0.0416, 0.0261, 0.0706],
       device='cuda:2'), in_proj_covar=tensor([0.0152, 0.0174, 0.0131, 0.0142, 0.0149, 0.0145, 0.0167, 0.0177],
       device='cuda:2'), out_proj_covar=tensor([1.1463e-04, 1.3107e-04, 9.6821e-05, 1.0392e-04, 1.0969e-04, 1.0928e-04,
        1.2657e-04, 1.3361e-04], device='cuda:2')
2022-12-24 00:04:47,165 WARNING [train.py:1060] (2/4) Exclude cut with ID 497-129325-0061-62254-0_sp1.1 from training. Duration: 0.97725
2022-12-24 00:04:54,588 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 2.240e+02 3.350e+02 3.895e+02 4.856e+02 7.828e+02, threshold=7.791e+02, percent-clipped=1.0
2022-12-24 00:04:55,136 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.3356, 1.4920, 1.2834, 1.7789, 1.6859, 1.4630, 0.9893, 1.2474],
       device='cuda:2'), covar=tensor([0.2045, 0.2025, 0.1779, 0.1201, 0.1282, 0.1123, 0.2399, 0.1650],
       device='cuda:2'), in_proj_covar=tensor([0.0249, 0.0230, 0.0220, 0.0203, 0.0263, 0.0199, 0.0227, 0.0203],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-24 00:04:56,396 INFO [zipformer.py:660] (2/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=88947.0, num_to_drop=1, layers_to_drop={1}
2022-12-24 00:05:04,822 INFO [train.py:894] (2/4) Epoch 26, batch 1300, loss[loss=0.1543, simple_loss=0.2395, pruned_loss=0.03452, over 18669.00 frames. ], tot_loss[loss=0.1695, simple_loss=0.2592, pruned_loss=0.03992, over 3708583.15 frames. ], batch size: 48, lr: 4.41e-03, grad_scale: 16.0
2022-12-24 00:05:27,582 WARNING [train.py:1060] (2/4) Exclude cut with ID 7395-89880-0031-39906-0_sp0.9 from training. Duration: 22.97225
2022-12-24 00:05:59,208 WARNING [train.py:1060] (2/4) Exclude cut with ID 7395-89880-0047-39922-0_sp0.9 from training. Duration: 21.97775
2022-12-24 00:06:08,029 INFO [zipformer.py:660] (2/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=88995.0, num_to_drop=1, layers_to_drop={0}
2022-12-24 00:06:11,926 WARNING [train.py:1060] (2/4) Exclude cut with ID 1112-1043-0006-89194-0_sp0.9 from training. Duration: 21.8333125
2022-12-24 00:06:19,454 INFO [train.py:894] (2/4) Epoch 26, batch 1350, loss[loss=0.1726, simple_loss=0.2691, pruned_loss=0.03804, over 18530.00 frames. ], tot_loss[loss=0.1694, simple_loss=0.2591, pruned_loss=0.03984, over 3709495.69 frames. ], batch size: 55, lr: 4.41e-03, grad_scale: 16.0
2022-12-24 00:06:22,564 WARNING [train.py:1060] (2/4) Exclude cut with ID 1914-133440-0031-94921-0 from training. Duration: 20.47
2022-12-24 00:06:53,217 INFO [scaling.py:679] (2/4) Whitening: num_groups=8, num_channels=96, metric=1.62 vs. limit=2.0
2022-12-24 00:06:56,709 INFO [zipformer.py:660] (2/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=89027.0, num_to_drop=0, layers_to_drop=set()
2022-12-24 00:07:00,044 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([2.3290, 1.6040, 2.0233, 1.9349, 2.2700, 2.3371, 2.1246, 2.0033],
       device='cuda:2'), covar=tensor([0.2296, 0.3407, 0.2641, 0.2916, 0.2123, 0.1038, 0.3293, 0.1377],
       device='cuda:2'), in_proj_covar=tensor([0.0273, 0.0300, 0.0287, 0.0324, 0.0317, 0.0258, 0.0353, 0.0248],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-24 00:07:24,339 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 1.964e+02 3.078e+02 3.769e+02 4.964e+02 1.403e+03, threshold=7.538e+02, percent-clipped=3.0
2022-12-24 00:07:28,718 WARNING [train.py:1060] (2/4) Exclude cut with ID 7395-89880-0037-39912-0_sp0.9 from training. Duration: 20.67225
2022-12-24 00:07:34,452 INFO [train.py:894] (2/4) Epoch 26, batch 1400, loss[loss=0.1633, simple_loss=0.2592, pruned_loss=0.0337, over 18519.00 frames. ], tot_loss[loss=0.1687, simple_loss=0.2584, pruned_loss=0.03946, over 3710180.77 frames. ], batch size: 52, lr: 4.41e-03, grad_scale: 16.0
2022-12-24 00:07:48,691 WARNING [train.py:1060] (2/4) Exclude cut with ID 1914-133440-0024-94914-0_sp0.9 from training. Duration: 25.2444375
2022-12-24 00:08:12,334 WARNING [train.py:1060] (2/4) Exclude cut with ID 3340-169293-0021-76797-0_sp0.9 from training. Duration: 21.1445
2022-12-24 00:08:27,505 INFO [zipformer.py:660] (2/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=89088.0, num_to_drop=0, layers_to_drop=set()
2022-12-24 00:08:48,469 INFO [train.py:894] (2/4) Epoch 26, batch 1450, loss[loss=0.1494, simple_loss=0.2318, pruned_loss=0.03347, over 18489.00 frames. ], tot_loss[loss=0.1685, simple_loss=0.2583, pruned_loss=0.03931, over 3710244.39 frames. ], batch size: 43, lr: 4.41e-03, grad_scale: 16.0
2022-12-24 00:09:27,557 WARNING [train.py:1060] (2/4) Exclude cut with ID 2929-85685-0079-62383-0_sp0.9 from training. Duration: 33.038875
2022-12-24 00:09:53,309 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 2.527e+02 3.198e+02 3.701e+02 4.884e+02 9.983e+02, threshold=7.403e+02, percent-clipped=2.0
2022-12-24 00:10:03,997 INFO [train.py:894] (2/4) Epoch 26, batch 1500, loss[loss=0.1545, simple_loss=0.2386, pruned_loss=0.03523, over 18492.00 frames. ], tot_loss[loss=0.1686, simple_loss=0.2585, pruned_loss=0.03934, over 3710944.36 frames. ], batch size: 43, lr: 4.41e-03, grad_scale: 16.0
2022-12-24 00:10:04,708 WARNING [train.py:1060] (2/4) Exclude cut with ID 6426-64291-0000-16059-0_sp0.9 from training. Duration: 20.0944375
2022-12-24 00:10:19,266 WARNING [train.py:1060] (2/4) Exclude cut with ID 2929-85685-0044-62348-0_sp1.1 from training. Duration: 20.4
2022-12-24 00:10:25,119 WARNING [train.py:1060] (2/4) Exclude cut with ID 6330-62851-0022-91297-0 from training. Duration: 20.085
2022-12-24 00:10:28,754 INFO [zipformer.py:660] (2/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=89169.0, num_to_drop=0, layers_to_drop=set()
2022-12-24 00:10:37,329 WARNING [train.py:1060] (2/4) Exclude cut with ID 4860-13185-0032-76709-0_sp0.9 from training. Duration: 23.07775
2022-12-24 00:11:18,030 INFO [train.py:894] (2/4) Epoch 26, batch 1550, loss[loss=0.1686, simple_loss=0.27, pruned_loss=0.03357, over 18679.00 frames. ], tot_loss[loss=0.1683, simple_loss=0.258, pruned_loss=0.03927, over 3711476.42 frames. ], batch size: 62, lr: 4.40e-03, grad_scale: 16.0
2022-12-24 00:11:26,719 WARNING [train.py:1060] (2/4) Exclude cut with ID 2929-85685-0044-62348-0_sp0.9 from training. Duration: 24.9333125
2022-12-24 00:11:32,865 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.9508, 1.9615, 2.2591, 1.1766, 2.2267, 2.3493, 1.5410, 2.6081],
       device='cuda:2'), covar=tensor([0.1265, 0.1969, 0.1327, 0.2180, 0.0835, 0.1101, 0.2530, 0.0575],
       device='cuda:2'), in_proj_covar=tensor([0.0195, 0.0213, 0.0206, 0.0193, 0.0171, 0.0215, 0.0213, 0.0191],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-24 00:11:59,405 INFO [zipformer.py:660] (2/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=89230.0, num_to_drop=0, layers_to_drop=set()
2022-12-24 00:12:13,864 WARNING [train.py:1060] (2/4) Exclude cut with ID 5118-111612-0016-124680-0_sp0.9 from training. Duration: 20.388875
2022-12-24 00:12:18,024 WARNING [train.py:1060] (2/4) Exclude cut with ID 432-122774-0017-62487-0_sp1.1 from training. Duration: 20.3590625
2022-12-24 00:12:22,211 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 2.063e+02 3.106e+02 4.096e+02 5.017e+02 1.539e+03, threshold=8.192e+02, percent-clipped=3.0
2022-12-24 00:12:32,777 INFO [train.py:894] (2/4) Epoch 26, batch 1600, loss[loss=0.1667, simple_loss=0.26, pruned_loss=0.03673, over 18572.00 frames. ], tot_loss[loss=0.1687, simple_loss=0.2584, pruned_loss=0.03956, over 3711428.71 frames. ], batch size: 57, lr: 4.40e-03, grad_scale: 16.0
2022-12-24 00:12:43,368 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([2.6579, 2.7020, 1.9394, 3.3946, 3.0001, 2.6129, 3.8213, 2.6292],
       device='cuda:2'), covar=tensor([0.0817, 0.1833, 0.2851, 0.1668, 0.1663, 0.0863, 0.0770, 0.1257],
       device='cuda:2'), in_proj_covar=tensor([0.0182, 0.0215, 0.0256, 0.0292, 0.0242, 0.0195, 0.0208, 0.0209],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-24 00:12:57,749 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([2.4428, 1.8859, 2.0617, 2.1196, 2.2826, 2.3606, 2.3579, 1.9144],
       device='cuda:2'), covar=tensor([0.2318, 0.3540, 0.2697, 0.3133, 0.2191, 0.1022, 0.3611, 0.1406],
       device='cuda:2'), in_proj_covar=tensor([0.0271, 0.0298, 0.0286, 0.0322, 0.0315, 0.0257, 0.0352, 0.0246],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-24 00:13:25,728 WARNING [train.py:1060] (2/4) Exclude cut with ID 3557-8342-0013-54691-0_sp1.1 from training. Duration: 0.836375
2022-12-24 00:13:47,651 INFO [train.py:894] (2/4) Epoch 26, batch 1650, loss[loss=0.2006, simple_loss=0.2921, pruned_loss=0.05458, over 18386.00 frames. ], tot_loss[loss=0.1695, simple_loss=0.2593, pruned_loss=0.03986, over 3711602.45 frames. ], batch size: 53, lr: 4.40e-03, grad_scale: 16.0
2022-12-24 00:14:07,852 WARNING [train.py:1060] (2/4) Exclude cut with ID 8565-290391-0049-67394-0_sp0.9 from training. Duration: 21.3166875
2022-12-24 00:14:37,529 WARNING [train.py:1060] (2/4) Exclude cut with ID 6533-399-0029-104863-0_sp0.9 from training. Duration: 22.1055625
2022-12-24 00:14:47,763 WARNING [train.py:1060] (2/4) Exclude cut with ID 7699-105389-0094-26379-0_sp1.1 from training. Duration: 21.77725
2022-12-24 00:14:50,292 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 2.450e+02 3.582e+02 4.478e+02 5.269e+02 1.215e+03, threshold=8.956e+02, percent-clipped=2.0
2022-12-24 00:15:01,876 INFO [train.py:894] (2/4) Epoch 26, batch 1700, loss[loss=0.1928, simple_loss=0.2806, pruned_loss=0.0525, over 18668.00 frames. ], tot_loss[loss=0.1708, simple_loss=0.2599, pruned_loss=0.04086, over 3713206.00 frames. ], batch size: 48, lr: 4.40e-03, grad_scale: 16.0
2022-12-24 00:15:10,189 WARNING [train.py:1060] (2/4) Exclude cut with ID 7255-291500-0005-134304-0_sp0.9 from training. Duration: 27.8166875
2022-12-24 00:15:33,026 INFO [zipformer.py:660] (2/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=89374.0, num_to_drop=0, layers_to_drop=set()
2022-12-24 00:15:35,465 WARNING [train.py:1060] (2/4) Exclude cut with ID 7357-94126-0021-15852-0_sp1.1 from training. Duration: 22.5090625
2022-12-24 00:15:41,305 WARNING [train.py:1060] (2/4) Exclude cut with ID 7255-291500-0005-134304-0 from training. Duration: 25.035
2022-12-24 00:15:45,769 INFO [zipformer.py:660] (2/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=89383.0, num_to_drop=0, layers_to_drop=set()
2022-12-24 00:15:59,547 WARNING [train.py:1060] (2/4) Exclude cut with ID 774-127930-0014-10412-0_sp1.1 from training. Duration: 0.95
2022-12-24 00:16:17,128 INFO [train.py:894] (2/4) Epoch 26, batch 1750, loss[loss=0.2426, simple_loss=0.3075, pruned_loss=0.08887, over 18690.00 frames. ], tot_loss[loss=0.1726, simple_loss=0.2607, pruned_loss=0.04221, over 3714373.13 frames. ], batch size: 62, lr: 4.40e-03, grad_scale: 16.0
2022-12-24 00:16:19,014 WARNING [train.py:1060] (2/4) Exclude cut with ID 3033-130750-0096-55598-0_sp0.9 from training. Duration: 0.92225
2022-12-24 00:16:45,558 WARNING [train.py:1060] (2/4) Exclude cut with ID 4511-76322-0006-80011-0 from training. Duration: 21.97
2022-12-24 00:16:50,104 INFO [zipformer.py:660] (2/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=89425.0, num_to_drop=0, layers_to_drop=set()
2022-12-24 00:17:04,655 INFO [zipformer.py:660] (2/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=89435.0, num_to_drop=0, layers_to_drop=set()
2022-12-24 00:17:05,837 WARNING [train.py:1060] (2/4) Exclude cut with ID 7492-105653-0055-62765-0_sp0.9 from training. Duration: 21.97225
2022-12-24 00:17:07,473 WARNING [train.py:1060] (2/4) Exclude cut with ID 453-131332-0000-47844-0_sp0.9 from training. Duration: 25.3333125
2022-12-24 00:17:10,626 INFO [zipformer.py:660] (2/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=89439.0, num_to_drop=1, layers_to_drop={0}
2022-12-24 00:17:16,239 WARNING [train.py:1060] (2/4) Exclude cut with ID 5172-29468-0015-19128-0_sp0.9 from training. Duration: 21.5055625
2022-12-24 00:17:21,236 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 2.550e+02 4.037e+02 4.930e+02 6.113e+02 1.176e+03, threshold=9.860e+02, percent-clipped=6.0
2022-12-24 00:17:27,637 WARNING [train.py:1060] (2/4) Exclude cut with ID 453-131332-0000-47844-0_sp1.1 from training. Duration: 20.72725
2022-12-24 00:17:32,061 INFO [train.py:894] (2/4) Epoch 26, batch 1800, loss[loss=0.1579, simple_loss=0.2388, pruned_loss=0.03846, over 18541.00 frames. ], tot_loss[loss=0.1745, simple_loss=0.2614, pruned_loss=0.0438, over 3713915.48 frames. ], batch size: 47, lr: 4.40e-03, grad_scale: 16.0
2022-12-24 00:17:59,440 WARNING [train.py:1060] (2/4) Exclude cut with ID 8631-249866-0030-130156-0_sp0.9 from training. Duration: 26.32775
2022-12-24 00:18:21,763 INFO [zipformer.py:660] (2/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=89486.0, num_to_drop=0, layers_to_drop=set()
2022-12-24 00:18:32,155 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.5206, 1.4044, 1.4713, 1.3652, 0.8173, 2.2615, 0.7639, 1.3282],
       device='cuda:2'), covar=tensor([0.3270, 0.2223, 0.2130, 0.2276, 0.1650, 0.0366, 0.1882, 0.0976],
       device='cuda:2'), in_proj_covar=tensor([0.0133, 0.0118, 0.0126, 0.0123, 0.0106, 0.0097, 0.0091, 0.0091],
       device='cuda:2'), out_proj_covar=tensor([0.0005, 0.0005, 0.0005, 0.0005, 0.0004, 0.0003, 0.0004, 0.0004],
       device='cuda:2')
2022-12-24 00:18:33,764 WARNING [train.py:1060] (2/4) Exclude cut with ID 3867-173237-0077-144769-0 from training. Duration: 20.025
2022-12-24 00:18:38,759 WARNING [train.py:1060] (2/4) Exclude cut with ID 6709-74022-0004-86860-0_sp1.1 from training. Duration: 0.9409375
2022-12-24 00:18:38,769 WARNING [train.py:1060] (2/4) Exclude cut with ID 4757-1811-0023-62229-0_sp0.9 from training. Duration: 21.37775
2022-12-24 00:18:43,285 INFO [zipformer.py:660] (2/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=89500.0, num_to_drop=1, layers_to_drop={1}
2022-12-24 00:18:46,963 INFO [train.py:894] (2/4) Epoch 26, batch 1850, loss[loss=0.1869, simple_loss=0.2707, pruned_loss=0.05158, over 18558.00 frames. ], tot_loss[loss=0.1752, simple_loss=0.2613, pruned_loss=0.04456, over 3713690.73 frames. ], batch size: 58, lr: 4.40e-03, grad_scale: 16.0
2022-12-24 00:18:56,037 INFO [zipformer.py:660] (2/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=89509.0, num_to_drop=0, layers_to_drop=set()
2022-12-24 00:18:59,987 WARNING [train.py:1060] (2/4) Exclude cut with ID 1250-135782-0004-25974-0_sp0.9 from training. Duration: 21.17225
2022-12-24 00:19:01,505 WARNING [train.py:1060] (2/4) Exclude cut with ID 7357-94126-0021-15852-0_sp0.9 from training. Duration: 27.511125
2022-12-24 00:19:19,541 INFO [zipformer.py:660] (2/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=89525.0, num_to_drop=0, layers_to_drop=set()
2022-12-24 00:19:27,121 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([0.1144, 1.0649, 1.2152, 0.6766, 0.7432, 1.2518, 1.2795, 1.1910],
       device='cuda:2'), covar=tensor([0.0774, 0.0357, 0.0350, 0.0410, 0.0456, 0.0528, 0.0283, 0.0605],
       device='cuda:2'), in_proj_covar=tensor([0.0152, 0.0175, 0.0131, 0.0143, 0.0151, 0.0146, 0.0167, 0.0179],
       device='cuda:2'), out_proj_covar=tensor([1.1485e-04, 1.3178e-04, 9.6975e-05, 1.0516e-04, 1.1059e-04, 1.1005e-04,
        1.2641e-04, 1.3478e-04], device='cuda:2')
2022-12-24 00:19:32,398 WARNING [train.py:1060] (2/4) Exclude cut with ID 453-131332-0000-47844-0 from training. Duration: 22.8
2022-12-24 00:19:32,846 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([2.7552, 2.4605, 2.0269, 0.7793, 1.8035, 2.2606, 1.9288, 2.1420],
       device='cuda:2'), covar=tensor([0.0657, 0.0565, 0.1049, 0.1754, 0.1231, 0.1377, 0.1522, 0.0832],
       device='cuda:2'), in_proj_covar=tensor([0.0175, 0.0187, 0.0205, 0.0190, 0.0210, 0.0203, 0.0216, 0.0202],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-24 00:19:36,717 WARNING [train.py:1060] (2/4) Exclude cut with ID 4964-30587-0040-44509-0 from training. Duration: 22.585
2022-12-24 00:19:51,658 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 2.439e+02 3.869e+02 4.361e+02 5.499e+02 1.093e+03, threshold=8.721e+02, percent-clipped=2.0
2022-12-24 00:20:01,913 INFO [train.py:894] (2/4) Epoch 26, batch 1900, loss[loss=0.1957, simple_loss=0.2805, pruned_loss=0.05542, over 18585.00 frames. ], tot_loss[loss=0.1771, simple_loss=0.2622, pruned_loss=0.04598, over 3714083.14 frames. ], batch size: 57, lr: 4.40e-03, grad_scale: 16.0
2022-12-24 00:20:04,962 WARNING [train.py:1060] (2/4) Exclude cut with ID 6978-92210-0001-146967-0_sp0.9 from training. Duration: 22.0166875
2022-12-24 00:20:05,818 INFO [scaling.py:679] (2/4) Whitening: num_groups=8, num_channels=192, metric=1.79 vs. limit=2.0
2022-12-24 00:20:21,383 WARNING [train.py:1060] (2/4) Exclude cut with ID 7255-291500-0003-134302-0_sp1.1 from training. Duration: 24.395375
2022-12-24 00:20:27,950 INFO [zipformer.py:660] (2/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=89570.0, num_to_drop=0, layers_to_drop=set()
2022-12-24 00:20:29,051 WARNING [train.py:1060] (2/4) Exclude cut with ID 3488-85273-0017-41203-0_sp0.9 from training. Duration: 27.47775
2022-12-24 00:20:32,024 WARNING [train.py:1060] (2/4) Exclude cut with ID 432-122774-0017-62487-0_sp0.9 from training. Duration: 24.8833125
2022-12-24 00:20:34,935 WARNING [train.py:1060] (2/4) Exclude cut with ID 6758-72288-0033-108368-0 from training. Duration: 23.39
2022-12-24 00:20:40,909 WARNING [train.py:1060] (2/4) Exclude cut with ID 8291-282929-0007-12994-0_sp0.9 from training. Duration: 28.72225
2022-12-24 00:20:49,618 WARNING [train.py:1060] (2/4) Exclude cut with ID 585-294811-0110-133686-0_sp0.9 from training. Duration: 20.8944375
2022-12-24 00:21:08,121 WARNING [train.py:1060] (2/4) Exclude cut with ID 5796-66357-0007-116447-0_sp0.9 from training. Duration: 23.8444375
2022-12-24 00:21:17,392 INFO [train.py:894] (2/4) Epoch 26, batch 1950, loss[loss=0.1563, simple_loss=0.2418, pruned_loss=0.03539, over 18703.00 frames. ], tot_loss[loss=0.1783, simple_loss=0.263, pruned_loss=0.04675, over 3715062.58 frames. ], batch size: 50, lr: 4.39e-03, grad_scale: 16.0
2022-12-24 00:21:29,282 WARNING [train.py:1060] (2/4) Exclude cut with ID 8291-282929-0007-12994-0 from training. Duration: 25.85
2022-12-24 00:21:29,292 WARNING [train.py:1060] (2/4) Exclude cut with ID 8291-282929-0023-13010-0 from training. Duration: 21.39
2022-12-24 00:21:40,887 WARNING [train.py:1060] (2/4) Exclude cut with ID 7357-94126-0014-15845-0 from training. Duration: 27.92
2022-12-24 00:22:09,293 WARNING [train.py:1060] (2/4) Exclude cut with ID 8631-249866-0039-130165-0_sp0.9 from training. Duration: 20.661125
2022-12-24 00:22:22,081 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 2.800e+02 4.317e+02 5.022e+02 6.376e+02 1.135e+03, threshold=1.004e+03, percent-clipped=6.0
2022-12-24 00:22:28,493 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([2.5605, 2.2028, 1.7228, 2.4708, 1.9842, 2.1473, 2.1101, 2.4294],
       device='cuda:2'), covar=tensor([0.2148, 0.3072, 0.2040, 0.2635, 0.3810, 0.1083, 0.3028, 0.1038],
       device='cuda:2'), in_proj_covar=tensor([0.0298, 0.0297, 0.0251, 0.0348, 0.0279, 0.0233, 0.0295, 0.0220],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0003, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-24 00:22:32,553 INFO [train.py:894] (2/4) Epoch 26, batch 2000, loss[loss=0.1775, simple_loss=0.2575, pruned_loss=0.0487, over 18413.00 frames. ], tot_loss[loss=0.1799, simple_loss=0.2642, pruned_loss=0.04782, over 3714586.31 frames. ], batch size: 46, lr: 4.39e-03, grad_scale: 16.0
2022-12-24 00:22:32,621 WARNING [train.py:1060] (2/4) Exclude cut with ID 7357-94126-0043-15874-0_sp0.9 from training. Duration: 20.07225
2022-12-24 00:22:39,749 WARNING [train.py:1060] (2/4) Exclude cut with ID 1085-156170-0017-128270-0 from training. Duration: 21.01
2022-12-24 00:22:54,300 INFO [scaling.py:679] (2/4) Whitening: num_groups=8, num_channels=192, metric=1.60 vs. limit=2.0
2022-12-24 00:23:05,508 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([0.5295, 1.7438, 2.0345, 1.0987, 1.3036, 2.1427, 1.9976, 1.6710],
       device='cuda:2'), covar=tensor([0.0889, 0.0417, 0.0333, 0.0463, 0.0428, 0.0496, 0.0242, 0.0785],
       device='cuda:2'), in_proj_covar=tensor([0.0152, 0.0175, 0.0132, 0.0143, 0.0151, 0.0146, 0.0168, 0.0179],
       device='cuda:2'), out_proj_covar=tensor([1.1481e-04, 1.3227e-04, 9.7484e-05, 1.0525e-04, 1.1056e-04, 1.1007e-04,
        1.2686e-04, 1.3515e-04], device='cuda:2')
2022-12-24 00:23:13,316 INFO [zipformer.py:660] (2/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=89680.0, num_to_drop=0, layers_to_drop=set()
2022-12-24 00:23:18,061 INFO [zipformer.py:660] (2/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=89683.0, num_to_drop=0, layers_to_drop=set()
2022-12-24 00:23:32,764 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.8334, 1.3161, 0.9001, 1.5140, 2.2982, 1.2620, 1.5322, 1.7283],
       device='cuda:2'), covar=tensor([0.1620, 0.2097, 0.2171, 0.1438, 0.1655, 0.1815, 0.1516, 0.1748],
       device='cuda:2'), in_proj_covar=tensor([0.0094, 0.0097, 0.0116, 0.0095, 0.0119, 0.0091, 0.0097, 0.0093],
       device='cuda:2'), out_proj_covar=tensor([0.0003, 0.0003, 0.0004, 0.0003, 0.0004, 0.0003, 0.0003, 0.0003],
       device='cuda:2')
2022-12-24 00:23:46,681 INFO [train.py:894] (2/4) Epoch 26, batch 2050, loss[loss=0.1994, simple_loss=0.2786, pruned_loss=0.06014, over 18667.00 frames. ], tot_loss[loss=0.18, simple_loss=0.264, pruned_loss=0.04801, over 3714788.79 frames. ], batch size: 60, lr: 4.39e-03, grad_scale: 16.0
2022-12-24 00:23:48,208 WARNING [train.py:1060] (2/4) Exclude cut with ID 2195-150901-0045-59933-0 from training. Duration: 20.65
2022-12-24 00:23:54,723 WARNING [train.py:1060] (2/4) Exclude cut with ID 5796-66357-0007-116447-0 from training. Duration: 21.46
2022-12-24 00:24:28,603 INFO [zipformer.py:660] (2/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=89730.0, num_to_drop=0, layers_to_drop=set()
2022-12-24 00:24:29,828 INFO [zipformer.py:660] (2/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=89731.0, num_to_drop=0, layers_to_drop=set()
2022-12-24 00:24:39,508 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.4949, 1.9056, 1.5323, 2.1596, 2.4495, 1.5540, 1.4035, 1.2858],
       device='cuda:2'), covar=tensor([0.2022, 0.1757, 0.1689, 0.1074, 0.1281, 0.1136, 0.2282, 0.1604],
       device='cuda:2'), in_proj_covar=tensor([0.0250, 0.0230, 0.0221, 0.0203, 0.0263, 0.0199, 0.0228, 0.0203],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-24 00:24:40,498 WARNING [train.py:1060] (2/4) Exclude cut with ID 3557-8342-0013-54691-0 from training. Duration: 0.92
2022-12-24 00:24:45,182 INFO [zipformer.py:660] (2/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=89741.0, num_to_drop=0, layers_to_drop=set()
2022-12-24 00:24:47,880 WARNING [train.py:1060] (2/4) Exclude cut with ID 8291-282929-0023-13010-0_sp0.9 from training. Duration: 23.7666875
2022-12-24 00:24:52,049 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 2.904e+02 4.283e+02 5.573e+02 6.773e+02 1.480e+03, threshold=1.115e+03, percent-clipped=3.0
2022-12-24 00:25:02,381 INFO [train.py:894] (2/4) Epoch 26, batch 2100, loss[loss=0.1816, simple_loss=0.2723, pruned_loss=0.04544, over 18543.00 frames. ], tot_loss[loss=0.1797, simple_loss=0.2635, pruned_loss=0.04794, over 3714200.60 frames. ], batch size: 55, lr: 4.39e-03, grad_scale: 16.0
2022-12-24 00:25:22,526 WARNING [train.py:1060] (2/4) Exclude cut with ID 8544-281189-0060-101339-0_sp0.9 from training. Duration: 20.861125
2022-12-24 00:25:31,658 WARNING [train.py:1060] (2/4) Exclude cut with ID 3488-65654-0031-41259-0_sp0.9 from training. Duration: 22.711125
2022-12-24 00:25:45,651 INFO [zipformer.py:660] (2/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=89781.0, num_to_drop=0, layers_to_drop=set()
2022-12-24 00:26:06,392 INFO [zipformer.py:660] (2/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=89795.0, num_to_drop=1, layers_to_drop={0}
2022-12-24 00:26:13,173 WARNING [train.py:1060] (2/4) Exclude cut with ID 6951-79737-0043-132310-0_sp1.1 from training. Duration: 22.986375
2022-12-24 00:26:18,516 INFO [train.py:894] (2/4) Epoch 26, batch 2150, loss[loss=0.1702, simple_loss=0.2418, pruned_loss=0.04935, over 18407.00 frames. ], tot_loss[loss=0.1789, simple_loss=0.2622, pruned_loss=0.04781, over 3713448.54 frames. ], batch size: 42, lr: 4.39e-03, grad_scale: 16.0
2022-12-24 00:26:30,307 WARNING [train.py:1060] (2/4) Exclude cut with ID 8040-260924-0003-80960-0_sp0.9 from training. Duration: 22.07225
2022-12-24 00:26:34,813 WARNING [train.py:1060] (2/4) Exclude cut with ID 7699-105389-0045-26330-0_sp0.9 from training. Duration: 20.3055625
2022-12-24 00:26:37,656 WARNING [train.py:1060] (2/4) Exclude cut with ID 6356-271890-0060-94317-0_sp0.9 from training. Duration: 20.72225
2022-12-24 00:26:52,363 INFO [zipformer.py:660] (2/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=89825.0, num_to_drop=0, layers_to_drop=set()
2022-12-24 00:26:56,612 WARNING [train.py:1060] (2/4) Exclude cut with ID 3488-85273-0017-41203-0_sp1.1 from training. Duration: 22.4818125
2022-12-24 00:27:21,013 WARNING [train.py:1060] (2/4) Exclude cut with ID 4964-30587-0040-44509-0_sp0.9 from training. Duration: 25.0944375
2022-12-24 00:27:24,472 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 2.879e+02 4.099e+02 4.650e+02 6.605e+02 1.804e+03, threshold=9.299e+02, percent-clipped=2.0
2022-12-24 00:27:24,553 WARNING [train.py:1060] (2/4) Exclude cut with ID 6533-399-0047-104881-0 from training. Duration: 21.515
2022-12-24 00:27:30,755 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([5.9032, 4.9114, 5.1877, 5.8465, 5.4278, 5.2013, 5.9965, 1.6451],
       device='cuda:2'), covar=tensor([0.0598, 0.0701, 0.0531, 0.0743, 0.1343, 0.1102, 0.0413, 0.5319],
       device='cuda:2'), in_proj_covar=tensor([0.0357, 0.0235, 0.0245, 0.0282, 0.0335, 0.0277, 0.0300, 0.0292],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0001, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-24 00:27:31,932 WARNING [train.py:1060] (2/4) Exclude cut with ID 7357-94126-0009-15840-0_sp0.9 from training. Duration: 27.02225
2022-12-24 00:27:35,064 INFO [train.py:894] (2/4) Epoch 26, batch 2200, loss[loss=0.1661, simple_loss=0.2488, pruned_loss=0.04174, over 18559.00 frames. ], tot_loss[loss=0.1792, simple_loss=0.2622, pruned_loss=0.04813, over 3712856.94 frames. ], batch size: 49, lr: 4.39e-03, grad_scale: 16.0
2022-12-24 00:27:36,484 WARNING [train.py:1060] (2/4) Exclude cut with ID 432-122774-0010-62480-0_sp0.9 from training. Duration: 22.22225
2022-12-24 00:27:43,675 WARNING [train.py:1060] (2/4) Exclude cut with ID 4964-30587-0085-44554-0_sp0.9 from training. Duration: 20.85
2022-12-24 00:27:52,482 INFO [zipformer.py:660] (2/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=89865.0, num_to_drop=0, layers_to_drop=set()
2022-12-24 00:27:52,709 INFO [zipformer.py:660] (2/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=89865.0, num_to_drop=0, layers_to_drop=set()
2022-12-24 00:28:01,319 INFO [scaling.py:679] (2/4) Whitening: num_groups=8, num_channels=192, metric=1.51 vs. limit=2.0
2022-12-24 00:28:05,388 INFO [zipformer.py:660] (2/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=89873.0, num_to_drop=0, layers_to_drop=set()
2022-12-24 00:28:15,529 WARNING [train.py:1060] (2/4) Exclude cut with ID 4295-39940-0007-92567-0 from training. Duration: 21.54
2022-12-24 00:28:19,964 WARNING [train.py:1060] (2/4) Exclude cut with ID 4964-30587-0040-44509-0_sp1.1 from training. Duration: 20.5318125
2022-12-24 00:28:29,866 WARNING [train.py:1060] (2/4) Exclude cut with ID 7255-291500-0012-134311-0_sp0.9 from training. Duration: 21.9333125
2022-12-24 00:28:49,711 INFO [train.py:894] (2/4) Epoch 26, batch 2250, loss[loss=0.1873, simple_loss=0.2694, pruned_loss=0.0526, over 18723.00 frames. ], tot_loss[loss=0.1796, simple_loss=0.2623, pruned_loss=0.04846, over 3713016.58 frames. ], batch size: 77, lr: 4.39e-03, grad_scale: 16.0
2022-12-24 00:29:10,528 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([2.0750, 1.8715, 1.8026, 1.0588, 2.3243, 2.1032, 1.9057, 1.6078],
       device='cuda:2'), covar=tensor([0.0385, 0.0508, 0.0482, 0.0798, 0.0358, 0.0447, 0.0461, 0.0962],
       device='cuda:2'), in_proj_covar=tensor([0.0124, 0.0131, 0.0130, 0.0120, 0.0104, 0.0127, 0.0134, 0.0161],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0001, 0.0001, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-24 00:29:19,354 WARNING [train.py:1060] (2/4) Exclude cut with ID 8631-249866-0025-130151-0_sp0.9 from training. Duration: 21.7944375
2022-12-24 00:29:25,267 INFO [zipformer.py:660] (2/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=89926.0, num_to_drop=1, layers_to_drop={2}
2022-12-24 00:29:29,392 WARNING [train.py:1060] (2/4) Exclude cut with ID 8291-282929-0002-12989-0_sp0.9 from training. Duration: 22.4666875
2022-12-24 00:29:35,989 WARNING [train.py:1060] (2/4) Exclude cut with ID 6121-9014-0076-24124-0 from training. Duration: 21.635
2022-12-24 00:29:39,856 WARNING [train.py:1060] (2/4) Exclude cut with ID 6121-9014-0076-24124-0_sp0.9 from training. Duration: 24.038875
2022-12-24 00:29:48,958 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.7910, 1.1379, 0.8411, 1.3584, 2.2104, 1.0157, 1.4635, 1.5535],
       device='cuda:2'), covar=tensor([0.1513, 0.2142, 0.2028, 0.1399, 0.1655, 0.1704, 0.1441, 0.1689],
       device='cuda:2'), in_proj_covar=tensor([0.0094, 0.0097, 0.0115, 0.0095, 0.0118, 0.0091, 0.0097, 0.0092],
       device='cuda:2'), out_proj_covar=tensor([0.0003, 0.0003, 0.0004, 0.0003, 0.0004, 0.0003, 0.0003, 0.0003],
       device='cuda:2')
2022-12-24 00:29:54,788 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 3.010e+02 4.413e+02 5.147e+02 7.329e+02 2.051e+03, threshold=1.029e+03, percent-clipped=6.0
2022-12-24 00:30:06,067 INFO [train.py:894] (2/4) Epoch 26, batch 2300, loss[loss=0.1674, simple_loss=0.256, pruned_loss=0.03939, over 18612.00 frames. ], tot_loss[loss=0.1799, simple_loss=0.2624, pruned_loss=0.04871, over 3713517.06 frames. ], batch size: 53, lr: 4.39e-03, grad_scale: 16.0
2022-12-24 00:30:23,818 WARNING [train.py:1060] (2/4) Exclude cut with ID 7255-291500-0009-134308-0_sp1.1 from training. Duration: 21.786375
2022-12-24 00:30:35,666 WARNING [train.py:1060] (2/4) Exclude cut with ID 8291-282929-0002-12989-0 from training. Duration: 20.22
2022-12-24 00:31:25,505 INFO [train.py:894] (2/4) Epoch 26, batch 2350, loss[loss=0.2071, simple_loss=0.2779, pruned_loss=0.06817, over 18667.00 frames. ], tot_loss[loss=0.1791, simple_loss=0.2618, pruned_loss=0.04825, over 3713968.89 frames. ], batch size: 60, lr: 4.38e-03, grad_scale: 16.0
2022-12-24 00:32:06,619 INFO [zipformer.py:660] (2/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=90030.0, num_to_drop=0, layers_to_drop=set()
2022-12-24 00:32:10,050 INFO [scaling.py:679] (2/4) Whitening: num_groups=8, num_channels=96, metric=1.53 vs. limit=2.0
2022-12-24 00:32:15,652 INFO [zipformer.py:660] (2/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=90036.0, num_to_drop=0, layers_to_drop=set()
2022-12-24 00:32:30,329 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 2.765e+02 4.074e+02 4.769e+02 5.976e+02 1.069e+03, threshold=9.538e+02, percent-clipped=1.0
2022-12-24 00:32:41,316 INFO [train.py:894] (2/4) Epoch 26, batch 2400, loss[loss=0.1587, simple_loss=0.2371, pruned_loss=0.04014, over 18585.00 frames. ], tot_loss[loss=0.1786, simple_loss=0.2615, pruned_loss=0.04783, over 3713284.81 frames. ], batch size: 45, lr: 4.38e-03, grad_scale: 16.0
2022-12-24 00:32:43,413 WARNING [train.py:1060] (2/4) Exclude cut with ID 6951-79737-0043-132310-0 from training. Duration: 25.285
2022-12-24 00:33:19,945 INFO [zipformer.py:660] (2/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=90078.0, num_to_drop=0, layers_to_drop=set()
2022-12-24 00:33:24,491 INFO [zipformer.py:660] (2/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=90081.0, num_to_drop=0, layers_to_drop=set()
2022-12-24 00:33:43,089 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.6900, 1.6570, 1.3326, 1.6802, 1.7513, 1.5679, 2.0345, 1.7963],
       device='cuda:2'), covar=tensor([0.1134, 0.1839, 0.2945, 0.1723, 0.2107, 0.1180, 0.1128, 0.1453],
       device='cuda:2'), in_proj_covar=tensor([0.0182, 0.0215, 0.0257, 0.0293, 0.0243, 0.0196, 0.0209, 0.0209],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-24 00:33:46,055 INFO [zipformer.py:660] (2/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=90095.0, num_to_drop=1, layers_to_drop={0}
2022-12-24 00:33:47,201 WARNING [train.py:1060] (2/4) Exclude cut with ID 811-130148-0001-63453-0_sp0.9 from training. Duration: 20.861125
2022-12-24 00:33:58,447 INFO [train.py:894] (2/4) Epoch 26, batch 2450, loss[loss=0.1884, simple_loss=0.2734, pruned_loss=0.05165, over 18623.00 frames. ], tot_loss[loss=0.1792, simple_loss=0.2619, pruned_loss=0.04822, over 3714090.29 frames. ], batch size: 69, lr: 4.38e-03, grad_scale: 8.0
2022-12-24 00:34:10,281 WARNING [train.py:1060] (2/4) Exclude cut with ID 6010-56788-0055-90261-0 from training. Duration: 20.88
2022-12-24 00:34:38,126 INFO [zipformer.py:660] (2/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=90129.0, num_to_drop=0, layers_to_drop=set()
2022-12-24 00:34:40,115 INFO [zipformer.py:660] (2/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=90130.0, num_to_drop=0, layers_to_drop=set()
2022-12-24 00:34:41,431 WARNING [train.py:1060] (2/4) Exclude cut with ID 7357-94126-0045-15876-0_sp0.9 from training. Duration: 23.4166875
2022-12-24 00:34:59,303 INFO [zipformer.py:660] (2/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=90143.0, num_to_drop=1, layers_to_drop={1}
2022-12-24 00:35:05,266 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 2.816e+02 4.059e+02 5.047e+02 6.343e+02 1.627e+03, threshold=1.009e+03, percent-clipped=6.0
2022-12-24 00:35:14,329 INFO [train.py:894] (2/4) Epoch 26, batch 2500, loss[loss=0.1665, simple_loss=0.2422, pruned_loss=0.04541, over 18437.00 frames. ], tot_loss[loss=0.1787, simple_loss=0.2617, pruned_loss=0.04785, over 3713398.60 frames. ], batch size: 42, lr: 4.38e-03, grad_scale: 8.0
2022-12-24 00:35:33,037 INFO [zipformer.py:660] (2/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=90165.0, num_to_drop=0, layers_to_drop=set()
2022-12-24 00:35:56,756 WARNING [train.py:1060] (2/4) Exclude cut with ID 7255-291500-0006-134305-0 from training. Duration: 21.24
2022-12-24 00:35:56,769 WARNING [train.py:1060] (2/4) Exclude cut with ID 6533-399-0047-104881-0_sp0.9 from training. Duration: 23.9055625
2022-12-24 00:36:07,854 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([3.4358, 3.1169, 2.3295, 1.8734, 3.8645, 3.8792, 3.3358, 2.7555],
       device='cuda:2'), covar=tensor([0.0360, 0.0373, 0.0561, 0.0716, 0.0226, 0.0339, 0.0387, 0.0721],
       device='cuda:2'), in_proj_covar=tensor([0.0126, 0.0132, 0.0132, 0.0121, 0.0104, 0.0128, 0.0135, 0.0162],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0001, 0.0001, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-24 00:36:12,590 INFO [zipformer.py:660] (2/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=90191.0, num_to_drop=0, layers_to_drop=set()
2022-12-24 00:36:12,804 INFO [scaling.py:679] (2/4) Whitening: num_groups=1, num_channels=384, metric=3.33 vs. limit=5.0
2022-12-24 00:36:21,009 INFO [scaling.py:679] (2/4) Whitening: num_groups=8, num_channels=96, metric=1.36 vs. limit=2.0
2022-12-24 00:36:21,800 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.1027, 1.0972, 0.9937, 1.2472, 1.2721, 1.1842, 1.1277, 1.0071],
       device='cuda:2'), covar=tensor([0.0278, 0.0230, 0.0528, 0.0202, 0.0234, 0.0373, 0.0266, 0.0317],
       device='cuda:2'), in_proj_covar=tensor([0.0095, 0.0127, 0.0152, 0.0121, 0.0117, 0.0121, 0.0099, 0.0127],
       device='cuda:2'), out_proj_covar=tensor([7.4891e-05, 1.0016e-04, 1.2416e-04, 9.6029e-05, 9.3707e-05, 9.2434e-05,
        7.7150e-05, 1.0022e-04], device='cuda:2')
2022-12-24 00:36:24,638 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([2.4910, 2.2466, 2.0350, 1.4284, 2.8826, 2.7166, 2.2828, 1.8220],
       device='cuda:2'), covar=tensor([0.0421, 0.0503, 0.0573, 0.0794, 0.0283, 0.0422, 0.0477, 0.0975],
       device='cuda:2'), in_proj_covar=tensor([0.0126, 0.0133, 0.0132, 0.0121, 0.0105, 0.0128, 0.0136, 0.0162],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0001, 0.0001, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-24 00:36:29,985 INFO [train.py:894] (2/4) Epoch 26, batch 2550, loss[loss=0.1876, simple_loss=0.2764, pruned_loss=0.04944, over 18662.00 frames. ], tot_loss[loss=0.1789, simple_loss=0.2619, pruned_loss=0.04795, over 3713971.47 frames. ], batch size: 62, lr: 4.38e-03, grad_scale: 8.0
2022-12-24 00:36:32,672 WARNING [train.py:1060] (2/4) Exclude cut with ID 6758-72288-0033-108368-0_sp0.9 from training. Duration: 25.988875
2022-12-24 00:36:41,446 WARNING [train.py:1060] (2/4) Exclude cut with ID 7255-291500-0001-134300-0_sp0.9 from training. Duration: 20.67225
2022-12-24 00:36:44,213 INFO [zipformer.py:660] (2/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=90213.0, num_to_drop=0, layers_to_drop=set()
2022-12-24 00:36:44,568 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.8847, 2.2561, 1.7883, 2.6626, 2.8843, 1.7959, 2.1550, 1.4402],
       device='cuda:2'), covar=tensor([0.1899, 0.1707, 0.1545, 0.0938, 0.1420, 0.1103, 0.1814, 0.1524],
       device='cuda:2'), in_proj_covar=tensor([0.0252, 0.0232, 0.0222, 0.0203, 0.0265, 0.0199, 0.0228, 0.0203],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-24 00:36:56,053 INFO [zipformer.py:660] (2/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=90221.0, num_to_drop=1, layers_to_drop={3}
2022-12-24 00:37:13,197 INFO [zipformer.py:660] (2/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=90232.0, num_to_drop=0, layers_to_drop=set()
2022-12-24 00:37:28,561 WARNING [train.py:1060] (2/4) Exclude cut with ID 3488-85273-0038-41224-0 from training. Duration: 20.34
2022-12-24 00:37:35,936 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 2.732e+02 4.133e+02 5.034e+02 6.191e+02 1.589e+03, threshold=1.007e+03, percent-clipped=2.0
2022-12-24 00:37:44,770 INFO [train.py:894] (2/4) Epoch 26, batch 2600, loss[loss=0.1482, simple_loss=0.2261, pruned_loss=0.03516, over 18483.00 frames. ], tot_loss[loss=0.1782, simple_loss=0.2608, pruned_loss=0.04783, over 3714875.51 frames. ], batch size: 43, lr: 4.38e-03, grad_scale: 8.0
2022-12-24 00:38:35,588 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([4.2552, 3.7142, 3.7183, 4.1213, 3.9349, 3.8036, 4.3753, 1.5038],
       device='cuda:2'), covar=tensor([0.0714, 0.0740, 0.0665, 0.0808, 0.1331, 0.1158, 0.0635, 0.4731],
       device='cuda:2'), in_proj_covar=tensor([0.0356, 0.0235, 0.0244, 0.0283, 0.0335, 0.0276, 0.0299, 0.0290],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0001, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-24 00:38:40,289 WARNING [train.py:1060] (2/4) Exclude cut with ID 7357-94126-0026-15857-0_sp0.9 from training. Duration: 25.061125
2022-12-24 00:38:46,745 INFO [zipformer.py:660] (2/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=90293.0, num_to_drop=0, layers_to_drop=set()
2022-12-24 00:38:52,333 WARNING [train.py:1060] (2/4) Exclude cut with ID 3033-130750-0096-55598-0 from training. Duration: 0.83
2022-12-24 00:39:01,102 INFO [train.py:894] (2/4) Epoch 26, batch 2650, loss[loss=0.1497, simple_loss=0.2333, pruned_loss=0.03303, over 18560.00 frames. ], tot_loss[loss=0.178, simple_loss=0.2609, pruned_loss=0.0476, over 3715148.35 frames. ], batch size: 49, lr: 4.38e-03, grad_scale: 8.0
2022-12-24 00:39:18,957 WARNING [train.py:1060] (2/4) Exclude cut with ID 3488-85273-0017-41203-0 from training. Duration: 24.73
2022-12-24 00:39:33,178 WARNING [train.py:1060] (2/4) Exclude cut with ID 7255-291500-0009-134308-0 from training. Duration: 23.965
2022-12-24 00:39:40,682 WARNING [train.py:1060] (2/4) Exclude cut with ID 6978-92210-0030-146996-0_sp0.9 from training. Duration: 22.088875
2022-12-24 00:39:51,452 INFO [zipformer.py:660] (2/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=90336.0, num_to_drop=0, layers_to_drop=set()
2022-12-24 00:39:55,529 WARNING [train.py:1060] (2/4) Exclude cut with ID 7255-291500-0006-134305-0_sp0.9 from training. Duration: 23.6
2022-12-24 00:40:07,444 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 2.288e+02 4.377e+02 5.254e+02 6.585e+02 1.615e+03, threshold=1.051e+03, percent-clipped=2.0
2022-12-24 00:40:08,228 INFO [scaling.py:679] (2/4) Whitening: num_groups=8, num_channels=96, metric=1.16 vs. limit=2.0
2022-12-24 00:40:13,375 INFO [zipformer.py:660] (2/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=90351.0, num_to_drop=0, layers_to_drop=set()
2022-12-24 00:40:15,944 INFO [train.py:894] (2/4) Epoch 26, batch 2700, loss[loss=0.1795, simple_loss=0.2648, pruned_loss=0.04712, over 18501.00 frames. ], tot_loss[loss=0.1774, simple_loss=0.2604, pruned_loss=0.04724, over 3714924.79 frames. ], batch size: 52, lr: 4.38e-03, grad_scale: 8.0
2022-12-24 00:41:05,168 INFO [zipformer.py:660] (2/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=90384.0, num_to_drop=0, layers_to_drop=set()
2022-12-24 00:41:33,746 INFO [train.py:894] (2/4) Epoch 26, batch 2750, loss[loss=0.1976, simple_loss=0.2747, pruned_loss=0.0602, over 18695.00 frames. ], tot_loss[loss=0.1774, simple_loss=0.2605, pruned_loss=0.04714, over 3715122.55 frames. ], batch size: 62, lr: 4.38e-03, grad_scale: 8.0
2022-12-24 00:41:33,811 WARNING [train.py:1060] (2/4) Exclude cut with ID 8291-282929-0024-13011-0 from training. Duration: 23.795
2022-12-24 00:41:48,556 INFO [zipformer.py:660] (2/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=90412.0, num_to_drop=0, layers_to_drop=set()
2022-12-24 00:41:49,641 WARNING [train.py:1060] (2/4) Exclude cut with ID 8631-249866-0030-130156-0_sp1.1 from training. Duration: 21.5409375
2022-12-24 00:41:52,748 WARNING [train.py:1060] (2/4) Exclude cut with ID 6978-92210-0019-146985-0_sp0.9 from training. Duration: 24.97775
2022-12-24 00:42:04,957 WARNING [train.py:1060] (2/4) Exclude cut with ID 1085-156170-0017-128270-0_sp0.9 from training. Duration: 23.3444375
2022-12-24 00:42:14,913 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.9816, 2.0347, 2.2494, 1.1965, 2.2457, 2.3017, 1.6548, 2.6307],
       device='cuda:2'), covar=tensor([0.1183, 0.1750, 0.1272, 0.2032, 0.0721, 0.1142, 0.2303, 0.0594],
       device='cuda:2'), in_proj_covar=tensor([0.0200, 0.0216, 0.0209, 0.0196, 0.0174, 0.0219, 0.0217, 0.0194],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-24 00:42:30,684 WARNING [train.py:1060] (2/4) Exclude cut with ID 6010-56788-0055-90261-0_sp0.9 from training. Duration: 23.2
2022-12-24 00:42:36,804 WARNING [train.py:1060] (2/4) Exclude cut with ID 5653-46179-0060-117930-0_sp0.9 from training. Duration: 21.17225
2022-12-24 00:42:41,037 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 2.721e+02 4.034e+02 5.147e+02 6.608e+02 1.665e+03, threshold=1.029e+03, percent-clipped=4.0
2022-12-24 00:42:50,350 INFO [train.py:894] (2/4) Epoch 26, batch 2800, loss[loss=0.1692, simple_loss=0.2494, pruned_loss=0.04449, over 18436.00 frames. ], tot_loss[loss=0.177, simple_loss=0.2603, pruned_loss=0.04689, over 3715188.24 frames. ], batch size: 48, lr: 4.37e-03, grad_scale: 8.0
2022-12-24 00:42:58,249 WARNING [train.py:1060] (2/4) Exclude cut with ID 7255-291500-0008-134307-0_sp0.9 from training. Duration: 24.6555625
2022-12-24 00:43:00,047 INFO [zipformer.py:660] (2/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=90459.0, num_to_drop=0, layers_to_drop=set()
2022-12-24 00:43:40,221 INFO [zipformer.py:660] (2/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=90486.0, num_to_drop=0, layers_to_drop=set()
2022-12-24 00:43:49,148 WARNING [train.py:1060] (2/4) Exclude cut with ID 3488-65654-0031-41259-0 from training. Duration: 20.44
2022-12-24 00:43:56,066 INFO [zipformer.py:660] (2/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=90496.0, num_to_drop=0, layers_to_drop=set()
2022-12-24 00:44:06,643 INFO [train.py:894] (2/4) Epoch 26, batch 2850, loss[loss=0.1827, simple_loss=0.2647, pruned_loss=0.05038, over 18597.00 frames. ], tot_loss[loss=0.1766, simple_loss=0.2597, pruned_loss=0.0468, over 3714346.06 frames. ], batch size: 51, lr: 4.37e-03, grad_scale: 8.0
2022-12-24 00:44:06,720 WARNING [train.py:1060] (2/4) Exclude cut with ID 6951-79737-0018-132285-0_sp0.9 from training. Duration: 23.45
2022-12-24 00:44:32,459 INFO [zipformer.py:660] (2/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=90520.0, num_to_drop=0, layers_to_drop=set()
2022-12-24 00:44:33,868 INFO [zipformer.py:660] (2/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=90521.0, num_to_drop=1, layers_to_drop={0}
2022-12-24 00:44:37,996 WARNING [train.py:1060] (2/4) Exclude cut with ID 6945-60535-0076-12784-0_sp0.9 from training. Duration: 20.52225
2022-12-24 00:44:46,625 WARNING [train.py:1060] (2/4) Exclude cut with ID 7255-291500-0008-134307-0 from training. Duration: 22.19
2022-12-24 00:44:58,338 WARNING [train.py:1060] (2/4) Exclude cut with ID 7357-94126-0014-15845-0_sp1.1 from training. Duration: 25.3818125
2022-12-24 00:45:11,639 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 2.952e+02 4.260e+02 4.866e+02 6.366e+02 1.571e+03, threshold=9.732e+02, percent-clipped=4.0
2022-12-24 00:45:15,310 WARNING [train.py:1060] (2/4) Exclude cut with ID 6951-79737-0043-132310-0_sp0.9 from training. Duration: 28.0944375
2022-12-24 00:45:21,396 INFO [train.py:894] (2/4) Epoch 26, batch 2900, loss[loss=0.1616, simple_loss=0.2496, pruned_loss=0.03678, over 18578.00 frames. ], tot_loss[loss=0.1765, simple_loss=0.2598, pruned_loss=0.04658, over 3714977.06 frames. ], batch size: 51, lr: 4.37e-03, grad_scale: 8.0
2022-12-24 00:45:22,899 WARNING [train.py:1060] (2/4) Exclude cut with ID 2195-150901-0045-59933-0_sp0.9 from training. Duration: 22.9444375
2022-12-24 00:45:27,728 INFO [zipformer.py:660] (2/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=90557.0, num_to_drop=0, layers_to_drop=set()
2022-12-24 00:45:30,070 WARNING [train.py:1060] (2/4) Exclude cut with ID 8291-282929-0024-13011-0_sp1.1 from training. Duration: 21.6318125
2022-12-24 00:45:41,004 INFO [scaling.py:679] (2/4) Whitening: num_groups=8, num_channels=192, metric=1.50 vs. limit=2.0
2022-12-24 00:45:46,092 INFO [zipformer.py:660] (2/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=90569.0, num_to_drop=0, layers_to_drop=set()
2022-12-24 00:45:47,358 WARNING [train.py:1060] (2/4) Exclude cut with ID 8631-249866-0030-130156-0 from training. Duration: 23.695
2022-12-24 00:46:05,053 INFO [scaling.py:679] (2/4) Whitening: num_groups=1, num_channels=384, metric=2.91 vs. limit=5.0
2022-12-24 00:46:12,801 WARNING [train.py:1060] (2/4) Exclude cut with ID 7699-105389-0094-26379-0 from training. Duration: 23.955
2022-12-24 00:46:14,483 INFO [zipformer.py:660] (2/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=90588.0, num_to_drop=0, layers_to_drop=set()
2022-12-24 00:46:38,265 INFO [train.py:894] (2/4) Epoch 26, batch 2950, loss[loss=0.1538, simple_loss=0.2376, pruned_loss=0.03506, over 18525.00 frames. ], tot_loss[loss=0.1764, simple_loss=0.2599, pruned_loss=0.0465, over 3715207.53 frames. ], batch size: 47, lr: 4.37e-03, grad_scale: 8.0
2022-12-24 00:46:48,239 WARNING [train.py:1060] (2/4) Exclude cut with ID 8291-282929-0024-13011-0_sp0.9 from training. Duration: 26.438875
2022-12-24 00:47:27,909 WARNING [train.py:1060] (2/4) Exclude cut with ID 7699-105389-0021-26306-0_sp0.9 from training. Duration: 21.2444375
2022-12-24 00:47:27,931 WARNING [train.py:1060] (2/4) Exclude cut with ID 7357-94126-0014-15845-0_sp0.9 from training. Duration: 31.02225
2022-12-24 00:47:33,719 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([2.3641, 1.7382, 1.0064, 1.8129, 2.5152, 1.5935, 1.9525, 2.1802],
       device='cuda:2'), covar=tensor([0.1472, 0.1947, 0.2182, 0.1405, 0.1654, 0.1917, 0.1418, 0.1618],
       device='cuda:2'), in_proj_covar=tensor([0.0094, 0.0097, 0.0116, 0.0095, 0.0119, 0.0092, 0.0098, 0.0093],
       device='cuda:2'), out_proj_covar=tensor([0.0003, 0.0003, 0.0004, 0.0003, 0.0004, 0.0003, 0.0003, 0.0003],
       device='cuda:2')
2022-12-24 00:47:40,092 WARNING [train.py:1060] (2/4) Exclude cut with ID 432-122774-0017-62487-0 from training. Duration: 22.395
2022-12-24 00:47:46,250 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 2.852e+02 3.883e+02 4.957e+02 5.801e+02 1.568e+03, threshold=9.915e+02, percent-clipped=5.0
2022-12-24 00:47:49,729 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([2.4710, 2.4515, 1.8355, 3.0893, 2.8625, 2.3390, 3.6165, 2.4623],
       device='cuda:2'), covar=tensor([0.0828, 0.1890, 0.2763, 0.1762, 0.1656, 0.0868, 0.0820, 0.1265],
       device='cuda:2'), in_proj_covar=tensor([0.0181, 0.0214, 0.0255, 0.0291, 0.0241, 0.0194, 0.0207, 0.0209],
       device='cuda:2'), out_proj_covar=tensor([0.0001, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-24 00:47:55,785 INFO [train.py:894] (2/4) Epoch 26, batch 3000, loss[loss=0.1535, simple_loss=0.2422, pruned_loss=0.03242, over 18660.00 frames. ], tot_loss[loss=0.1777, simple_loss=0.261, pruned_loss=0.04719, over 3714864.40 frames. ], batch size: 48, lr: 4.37e-03, grad_scale: 8.0
2022-12-24 00:47:55,785 INFO [train.py:919] (2/4) Computing validation loss
2022-12-24 00:48:06,743 INFO [train.py:928] (2/4) Epoch 26, validation: loss=0.1627, simple_loss=0.2592, pruned_loss=0.03305, over 944034.00 frames. 
2022-12-24 00:48:06,744 INFO [train.py:929] (2/4) Maximum memory allocated so far is 24680MB
2022-12-24 00:48:08,371 WARNING [train.py:1060] (2/4) Exclude cut with ID 7357-94126-0045-15876-0 from training. Duration: 21.075
2022-12-24 00:48:14,137 WARNING [train.py:1060] (2/4) Exclude cut with ID 6482-98857-0025-147532-0_sp0.9 from training. Duration: 20.0055625
2022-12-24 00:48:15,714 WARNING [train.py:1060] (2/4) Exclude cut with ID 6951-79737-0037-132304-0_sp0.9 from training. Duration: 22.05
2022-12-24 00:48:15,727 WARNING [train.py:1060] (2/4) Exclude cut with ID 7255-291500-0003-134302-0 from training. Duration: 26.8349375
2022-12-24 00:48:18,535 WARNING [train.py:1060] (2/4) Exclude cut with ID 7357-94126-0009-15840-0_sp1.1 from training. Duration: 22.1090625
2022-12-24 00:48:25,606 WARNING [train.py:1060] (2/4) Exclude cut with ID 7699-105389-0094-26379-0_sp0.9 from training. Duration: 26.6166875
2022-12-24 00:48:44,394 WARNING [train.py:1060] (2/4) Exclude cut with ID 2046-178027-0000-53705-0_sp0.9 from training. Duration: 20.3055625
2022-12-24 00:49:07,318 WARNING [train.py:1060] (2/4) Exclude cut with ID 7205-50138-0008-5373-0_sp0.9 from training. Duration: 20.7
2022-12-24 00:49:22,262 INFO [train.py:894] (2/4) Epoch 26, batch 3050, loss[loss=0.1844, simple_loss=0.2671, pruned_loss=0.05091, over 18471.00 frames. ], tot_loss[loss=0.1766, simple_loss=0.2599, pruned_loss=0.04669, over 3714032.64 frames. ], batch size: 54, lr: 4.37e-03, grad_scale: 8.0
2022-12-24 00:49:28,513 INFO [zipformer.py:660] (2/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=90707.0, num_to_drop=0, layers_to_drop=set()
2022-12-24 00:49:50,718 WARNING [train.py:1060] (2/4) Exclude cut with ID 6978-92210-0019-146985-0 from training. Duration: 22.48
2022-12-24 00:50:05,807 WARNING [train.py:1060] (2/4) Exclude cut with ID 7255-291500-0003-134302-0_sp0.9 from training. Duration: 29.816625
2022-12-24 00:50:25,970 WARNING [train.py:1060] (2/4) Exclude cut with ID 7255-291500-0005-134304-0_sp1.1 from training. Duration: 22.7590625
2022-12-24 00:50:28,925 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 2.601e+02 4.164e+02 4.877e+02 6.267e+02 1.480e+03, threshold=9.754e+02, percent-clipped=2.0
2022-12-24 00:50:32,162 WARNING [train.py:1060] (2/4) Exclude cut with ID 7357-94126-0026-15857-0 from training. Duration: 22.555
2022-12-24 00:50:37,989 INFO [train.py:894] (2/4) Epoch 26, batch 3100, loss[loss=0.17, simple_loss=0.2424, pruned_loss=0.04879, over 18685.00 frames. ], tot_loss[loss=0.1762, simple_loss=0.2595, pruned_loss=0.04645, over 3714370.63 frames. ], batch size: 46, lr: 4.37e-03, grad_scale: 8.0
2022-12-24 00:50:39,834 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([2.3262, 2.0519, 1.4980, 0.5732, 1.4710, 2.0512, 1.7874, 1.8827],
       device='cuda:2'), covar=tensor([0.0575, 0.0515, 0.1078, 0.1544, 0.1103, 0.1417, 0.1531, 0.0641],
       device='cuda:2'), in_proj_covar=tensor([0.0175, 0.0189, 0.0209, 0.0190, 0.0209, 0.0204, 0.0217, 0.0204],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-24 00:50:41,123 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.7734, 2.0911, 1.9609, 1.0039, 2.1378, 1.6099, 0.5812, 1.2985],
       device='cuda:2'), covar=tensor([0.1865, 0.1308, 0.1349, 0.2604, 0.0987, 0.0838, 0.3174, 0.1345],
       device='cuda:2'), in_proj_covar=tensor([0.0149, 0.0146, 0.0159, 0.0124, 0.0151, 0.0115, 0.0143, 0.0114],
       device='cuda:2'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0002, 0.0003, 0.0002, 0.0003, 0.0002],
       device='cuda:2')
2022-12-24 00:50:52,849 WARNING [train.py:1060] (2/4) Exclude cut with ID 1250-135782-0005-25975-0_sp0.9 from training. Duration: 21.688875
2022-12-24 00:51:28,610 WARNING [train.py:1060] (2/4) Exclude cut with ID 3488-85273-0038-41224-0_sp0.9 from training. Duration: 22.6
2022-12-24 00:51:28,900 INFO [zipformer.py:660] (2/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=90786.0, num_to_drop=0, layers_to_drop=set()
2022-12-24 00:51:55,243 INFO [train.py:894] (2/4) Epoch 26, batch 3150, loss[loss=0.1782, simple_loss=0.2614, pruned_loss=0.04746, over 18652.00 frames. ], tot_loss[loss=0.1774, simple_loss=0.2605, pruned_loss=0.04711, over 3715699.22 frames. ], batch size: 99, lr: 4.37e-03, grad_scale: 8.0
2022-12-24 00:52:05,858 WARNING [train.py:1060] (2/4) Exclude cut with ID 7357-94126-0009-15840-0 from training. Duration: 24.32
2022-12-24 00:52:13,179 INFO [zipformer.py:660] (2/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=90815.0, num_to_drop=0, layers_to_drop=set()
2022-12-24 00:52:42,601 INFO [zipformer.py:660] (2/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=90834.0, num_to_drop=0, layers_to_drop=set()
2022-12-24 00:53:01,674 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 2.469e+02 4.405e+02 5.375e+02 6.469e+02 1.692e+03, threshold=1.075e+03, percent-clipped=5.0
2022-12-24 00:53:03,341 WARNING [train.py:1060] (2/4) Exclude cut with ID 8291-276745-0093-13116-0_sp0.9 from training. Duration: 21.061125
2022-12-24 00:53:05,137 INFO [zipformer.py:660] (2/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=90849.0, num_to_drop=0, layers_to_drop=set()
2022-12-24 00:53:06,589 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.7081, 1.2124, 0.7535, 1.2868, 2.1015, 0.9675, 1.3119, 1.5725],
       device='cuda:2'), covar=tensor([0.1675, 0.2239, 0.2200, 0.1545, 0.1799, 0.1883, 0.1548, 0.1725],
       device='cuda:2'), in_proj_covar=tensor([0.0094, 0.0097, 0.0116, 0.0096, 0.0119, 0.0092, 0.0098, 0.0093],
       device='cuda:2'), out_proj_covar=tensor([0.0003, 0.0003, 0.0004, 0.0003, 0.0004, 0.0003, 0.0003, 0.0003],
       device='cuda:2')
2022-12-24 00:53:09,161 INFO [zipformer.py:660] (2/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=90852.0, num_to_drop=0, layers_to_drop=set()
2022-12-24 00:53:10,234 INFO [train.py:894] (2/4) Epoch 26, batch 3200, loss[loss=0.1697, simple_loss=0.2477, pruned_loss=0.04586, over 18509.00 frames. ], tot_loss[loss=0.1765, simple_loss=0.26, pruned_loss=0.04652, over 3715731.54 frames. ], batch size: 47, lr: 4.36e-03, grad_scale: 8.0
2022-12-24 00:53:17,354 INFO [scaling.py:679] (2/4) Whitening: num_groups=8, num_channels=96, metric=1.26 vs. limit=2.0
2022-12-24 00:53:19,170 WARNING [train.py:1060] (2/4) Exclude cut with ID 7357-94126-0024-15855-0_sp0.9 from training. Duration: 20.32225
2022-12-24 00:53:32,716 WARNING [train.py:1060] (2/4) Exclude cut with ID 3033-130750-0096-55598-0_sp1.1 from training. Duration: 0.7545625
2022-12-24 00:53:48,852 WARNING [train.py:1060] (2/4) Exclude cut with ID 4295-39940-0007-92567-0_sp0.9 from training. Duration: 23.9333125
2022-12-24 00:54:04,112 INFO [zipformer.py:660] (2/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=90888.0, num_to_drop=0, layers_to_drop=set()
2022-12-24 00:54:15,009 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.8505, 1.4862, 1.6598, 2.0216, 1.7816, 3.6146, 1.4600, 1.6858],
       device='cuda:2'), covar=tensor([0.0811, 0.1821, 0.1041, 0.0873, 0.1481, 0.0306, 0.1457, 0.1582],
       device='cuda:2'), in_proj_covar=tensor([0.0073, 0.0083, 0.0072, 0.0075, 0.0092, 0.0076, 0.0085, 0.0077],
       device='cuda:2'), out_proj_covar=tensor([0.0004, 0.0004, 0.0004, 0.0004, 0.0004, 0.0004, 0.0004, 0.0004],
       device='cuda:2')
2022-12-24 00:54:20,520 WARNING [train.py:1060] (2/4) Exclude cut with ID 7255-291500-0008-134307-0_sp1.1 from training. Duration: 20.17275
2022-12-24 00:54:24,724 WARNING [train.py:1060] (2/4) Exclude cut with ID 6978-92210-0019-146985-0_sp1.1 from training. Duration: 20.436375
2022-12-24 00:54:26,602 INFO [train.py:894] (2/4) Epoch 26, batch 3250, loss[loss=0.1931, simple_loss=0.2807, pruned_loss=0.05275, over 18594.00 frames. ], tot_loss[loss=0.1755, simple_loss=0.2593, pruned_loss=0.04586, over 3715038.39 frames. ], batch size: 69, lr: 4.36e-03, grad_scale: 8.0
2022-12-24 00:54:38,348 INFO [zipformer.py:660] (2/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=90910.0, num_to_drop=0, layers_to_drop=set()
2022-12-24 00:55:16,729 INFO [zipformer.py:660] (2/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=90936.0, num_to_drop=0, layers_to_drop=set()
2022-12-24 00:55:32,538 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 2.717e+02 4.387e+02 5.223e+02 6.052e+02 1.643e+03, threshold=1.045e+03, percent-clipped=3.0
2022-12-24 00:55:42,432 INFO [train.py:894] (2/4) Epoch 26, batch 3300, loss[loss=0.1609, simple_loss=0.2361, pruned_loss=0.04285, over 18497.00 frames. ], tot_loss[loss=0.176, simple_loss=0.2596, pruned_loss=0.04617, over 3715072.60 frames. ], batch size: 43, lr: 4.36e-03, grad_scale: 8.0
2022-12-24 00:55:43,978 WARNING [train.py:1060] (2/4) Exclude cut with ID 4234-40345-0022-142709-0_sp0.9 from training. Duration: 23.1055625
2022-12-24 00:55:45,524 WARNING [train.py:1060] (2/4) Exclude cut with ID 8291-282929-0007-12994-0_sp1.1 from training. Duration: 23.5
2022-12-24 00:55:56,987 WARNING [train.py:1060] (2/4) Exclude cut with ID 7255-291500-0009-134308-0_sp0.9 from training. Duration: 26.62775
2022-12-24 00:56:09,679 WARNING [train.py:1060] (2/4) Exclude cut with ID 6951-79737-0018-132285-0 from training. Duration: 21.105
2022-12-24 00:56:14,113 WARNING [train.py:1060] (2/4) Exclude cut with ID 4511-76322-0006-80011-0_sp0.9 from training. Duration: 24.411125
2022-12-24 00:56:43,159 WARNING [train.py:1060] (2/4) Exclude cut with ID 6758-72288-0033-108368-0_sp1.1 from training. Duration: 21.263625
2022-12-24 00:57:00,715 INFO [train.py:894] (2/4) Epoch 26, batch 3350, loss[loss=0.1787, simple_loss=0.2706, pruned_loss=0.04343, over 18621.00 frames. ], tot_loss[loss=0.1759, simple_loss=0.2595, pruned_loss=0.04612, over 3714674.95 frames. ], batch size: 69, lr: 4.36e-03, grad_scale: 8.0
2022-12-24 00:57:07,493 INFO [zipformer.py:660] (2/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=91007.0, num_to_drop=0, layers_to_drop=set()
2022-12-24 00:57:17,108 WARNING [train.py:1060] (2/4) Exclude cut with ID 4234-40345-0022-142709-0 from training. Duration: 20.795
2022-12-24 00:57:25,839 WARNING [train.py:1060] (2/4) Exclude cut with ID 7357-94126-0021-15852-0 from training. Duration: 24.76
2022-12-24 00:57:27,399 WARNING [train.py:1060] (2/4) Exclude cut with ID 3867-173237-0077-144769-0_sp0.9 from training. Duration: 22.25
2022-12-24 00:57:53,742 WARNING [train.py:1060] (2/4) Exclude cut with ID 7357-94126-0026-15857-0_sp1.1 from training. Duration: 20.5045625
2022-12-24 00:58:07,449 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 2.659e+02 3.619e+02 4.459e+02 5.984e+02 1.100e+03, threshold=8.918e+02, percent-clipped=1.0
2022-12-24 00:58:17,685 INFO [train.py:894] (2/4) Epoch 26, batch 3400, loss[loss=0.1612, simple_loss=0.2409, pruned_loss=0.04074, over 18552.00 frames. ], tot_loss[loss=0.1756, simple_loss=0.2592, pruned_loss=0.046, over 3713821.59 frames. ], batch size: 49, lr: 4.36e-03, grad_scale: 8.0
2022-12-24 00:58:20,582 INFO [zipformer.py:660] (2/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=91055.0, num_to_drop=0, layers_to_drop=set()
2022-12-24 00:58:53,312 INFO [scaling.py:679] (2/4) Whitening: num_groups=8, num_channels=192, metric=1.49 vs. limit=2.0
2022-12-24 00:59:29,991 INFO [train.py:894] (2/4) Epoch 26, batch 3450, loss[loss=0.1807, simple_loss=0.2626, pruned_loss=0.04944, over 18583.00 frames. ], tot_loss[loss=0.1769, simple_loss=0.2605, pruned_loss=0.04666, over 3713399.13 frames. ], batch size: 51, lr: 4.36e-03, grad_scale: 8.0
2022-12-24 00:59:47,537 INFO [zipformer.py:660] (2/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=91115.0, num_to_drop=0, layers_to_drop=set()
2022-12-24 01:00:33,638 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 2.302e+02 4.218e+02 5.018e+02 6.316e+02 1.663e+03, threshold=1.004e+03, percent-clipped=6.0
2022-12-24 01:00:41,924 INFO [zipformer.py:660] (2/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=91152.0, num_to_drop=0, layers_to_drop=set()
2022-12-24 01:00:43,281 INFO [train.py:894] (2/4) Epoch 26, batch 3500, loss[loss=0.2307, simple_loss=0.2971, pruned_loss=0.08214, over 18665.00 frames. ], tot_loss[loss=0.1776, simple_loss=0.2609, pruned_loss=0.04716, over 3713947.27 frames. ], batch size: 172, lr: 4.36e-03, grad_scale: 8.0
2022-12-24 01:01:05,554 WARNING [train.py:1060] (2/4) Exclude cut with ID 7859-102521-0017-7548-0_sp1.1 from training. Duration: 22.2954375
2022-12-24 01:01:14,984 INFO [train.py:894] (2/4) Epoch 27, batch 0, loss[loss=0.1953, simple_loss=0.2807, pruned_loss=0.05495, over 18689.00 frames. ], tot_loss[loss=0.1953, simple_loss=0.2807, pruned_loss=0.05495, over 18689.00 frames. ], batch size: 60, lr: 4.27e-03, grad_scale: 8.0
2022-12-24 01:01:14,985 INFO [train.py:919] (2/4) Computing validation loss
2022-12-24 01:01:20,252 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.9452, 2.0253, 2.2717, 1.4356, 2.4598, 2.4082, 1.6965, 2.7974],
       device='cuda:2'), covar=tensor([0.1355, 0.1952, 0.1511, 0.2290, 0.0736, 0.1190, 0.2613, 0.0561],
       device='cuda:2'), in_proj_covar=tensor([0.0199, 0.0217, 0.0209, 0.0195, 0.0173, 0.0219, 0.0218, 0.0193],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-24 01:01:21,726 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.4084, 1.0532, 0.7222, 1.0712, 1.8196, 0.6000, 1.1224, 1.1749],
       device='cuda:2'), covar=tensor([0.1836, 0.2187, 0.2079, 0.1600, 0.1914, 0.1982, 0.1585, 0.2025],
       device='cuda:2'), in_proj_covar=tensor([0.0094, 0.0097, 0.0116, 0.0096, 0.0120, 0.0092, 0.0098, 0.0093],
       device='cuda:2'), out_proj_covar=tensor([0.0003, 0.0003, 0.0004, 0.0003, 0.0004, 0.0003, 0.0003, 0.0003],
       device='cuda:2')
2022-12-24 01:01:25,754 INFO [train.py:928] (2/4) Epoch 27, validation: loss=0.1651, simple_loss=0.2615, pruned_loss=0.03432, over 944034.00 frames. 
2022-12-24 01:01:25,754 INFO [train.py:929] (2/4) Maximum memory allocated so far is 24680MB
2022-12-24 01:01:30,180 INFO [zipformer.py:660] (2/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=91163.0, num_to_drop=0, layers_to_drop=set()
2022-12-24 01:01:40,814 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([0.7652, 0.7141, 0.6961, 0.8239, 0.8928, 0.8213, 0.7945, 0.7112],
       device='cuda:2'), covar=tensor([0.0229, 0.0218, 0.0424, 0.0196, 0.0224, 0.0330, 0.0229, 0.0271],
       device='cuda:2'), in_proj_covar=tensor([0.0096, 0.0127, 0.0153, 0.0122, 0.0117, 0.0122, 0.0101, 0.0129],
       device='cuda:2'), out_proj_covar=tensor([7.5616e-05, 1.0051e-04, 1.2533e-04, 9.6618e-05, 9.3855e-05, 9.3386e-05,
        7.8099e-05, 1.0135e-04], device='cuda:2')
2022-12-24 01:02:01,340 INFO [scaling.py:679] (2/4) Whitening: num_groups=8, num_channels=96, metric=1.54 vs. limit=2.0
2022-12-24 01:02:14,813 WARNING [train.py:1060] (2/4) Exclude cut with ID 298-126791-0067-24026-0_sp0.9 from training. Duration: 21.438875
2022-12-24 01:02:20,248 WARNING [train.py:1060] (2/4) Exclude cut with ID 5652-39938-0025-23684-0_sp0.9 from training. Duration: 22.2055625
2022-12-24 01:02:26,709 INFO [zipformer.py:660] (2/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=91200.0, num_to_drop=0, layers_to_drop=set()
2022-12-24 01:02:34,661 INFO [zipformer.py:660] (2/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=91205.0, num_to_drop=0, layers_to_drop=set()
2022-12-24 01:02:41,925 INFO [train.py:894] (2/4) Epoch 27, batch 50, loss[loss=0.1746, simple_loss=0.2647, pruned_loss=0.04218, over 18553.00 frames. ], tot_loss[loss=0.1738, simple_loss=0.2634, pruned_loss=0.04214, over 837795.21 frames. ], batch size: 55, lr: 4.27e-03, grad_scale: 8.0
2022-12-24 01:03:36,501 INFO [scaling.py:679] (2/4) Whitening: num_groups=8, num_channels=96, metric=1.31 vs. limit=2.0
2022-12-24 01:03:38,310 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 2.308e+02 3.518e+02 4.263e+02 5.371e+02 1.025e+03, threshold=8.526e+02, percent-clipped=1.0
2022-12-24 01:03:57,065 INFO [train.py:894] (2/4) Epoch 27, batch 100, loss[loss=0.1375, simple_loss=0.2232, pruned_loss=0.02588, over 18549.00 frames. ], tot_loss[loss=0.1695, simple_loss=0.2586, pruned_loss=0.04021, over 1475423.19 frames. ], batch size: 44, lr: 4.27e-03, grad_scale: 8.0
2022-12-24 01:04:27,033 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.7439, 1.5215, 1.5514, 1.9699, 1.9142, 3.6106, 1.6549, 1.7612],
       device='cuda:2'), covar=tensor([0.0767, 0.1694, 0.0978, 0.0840, 0.1284, 0.0191, 0.1231, 0.1383],
       device='cuda:2'), in_proj_covar=tensor([0.0073, 0.0083, 0.0072, 0.0075, 0.0091, 0.0076, 0.0085, 0.0077],
       device='cuda:2'), out_proj_covar=tensor([0.0004, 0.0004, 0.0004, 0.0004, 0.0004, 0.0004, 0.0004, 0.0004],
       device='cuda:2')
2022-12-24 01:04:35,375 INFO [zipformer.py:660] (2/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=91285.0, num_to_drop=0, layers_to_drop=set()
2022-12-24 01:05:11,680 INFO [train.py:894] (2/4) Epoch 27, batch 150, loss[loss=0.148, simple_loss=0.2448, pruned_loss=0.02563, over 18669.00 frames. ], tot_loss[loss=0.1683, simple_loss=0.2574, pruned_loss=0.03965, over 1972292.55 frames. ], batch size: 48, lr: 4.27e-03, grad_scale: 8.0
2022-12-24 01:05:15,466 INFO [scaling.py:679] (2/4) Whitening: num_groups=8, num_channels=96, metric=1.30 vs. limit=2.0
2022-12-24 01:05:21,815 WARNING [train.py:1060] (2/4) Exclude cut with ID 7859-102521-0017-7548-0 from training. Duration: 24.525
2022-12-24 01:05:53,862 WARNING [train.py:1060] (2/4) Exclude cut with ID 3699-47246-0007-3408-0_sp0.9 from training. Duration: 20.26675
2022-12-24 01:06:02,736 INFO [scaling.py:679] (2/4) Whitening: num_groups=1, num_channels=384, metric=4.78 vs. limit=5.0
2022-12-24 01:06:06,746 INFO [zipformer.py:660] (2/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=91346.0, num_to_drop=0, layers_to_drop=set()
2022-12-24 01:06:07,770 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 2.293e+02 3.105e+02 3.683e+02 4.423e+02 8.436e+02, threshold=7.367e+02, percent-clipped=0.0
2022-12-24 01:06:07,787 WARNING [train.py:1060] (2/4) Exclude cut with ID 7859-102521-0017-7548-0_sp0.9 from training. Duration: 27.25
2022-12-24 01:06:27,652 INFO [train.py:894] (2/4) Epoch 27, batch 200, loss[loss=0.1691, simple_loss=0.2647, pruned_loss=0.03671, over 18525.00 frames. ], tot_loss[loss=0.1676, simple_loss=0.2571, pruned_loss=0.0391, over 2357597.79 frames. ], batch size: 55, lr: 4.27e-03, grad_scale: 8.0
2022-12-24 01:07:19,293 WARNING [train.py:1060] (2/4) Exclude cut with ID 6426-64292-0017-15984-0 from training. Duration: 21.68
2022-12-24 01:07:31,731 WARNING [train.py:1060] (2/4) Exclude cut with ID 4278-13270-0007-59342-0 from training. Duration: 21.6300625
2022-12-24 01:07:35,254 INFO [scaling.py:679] (2/4) Whitening: num_groups=8, num_channels=96, metric=1.39 vs. limit=2.0
2022-12-24 01:07:41,616 INFO [train.py:894] (2/4) Epoch 27, batch 250, loss[loss=0.1521, simple_loss=0.2482, pruned_loss=0.02801, over 18523.00 frames. ], tot_loss[loss=0.1662, simple_loss=0.2557, pruned_loss=0.0383, over 2658321.73 frames. ], batch size: 52, lr: 4.27e-03, grad_scale: 8.0
2022-12-24 01:07:42,070 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([2.4563, 2.9687, 3.0984, 1.8203, 3.1465, 3.5411, 2.4608, 3.6360],
       device='cuda:2'), covar=tensor([0.1444, 0.1640, 0.1438, 0.2150, 0.0795, 0.1006, 0.2039, 0.0492],
       device='cuda:2'), in_proj_covar=tensor([0.0198, 0.0216, 0.0208, 0.0194, 0.0172, 0.0218, 0.0217, 0.0192],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-24 01:07:56,698 WARNING [train.py:1060] (2/4) Exclude cut with ID 4278-13270-0007-59342-0_sp0.9 from training. Duration: 24.033375
2022-12-24 01:08:35,871 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 1.808e+02 3.295e+02 4.086e+02 4.831e+02 1.349e+03, threshold=8.172e+02, percent-clipped=3.0
2022-12-24 01:08:53,988 WARNING [train.py:1060] (2/4) Exclude cut with ID 4278-13270-0009-59344-0 from training. Duration: 22.905
2022-12-24 01:08:55,348 INFO [train.py:894] (2/4) Epoch 27, batch 300, loss[loss=0.1616, simple_loss=0.248, pruned_loss=0.03761, over 18640.00 frames. ], tot_loss[loss=0.1664, simple_loss=0.2561, pruned_loss=0.03835, over 2891189.94 frames. ], batch size: 53, lr: 4.27e-03, grad_scale: 8.0
2022-12-24 01:08:55,439 WARNING [train.py:1060] (2/4) Exclude cut with ID 5622-44585-0006-90525-0_sp1.1 from training. Duration: 23.4318125
2022-12-24 01:10:04,557 INFO [zipformer.py:660] (2/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=91505.0, num_to_drop=0, layers_to_drop=set()
2022-12-24 01:10:10,315 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([2.6354, 2.4363, 2.0161, 1.0511, 2.0384, 2.0088, 1.4890, 2.2408],
       device='cuda:2'), covar=tensor([0.0716, 0.0706, 0.1428, 0.1825, 0.1460, 0.1508, 0.1955, 0.0928],
       device='cuda:2'), in_proj_covar=tensor([0.0176, 0.0188, 0.0209, 0.0191, 0.0210, 0.0205, 0.0219, 0.0205],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-24 01:10:11,307 INFO [train.py:894] (2/4) Epoch 27, batch 350, loss[loss=0.2106, simple_loss=0.2925, pruned_loss=0.06431, over 18507.00 frames. ], tot_loss[loss=0.1669, simple_loss=0.2566, pruned_loss=0.03857, over 3073541.32 frames. ], batch size: 52, lr: 4.27e-03, grad_scale: 8.0
2022-12-24 01:10:55,829 WARNING [train.py:1060] (2/4) Exclude cut with ID 4278-13270-0009-59344-0_sp1.1 from training. Duration: 20.82275
2022-12-24 01:10:57,287 WARNING [train.py:1060] (2/4) Exclude cut with ID 4278-13270-0009-59344-0_sp0.9 from training. Duration: 25.45
2022-12-24 01:11:00,268 INFO [zipformer.py:660] (2/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=91542.0, num_to_drop=1, layers_to_drop={0}
2022-12-24 01:11:07,144 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 2.079e+02 3.172e+02 3.767e+02 4.627e+02 1.022e+03, threshold=7.535e+02, percent-clipped=2.0
2022-12-24 01:11:16,307 INFO [zipformer.py:660] (2/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=91553.0, num_to_drop=0, layers_to_drop=set()
2022-12-24 01:11:25,878 INFO [train.py:894] (2/4) Epoch 27, batch 400, loss[loss=0.1343, simple_loss=0.2213, pruned_loss=0.02365, over 18478.00 frames. ], tot_loss[loss=0.1673, simple_loss=0.257, pruned_loss=0.03878, over 3214446.91 frames. ], batch size: 43, lr: 4.26e-03, grad_scale: 8.0
2022-12-24 01:11:59,300 WARNING [train.py:1060] (2/4) Exclude cut with ID 5622-44585-0006-90525-0 from training. Duration: 25.775
2022-12-24 01:12:19,944 WARNING [train.py:1060] (2/4) Exclude cut with ID 2929-85685-0071-62375-0_sp0.9 from training. Duration: 22.25
2022-12-24 01:12:31,436 INFO [zipformer.py:660] (2/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=91603.0, num_to_drop=1, layers_to_drop={0}
2022-12-24 01:12:31,511 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([2.0750, 1.8701, 1.4825, 1.4094, 1.7548, 1.9025, 1.7107, 1.8644],
       device='cuda:2'), covar=tensor([0.2396, 0.3210, 0.2267, 0.2804, 0.3401, 0.1195, 0.3189, 0.1054],
       device='cuda:2'), in_proj_covar=tensor([0.0301, 0.0300, 0.0254, 0.0351, 0.0282, 0.0236, 0.0298, 0.0224],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0003, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-24 01:12:41,283 INFO [train.py:894] (2/4) Epoch 27, batch 450, loss[loss=0.1789, simple_loss=0.2674, pruned_loss=0.04514, over 18684.00 frames. ], tot_loss[loss=0.1674, simple_loss=0.2573, pruned_loss=0.03872, over 3325016.26 frames. ], batch size: 62, lr: 4.26e-03, grad_scale: 8.0
2022-12-24 01:12:45,267 WARNING [train.py:1060] (2/4) Exclude cut with ID 3972-170212-0014-23379-0 from training. Duration: 26.205
2022-12-24 01:13:02,076 WARNING [train.py:1060] (2/4) Exclude cut with ID 5239-32139-0047-9341-0_sp0.9 from training. Duration: 30.1555625
2022-12-24 01:13:07,880 WARNING [train.py:1060] (2/4) Exclude cut with ID 1265-135635-0050-6781-0_sp0.9 from training. Duration: 21.8333125
2022-12-24 01:13:17,138 WARNING [train.py:1060] (2/4) Exclude cut with ID 1914-133440-0024-94914-0_sp1.1 from training. Duration: 20.6545625
2022-12-24 01:13:22,129 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.9102, 2.1467, 1.8419, 2.4406, 3.2221, 1.7736, 1.7757, 1.4458],
       device='cuda:2'), covar=tensor([0.1812, 0.1779, 0.1496, 0.1018, 0.1179, 0.1066, 0.2056, 0.1560],
       device='cuda:2'), in_proj_covar=tensor([0.0250, 0.0230, 0.0221, 0.0203, 0.0264, 0.0198, 0.0228, 0.0203],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-24 01:13:27,191 INFO [zipformer.py:660] (2/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=91641.0, num_to_drop=0, layers_to_drop=set()
2022-12-24 01:13:36,641 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 2.293e+02 3.165e+02 3.798e+02 4.275e+02 8.978e+02, threshold=7.596e+02, percent-clipped=3.0
2022-12-24 01:13:55,161 INFO [train.py:894] (2/4) Epoch 27, batch 500, loss[loss=0.1522, simple_loss=0.2363, pruned_loss=0.03401, over 18518.00 frames. ], tot_loss[loss=0.1682, simple_loss=0.2583, pruned_loss=0.03906, over 3411025.43 frames. ], batch size: 44, lr: 4.26e-03, grad_scale: 8.0
2022-12-24 01:13:59,259 WARNING [train.py:1060] (2/4) Exclude cut with ID 7395-89880-0045-39920-0_sp0.9 from training. Duration: 20.52225
2022-12-24 01:14:20,059 WARNING [train.py:1060] (2/4) Exclude cut with ID 3972-170212-0014-23379-0_sp0.9 from training. Duration: 29.1166875
2022-12-24 01:14:28,638 INFO [scaling.py:679] (2/4) Whitening: num_groups=8, num_channels=96, metric=1.11 vs. limit=2.0
2022-12-24 01:15:09,272 INFO [train.py:894] (2/4) Epoch 27, batch 550, loss[loss=0.1712, simple_loss=0.2628, pruned_loss=0.03975, over 18518.00 frames. ], tot_loss[loss=0.168, simple_loss=0.2578, pruned_loss=0.03911, over 3477521.13 frames. ], batch size: 52, lr: 4.26e-03, grad_scale: 8.0
2022-12-24 01:15:22,447 WARNING [train.py:1060] (2/4) Exclude cut with ID 543-133211-0007-59831-0_sp0.9 from training. Duration: 21.388875
2022-12-24 01:15:59,480 WARNING [train.py:1060] (2/4) Exclude cut with ID 1914-133440-0024-94914-0 from training. Duration: 22.72
2022-12-24 01:16:00,892 WARNING [train.py:1060] (2/4) Exclude cut with ID 1914-133440-0031-94921-0_sp0.9 from training. Duration: 22.7444375
2022-12-24 01:16:03,725 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 2.126e+02 3.350e+02 3.947e+02 4.704e+02 8.228e+02, threshold=7.893e+02, percent-clipped=1.0
2022-12-24 01:16:20,771 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.9322, 1.9218, 1.4661, 2.1944, 2.1284, 1.8547, 2.6089, 2.0118],
       device='cuda:2'), covar=tensor([0.0934, 0.1753, 0.2880, 0.1546, 0.1830, 0.0964, 0.0940, 0.1356],
       device='cuda:2'), in_proj_covar=tensor([0.0182, 0.0215, 0.0254, 0.0291, 0.0241, 0.0193, 0.0207, 0.0209],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-24 01:16:23,307 INFO [train.py:894] (2/4) Epoch 27, batch 600, loss[loss=0.1655, simple_loss=0.2427, pruned_loss=0.04415, over 18546.00 frames. ], tot_loss[loss=0.1675, simple_loss=0.2571, pruned_loss=0.03898, over 3530093.08 frames. ], batch size: 41, lr: 4.26e-03, grad_scale: 8.0
2022-12-24 01:16:44,497 WARNING [train.py:1060] (2/4) Exclude cut with ID 4133-6541-0027-40495-0_sp1.1 from training. Duration: 0.9681875
2022-12-24 01:16:47,290 WARNING [train.py:1060] (2/4) Exclude cut with ID 6330-62851-0022-91297-0_sp0.9 from training. Duration: 22.3166875
2022-12-24 01:16:52,429 WARNING [train.py:1060] (2/4) Exclude cut with ID 543-133212-0015-59917-0_sp0.9 from training. Duration: 21.8166875
2022-12-24 01:17:39,239 INFO [train.py:894] (2/4) Epoch 27, batch 650, loss[loss=0.1726, simple_loss=0.2704, pruned_loss=0.03738, over 18525.00 frames. ], tot_loss[loss=0.1671, simple_loss=0.2569, pruned_loss=0.0387, over 3570537.36 frames. ], batch size: 58, lr: 4.26e-03, grad_scale: 8.0
2022-12-24 01:18:34,279 WARNING [train.py:1060] (2/4) Exclude cut with ID 4957-30119-0041-23990-0_sp0.9 from training. Duration: 20.22775
2022-12-24 01:18:35,749 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 2.062e+02 3.215e+02 3.943e+02 4.935e+02 9.655e+02, threshold=7.887e+02, percent-clipped=2.0
2022-12-24 01:18:55,722 INFO [train.py:894] (2/4) Epoch 27, batch 700, loss[loss=0.1475, simple_loss=0.236, pruned_loss=0.02952, over 18461.00 frames. ], tot_loss[loss=0.1669, simple_loss=0.2566, pruned_loss=0.03861, over 3602182.00 frames. ], batch size: 42, lr: 4.26e-03, grad_scale: 8.0
2022-12-24 01:19:10,958 INFO [scaling.py:679] (2/4) Whitening: num_groups=8, num_channels=192, metric=1.93 vs. limit=2.0
2022-12-24 01:19:15,961 WARNING [train.py:1060] (2/4) Exclude cut with ID 5239-32139-0047-9341-0_sp1.1 from training. Duration: 24.67275
2022-12-24 01:19:37,224 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([2.7735, 2.3493, 1.8431, 2.6589, 2.1644, 2.3406, 2.2187, 2.8130],
       device='cuda:2'), covar=tensor([0.2096, 0.3464, 0.1966, 0.2710, 0.3698, 0.1058, 0.3159, 0.0886],
       device='cuda:2'), in_proj_covar=tensor([0.0302, 0.0299, 0.0254, 0.0349, 0.0281, 0.0236, 0.0298, 0.0222],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0003, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-24 01:19:41,313 INFO [zipformer.py:660] (2/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=91891.0, num_to_drop=0, layers_to_drop=set()
2022-12-24 01:19:42,970 WARNING [train.py:1060] (2/4) Exclude cut with ID 3082-165428-0081-50734-0_sp0.9 from training. Duration: 21.8055625
2022-12-24 01:19:51,792 INFO [zipformer.py:660] (2/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=91898.0, num_to_drop=1, layers_to_drop={3}
2022-12-24 01:19:59,429 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([2.0459, 2.5850, 2.1552, 3.1309, 3.1254, 1.9196, 2.1845, 1.4879],
       device='cuda:2'), covar=tensor([0.1682, 0.1497, 0.1286, 0.0805, 0.1183, 0.1002, 0.1722, 0.1533],
       device='cuda:2'), in_proj_covar=tensor([0.0250, 0.0230, 0.0221, 0.0203, 0.0264, 0.0198, 0.0228, 0.0204],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-24 01:20:09,752 INFO [train.py:894] (2/4) Epoch 27, batch 750, loss[loss=0.1319, simple_loss=0.218, pruned_loss=0.02294, over 18408.00 frames. ], tot_loss[loss=0.1676, simple_loss=0.2573, pruned_loss=0.03894, over 3627390.20 frames. ], batch size: 42, lr: 4.26e-03, grad_scale: 8.0
2022-12-24 01:20:19,989 WARNING [train.py:1060] (2/4) Exclude cut with ID 3340-169293-0054-76830-0_sp0.9 from training. Duration: 22.6666875
2022-12-24 01:20:57,088 INFO [zipformer.py:660] (2/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=91941.0, num_to_drop=0, layers_to_drop=set()
2022-12-24 01:21:05,129 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 1.943e+02 3.317e+02 4.158e+02 4.971e+02 2.653e+03, threshold=8.315e+02, percent-clipped=1.0
2022-12-24 01:21:13,052 INFO [zipformer.py:660] (2/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=91952.0, num_to_drop=0, layers_to_drop=set()
2022-12-24 01:21:14,328 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.8453, 1.2416, 0.9211, 1.5224, 2.2930, 1.4978, 1.5005, 1.7073],
       device='cuda:2'), covar=tensor([0.1690, 0.2423, 0.2237, 0.1601, 0.1617, 0.1841, 0.1678, 0.1953],
       device='cuda:2'), in_proj_covar=tensor([0.0094, 0.0096, 0.0115, 0.0095, 0.0119, 0.0092, 0.0098, 0.0093],
       device='cuda:2'), out_proj_covar=tensor([0.0003, 0.0003, 0.0004, 0.0003, 0.0004, 0.0003, 0.0003, 0.0003],
       device='cuda:2')
2022-12-24 01:21:17,940 INFO [zipformer.py:660] (2/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=91955.0, num_to_drop=0, layers_to_drop=set()
2022-12-24 01:21:20,506 WARNING [train.py:1060] (2/4) Exclude cut with ID 2411-132532-0017-82279-0_sp1.1 from training. Duration: 0.9681875
2022-12-24 01:21:24,660 INFO [train.py:894] (2/4) Epoch 27, batch 800, loss[loss=0.1812, simple_loss=0.2738, pruned_loss=0.0443, over 18509.00 frames. ], tot_loss[loss=0.1681, simple_loss=0.2583, pruned_loss=0.03894, over 3645996.10 frames. ], batch size: 52, lr: 4.26e-03, grad_scale: 8.0
2022-12-24 01:21:42,764 WARNING [train.py:1060] (2/4) Exclude cut with ID 6330-62850-0007-91323-0 from training. Duration: 22.485
2022-12-24 01:21:57,228 INFO [zipformer.py:660] (2/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=91981.0, num_to_drop=1, layers_to_drop={0}
2022-12-24 01:22:09,359 INFO [zipformer.py:660] (2/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=91989.0, num_to_drop=0, layers_to_drop=set()
2022-12-24 01:22:19,916 WARNING [train.py:1060] (2/4) Exclude cut with ID 3972-170212-0014-23379-0_sp1.1 from training. Duration: 23.82275
2022-12-24 01:22:35,479 WARNING [train.py:1060] (2/4) Exclude cut with ID 4860-13185-0032-76709-0 from training. Duration: 20.77
2022-12-24 01:22:43,243 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.8478, 1.8025, 1.7045, 1.8659, 1.3412, 4.9582, 1.8005, 2.2725],
       device='cuda:2'), covar=tensor([0.3141, 0.2061, 0.2042, 0.2029, 0.1463, 0.0086, 0.1561, 0.0870],
       device='cuda:2'), in_proj_covar=tensor([0.0134, 0.0119, 0.0126, 0.0123, 0.0107, 0.0097, 0.0091, 0.0090],
       device='cuda:2'), out_proj_covar=tensor([0.0005, 0.0005, 0.0005, 0.0005, 0.0004, 0.0003, 0.0004, 0.0004],
       device='cuda:2')
2022-12-24 01:22:44,322 INFO [train.py:894] (2/4) Epoch 27, batch 850, loss[loss=0.1743, simple_loss=0.2703, pruned_loss=0.0391, over 18579.00 frames. ], tot_loss[loss=0.1676, simple_loss=0.2575, pruned_loss=0.03883, over 3661003.18 frames. ], batch size: 57, lr: 4.25e-03, grad_scale: 8.0
2022-12-24 01:22:44,384 WARNING [train.py:1060] (2/4) Exclude cut with ID 6426-64292-0017-15984-0_sp0.9 from training. Duration: 24.088875
2022-12-24 01:22:50,500 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([2.0880, 1.4037, 1.8155, 1.7464, 2.1211, 2.1596, 1.9489, 1.8364],
       device='cuda:2'), covar=tensor([0.2323, 0.3435, 0.2729, 0.3124, 0.2169, 0.1011, 0.3401, 0.1385],
       device='cuda:2'), in_proj_covar=tensor([0.0271, 0.0299, 0.0286, 0.0325, 0.0315, 0.0257, 0.0351, 0.0247],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-24 01:22:53,408 INFO [zipformer.py:660] (2/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=92016.0, num_to_drop=0, layers_to_drop=set()
2022-12-24 01:23:15,948 WARNING [train.py:1060] (2/4) Exclude cut with ID 6330-62850-0007-91323-0_sp1.1 from training. Duration: 20.4409375
2022-12-24 01:23:32,584 INFO [zipformer.py:660] (2/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=92042.0, num_to_drop=1, layers_to_drop={0}
2022-12-24 01:23:40,047 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 1.966e+02 3.267e+02 4.027e+02 4.878e+02 1.049e+03, threshold=8.054e+02, percent-clipped=5.0
2022-12-24 01:23:51,088 INFO [scaling.py:679] (2/4) Whitening: num_groups=8, num_channels=96, metric=1.16 vs. limit=2.0
2022-12-24 01:23:58,930 INFO [train.py:894] (2/4) Epoch 27, batch 900, loss[loss=0.1607, simple_loss=0.2564, pruned_loss=0.0325, over 18631.00 frames. ], tot_loss[loss=0.1676, simple_loss=0.2575, pruned_loss=0.03887, over 3672770.12 frames. ], batch size: 53, lr: 4.25e-03, grad_scale: 8.0
2022-12-24 01:24:23,292 INFO [scaling.py:679] (2/4) Whitening: num_groups=8, num_channels=96, metric=1.59 vs. limit=2.0
2022-12-24 01:24:32,540 WARNING [train.py:1060] (2/4) Exclude cut with ID 2929-85685-0062-62366-0_sp0.9 from training. Duration: 22.511125
2022-12-24 01:24:32,564 WARNING [train.py:1060] (2/4) Exclude cut with ID 7395-89880-0031-39906-0 from training. Duration: 20.675
2022-12-24 01:24:41,185 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([3.3600, 4.3729, 4.0429, 2.6644, 4.4113, 3.4724, 1.7090, 3.0853],
       device='cuda:2'), covar=tensor([0.2483, 0.1148, 0.1326, 0.2589, 0.0705, 0.0710, 0.3715, 0.1213],
       device='cuda:2'), in_proj_covar=tensor([0.0151, 0.0147, 0.0160, 0.0124, 0.0151, 0.0116, 0.0144, 0.0115],
       device='cuda:2'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0002, 0.0003, 0.0002, 0.0003, 0.0002],
       device='cuda:2')
2022-12-24 01:25:09,334 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.6804, 1.6628, 1.5221, 1.6437, 1.9396, 1.9056, 1.9443, 1.2834],
       device='cuda:2'), covar=tensor([0.0355, 0.0299, 0.0537, 0.0215, 0.0199, 0.0461, 0.0310, 0.0377],
       device='cuda:2'), in_proj_covar=tensor([0.0095, 0.0129, 0.0155, 0.0123, 0.0118, 0.0123, 0.0101, 0.0129],
       device='cuda:2'), out_proj_covar=tensor([7.5475e-05, 1.0145e-04, 1.2667e-04, 9.7036e-05, 9.4168e-05, 9.4031e-05,
        7.8501e-05, 1.0151e-04], device='cuda:2')
2022-12-24 01:25:11,520 INFO [train.py:894] (2/4) Epoch 27, batch 950, loss[loss=0.1899, simple_loss=0.2811, pruned_loss=0.04931, over 18636.00 frames. ], tot_loss[loss=0.1681, simple_loss=0.2579, pruned_loss=0.03914, over 3682572.85 frames. ], batch size: 53, lr: 4.25e-03, grad_scale: 16.0
2022-12-24 01:26:09,183 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 2.451e+02 3.241e+02 3.825e+02 4.834e+02 9.407e+02, threshold=7.649e+02, percent-clipped=3.0
2022-12-24 01:26:09,262 WARNING [train.py:1060] (2/4) Exclude cut with ID 6330-62850-0007-91323-0_sp0.9 from training. Duration: 24.9833125
2022-12-24 01:26:28,198 INFO [train.py:894] (2/4) Epoch 27, batch 1000, loss[loss=0.1758, simple_loss=0.2665, pruned_loss=0.04252, over 18638.00 frames. ], tot_loss[loss=0.1692, simple_loss=0.259, pruned_loss=0.03973, over 3688856.57 frames. ], batch size: 98, lr: 4.25e-03, grad_scale: 16.0
2022-12-24 01:26:38,564 WARNING [train.py:1060] (2/4) Exclude cut with ID 5239-32139-0047-9341-0 from training. Duration: 27.14
2022-12-24 01:26:55,347 WARNING [train.py:1060] (2/4) Exclude cut with ID 2929-85685-0044-62348-0 from training. Duration: 22.44
2022-12-24 01:27:16,166 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([2.3216, 2.0158, 1.6360, 2.0055, 1.8257, 2.0765, 1.8917, 2.1867],
       device='cuda:2'), covar=tensor([0.2514, 0.3486, 0.2182, 0.2922, 0.3939, 0.1168, 0.3358, 0.1145],
       device='cuda:2'), in_proj_covar=tensor([0.0301, 0.0299, 0.0254, 0.0349, 0.0281, 0.0235, 0.0298, 0.0222],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0003, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-24 01:27:24,996 INFO [zipformer.py:660] (2/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=92198.0, num_to_drop=1, layers_to_drop={1}
2022-12-24 01:27:42,726 INFO [train.py:894] (2/4) Epoch 27, batch 1050, loss[loss=0.1257, simple_loss=0.2081, pruned_loss=0.02162, over 18496.00 frames. ], tot_loss[loss=0.1693, simple_loss=0.2591, pruned_loss=0.03976, over 3694304.17 frames. ], batch size: 43, lr: 4.25e-03, grad_scale: 16.0
2022-12-24 01:28:11,485 WARNING [train.py:1060] (2/4) Exclude cut with ID 2929-85685-0060-62364-0_sp0.9 from training. Duration: 21.361125
2022-12-24 01:28:17,143 WARNING [train.py:1060] (2/4) Exclude cut with ID 2929-85685-0079-62383-0_sp1.1 from training. Duration: 27.0318125
2022-12-24 01:28:27,878 WARNING [train.py:1060] (2/4) Exclude cut with ID 5622-44585-0006-90525-0_sp0.9 from training. Duration: 28.638875
2022-12-24 01:28:36,239 INFO [zipformer.py:660] (2/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=92246.0, num_to_drop=1, layers_to_drop={1}
2022-12-24 01:28:37,440 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 2.257e+02 3.236e+02 3.952e+02 5.116e+02 1.564e+03, threshold=7.904e+02, percent-clipped=6.0
2022-12-24 01:28:37,641 INFO [zipformer.py:660] (2/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=92247.0, num_to_drop=0, layers_to_drop=set()
2022-12-24 01:28:43,267 WARNING [train.py:1060] (2/4) Exclude cut with ID 3340-169293-0054-76830-0 from training. Duration: 20.4
2022-12-24 01:28:56,545 INFO [train.py:894] (2/4) Epoch 27, batch 1100, loss[loss=0.159, simple_loss=0.2321, pruned_loss=0.04295, over 18545.00 frames. ], tot_loss[loss=0.169, simple_loss=0.259, pruned_loss=0.03952, over 3698302.83 frames. ], batch size: 44, lr: 4.25e-03, grad_scale: 16.0
2022-12-24 01:29:17,733 WARNING [train.py:1060] (2/4) Exclude cut with ID 2929-85685-0071-62375-0 from training. Duration: 20.025
2022-12-24 01:29:17,747 WARNING [train.py:1060] (2/4) Exclude cut with ID 2364-131735-0112-64612-0_sp0.9 from training. Duration: 20.488875
2022-12-24 01:29:23,654 WARNING [train.py:1060] (2/4) Exclude cut with ID 2929-85685-0079-62383-0 from training. Duration: 29.735
2022-12-24 01:29:51,387 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([2.2439, 2.2008, 1.6199, 2.4742, 2.3601, 2.1375, 2.8974, 2.2184],
       device='cuda:2'), covar=tensor([0.0920, 0.1717, 0.2834, 0.1750, 0.1783, 0.0880, 0.0933, 0.1311],
       device='cuda:2'), in_proj_covar=tensor([0.0183, 0.0216, 0.0256, 0.0292, 0.0243, 0.0195, 0.0208, 0.0210],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-24 01:30:11,170 INFO [scaling.py:679] (2/4) Whitening: num_groups=1, num_channels=384, metric=2.48 vs. limit=5.0
2022-12-24 01:30:11,848 INFO [train.py:894] (2/4) Epoch 27, batch 1150, loss[loss=0.1794, simple_loss=0.2788, pruned_loss=0.04005, over 18567.00 frames. ], tot_loss[loss=0.1692, simple_loss=0.2591, pruned_loss=0.03967, over 3701543.25 frames. ], batch size: 57, lr: 4.25e-03, grad_scale: 8.0
2022-12-24 01:30:13,481 INFO [zipformer.py:660] (2/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=92311.0, num_to_drop=0, layers_to_drop=set()
2022-12-24 01:30:14,813 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([2.8837, 3.9827, 3.7235, 1.4818, 4.0435, 2.9536, 0.6126, 2.5939],
       device='cuda:2'), covar=tensor([0.2025, 0.1080, 0.1469, 0.3843, 0.0740, 0.0943, 0.5091, 0.1590],
       device='cuda:2'), in_proj_covar=tensor([0.0150, 0.0146, 0.0159, 0.0125, 0.0150, 0.0116, 0.0144, 0.0114],
       device='cuda:2'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0002, 0.0003, 0.0002, 0.0003, 0.0002],
       device='cuda:2')
2022-12-24 01:30:42,507 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.9936, 1.9788, 1.3292, 2.2195, 2.0889, 1.7820, 2.9066, 2.0256],
       device='cuda:2'), covar=tensor([0.1042, 0.1855, 0.3230, 0.2030, 0.2044, 0.1137, 0.0896, 0.1505],
       device='cuda:2'), in_proj_covar=tensor([0.0184, 0.0218, 0.0258, 0.0295, 0.0245, 0.0196, 0.0210, 0.0211],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-24 01:30:43,480 WARNING [train.py:1060] (2/4) Exclude cut with ID 7276-92427-0014-12983-0_sp0.9 from training. Duration: 21.3055625
2022-12-24 01:30:44,933 WARNING [train.py:1060] (2/4) Exclude cut with ID 1025-75365-0008-79168-0_sp0.9 from training. Duration: 22.0666875
2022-12-24 01:30:53,162 INFO [zipformer.py:660] (2/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=92337.0, num_to_drop=1, layers_to_drop={3}
2022-12-24 01:30:56,798 INFO [scaling.py:679] (2/4) Whitening: num_groups=8, num_channels=96, metric=1.30 vs. limit=2.0
2022-12-24 01:31:08,644 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 2.090e+02 3.040e+02 3.627e+02 4.393e+02 8.345e+02, threshold=7.254e+02, percent-clipped=1.0
2022-12-24 01:31:26,210 INFO [train.py:894] (2/4) Epoch 27, batch 1200, loss[loss=0.1381, simple_loss=0.2267, pruned_loss=0.02471, over 18684.00 frames. ], tot_loss[loss=0.168, simple_loss=0.258, pruned_loss=0.03898, over 3703997.42 frames. ], batch size: 46, lr: 4.25e-03, grad_scale: 8.0
2022-12-24 01:32:35,672 WARNING [train.py:1060] (2/4) Exclude cut with ID 2929-85685-0062-62366-0 from training. Duration: 20.26
2022-12-24 01:32:42,010 INFO [train.py:894] (2/4) Epoch 27, batch 1250, loss[loss=0.2045, simple_loss=0.2912, pruned_loss=0.05885, over 18615.00 frames. ], tot_loss[loss=0.1672, simple_loss=0.257, pruned_loss=0.03864, over 3706331.51 frames. ], batch size: 181, lr: 4.24e-03, grad_scale: 8.0
2022-12-24 01:32:50,831 WARNING [train.py:1060] (2/4) Exclude cut with ID 5239-32139-0030-9324-0_sp0.9 from training. Duration: 21.3444375
2022-12-24 01:33:39,384 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 1.696e+02 3.119e+02 3.679e+02 4.523e+02 1.291e+03, threshold=7.358e+02, percent-clipped=5.0
2022-12-24 01:33:45,931 WARNING [train.py:1060] (2/4) Exclude cut with ID 497-129325-0061-62254-0_sp1.1 from training. Duration: 0.97725
2022-12-24 01:33:57,909 INFO [train.py:894] (2/4) Epoch 27, batch 1300, loss[loss=0.1881, simple_loss=0.2839, pruned_loss=0.0462, over 18596.00 frames. ], tot_loss[loss=0.1677, simple_loss=0.2574, pruned_loss=0.03901, over 3707933.54 frames. ], batch size: 56, lr: 4.24e-03, grad_scale: 8.0
2022-12-24 01:34:06,085 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.7110, 1.2458, 2.1866, 3.4689, 2.6595, 2.6463, 1.1722, 2.4622],
       device='cuda:2'), covar=tensor([0.1839, 0.1698, 0.1361, 0.0524, 0.0874, 0.1074, 0.2025, 0.0970],
       device='cuda:2'), in_proj_covar=tensor([0.0103, 0.0119, 0.0137, 0.0155, 0.0106, 0.0144, 0.0129, 0.0116],
       device='cuda:2'), out_proj_covar=tensor([0.0003, 0.0003, 0.0004, 0.0004, 0.0003, 0.0004, 0.0004, 0.0003],
       device='cuda:2')
2022-12-24 01:34:26,262 INFO [scaling.py:679] (2/4) Whitening: num_groups=8, num_channels=192, metric=1.84 vs. limit=2.0
2022-12-24 01:34:26,812 WARNING [train.py:1060] (2/4) Exclude cut with ID 7395-89880-0031-39906-0_sp0.9 from training. Duration: 22.97225
2022-12-24 01:34:45,556 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([3.1076, 2.4361, 1.8701, 2.7429, 2.2571, 2.5627, 2.4805, 3.2881],
       device='cuda:2'), covar=tensor([0.1993, 0.3384, 0.1951, 0.3238, 0.3984, 0.0995, 0.3232, 0.0844],
       device='cuda:2'), in_proj_covar=tensor([0.0298, 0.0297, 0.0251, 0.0345, 0.0278, 0.0233, 0.0295, 0.0219],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-24 01:34:57,190 INFO [scaling.py:679] (2/4) Whitening: num_groups=8, num_channels=192, metric=1.83 vs. limit=2.0
2022-12-24 01:34:59,201 WARNING [train.py:1060] (2/4) Exclude cut with ID 7395-89880-0047-39922-0_sp0.9 from training. Duration: 21.97775
2022-12-24 01:35:10,055 INFO [scaling.py:679] (2/4) Whitening: num_groups=1, num_channels=384, metric=2.72 vs. limit=5.0
2022-12-24 01:35:12,397 WARNING [train.py:1060] (2/4) Exclude cut with ID 1112-1043-0006-89194-0_sp0.9 from training. Duration: 21.8333125
2022-12-24 01:35:14,357 INFO [train.py:894] (2/4) Epoch 27, batch 1350, loss[loss=0.1735, simple_loss=0.2662, pruned_loss=0.04042, over 18649.00 frames. ], tot_loss[loss=0.1674, simple_loss=0.2574, pruned_loss=0.03871, over 3708989.45 frames. ], batch size: 62, lr: 4.24e-03, grad_scale: 8.0
2022-12-24 01:35:22,690 WARNING [train.py:1060] (2/4) Exclude cut with ID 1914-133440-0031-94921-0 from training. Duration: 20.47
2022-12-24 01:36:11,730 INFO [zipformer.py:660] (2/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=92547.0, num_to_drop=0, layers_to_drop=set()
2022-12-24 01:36:12,950 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 1.794e+02 3.181e+02 4.006e+02 5.114e+02 1.464e+03, threshold=8.011e+02, percent-clipped=2.0
2022-12-24 01:36:27,936 WARNING [train.py:1060] (2/4) Exclude cut with ID 7395-89880-0037-39912-0_sp0.9 from training. Duration: 20.67225
2022-12-24 01:36:30,561 INFO [train.py:894] (2/4) Epoch 27, batch 1400, loss[loss=0.2013, simple_loss=0.2752, pruned_loss=0.06364, over 18525.00 frames. ], tot_loss[loss=0.1687, simple_loss=0.2586, pruned_loss=0.03943, over 3710965.73 frames. ], batch size: 47, lr: 4.24e-03, grad_scale: 8.0
2022-12-24 01:36:46,349 WARNING [train.py:1060] (2/4) Exclude cut with ID 1914-133440-0024-94914-0_sp0.9 from training. Duration: 25.2444375
2022-12-24 01:37:07,445 WARNING [train.py:1060] (2/4) Exclude cut with ID 3340-169293-0021-76797-0_sp0.9 from training. Duration: 21.1445
2022-12-24 01:37:22,424 INFO [zipformer.py:660] (2/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=92595.0, num_to_drop=0, layers_to_drop=set()
2022-12-24 01:37:34,322 INFO [scaling.py:679] (2/4) Whitening: num_groups=8, num_channels=96, metric=1.30 vs. limit=2.0
2022-12-24 01:37:44,832 INFO [train.py:894] (2/4) Epoch 27, batch 1450, loss[loss=0.1813, simple_loss=0.2729, pruned_loss=0.04486, over 18658.00 frames. ], tot_loss[loss=0.1692, simple_loss=0.2593, pruned_loss=0.03955, over 3712161.87 frames. ], batch size: 62, lr: 4.24e-03, grad_scale: 8.0
2022-12-24 01:37:46,744 INFO [zipformer.py:660] (2/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=92611.0, num_to_drop=0, layers_to_drop=set()
2022-12-24 01:38:08,369 INFO [zipformer.py:660] (2/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=92625.0, num_to_drop=0, layers_to_drop=set()
2022-12-24 01:38:22,189 WARNING [train.py:1060] (2/4) Exclude cut with ID 2929-85685-0079-62383-0_sp0.9 from training. Duration: 33.038875
2022-12-24 01:38:25,140 INFO [zipformer.py:660] (2/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=92637.0, num_to_drop=1, layers_to_drop={1}
2022-12-24 01:38:41,491 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 2.186e+02 3.217e+02 4.081e+02 5.000e+02 1.032e+03, threshold=8.163e+02, percent-clipped=4.0
2022-12-24 01:38:48,152 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([2.0483, 2.0866, 2.4101, 1.3918, 2.4209, 2.3934, 1.7381, 2.8061],
       device='cuda:2'), covar=tensor([0.1235, 0.1847, 0.1342, 0.2060, 0.0699, 0.1126, 0.2342, 0.0533],
       device='cuda:2'), in_proj_covar=tensor([0.0199, 0.0216, 0.0208, 0.0196, 0.0171, 0.0219, 0.0218, 0.0193],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-24 01:38:58,298 INFO [zipformer.py:660] (2/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=92659.0, num_to_drop=0, layers_to_drop=set()
2022-12-24 01:38:59,530 INFO [train.py:894] (2/4) Epoch 27, batch 1500, loss[loss=0.1805, simple_loss=0.2804, pruned_loss=0.04029, over 18545.00 frames. ], tot_loss[loss=0.1686, simple_loss=0.2589, pruned_loss=0.03909, over 3712236.26 frames. ], batch size: 55, lr: 4.24e-03, grad_scale: 8.0
2022-12-24 01:38:59,626 WARNING [train.py:1060] (2/4) Exclude cut with ID 6426-64291-0000-16059-0_sp0.9 from training. Duration: 20.0944375
2022-12-24 01:39:12,820 WARNING [train.py:1060] (2/4) Exclude cut with ID 2929-85685-0044-62348-0_sp1.1 from training. Duration: 20.4
2022-12-24 01:39:20,663 WARNING [train.py:1060] (2/4) Exclude cut with ID 6330-62851-0022-91297-0 from training. Duration: 20.085
2022-12-24 01:39:32,586 WARNING [train.py:1060] (2/4) Exclude cut with ID 4860-13185-0032-76709-0_sp0.9 from training. Duration: 23.07775
2022-12-24 01:39:36,985 INFO [zipformer.py:660] (2/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=92685.0, num_to_drop=1, layers_to_drop={1}
2022-12-24 01:39:38,442 INFO [zipformer.py:660] (2/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=92686.0, num_to_drop=0, layers_to_drop=set()
2022-12-24 01:40:14,661 INFO [train.py:894] (2/4) Epoch 27, batch 1550, loss[loss=0.1826, simple_loss=0.2776, pruned_loss=0.04382, over 18674.00 frames. ], tot_loss[loss=0.168, simple_loss=0.2583, pruned_loss=0.03882, over 3711927.88 frames. ], batch size: 62, lr: 4.24e-03, grad_scale: 8.0
2022-12-24 01:40:17,477 WARNING [train.py:1060] (2/4) Exclude cut with ID 2929-85685-0044-62348-0_sp0.9 from training. Duration: 24.9333125
2022-12-24 01:40:29,838 INFO [zipformer.py:660] (2/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=92720.0, num_to_drop=0, layers_to_drop=set()
2022-12-24 01:41:00,888 WARNING [train.py:1060] (2/4) Exclude cut with ID 5118-111612-0016-124680-0_sp0.9 from training. Duration: 20.388875
2022-12-24 01:41:08,071 WARNING [train.py:1060] (2/4) Exclude cut with ID 432-122774-0017-62487-0_sp1.1 from training. Duration: 20.3590625
2022-12-24 01:41:11,048 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 2.280e+02 3.341e+02 4.064e+02 5.045e+02 9.351e+02, threshold=8.127e+02, percent-clipped=1.0
2022-12-24 01:41:29,804 INFO [train.py:894] (2/4) Epoch 27, batch 1600, loss[loss=0.1695, simple_loss=0.2588, pruned_loss=0.04012, over 18663.00 frames. ], tot_loss[loss=0.1669, simple_loss=0.2571, pruned_loss=0.03835, over 3712114.06 frames. ], batch size: 60, lr: 4.24e-03, grad_scale: 8.0
2022-12-24 01:42:02,896 INFO [zipformer.py:660] (2/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=92781.0, num_to_drop=0, layers_to_drop=set()
2022-12-24 01:42:07,279 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([2.3548, 2.5288, 2.8945, 1.8688, 2.8061, 2.7563, 2.0124, 3.0853],
       device='cuda:2'), covar=tensor([0.1267, 0.1607, 0.1538, 0.2128, 0.0670, 0.1168, 0.2219, 0.0547],
       device='cuda:2'), in_proj_covar=tensor([0.0196, 0.0214, 0.0206, 0.0194, 0.0169, 0.0216, 0.0215, 0.0191],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-24 01:42:15,973 WARNING [train.py:1060] (2/4) Exclude cut with ID 3557-8342-0013-54691-0_sp1.1 from training. Duration: 0.836375
2022-12-24 01:42:46,229 INFO [train.py:894] (2/4) Epoch 27, batch 1650, loss[loss=0.1672, simple_loss=0.2564, pruned_loss=0.03903, over 18523.00 frames. ], tot_loss[loss=0.168, simple_loss=0.2579, pruned_loss=0.039, over 3713585.72 frames. ], batch size: 47, lr: 4.24e-03, grad_scale: 8.0
2022-12-24 01:42:59,483 WARNING [train.py:1060] (2/4) Exclude cut with ID 8565-290391-0049-67394-0_sp0.9 from training. Duration: 21.3166875
2022-12-24 01:43:06,810 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.6750, 1.3623, 2.3594, 3.4525, 2.6019, 2.5705, 1.1857, 2.5479],
       device='cuda:2'), covar=tensor([0.1778, 0.1647, 0.1267, 0.0609, 0.0888, 0.1165, 0.2036, 0.0949],
       device='cuda:2'), in_proj_covar=tensor([0.0102, 0.0118, 0.0136, 0.0155, 0.0106, 0.0142, 0.0128, 0.0115],
       device='cuda:2'), out_proj_covar=tensor([0.0003, 0.0003, 0.0004, 0.0004, 0.0003, 0.0004, 0.0003, 0.0003],
       device='cuda:2')
2022-12-24 01:43:17,866 INFO [zipformer.py:660] (2/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=92831.0, num_to_drop=0, layers_to_drop=set()
2022-12-24 01:43:28,935 WARNING [train.py:1060] (2/4) Exclude cut with ID 6533-399-0029-104863-0_sp0.9 from training. Duration: 22.1055625
2022-12-24 01:43:42,346 WARNING [train.py:1060] (2/4) Exclude cut with ID 7699-105389-0094-26379-0_sp1.1 from training. Duration: 21.77725
2022-12-24 01:43:43,754 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 1.715e+02 3.368e+02 3.937e+02 4.829e+02 1.412e+03, threshold=7.874e+02, percent-clipped=4.0
2022-12-24 01:44:01,292 INFO [train.py:894] (2/4) Epoch 27, batch 1700, loss[loss=0.1339, simple_loss=0.2144, pruned_loss=0.02676, over 18483.00 frames. ], tot_loss[loss=0.1684, simple_loss=0.2571, pruned_loss=0.03986, over 3712972.18 frames. ], batch size: 43, lr: 4.23e-03, grad_scale: 8.0
2022-12-24 01:44:02,838 WARNING [train.py:1060] (2/4) Exclude cut with ID 7255-291500-0005-134304-0_sp0.9 from training. Duration: 27.8166875
2022-12-24 01:44:28,025 WARNING [train.py:1060] (2/4) Exclude cut with ID 7357-94126-0021-15852-0_sp1.1 from training. Duration: 22.5090625
2022-12-24 01:44:35,981 WARNING [train.py:1060] (2/4) Exclude cut with ID 7255-291500-0005-134304-0 from training. Duration: 25.035
2022-12-24 01:44:50,123 INFO [zipformer.py:660] (2/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=92892.0, num_to_drop=0, layers_to_drop=set()
2022-12-24 01:44:53,837 WARNING [train.py:1060] (2/4) Exclude cut with ID 774-127930-0014-10412-0_sp1.1 from training. Duration: 0.95
2022-12-24 01:45:00,617 INFO [zipformer.py:660] (2/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=92899.0, num_to_drop=0, layers_to_drop=set()
2022-12-24 01:45:12,260 WARNING [train.py:1060] (2/4) Exclude cut with ID 3033-130750-0096-55598-0_sp0.9 from training. Duration: 0.92225
2022-12-24 01:45:16,471 INFO [train.py:894] (2/4) Epoch 27, batch 1750, loss[loss=0.1978, simple_loss=0.2814, pruned_loss=0.05716, over 18587.00 frames. ], tot_loss[loss=0.1701, simple_loss=0.258, pruned_loss=0.04113, over 3712787.05 frames. ], batch size: 56, lr: 4.23e-03, grad_scale: 8.0
2022-12-24 01:45:38,282 WARNING [train.py:1060] (2/4) Exclude cut with ID 4511-76322-0006-80011-0 from training. Duration: 21.97
2022-12-24 01:45:47,643 INFO [zipformer.py:660] (2/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=92931.0, num_to_drop=0, layers_to_drop=set()
2022-12-24 01:45:56,571 WARNING [train.py:1060] (2/4) Exclude cut with ID 7492-105653-0055-62765-0_sp0.9 from training. Duration: 21.97225
2022-12-24 01:45:57,743 WARNING [train.py:1060] (2/4) Exclude cut with ID 453-131332-0000-47844-0_sp0.9 from training. Duration: 25.3333125
2022-12-24 01:46:09,062 WARNING [train.py:1060] (2/4) Exclude cut with ID 5172-29468-0015-19128-0_sp0.9 from training. Duration: 21.5055625
2022-12-24 01:46:13,294 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 2.049e+02 3.881e+02 4.504e+02 6.102e+02 9.952e+02, threshold=9.008e+02, percent-clipped=7.0
2022-12-24 01:46:17,522 WARNING [train.py:1060] (2/4) Exclude cut with ID 453-131332-0000-47844-0_sp1.1 from training. Duration: 20.72725
2022-12-24 01:46:30,517 INFO [train.py:894] (2/4) Epoch 27, batch 1800, loss[loss=0.1937, simple_loss=0.2823, pruned_loss=0.05255, over 18596.00 frames. ], tot_loss[loss=0.1726, simple_loss=0.2598, pruned_loss=0.04272, over 3713376.13 frames. ], batch size: 51, lr: 4.23e-03, grad_scale: 8.0
2022-12-24 01:46:30,996 INFO [zipformer.py:660] (2/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=92960.0, num_to_drop=0, layers_to_drop=set()
2022-12-24 01:46:43,968 INFO [scaling.py:679] (2/4) Whitening: num_groups=1, num_channels=384, metric=4.41 vs. limit=5.0
2022-12-24 01:46:47,748 WARNING [train.py:1060] (2/4) Exclude cut with ID 8631-249866-0030-130156-0_sp0.9 from training. Duration: 26.32775
2022-12-24 01:46:54,578 INFO [zipformer.py:660] (2/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=92975.0, num_to_drop=0, layers_to_drop=set()
2022-12-24 01:46:56,212 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([2.3384, 2.0474, 1.5664, 1.8535, 1.8935, 1.9967, 1.9203, 2.1450],
       device='cuda:2'), covar=tensor([0.2368, 0.3155, 0.2306, 0.2784, 0.3434, 0.1283, 0.3087, 0.1111],
       device='cuda:2'), in_proj_covar=tensor([0.0299, 0.0298, 0.0252, 0.0347, 0.0280, 0.0233, 0.0297, 0.0220],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-24 01:47:03,249 INFO [zipformer.py:660] (2/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=92981.0, num_to_drop=0, layers_to_drop=set()
2022-12-24 01:47:20,513 INFO [zipformer.py:660] (2/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=92992.0, num_to_drop=0, layers_to_drop=set()
2022-12-24 01:47:21,536 WARNING [train.py:1060] (2/4) Exclude cut with ID 3867-173237-0077-144769-0 from training. Duration: 20.025
2022-12-24 01:47:26,910 WARNING [train.py:1060] (2/4) Exclude cut with ID 6709-74022-0004-86860-0_sp1.1 from training. Duration: 0.9409375
2022-12-24 01:47:26,916 WARNING [train.py:1060] (2/4) Exclude cut with ID 4757-1811-0023-62229-0_sp0.9 from training. Duration: 21.37775
2022-12-24 01:47:46,308 INFO [train.py:894] (2/4) Epoch 27, batch 1850, loss[loss=0.1786, simple_loss=0.2639, pruned_loss=0.04663, over 18575.00 frames. ], tot_loss[loss=0.1739, simple_loss=0.2603, pruned_loss=0.04371, over 3713032.53 frames. ], batch size: 57, lr: 4.23e-03, grad_scale: 8.0
2022-12-24 01:47:47,363 INFO [zipformer.py:660] (2/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=93010.0, num_to_drop=0, layers_to_drop=set()
2022-12-24 01:47:48,561 WARNING [train.py:1060] (2/4) Exclude cut with ID 1250-135782-0004-25974-0_sp0.9 from training. Duration: 21.17225
2022-12-24 01:47:48,572 WARNING [train.py:1060] (2/4) Exclude cut with ID 7357-94126-0021-15852-0_sp0.9 from training. Duration: 27.511125
2022-12-24 01:48:21,114 WARNING [train.py:1060] (2/4) Exclude cut with ID 453-131332-0000-47844-0 from training. Duration: 22.8
2022-12-24 01:48:25,657 INFO [zipformer.py:660] (2/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=93036.0, num_to_drop=0, layers_to_drop=set()
2022-12-24 01:48:25,706 INFO [zipformer.py:660] (2/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=93036.0, num_to_drop=0, layers_to_drop=set()
2022-12-24 01:48:27,852 WARNING [train.py:1060] (2/4) Exclude cut with ID 4964-30587-0040-44509-0 from training. Duration: 22.585
2022-12-24 01:48:44,105 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 2.543e+02 3.991e+02 4.524e+02 5.614e+02 9.322e+02, threshold=9.048e+02, percent-clipped=2.0
2022-12-24 01:48:55,723 WARNING [train.py:1060] (2/4) Exclude cut with ID 6978-92210-0001-146967-0_sp0.9 from training. Duration: 22.0166875
2022-12-24 01:49:02,375 INFO [train.py:894] (2/4) Epoch 27, batch 1900, loss[loss=0.149, simple_loss=0.2292, pruned_loss=0.03437, over 18384.00 frames. ], tot_loss[loss=0.175, simple_loss=0.2606, pruned_loss=0.04465, over 3713019.29 frames. ], batch size: 46, lr: 4.23e-03, grad_scale: 8.0
2022-12-24 01:49:13,072 WARNING [train.py:1060] (2/4) Exclude cut with ID 7255-291500-0003-134302-0_sp1.1 from training. Duration: 24.395375
2022-12-24 01:49:14,923 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([2.5175, 2.1836, 1.7890, 2.2583, 1.9177, 2.1618, 2.0407, 2.3549],
       device='cuda:2'), covar=tensor([0.2062, 0.3167, 0.1937, 0.2546, 0.3609, 0.1064, 0.2918, 0.1001],
       device='cuda:2'), in_proj_covar=tensor([0.0299, 0.0297, 0.0252, 0.0346, 0.0279, 0.0233, 0.0296, 0.0219],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-24 01:49:19,707 INFO [zipformer.py:660] (2/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=93071.0, num_to_drop=0, layers_to_drop=set()
2022-12-24 01:49:22,232 WARNING [train.py:1060] (2/4) Exclude cut with ID 3488-85273-0017-41203-0_sp0.9 from training. Duration: 27.47775
2022-12-24 01:49:26,558 WARNING [train.py:1060] (2/4) Exclude cut with ID 432-122774-0017-62487-0_sp0.9 from training. Duration: 24.8833125
2022-12-24 01:49:26,692 INFO [zipformer.py:660] (2/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=93076.0, num_to_drop=0, layers_to_drop=set()
2022-12-24 01:49:29,430 WARNING [train.py:1060] (2/4) Exclude cut with ID 6758-72288-0033-108368-0 from training. Duration: 23.39
2022-12-24 01:49:35,181 WARNING [train.py:1060] (2/4) Exclude cut with ID 8291-282929-0007-12994-0_sp0.9 from training. Duration: 28.72225
2022-12-24 01:49:44,432 WARNING [train.py:1060] (2/4) Exclude cut with ID 585-294811-0110-133686-0_sp0.9 from training. Duration: 20.8944375
2022-12-24 01:49:58,055 INFO [zipformer.py:660] (2/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=93097.0, num_to_drop=1, layers_to_drop={3}
2022-12-24 01:49:59,128 WARNING [train.py:1060] (2/4) Exclude cut with ID 5796-66357-0007-116447-0_sp0.9 from training. Duration: 23.8444375
2022-12-24 01:50:16,857 INFO [train.py:894] (2/4) Epoch 27, batch 1950, loss[loss=0.1752, simple_loss=0.2632, pruned_loss=0.04364, over 18569.00 frames. ], tot_loss[loss=0.1767, simple_loss=0.2618, pruned_loss=0.04576, over 3713513.28 frames. ], batch size: 49, lr: 4.23e-03, grad_scale: 8.0
2022-12-24 01:50:22,626 WARNING [train.py:1060] (2/4) Exclude cut with ID 8291-282929-0007-12994-0 from training. Duration: 25.85
2022-12-24 01:50:22,633 WARNING [train.py:1060] (2/4) Exclude cut with ID 8291-282929-0023-13010-0 from training. Duration: 21.39
2022-12-24 01:50:24,342 INFO [zipformer.py:660] (2/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=93115.0, num_to_drop=0, layers_to_drop=set()
2022-12-24 01:50:33,437 WARNING [train.py:1060] (2/4) Exclude cut with ID 7357-94126-0014-15845-0 from training. Duration: 27.92
2022-12-24 01:51:02,998 WARNING [train.py:1060] (2/4) Exclude cut with ID 8631-249866-0039-130165-0_sp0.9 from training. Duration: 20.661125
2022-12-24 01:51:12,675 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 2.455e+02 4.181e+02 4.936e+02 5.974e+02 1.019e+03, threshold=9.872e+02, percent-clipped=2.0
2022-12-24 01:51:26,034 WARNING [train.py:1060] (2/4) Exclude cut with ID 7357-94126-0043-15874-0_sp0.9 from training. Duration: 20.07225
2022-12-24 01:51:31,590 INFO [train.py:894] (2/4) Epoch 27, batch 2000, loss[loss=0.2137, simple_loss=0.2943, pruned_loss=0.06656, over 18732.00 frames. ], tot_loss[loss=0.1777, simple_loss=0.2625, pruned_loss=0.0465, over 3713926.84 frames. ], batch size: 54, lr: 4.23e-03, grad_scale: 8.0
2022-12-24 01:51:31,616 WARNING [train.py:1060] (2/4) Exclude cut with ID 1085-156170-0017-128270-0 from training. Duration: 21.01
2022-12-24 01:51:56,030 INFO [zipformer.py:660] (2/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=93176.0, num_to_drop=0, layers_to_drop=set()
2022-12-24 01:52:11,692 INFO [zipformer.py:660] (2/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=93187.0, num_to_drop=0, layers_to_drop=set()
2022-12-24 01:52:42,716 WARNING [train.py:1060] (2/4) Exclude cut with ID 2195-150901-0045-59933-0 from training. Duration: 20.65
2022-12-24 01:52:45,662 INFO [train.py:894] (2/4) Epoch 27, batch 2050, loss[loss=0.1592, simple_loss=0.2435, pruned_loss=0.03744, over 18455.00 frames. ], tot_loss[loss=0.178, simple_loss=0.2624, pruned_loss=0.04675, over 3713991.75 frames. ], batch size: 50, lr: 4.23e-03, grad_scale: 8.0
2022-12-24 01:52:49,555 WARNING [train.py:1060] (2/4) Exclude cut with ID 5796-66357-0007-116447-0 from training. Duration: 21.46
2022-12-24 01:53:34,539 WARNING [train.py:1060] (2/4) Exclude cut with ID 3557-8342-0013-54691-0 from training. Duration: 0.92
2022-12-24 01:53:41,485 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 2.802e+02 4.249e+02 4.843e+02 5.680e+02 1.100e+03, threshold=9.685e+02, percent-clipped=2.0
2022-12-24 01:53:41,542 WARNING [train.py:1060] (2/4) Exclude cut with ID 8291-282929-0023-13010-0_sp0.9 from training. Duration: 23.7666875
2022-12-24 01:53:52,231 INFO [zipformer.py:660] (2/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=93255.0, num_to_drop=0, layers_to_drop=set()
2022-12-24 01:53:59,920 INFO [train.py:894] (2/4) Epoch 27, batch 2100, loss[loss=0.1523, simple_loss=0.2288, pruned_loss=0.03788, over 18485.00 frames. ], tot_loss[loss=0.178, simple_loss=0.2622, pruned_loss=0.04689, over 3714450.43 frames. ], batch size: 43, lr: 4.23e-03, grad_scale: 8.0
2022-12-24 01:54:18,163 WARNING [train.py:1060] (2/4) Exclude cut with ID 8544-281189-0060-101339-0_sp0.9 from training. Duration: 20.861125
2022-12-24 01:54:28,809 WARNING [train.py:1060] (2/4) Exclude cut with ID 3488-65654-0031-41259-0_sp0.9 from training. Duration: 22.711125
2022-12-24 01:54:31,920 INFO [zipformer.py:660] (2/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=93281.0, num_to_drop=0, layers_to_drop=set()
2022-12-24 01:54:40,835 INFO [zipformer.py:660] (2/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=93287.0, num_to_drop=0, layers_to_drop=set()
2022-12-24 01:55:10,156 WARNING [train.py:1060] (2/4) Exclude cut with ID 6951-79737-0043-132310-0_sp1.1 from training. Duration: 22.986375
2022-12-24 01:55:15,632 INFO [train.py:894] (2/4) Epoch 27, batch 2150, loss[loss=0.1965, simple_loss=0.2719, pruned_loss=0.06052, over 18457.00 frames. ], tot_loss[loss=0.177, simple_loss=0.2612, pruned_loss=0.04643, over 3714291.91 frames. ], batch size: 64, lr: 4.22e-03, grad_scale: 8.0
2022-12-24 01:55:22,473 INFO [scaling.py:679] (2/4) Whitening: num_groups=8, num_channels=192, metric=1.88 vs. limit=2.0
2022-12-24 01:55:27,399 WARNING [train.py:1060] (2/4) Exclude cut with ID 8040-260924-0003-80960-0_sp0.9 from training. Duration: 22.07225
2022-12-24 01:55:32,669 WARNING [train.py:1060] (2/4) Exclude cut with ID 7699-105389-0045-26330-0_sp0.9 from training. Duration: 20.3055625
2022-12-24 01:55:35,669 WARNING [train.py:1060] (2/4) Exclude cut with ID 6356-271890-0060-94317-0_sp0.9 from training. Duration: 20.72225
2022-12-24 01:55:44,491 INFO [zipformer.py:660] (2/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=93329.0, num_to_drop=0, layers_to_drop=set()
2022-12-24 01:55:47,317 INFO [zipformer.py:660] (2/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=93331.0, num_to_drop=0, layers_to_drop=set()
2022-12-24 01:55:54,887 WARNING [train.py:1060] (2/4) Exclude cut with ID 3488-85273-0017-41203-0_sp1.1 from training. Duration: 22.4818125
2022-12-24 01:56:11,860 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 2.853e+02 4.006e+02 4.754e+02 5.820e+02 9.095e+02, threshold=9.509e+02, percent-clipped=0.0
2022-12-24 01:56:20,577 WARNING [train.py:1060] (2/4) Exclude cut with ID 4964-30587-0040-44509-0_sp0.9 from training. Duration: 25.0944375
2022-12-24 01:56:23,595 WARNING [train.py:1060] (2/4) Exclude cut with ID 6533-399-0047-104881-0 from training. Duration: 21.515
2022-12-24 01:56:29,909 WARNING [train.py:1060] (2/4) Exclude cut with ID 7357-94126-0009-15840-0_sp0.9 from training. Duration: 27.02225
2022-12-24 01:56:31,364 INFO [train.py:894] (2/4) Epoch 27, batch 2200, loss[loss=0.169, simple_loss=0.2474, pruned_loss=0.04535, over 18364.00 frames. ], tot_loss[loss=0.1767, simple_loss=0.2607, pruned_loss=0.04631, over 3714379.68 frames. ], batch size: 46, lr: 4.22e-03, grad_scale: 8.0
2022-12-24 01:56:35,888 WARNING [train.py:1060] (2/4) Exclude cut with ID 432-122774-0010-62480-0_sp0.9 from training. Duration: 22.22225
2022-12-24 01:56:41,215 INFO [zipformer.py:660] (2/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=93366.0, num_to_drop=0, layers_to_drop=set()
2022-12-24 01:56:42,985 WARNING [train.py:1060] (2/4) Exclude cut with ID 4964-30587-0085-44554-0_sp0.9 from training. Duration: 20.85
2022-12-24 01:56:56,613 INFO [zipformer.py:660] (2/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=93376.0, num_to_drop=0, layers_to_drop=set()
2022-12-24 01:57:00,761 INFO [scaling.py:679] (2/4) Whitening: num_groups=8, num_channels=192, metric=1.80 vs. limit=2.0
2022-12-24 01:57:12,939 WARNING [train.py:1060] (2/4) Exclude cut with ID 4295-39940-0007-92567-0 from training. Duration: 21.54
2022-12-24 01:57:17,517 WARNING [train.py:1060] (2/4) Exclude cut with ID 4964-30587-0040-44509-0_sp1.1 from training. Duration: 20.5318125
2022-12-24 01:57:20,772 INFO [zipformer.py:660] (2/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=93392.0, num_to_drop=1, layers_to_drop={1}
2022-12-24 01:57:29,595 WARNING [train.py:1060] (2/4) Exclude cut with ID 7255-291500-0012-134311-0_sp0.9 from training. Duration: 21.9333125
2022-12-24 01:57:48,376 INFO [train.py:894] (2/4) Epoch 27, batch 2250, loss[loss=0.1755, simple_loss=0.2684, pruned_loss=0.04125, over 18461.00 frames. ], tot_loss[loss=0.1762, simple_loss=0.2604, pruned_loss=0.04604, over 3714368.83 frames. ], batch size: 54, lr: 4.22e-03, grad_scale: 8.0
2022-12-24 01:58:10,555 INFO [zipformer.py:660] (2/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=93424.0, num_to_drop=0, layers_to_drop=set()
2022-12-24 01:58:19,112 WARNING [train.py:1060] (2/4) Exclude cut with ID 8631-249866-0025-130151-0_sp0.9 from training. Duration: 21.7944375
2022-12-24 01:58:30,613 WARNING [train.py:1060] (2/4) Exclude cut with ID 8291-282929-0002-12989-0_sp0.9 from training. Duration: 22.4666875
2022-12-24 01:58:37,452 WARNING [train.py:1060] (2/4) Exclude cut with ID 6121-9014-0076-24124-0 from training. Duration: 21.635
2022-12-24 01:58:42,379 WARNING [train.py:1060] (2/4) Exclude cut with ID 6121-9014-0076-24124-0_sp0.9 from training. Duration: 24.038875
2022-12-24 01:58:47,235 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 2.729e+02 4.386e+02 5.059e+02 6.164e+02 8.969e+02, threshold=1.012e+03, percent-clipped=0.0
2022-12-24 01:59:05,746 INFO [train.py:894] (2/4) Epoch 27, batch 2300, loss[loss=0.1448, simple_loss=0.2178, pruned_loss=0.03589, over 18414.00 frames. ], tot_loss[loss=0.1767, simple_loss=0.2606, pruned_loss=0.04636, over 3715684.64 frames. ], batch size: 42, lr: 4.22e-03, grad_scale: 8.0
2022-12-24 01:59:22,025 INFO [zipformer.py:660] (2/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=93471.0, num_to_drop=0, layers_to_drop=set()
2022-12-24 01:59:25,059 WARNING [train.py:1060] (2/4) Exclude cut with ID 7255-291500-0009-134308-0_sp1.1 from training. Duration: 21.786375
2022-12-24 01:59:36,980 WARNING [train.py:1060] (2/4) Exclude cut with ID 8291-282929-0002-12989-0 from training. Duration: 20.22
2022-12-24 01:59:46,489 INFO [zipformer.py:660] (2/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=93487.0, num_to_drop=0, layers_to_drop=set()
2022-12-24 02:00:21,177 INFO [train.py:894] (2/4) Epoch 27, batch 2350, loss[loss=0.1656, simple_loss=0.2482, pruned_loss=0.04152, over 18722.00 frames. ], tot_loss[loss=0.177, simple_loss=0.2608, pruned_loss=0.04658, over 3717094.02 frames. ], batch size: 52, lr: 4.22e-03, grad_scale: 8.0
2022-12-24 02:00:57,253 INFO [zipformer.py:660] (2/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=93535.0, num_to_drop=0, layers_to_drop=set()
2022-12-24 02:01:17,327 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 2.947e+02 4.125e+02 5.226e+02 6.222e+02 1.366e+03, threshold=1.045e+03, percent-clipped=5.0
2022-12-24 02:01:27,669 INFO [zipformer.py:660] (2/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=93555.0, num_to_drop=0, layers_to_drop=set()
2022-12-24 02:01:35,147 INFO [train.py:894] (2/4) Epoch 27, batch 2400, loss[loss=0.1791, simple_loss=0.2737, pruned_loss=0.04227, over 18728.00 frames. ], tot_loss[loss=0.1761, simple_loss=0.26, pruned_loss=0.04614, over 3716661.01 frames. ], batch size: 54, lr: 4.22e-03, grad_scale: 8.0
2022-12-24 02:01:36,765 WARNING [train.py:1060] (2/4) Exclude cut with ID 6951-79737-0043-132310-0 from training. Duration: 25.285
2022-12-24 02:01:43,093 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.7087, 1.8329, 2.0575, 1.1638, 2.0159, 2.0959, 1.5624, 2.4454],
       device='cuda:2'), covar=tensor([0.1159, 0.1813, 0.1186, 0.1751, 0.0685, 0.1055, 0.2275, 0.0516],
       device='cuda:2'), in_proj_covar=tensor([0.0199, 0.0216, 0.0207, 0.0194, 0.0172, 0.0218, 0.0216, 0.0193],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-24 02:02:16,175 INFO [zipformer.py:660] (2/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=93587.0, num_to_drop=0, layers_to_drop=set()
2022-12-24 02:02:16,913 INFO [scaling.py:679] (2/4) Whitening: num_groups=8, num_channels=96, metric=1.29 vs. limit=2.0
2022-12-24 02:02:37,710 WARNING [train.py:1060] (2/4) Exclude cut with ID 811-130148-0001-63453-0_sp0.9 from training. Duration: 20.861125
2022-12-24 02:02:41,088 INFO [zipformer.py:660] (2/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=93603.0, num_to_drop=0, layers_to_drop=set()
2022-12-24 02:02:51,435 INFO [train.py:894] (2/4) Epoch 27, batch 2450, loss[loss=0.1988, simple_loss=0.2801, pruned_loss=0.05877, over 18589.00 frames. ], tot_loss[loss=0.1774, simple_loss=0.2613, pruned_loss=0.04678, over 3717054.17 frames. ], batch size: 57, lr: 4.22e-03, grad_scale: 8.0
2022-12-24 02:03:00,444 WARNING [train.py:1060] (2/4) Exclude cut with ID 6010-56788-0055-90261-0 from training. Duration: 20.88
2022-12-24 02:03:23,484 INFO [zipformer.py:660] (2/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=93631.0, num_to_drop=0, layers_to_drop=set()
2022-12-24 02:03:29,062 INFO [zipformer.py:660] (2/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=93635.0, num_to_drop=0, layers_to_drop=set()
2022-12-24 02:03:30,388 WARNING [train.py:1060] (2/4) Exclude cut with ID 7357-94126-0045-15876-0_sp0.9 from training. Duration: 23.4166875
2022-12-24 02:03:49,293 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 2.298e+02 3.850e+02 4.685e+02 5.940e+02 1.003e+03, threshold=9.370e+02, percent-clipped=0.0
2022-12-24 02:04:03,629 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.8339, 1.2694, 0.7569, 1.4828, 2.1574, 1.4619, 1.5519, 1.7464],
       device='cuda:2'), covar=tensor([0.1546, 0.2182, 0.2331, 0.1471, 0.1835, 0.1763, 0.1517, 0.1689],
       device='cuda:2'), in_proj_covar=tensor([0.0094, 0.0097, 0.0116, 0.0096, 0.0120, 0.0092, 0.0098, 0.0094],
       device='cuda:2'), out_proj_covar=tensor([0.0003, 0.0003, 0.0004, 0.0003, 0.0004, 0.0003, 0.0003, 0.0003],
       device='cuda:2')
2022-12-24 02:04:07,630 INFO [train.py:894] (2/4) Epoch 27, batch 2500, loss[loss=0.1729, simple_loss=0.2594, pruned_loss=0.04314, over 18597.00 frames. ], tot_loss[loss=0.1763, simple_loss=0.2602, pruned_loss=0.0462, over 3717335.21 frames. ], batch size: 99, lr: 4.22e-03, grad_scale: 8.0
2022-12-24 02:04:16,391 INFO [zipformer.py:660] (2/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=93666.0, num_to_drop=0, layers_to_drop=set()
2022-12-24 02:04:36,365 INFO [zipformer.py:660] (2/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=93679.0, num_to_drop=0, layers_to_drop=set()
2022-12-24 02:04:47,263 WARNING [train.py:1060] (2/4) Exclude cut with ID 7255-291500-0006-134305-0 from training. Duration: 21.24
2022-12-24 02:04:47,276 WARNING [train.py:1060] (2/4) Exclude cut with ID 6533-399-0047-104881-0_sp0.9 from training. Duration: 23.9055625
2022-12-24 02:04:57,013 INFO [zipformer.py:660] (2/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=93692.0, num_to_drop=0, layers_to_drop=set()
2022-12-24 02:05:20,532 WARNING [train.py:1060] (2/4) Exclude cut with ID 6758-72288-0033-108368-0_sp0.9 from training. Duration: 25.988875
2022-12-24 02:05:23,668 INFO [train.py:894] (2/4) Epoch 27, batch 2550, loss[loss=0.1766, simple_loss=0.2468, pruned_loss=0.0532, over 18467.00 frames. ], tot_loss[loss=0.1755, simple_loss=0.259, pruned_loss=0.04601, over 3716686.72 frames. ], batch size: 43, lr: 4.22e-03, grad_scale: 8.0
2022-12-24 02:05:26,883 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.6247, 1.4851, 1.0866, 0.2778, 1.1444, 1.5413, 1.3549, 1.3865],
       device='cuda:2'), covar=tensor([0.0690, 0.0593, 0.1124, 0.1742, 0.1148, 0.1617, 0.1776, 0.0787],
       device='cuda:2'), in_proj_covar=tensor([0.0176, 0.0191, 0.0212, 0.0191, 0.0212, 0.0206, 0.0219, 0.0206],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-24 02:05:27,922 WARNING [train.py:1060] (2/4) Exclude cut with ID 7255-291500-0001-134300-0_sp0.9 from training. Duration: 20.67225
2022-12-24 02:05:29,490 INFO [zipformer.py:660] (2/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=93714.0, num_to_drop=0, layers_to_drop=set()
2022-12-24 02:06:09,772 INFO [zipformer.py:660] (2/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=93740.0, num_to_drop=0, layers_to_drop=set()
2022-12-24 02:06:17,904 WARNING [train.py:1060] (2/4) Exclude cut with ID 3488-85273-0038-41224-0 from training. Duration: 20.34
2022-12-24 02:06:19,555 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.8548, 1.3326, 0.9662, 1.4788, 2.1764, 1.3428, 1.5100, 1.7948],
       device='cuda:2'), covar=tensor([0.1514, 0.2033, 0.2051, 0.1382, 0.1739, 0.1785, 0.1461, 0.1583],
       device='cuda:2'), in_proj_covar=tensor([0.0095, 0.0097, 0.0117, 0.0096, 0.0120, 0.0093, 0.0099, 0.0094],
       device='cuda:2'), out_proj_covar=tensor([0.0003, 0.0003, 0.0004, 0.0003, 0.0004, 0.0003, 0.0003, 0.0003],
       device='cuda:2')
2022-12-24 02:06:20,809 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 2.049e+02 3.985e+02 4.763e+02 6.217e+02 1.238e+03, threshold=9.527e+02, percent-clipped=6.0
2022-12-24 02:06:37,890 INFO [train.py:894] (2/4) Epoch 27, batch 2600, loss[loss=0.1792, simple_loss=0.2733, pruned_loss=0.04257, over 18500.00 frames. ], tot_loss[loss=0.1759, simple_loss=0.2592, pruned_loss=0.04626, over 3715791.39 frames. ], batch size: 52, lr: 4.21e-03, grad_scale: 8.0
2022-12-24 02:06:54,033 INFO [zipformer.py:660] (2/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=93770.0, num_to_drop=0, layers_to_drop=set()
2022-12-24 02:06:55,984 INFO [zipformer.py:660] (2/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=93771.0, num_to_drop=0, layers_to_drop=set()
2022-12-24 02:07:29,072 WARNING [train.py:1060] (2/4) Exclude cut with ID 7357-94126-0026-15857-0_sp0.9 from training. Duration: 25.061125
2022-12-24 02:07:41,148 WARNING [train.py:1060] (2/4) Exclude cut with ID 3033-130750-0096-55598-0 from training. Duration: 0.83
2022-12-24 02:07:53,328 INFO [train.py:894] (2/4) Epoch 27, batch 2650, loss[loss=0.1728, simple_loss=0.2592, pruned_loss=0.04326, over 18720.00 frames. ], tot_loss[loss=0.1763, simple_loss=0.2593, pruned_loss=0.04664, over 3715318.85 frames. ], batch size: 50, lr: 4.21e-03, grad_scale: 8.0
2022-12-24 02:08:07,511 INFO [zipformer.py:660] (2/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=93819.0, num_to_drop=0, layers_to_drop=set()
2022-12-24 02:08:08,784 WARNING [train.py:1060] (2/4) Exclude cut with ID 3488-85273-0017-41203-0 from training. Duration: 24.73
2022-12-24 02:08:21,915 WARNING [train.py:1060] (2/4) Exclude cut with ID 7255-291500-0009-134308-0 from training. Duration: 23.965
2022-12-24 02:08:25,083 INFO [zipformer.py:660] (2/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=93831.0, num_to_drop=0, layers_to_drop=set()
2022-12-24 02:08:31,203 WARNING [train.py:1060] (2/4) Exclude cut with ID 6978-92210-0030-146996-0_sp0.9 from training. Duration: 22.088875
2022-12-24 02:08:45,821 WARNING [train.py:1060] (2/4) Exclude cut with ID 7255-291500-0006-134305-0_sp0.9 from training. Duration: 23.6
2022-12-24 02:08:50,009 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 2.431e+02 4.153e+02 4.837e+02 6.025e+02 1.293e+03, threshold=9.673e+02, percent-clipped=4.0
2022-12-24 02:09:00,625 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.7618, 2.2943, 1.6276, 2.4583, 3.1392, 1.6988, 2.0122, 1.4588],
       device='cuda:2'), covar=tensor([0.1976, 0.1751, 0.1712, 0.1119, 0.1355, 0.1206, 0.1965, 0.1690],
       device='cuda:2'), in_proj_covar=tensor([0.0251, 0.0231, 0.0223, 0.0204, 0.0267, 0.0199, 0.0230, 0.0205],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-24 02:09:07,694 INFO [train.py:894] (2/4) Epoch 27, batch 2700, loss[loss=0.1687, simple_loss=0.257, pruned_loss=0.04017, over 18497.00 frames. ], tot_loss[loss=0.1766, simple_loss=0.2597, pruned_loss=0.04679, over 3715294.44 frames. ], batch size: 52, lr: 4.21e-03, grad_scale: 8.0
2022-12-24 02:10:22,945 INFO [train.py:894] (2/4) Epoch 27, batch 2750, loss[loss=0.1552, simple_loss=0.2312, pruned_loss=0.03959, over 18478.00 frames. ], tot_loss[loss=0.1755, simple_loss=0.2588, pruned_loss=0.04604, over 3714730.62 frames. ], batch size: 43, lr: 4.21e-03, grad_scale: 8.0
2022-12-24 02:10:26,393 WARNING [train.py:1060] (2/4) Exclude cut with ID 8291-282929-0024-13011-0 from training. Duration: 23.795
2022-12-24 02:10:41,399 WARNING [train.py:1060] (2/4) Exclude cut with ID 8631-249866-0030-130156-0_sp1.1 from training. Duration: 21.5409375
2022-12-24 02:10:44,823 WARNING [train.py:1060] (2/4) Exclude cut with ID 6978-92210-0019-146985-0_sp0.9 from training. Duration: 24.97775
2022-12-24 02:10:54,915 WARNING [train.py:1060] (2/4) Exclude cut with ID 1085-156170-0017-128270-0_sp0.9 from training. Duration: 23.3444375
2022-12-24 02:11:21,037 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 2.780e+02 4.032e+02 4.691e+02 5.983e+02 1.414e+03, threshold=9.382e+02, percent-clipped=4.0
2022-12-24 02:11:22,550 WARNING [train.py:1060] (2/4) Exclude cut with ID 6010-56788-0055-90261-0_sp0.9 from training. Duration: 23.2
2022-12-24 02:11:28,583 WARNING [train.py:1060] (2/4) Exclude cut with ID 5653-46179-0060-117930-0_sp0.9 from training. Duration: 21.17225
2022-12-24 02:11:39,468 INFO [train.py:894] (2/4) Epoch 27, batch 2800, loss[loss=0.2021, simple_loss=0.2819, pruned_loss=0.06114, over 18605.00 frames. ], tot_loss[loss=0.176, simple_loss=0.2596, pruned_loss=0.04619, over 3714477.72 frames. ], batch size: 56, lr: 4.21e-03, grad_scale: 8.0
2022-12-24 02:11:48,834 WARNING [train.py:1060] (2/4) Exclude cut with ID 7255-291500-0008-134307-0_sp0.9 from training. Duration: 24.6555625
2022-12-24 02:11:50,845 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([2.5398, 2.1346, 1.9978, 1.2729, 2.6919, 2.4963, 2.2080, 1.7446],
       device='cuda:2'), covar=tensor([0.0350, 0.0485, 0.0519, 0.0807, 0.0311, 0.0420, 0.0470, 0.0958],
       device='cuda:2'), in_proj_covar=tensor([0.0125, 0.0131, 0.0129, 0.0118, 0.0104, 0.0127, 0.0133, 0.0162],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0001, 0.0001, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-24 02:12:11,974 INFO [zipformer.py:660] (2/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=93981.0, num_to_drop=0, layers_to_drop=set()
2022-12-24 02:12:37,886 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.5074, 1.2572, 1.7541, 2.7525, 2.0414, 2.3623, 0.9004, 2.1303],
       device='cuda:2'), covar=tensor([0.1873, 0.1637, 0.1397, 0.0725, 0.1060, 0.1041, 0.2033, 0.1122],
       device='cuda:2'), in_proj_covar=tensor([0.0103, 0.0119, 0.0136, 0.0156, 0.0107, 0.0144, 0.0128, 0.0116],
       device='cuda:2'), out_proj_covar=tensor([0.0003, 0.0003, 0.0004, 0.0004, 0.0003, 0.0004, 0.0004, 0.0003],
       device='cuda:2')
2022-12-24 02:12:47,333 WARNING [train.py:1060] (2/4) Exclude cut with ID 3488-65654-0031-41259-0 from training. Duration: 20.44
2022-12-24 02:12:59,239 INFO [train.py:894] (2/4) Epoch 27, batch 2850, loss[loss=0.1488, simple_loss=0.225, pruned_loss=0.0363, over 18473.00 frames. ], tot_loss[loss=0.1763, simple_loss=0.2599, pruned_loss=0.04635, over 3714049.72 frames. ], batch size: 43, lr: 4.21e-03, grad_scale: 8.0
2022-12-24 02:13:02,851 WARNING [train.py:1060] (2/4) Exclude cut with ID 6951-79737-0018-132285-0_sp0.9 from training. Duration: 23.45
2022-12-24 02:13:30,463 WARNING [train.py:1060] (2/4) Exclude cut with ID 6945-60535-0076-12784-0_sp0.9 from training. Duration: 20.52225
2022-12-24 02:13:39,322 WARNING [train.py:1060] (2/4) Exclude cut with ID 7255-291500-0008-134307-0 from training. Duration: 22.19
2022-12-24 02:13:47,334 INFO [zipformer.py:660] (2/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=94042.0, num_to_drop=0, layers_to_drop=set()
2022-12-24 02:13:51,270 WARNING [train.py:1060] (2/4) Exclude cut with ID 7357-94126-0014-15845-0_sp1.1 from training. Duration: 25.3818125
2022-12-24 02:13:55,592 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 2.597e+02 3.768e+02 4.678e+02 5.604e+02 1.868e+03, threshold=9.356e+02, percent-clipped=5.0
2022-12-24 02:14:02,235 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.9043, 1.4685, 0.8576, 1.5575, 2.2411, 1.4678, 1.6600, 1.9384],
       device='cuda:2'), covar=tensor([0.1535, 0.2014, 0.2269, 0.1425, 0.1808, 0.1837, 0.1417, 0.1548],
       device='cuda:2'), in_proj_covar=tensor([0.0094, 0.0097, 0.0116, 0.0096, 0.0119, 0.0092, 0.0098, 0.0094],
       device='cuda:2'), out_proj_covar=tensor([0.0003, 0.0003, 0.0004, 0.0003, 0.0004, 0.0003, 0.0003, 0.0003],
       device='cuda:2')
2022-12-24 02:14:09,570 WARNING [train.py:1060] (2/4) Exclude cut with ID 6951-79737-0043-132310-0_sp0.9 from training. Duration: 28.0944375
2022-12-24 02:14:14,009 INFO [train.py:894] (2/4) Epoch 27, batch 2900, loss[loss=0.1594, simple_loss=0.245, pruned_loss=0.03696, over 18533.00 frames. ], tot_loss[loss=0.1761, simple_loss=0.2598, pruned_loss=0.04624, over 3713663.80 frames. ], batch size: 47, lr: 4.21e-03, grad_scale: 8.0
2022-12-24 02:14:15,408 WARNING [train.py:1060] (2/4) Exclude cut with ID 2195-150901-0045-59933-0_sp0.9 from training. Duration: 22.9444375
2022-12-24 02:14:24,294 WARNING [train.py:1060] (2/4) Exclude cut with ID 8291-282929-0024-13011-0_sp1.1 from training. Duration: 21.6318125
2022-12-24 02:14:39,641 WARNING [train.py:1060] (2/4) Exclude cut with ID 8631-249866-0030-130156-0 from training. Duration: 23.695
2022-12-24 02:15:05,809 WARNING [train.py:1060] (2/4) Exclude cut with ID 7699-105389-0094-26379-0 from training. Duration: 23.955
2022-12-24 02:15:10,105 INFO [scaling.py:679] (2/4) Whitening: num_groups=8, num_channels=96, metric=1.35 vs. limit=2.0
2022-12-24 02:15:29,439 INFO [train.py:894] (2/4) Epoch 27, batch 2950, loss[loss=0.1909, simple_loss=0.2745, pruned_loss=0.0537, over 18460.00 frames. ], tot_loss[loss=0.1767, simple_loss=0.2602, pruned_loss=0.04663, over 3713382.94 frames. ], batch size: 64, lr: 4.21e-03, grad_scale: 8.0
2022-12-24 02:15:40,132 WARNING [train.py:1060] (2/4) Exclude cut with ID 8291-282929-0024-13011-0_sp0.9 from training. Duration: 26.438875
2022-12-24 02:15:53,337 INFO [zipformer.py:660] (2/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=94126.0, num_to_drop=0, layers_to_drop=set()
2022-12-24 02:16:20,820 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.6803, 1.3213, 0.8392, 1.3880, 2.2309, 0.9879, 1.4966, 1.5250],
       device='cuda:2'), covar=tensor([0.1600, 0.2029, 0.2119, 0.1441, 0.1647, 0.1818, 0.1405, 0.1827],
       device='cuda:2'), in_proj_covar=tensor([0.0095, 0.0097, 0.0117, 0.0096, 0.0120, 0.0093, 0.0098, 0.0094],
       device='cuda:2'), out_proj_covar=tensor([0.0003, 0.0003, 0.0004, 0.0003, 0.0004, 0.0003, 0.0003, 0.0003],
       device='cuda:2')
2022-12-24 02:16:22,101 WARNING [train.py:1060] (2/4) Exclude cut with ID 7699-105389-0021-26306-0_sp0.9 from training. Duration: 21.2444375
2022-12-24 02:16:23,543 WARNING [train.py:1060] (2/4) Exclude cut with ID 7357-94126-0014-15845-0_sp0.9 from training. Duration: 31.02225
2022-12-24 02:16:28,311 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 2.796e+02 4.001e+02 4.676e+02 5.436e+02 1.154e+03, threshold=9.352e+02, percent-clipped=2.0
2022-12-24 02:16:33,390 WARNING [train.py:1060] (2/4) Exclude cut with ID 432-122774-0017-62487-0 from training. Duration: 22.395
2022-12-24 02:16:45,794 INFO [train.py:894] (2/4) Epoch 27, batch 3000, loss[loss=0.1437, simple_loss=0.2271, pruned_loss=0.03017, over 18378.00 frames. ], tot_loss[loss=0.1757, simple_loss=0.2592, pruned_loss=0.0461, over 3713233.97 frames. ], batch size: 46, lr: 4.21e-03, grad_scale: 8.0
2022-12-24 02:16:45,795 INFO [train.py:919] (2/4) Computing validation loss
2022-12-24 02:16:56,402 INFO [train.py:928] (2/4) Epoch 27, validation: loss=0.1648, simple_loss=0.2604, pruned_loss=0.03454, over 944034.00 frames. 
2022-12-24 02:16:56,402 INFO [train.py:929] (2/4) Maximum memory allocated so far is 24680MB
2022-12-24 02:16:59,705 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.5883, 1.8180, 1.5328, 2.1244, 2.5053, 1.6131, 1.5193, 1.3644],
       device='cuda:2'), covar=tensor([0.2007, 0.1846, 0.1741, 0.1130, 0.1296, 0.1166, 0.2248, 0.1581],
       device='cuda:2'), in_proj_covar=tensor([0.0253, 0.0233, 0.0225, 0.0205, 0.0268, 0.0201, 0.0231, 0.0206],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0003, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-24 02:17:00,723 WARNING [train.py:1060] (2/4) Exclude cut with ID 7357-94126-0045-15876-0 from training. Duration: 21.075
2022-12-24 02:17:05,909 WARNING [train.py:1060] (2/4) Exclude cut with ID 6482-98857-0025-147532-0_sp0.9 from training. Duration: 20.0055625
2022-12-24 02:17:07,304 WARNING [train.py:1060] (2/4) Exclude cut with ID 6951-79737-0037-132304-0_sp0.9 from training. Duration: 22.05
2022-12-24 02:17:07,317 WARNING [train.py:1060] (2/4) Exclude cut with ID 7255-291500-0003-134302-0 from training. Duration: 26.8349375
2022-12-24 02:17:10,106 WARNING [train.py:1060] (2/4) Exclude cut with ID 7357-94126-0009-15840-0_sp1.1 from training. Duration: 22.1090625
2022-12-24 02:17:18,280 WARNING [train.py:1060] (2/4) Exclude cut with ID 7699-105389-0094-26379-0_sp0.9 from training. Duration: 26.6166875
2022-12-24 02:17:31,415 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.5327, 1.1499, 1.9757, 3.1725, 2.2524, 2.5154, 0.7398, 2.3555],
       device='cuda:2'), covar=tensor([0.1873, 0.1877, 0.1466, 0.0656, 0.1105, 0.1165, 0.2398, 0.1043],
       device='cuda:2'), in_proj_covar=tensor([0.0104, 0.0120, 0.0137, 0.0157, 0.0108, 0.0145, 0.0130, 0.0117],
       device='cuda:2'), out_proj_covar=tensor([0.0003, 0.0003, 0.0004, 0.0004, 0.0003, 0.0004, 0.0004, 0.0003],
       device='cuda:2')
2022-12-24 02:17:35,687 WARNING [train.py:1060] (2/4) Exclude cut with ID 2046-178027-0000-53705-0_sp0.9 from training. Duration: 20.3055625
2022-12-24 02:17:36,082 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.7936, 2.1945, 1.6464, 2.4467, 3.2034, 1.7929, 2.0822, 1.4580],
       device='cuda:2'), covar=tensor([0.1883, 0.1734, 0.1629, 0.1037, 0.1205, 0.1087, 0.1893, 0.1533],
       device='cuda:2'), in_proj_covar=tensor([0.0251, 0.0232, 0.0223, 0.0204, 0.0266, 0.0199, 0.0230, 0.0204],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-24 02:17:58,932 WARNING [train.py:1060] (2/4) Exclude cut with ID 7205-50138-0008-5373-0_sp0.9 from training. Duration: 20.7
2022-12-24 02:18:12,659 INFO [train.py:894] (2/4) Epoch 27, batch 3050, loss[loss=0.166, simple_loss=0.2526, pruned_loss=0.03966, over 18499.00 frames. ], tot_loss[loss=0.1757, simple_loss=0.2595, pruned_loss=0.04597, over 3714032.50 frames. ], batch size: 52, lr: 4.20e-03, grad_scale: 8.0
2022-12-24 02:18:32,170 INFO [zipformer.py:660] (2/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=94223.0, num_to_drop=1, layers_to_drop={1}
2022-12-24 02:18:43,444 WARNING [train.py:1060] (2/4) Exclude cut with ID 6978-92210-0019-146985-0 from training. Duration: 22.48
2022-12-24 02:18:57,312 WARNING [train.py:1060] (2/4) Exclude cut with ID 7255-291500-0003-134302-0_sp0.9 from training. Duration: 29.816625
2022-12-24 02:19:11,006 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 2.668e+02 4.124e+02 4.922e+02 6.064e+02 1.653e+03, threshold=9.844e+02, percent-clipped=4.0
2022-12-24 02:19:17,871 WARNING [train.py:1060] (2/4) Exclude cut with ID 7255-291500-0005-134304-0_sp1.1 from training. Duration: 22.7590625
2022-12-24 02:19:26,041 WARNING [train.py:1060] (2/4) Exclude cut with ID 7357-94126-0026-15857-0 from training. Duration: 22.555
2022-12-24 02:19:28,707 INFO [train.py:894] (2/4) Epoch 27, batch 3100, loss[loss=0.1498, simple_loss=0.231, pruned_loss=0.03428, over 18535.00 frames. ], tot_loss[loss=0.1754, simple_loss=0.2591, pruned_loss=0.04583, over 3713636.39 frames. ], batch size: 44, lr: 4.20e-03, grad_scale: 8.0
2022-12-24 02:19:46,358 WARNING [train.py:1060] (2/4) Exclude cut with ID 1250-135782-0005-25975-0_sp0.9 from training. Duration: 21.688875
2022-12-24 02:19:56,550 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([3.5167, 2.5860, 2.1093, 3.2569, 2.3968, 2.5965, 2.5573, 3.5108],
       device='cuda:2'), covar=tensor([0.1797, 0.3406, 0.2040, 0.2865, 0.4094, 0.1080, 0.3309, 0.0793],
       device='cuda:2'), in_proj_covar=tensor([0.0298, 0.0299, 0.0253, 0.0349, 0.0281, 0.0234, 0.0296, 0.0220],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0003, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-24 02:20:02,962 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.9872, 1.8615, 1.5631, 1.5504, 1.7136, 1.7910, 1.6870, 1.7053],
       device='cuda:2'), covar=tensor([0.2306, 0.3014, 0.2137, 0.2528, 0.3317, 0.1176, 0.2773, 0.1113],
       device='cuda:2'), in_proj_covar=tensor([0.0298, 0.0299, 0.0253, 0.0349, 0.0281, 0.0234, 0.0296, 0.0220],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0003, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-24 02:20:05,590 INFO [zipformer.py:660] (2/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=94284.0, num_to_drop=1, layers_to_drop={1}
2022-12-24 02:20:18,850 WARNING [train.py:1060] (2/4) Exclude cut with ID 3488-85273-0038-41224-0_sp0.9 from training. Duration: 22.6
2022-12-24 02:20:43,868 INFO [train.py:894] (2/4) Epoch 27, batch 3150, loss[loss=0.1578, simple_loss=0.2534, pruned_loss=0.03106, over 18547.00 frames. ], tot_loss[loss=0.1746, simple_loss=0.258, pruned_loss=0.04561, over 3711897.27 frames. ], batch size: 55, lr: 4.20e-03, grad_scale: 16.0
2022-12-24 02:20:56,204 WARNING [train.py:1060] (2/4) Exclude cut with ID 7357-94126-0009-15840-0 from training. Duration: 24.32
2022-12-24 02:21:25,063 INFO [zipformer.py:660] (2/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=94337.0, num_to_drop=0, layers_to_drop=set()
2022-12-24 02:21:41,693 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 2.425e+02 3.935e+02 4.692e+02 5.963e+02 1.108e+03, threshold=9.384e+02, percent-clipped=1.0
2022-12-24 02:21:42,119 INFO [zipformer.py:660] (2/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=94348.0, num_to_drop=0, layers_to_drop=set()
2022-12-24 02:21:50,682 WARNING [train.py:1060] (2/4) Exclude cut with ID 8291-276745-0093-13116-0_sp0.9 from training. Duration: 21.061125
2022-12-24 02:22:00,331 INFO [train.py:894] (2/4) Epoch 27, batch 3200, loss[loss=0.2471, simple_loss=0.3146, pruned_loss=0.08985, over 18630.00 frames. ], tot_loss[loss=0.1758, simple_loss=0.259, pruned_loss=0.04623, over 3712686.24 frames. ], batch size: 181, lr: 4.20e-03, grad_scale: 16.0
2022-12-24 02:22:04,771 WARNING [train.py:1060] (2/4) Exclude cut with ID 7357-94126-0024-15855-0_sp0.9 from training. Duration: 20.32225
2022-12-24 02:22:18,567 WARNING [train.py:1060] (2/4) Exclude cut with ID 3033-130750-0096-55598-0_sp1.1 from training. Duration: 0.7545625
2022-12-24 02:22:33,458 WARNING [train.py:1060] (2/4) Exclude cut with ID 4295-39940-0007-92567-0_sp0.9 from training. Duration: 23.9333125
2022-12-24 02:23:03,404 WARNING [train.py:1060] (2/4) Exclude cut with ID 7255-291500-0008-134307-0_sp1.1 from training. Duration: 20.17275
2022-12-24 02:23:09,846 WARNING [train.py:1060] (2/4) Exclude cut with ID 6978-92210-0019-146985-0_sp1.1 from training. Duration: 20.436375
2022-12-24 02:23:14,700 INFO [zipformer.py:660] (2/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=94409.0, num_to_drop=1, layers_to_drop={1}
2022-12-24 02:23:15,658 INFO [train.py:894] (2/4) Epoch 27, batch 3250, loss[loss=0.196, simple_loss=0.2841, pruned_loss=0.054, over 18496.00 frames. ], tot_loss[loss=0.1755, simple_loss=0.2594, pruned_loss=0.0458, over 3713017.47 frames. ], batch size: 52, lr: 4.20e-03, grad_scale: 16.0
2022-12-24 02:23:39,593 INFO [zipformer.py:660] (2/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=94426.0, num_to_drop=0, layers_to_drop=set()
2022-12-24 02:24:11,648 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 3.045e+02 4.265e+02 5.105e+02 6.397e+02 1.288e+03, threshold=1.021e+03, percent-clipped=3.0
2022-12-24 02:24:29,893 INFO [train.py:894] (2/4) Epoch 27, batch 3300, loss[loss=0.2048, simple_loss=0.2834, pruned_loss=0.06305, over 18576.00 frames. ], tot_loss[loss=0.1753, simple_loss=0.2592, pruned_loss=0.04567, over 3713427.77 frames. ], batch size: 97, lr: 4.20e-03, grad_scale: 16.0
2022-12-24 02:24:29,918 WARNING [train.py:1060] (2/4) Exclude cut with ID 4234-40345-0022-142709-0_sp0.9 from training. Duration: 23.1055625
2022-12-24 02:24:33,586 WARNING [train.py:1060] (2/4) Exclude cut with ID 8291-282929-0007-12994-0_sp1.1 from training. Duration: 23.5
2022-12-24 02:24:43,630 WARNING [train.py:1060] (2/4) Exclude cut with ID 7255-291500-0009-134308-0_sp0.9 from training. Duration: 26.62775
2022-12-24 02:24:51,271 INFO [zipformer.py:660] (2/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=94474.0, num_to_drop=0, layers_to_drop=set()
2022-12-24 02:24:55,636 WARNING [train.py:1060] (2/4) Exclude cut with ID 6951-79737-0018-132285-0 from training. Duration: 21.105
2022-12-24 02:25:00,585 WARNING [train.py:1060] (2/4) Exclude cut with ID 4511-76322-0006-80011-0_sp0.9 from training. Duration: 24.411125
2022-12-24 02:25:27,361 WARNING [train.py:1060] (2/4) Exclude cut with ID 6758-72288-0033-108368-0_sp1.1 from training. Duration: 21.263625
2022-12-24 02:25:46,681 INFO [train.py:894] (2/4) Epoch 27, batch 3350, loss[loss=0.1858, simple_loss=0.2703, pruned_loss=0.05063, over 18523.00 frames. ], tot_loss[loss=0.176, simple_loss=0.2599, pruned_loss=0.04598, over 3713832.36 frames. ], batch size: 55, lr: 4.20e-03, grad_scale: 16.0
2022-12-24 02:25:49,911 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([0.6970, 0.6945, 0.6416, 0.7981, 0.8619, 0.8022, 0.7567, 0.7065],
       device='cuda:2'), covar=tensor([0.0246, 0.0223, 0.0419, 0.0185, 0.0210, 0.0294, 0.0222, 0.0261],
       device='cuda:2'), in_proj_covar=tensor([0.0096, 0.0127, 0.0153, 0.0119, 0.0116, 0.0121, 0.0100, 0.0128],
       device='cuda:2'), out_proj_covar=tensor([7.5804e-05, 1.0049e-04, 1.2501e-04, 9.4444e-05, 9.2757e-05, 9.2678e-05,
        7.7746e-05, 1.0093e-04], device='cuda:2')
2022-12-24 02:25:51,494 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([2.7989, 2.5791, 2.0629, 0.9198, 1.9298, 2.1707, 1.8984, 2.1288],
       device='cuda:2'), covar=tensor([0.0621, 0.0578, 0.1291, 0.1892, 0.1386, 0.1614, 0.1702, 0.0929],
       device='cuda:2'), in_proj_covar=tensor([0.0175, 0.0188, 0.0210, 0.0189, 0.0209, 0.0204, 0.0216, 0.0203],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-24 02:26:00,103 WARNING [train.py:1060] (2/4) Exclude cut with ID 4234-40345-0022-142709-0 from training. Duration: 20.795
2022-12-24 02:26:10,839 WARNING [train.py:1060] (2/4) Exclude cut with ID 7357-94126-0021-15852-0 from training. Duration: 24.76
2022-12-24 02:26:10,855 WARNING [train.py:1060] (2/4) Exclude cut with ID 3867-173237-0077-144769-0_sp0.9 from training. Duration: 22.25
2022-12-24 02:26:34,768 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.7565, 1.7263, 1.8050, 1.7291, 1.3821, 3.8162, 1.6361, 2.1351],
       device='cuda:2'), covar=tensor([0.3073, 0.1970, 0.1867, 0.2037, 0.1411, 0.0184, 0.1541, 0.0823],
       device='cuda:2'), in_proj_covar=tensor([0.0133, 0.0117, 0.0125, 0.0122, 0.0106, 0.0097, 0.0090, 0.0090],
       device='cuda:2'), out_proj_covar=tensor([0.0005, 0.0005, 0.0005, 0.0005, 0.0004, 0.0003, 0.0004, 0.0004],
       device='cuda:2')
2022-12-24 02:26:37,539 WARNING [train.py:1060] (2/4) Exclude cut with ID 7357-94126-0026-15857-0_sp1.1 from training. Duration: 20.5045625
2022-12-24 02:26:43,828 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 2.518e+02 3.960e+02 4.769e+02 6.084e+02 1.782e+03, threshold=9.538e+02, percent-clipped=2.0
2022-12-24 02:27:02,427 INFO [train.py:894] (2/4) Epoch 27, batch 3400, loss[loss=0.2092, simple_loss=0.2868, pruned_loss=0.06582, over 18437.00 frames. ], tot_loss[loss=0.1754, simple_loss=0.2593, pruned_loss=0.04576, over 3713385.42 frames. ], batch size: 64, lr: 4.20e-03, grad_scale: 16.0
2022-12-24 02:27:18,725 INFO [scaling.py:679] (2/4) Whitening: num_groups=1, num_channels=384, metric=4.38 vs. limit=5.0
2022-12-24 02:27:29,618 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([4.2418, 3.6480, 3.6886, 4.1530, 3.8732, 3.7279, 4.3588, 1.3486],
       device='cuda:2'), covar=tensor([0.0776, 0.0763, 0.0710, 0.0846, 0.1366, 0.1161, 0.0649, 0.5099],
       device='cuda:2'), in_proj_covar=tensor([0.0361, 0.0238, 0.0247, 0.0285, 0.0339, 0.0276, 0.0304, 0.0293],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0001, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-24 02:27:29,620 INFO [zipformer.py:660] (2/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=94579.0, num_to_drop=1, layers_to_drop={3}
2022-12-24 02:27:30,222 INFO [scaling.py:679] (2/4) Whitening: num_groups=8, num_channels=96, metric=1.12 vs. limit=2.0
2022-12-24 02:28:13,974 INFO [train.py:894] (2/4) Epoch 27, batch 3450, loss[loss=0.1972, simple_loss=0.283, pruned_loss=0.05575, over 18687.00 frames. ], tot_loss[loss=0.1752, simple_loss=0.2591, pruned_loss=0.04569, over 3712674.19 frames. ], batch size: 69, lr: 4.20e-03, grad_scale: 16.0
2022-12-24 02:28:15,525 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([2.5592, 3.7344, 3.6054, 1.3899, 3.8024, 2.7686, 0.6622, 2.4854],
       device='cuda:2'), covar=tensor([0.2030, 0.1201, 0.1461, 0.3586, 0.0899, 0.0916, 0.4752, 0.1457],
       device='cuda:2'), in_proj_covar=tensor([0.0150, 0.0150, 0.0162, 0.0125, 0.0153, 0.0117, 0.0146, 0.0116],
       device='cuda:2'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0002, 0.0003, 0.0002, 0.0003, 0.0002],
       device='cuda:2')
2022-12-24 02:28:52,972 INFO [zipformer.py:660] (2/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=94637.0, num_to_drop=0, layers_to_drop=set()
2022-12-24 02:29:07,392 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 2.881e+02 4.475e+02 5.396e+02 6.310e+02 1.606e+03, threshold=1.079e+03, percent-clipped=4.0
2022-12-24 02:29:13,864 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.9468, 1.5581, 1.8220, 1.6703, 1.9132, 1.8911, 1.7767, 1.7859],
       device='cuda:2'), covar=tensor([0.1925, 0.2702, 0.2119, 0.2508, 0.1901, 0.0958, 0.2955, 0.1154],
       device='cuda:2'), in_proj_covar=tensor([0.0274, 0.0304, 0.0290, 0.0329, 0.0321, 0.0262, 0.0357, 0.0251],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-24 02:29:25,870 INFO [train.py:894] (2/4) Epoch 27, batch 3500, loss[loss=0.1859, simple_loss=0.2724, pruned_loss=0.04976, over 18572.00 frames. ], tot_loss[loss=0.1765, simple_loss=0.2601, pruned_loss=0.04642, over 3714086.93 frames. ], batch size: 171, lr: 4.19e-03, grad_scale: 16.0
2022-12-24 02:29:46,878 WARNING [train.py:1060] (2/4) Exclude cut with ID 7859-102521-0017-7548-0_sp1.1 from training. Duration: 22.2954375
2022-12-24 02:29:55,608 INFO [train.py:894] (2/4) Epoch 28, batch 0, loss[loss=0.16, simple_loss=0.2445, pruned_loss=0.0377, over 18527.00 frames. ], tot_loss[loss=0.16, simple_loss=0.2445, pruned_loss=0.0377, over 18527.00 frames. ], batch size: 47, lr: 4.12e-03, grad_scale: 16.0
2022-12-24 02:29:55,608 INFO [train.py:919] (2/4) Computing validation loss
2022-12-24 02:30:06,233 INFO [train.py:928] (2/4) Epoch 28, validation: loss=0.1635, simple_loss=0.2596, pruned_loss=0.03369, over 944034.00 frames. 
2022-12-24 02:30:06,234 INFO [train.py:929] (2/4) Maximum memory allocated so far is 24680MB
2022-12-24 02:30:34,547 INFO [zipformer.py:660] (2/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=94685.0, num_to_drop=0, layers_to_drop=set()
2022-12-24 02:30:57,083 WARNING [train.py:1060] (2/4) Exclude cut with ID 298-126791-0067-24026-0_sp0.9 from training. Duration: 21.438875
2022-12-24 02:31:01,652 WARNING [train.py:1060] (2/4) Exclude cut with ID 5652-39938-0025-23684-0_sp0.9 from training. Duration: 22.2055625
2022-12-24 02:31:03,305 INFO [zipformer.py:660] (2/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=94704.0, num_to_drop=1, layers_to_drop={3}
2022-12-24 02:31:20,488 INFO [train.py:894] (2/4) Epoch 28, batch 50, loss[loss=0.1656, simple_loss=0.266, pruned_loss=0.03256, over 18470.00 frames. ], tot_loss[loss=0.1719, simple_loss=0.2616, pruned_loss=0.04113, over 837189.85 frames. ], batch size: 54, lr: 4.12e-03, grad_scale: 16.0
2022-12-24 02:32:09,506 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 1.789e+02 3.263e+02 4.050e+02 5.069e+02 1.061e+03, threshold=8.099e+02, percent-clipped=0.0
2022-12-24 02:32:34,822 INFO [train.py:894] (2/4) Epoch 28, batch 100, loss[loss=0.1736, simple_loss=0.2668, pruned_loss=0.04024, over 18569.00 frames. ], tot_loss[loss=0.1688, simple_loss=0.2592, pruned_loss=0.03923, over 1475649.18 frames. ], batch size: 57, lr: 4.11e-03, grad_scale: 16.0
2022-12-24 02:33:48,264 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.8078, 1.2122, 0.8431, 1.3836, 2.0895, 1.2311, 1.4249, 1.6433],
       device='cuda:2'), covar=tensor([0.1600, 0.2235, 0.2199, 0.1567, 0.1765, 0.1789, 0.1503, 0.1766],
       device='cuda:2'), in_proj_covar=tensor([0.0094, 0.0097, 0.0116, 0.0096, 0.0120, 0.0093, 0.0099, 0.0094],
       device='cuda:2'), out_proj_covar=tensor([0.0003, 0.0003, 0.0004, 0.0003, 0.0004, 0.0003, 0.0003, 0.0003],
       device='cuda:2')
2022-12-24 02:33:50,519 INFO [train.py:894] (2/4) Epoch 28, batch 150, loss[loss=0.1762, simple_loss=0.266, pruned_loss=0.04317, over 18721.00 frames. ], tot_loss[loss=0.1671, simple_loss=0.2572, pruned_loss=0.03847, over 1972289.79 frames. ], batch size: 65, lr: 4.11e-03, grad_scale: 16.0
2022-12-24 02:33:54,729 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([2.3529, 2.0812, 1.6981, 1.9761, 1.8450, 2.0667, 1.8710, 2.1907],
       device='cuda:2'), covar=tensor([0.2317, 0.3320, 0.2075, 0.2841, 0.3775, 0.1133, 0.3176, 0.1104],
       device='cuda:2'), in_proj_covar=tensor([0.0298, 0.0298, 0.0253, 0.0348, 0.0280, 0.0235, 0.0296, 0.0221],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-24 02:33:58,498 WARNING [train.py:1060] (2/4) Exclude cut with ID 7859-102521-0017-7548-0 from training. Duration: 24.525
2022-12-24 02:34:35,127 WARNING [train.py:1060] (2/4) Exclude cut with ID 3699-47246-0007-3408-0_sp0.9 from training. Duration: 20.26675
2022-12-24 02:34:39,510 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 2.175e+02 2.935e+02 3.702e+02 4.603e+02 1.159e+03, threshold=7.404e+02, percent-clipped=1.0
2022-12-24 02:34:47,221 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([2.3221, 2.0658, 1.5574, 0.5207, 1.5375, 2.1001, 1.7522, 1.8938],
       device='cuda:2'), covar=tensor([0.0695, 0.0583, 0.1152, 0.1689, 0.1125, 0.1508, 0.1692, 0.0723],
       device='cuda:2'), in_proj_covar=tensor([0.0176, 0.0190, 0.0211, 0.0191, 0.0211, 0.0205, 0.0217, 0.0204],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-24 02:34:48,368 WARNING [train.py:1060] (2/4) Exclude cut with ID 7859-102521-0017-7548-0_sp0.9 from training. Duration: 27.25
2022-12-24 02:35:05,998 INFO [train.py:894] (2/4) Epoch 28, batch 200, loss[loss=0.1524, simple_loss=0.2286, pruned_loss=0.03809, over 18480.00 frames. ], tot_loss[loss=0.1638, simple_loss=0.2536, pruned_loss=0.03706, over 2358070.50 frames. ], batch size: 43, lr: 4.11e-03, grad_scale: 16.0
2022-12-24 02:35:25,636 INFO [zipformer.py:660] (2/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=94879.0, num_to_drop=1, layers_to_drop={2}
2022-12-24 02:35:59,922 WARNING [train.py:1060] (2/4) Exclude cut with ID 6426-64292-0017-15984-0 from training. Duration: 21.68
2022-12-24 02:36:09,903 WARNING [train.py:1060] (2/4) Exclude cut with ID 4278-13270-0007-59342-0 from training. Duration: 21.6300625
2022-12-24 02:36:20,720 INFO [train.py:894] (2/4) Epoch 28, batch 250, loss[loss=0.1469, simple_loss=0.2251, pruned_loss=0.03439, over 18446.00 frames. ], tot_loss[loss=0.1644, simple_loss=0.2537, pruned_loss=0.03751, over 2658369.23 frames. ], batch size: 43, lr: 4.11e-03, grad_scale: 16.0
2022-12-24 02:36:32,651 WARNING [train.py:1060] (2/4) Exclude cut with ID 4278-13270-0007-59342-0_sp0.9 from training. Duration: 24.033375
2022-12-24 02:36:37,042 INFO [zipformer.py:660] (2/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=94927.0, num_to_drop=1, layers_to_drop={0}
2022-12-24 02:37:07,818 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 1.987e+02 2.888e+02 3.443e+02 4.147e+02 1.178e+03, threshold=6.886e+02, percent-clipped=2.0
2022-12-24 02:37:31,407 WARNING [train.py:1060] (2/4) Exclude cut with ID 4278-13270-0009-59344-0 from training. Duration: 22.905
2022-12-24 02:37:31,441 WARNING [train.py:1060] (2/4) Exclude cut with ID 5622-44585-0006-90525-0_sp1.1 from training. Duration: 23.4318125
2022-12-24 02:37:34,421 INFO [train.py:894] (2/4) Epoch 28, batch 300, loss[loss=0.1483, simple_loss=0.2262, pruned_loss=0.03517, over 18396.00 frames. ], tot_loss[loss=0.1628, simple_loss=0.2518, pruned_loss=0.03686, over 2892822.36 frames. ], batch size: 42, lr: 4.11e-03, grad_scale: 16.0
2022-12-24 02:38:31,508 INFO [zipformer.py:660] (2/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=95004.0, num_to_drop=0, layers_to_drop=set()
2022-12-24 02:38:48,899 INFO [train.py:894] (2/4) Epoch 28, batch 350, loss[loss=0.1624, simple_loss=0.2554, pruned_loss=0.03465, over 18583.00 frames. ], tot_loss[loss=0.1653, simple_loss=0.2545, pruned_loss=0.03805, over 3075805.33 frames. ], batch size: 51, lr: 4.11e-03, grad_scale: 16.0
2022-12-24 02:39:26,968 WARNING [train.py:1060] (2/4) Exclude cut with ID 4278-13270-0009-59344-0_sp1.1 from training. Duration: 20.82275
2022-12-24 02:39:28,518 WARNING [train.py:1060] (2/4) Exclude cut with ID 4278-13270-0009-59344-0_sp0.9 from training. Duration: 25.45
2022-12-24 02:39:35,531 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 2.241e+02 3.092e+02 3.846e+02 4.799e+02 1.160e+03, threshold=7.692e+02, percent-clipped=9.0
2022-12-24 02:39:41,467 INFO [zipformer.py:660] (2/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=95052.0, num_to_drop=0, layers_to_drop=set()
2022-12-24 02:40:01,311 INFO [scaling.py:679] (2/4) Whitening: num_groups=8, num_channels=96, metric=1.54 vs. limit=2.0
2022-12-24 02:40:01,455 INFO [train.py:894] (2/4) Epoch 28, batch 400, loss[loss=0.1658, simple_loss=0.257, pruned_loss=0.03733, over 18471.00 frames. ], tot_loss[loss=0.1668, simple_loss=0.2563, pruned_loss=0.03864, over 3216411.96 frames. ], batch size: 54, lr: 4.11e-03, grad_scale: 16.0
2022-12-24 02:40:28,919 WARNING [train.py:1060] (2/4) Exclude cut with ID 5622-44585-0006-90525-0 from training. Duration: 25.775
2022-12-24 02:40:29,490 INFO [scaling.py:679] (2/4) Whitening: num_groups=1, num_channels=384, metric=2.59 vs. limit=5.0
2022-12-24 02:40:51,900 WARNING [train.py:1060] (2/4) Exclude cut with ID 2929-85685-0071-62375-0_sp0.9 from training. Duration: 22.25
2022-12-24 02:40:59,066 INFO [zipformer.py:660] (2/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=95105.0, num_to_drop=1, layers_to_drop={0}
2022-12-24 02:41:15,544 INFO [train.py:894] (2/4) Epoch 28, batch 450, loss[loss=0.1683, simple_loss=0.2495, pruned_loss=0.04357, over 18685.00 frames. ], tot_loss[loss=0.1664, simple_loss=0.2558, pruned_loss=0.03848, over 3326853.49 frames. ], batch size: 46, lr: 4.11e-03, grad_scale: 16.0
2022-12-24 02:41:16,247 WARNING [train.py:1060] (2/4) Exclude cut with ID 3972-170212-0014-23379-0 from training. Duration: 26.205
2022-12-24 02:41:32,771 WARNING [train.py:1060] (2/4) Exclude cut with ID 5239-32139-0047-9341-0_sp0.9 from training. Duration: 30.1555625
2022-12-24 02:41:36,649 INFO [scaling.py:679] (2/4) Whitening: num_groups=8, num_channels=96, metric=1.18 vs. limit=2.0
2022-12-24 02:41:39,034 WARNING [train.py:1060] (2/4) Exclude cut with ID 1265-135635-0050-6781-0_sp0.9 from training. Duration: 21.8333125
2022-12-24 02:41:47,793 WARNING [train.py:1060] (2/4) Exclude cut with ID 1914-133440-0024-94914-0_sp1.1 from training. Duration: 20.6545625
2022-12-24 02:42:03,578 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 2.356e+02 3.308e+02 4.017e+02 5.104e+02 8.552e+02, threshold=8.035e+02, percent-clipped=1.0
2022-12-24 02:42:27,686 WARNING [train.py:1060] (2/4) Exclude cut with ID 7395-89880-0045-39920-0_sp0.9 from training. Duration: 20.52225
2022-12-24 02:42:30,400 INFO [train.py:894] (2/4) Epoch 28, batch 500, loss[loss=0.1591, simple_loss=0.2537, pruned_loss=0.03223, over 18507.00 frames. ], tot_loss[loss=0.1674, simple_loss=0.2571, pruned_loss=0.03887, over 3412077.72 frames. ], batch size: 52, lr: 4.11e-03, grad_scale: 16.0
2022-12-24 02:42:30,846 INFO [zipformer.py:660] (2/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=95166.0, num_to_drop=1, layers_to_drop={0}
2022-12-24 02:42:47,791 WARNING [train.py:1060] (2/4) Exclude cut with ID 3972-170212-0014-23379-0_sp0.9 from training. Duration: 29.1166875
2022-12-24 02:42:55,281 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([2.1249, 1.3061, 1.8664, 1.8019, 2.1339, 2.1677, 1.9207, 1.8903],
       device='cuda:2'), covar=tensor([0.2332, 0.3471, 0.2613, 0.2813, 0.2153, 0.1063, 0.3201, 0.1389],
       device='cuda:2'), in_proj_covar=tensor([0.0270, 0.0300, 0.0286, 0.0325, 0.0318, 0.0259, 0.0352, 0.0248],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-24 02:43:40,621 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([2.5530, 2.3932, 1.9936, 1.4334, 2.8951, 2.7107, 2.4032, 1.9824],
       device='cuda:2'), covar=tensor([0.0359, 0.0409, 0.0533, 0.0743, 0.0277, 0.0362, 0.0423, 0.0819],
       device='cuda:2'), in_proj_covar=tensor([0.0125, 0.0131, 0.0130, 0.0118, 0.0104, 0.0127, 0.0133, 0.0162],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0001, 0.0001, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-24 02:43:45,713 INFO [train.py:894] (2/4) Epoch 28, batch 550, loss[loss=0.1577, simple_loss=0.2486, pruned_loss=0.03337, over 18450.00 frames. ], tot_loss[loss=0.1681, simple_loss=0.2576, pruned_loss=0.0393, over 3478740.51 frames. ], batch size: 43, lr: 4.11e-03, grad_scale: 16.0
2022-12-24 02:43:46,075 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.4223, 1.3688, 1.3665, 1.2718, 1.2147, 3.0957, 1.3327, 1.7262],
       device='cuda:2'), covar=tensor([0.4673, 0.3030, 0.2893, 0.3208, 0.1747, 0.0290, 0.1871, 0.1069],
       device='cuda:2'), in_proj_covar=tensor([0.0133, 0.0117, 0.0125, 0.0123, 0.0107, 0.0097, 0.0090, 0.0090],
       device='cuda:2'), out_proj_covar=tensor([0.0005, 0.0005, 0.0005, 0.0005, 0.0004, 0.0003, 0.0004, 0.0004],
       device='cuda:2')
2022-12-24 02:43:47,093 WARNING [train.py:1060] (2/4) Exclude cut with ID 543-133211-0007-59831-0_sp0.9 from training. Duration: 21.388875
2022-12-24 02:44:21,342 WARNING [train.py:1060] (2/4) Exclude cut with ID 1914-133440-0024-94914-0 from training. Duration: 22.72
2022-12-24 02:44:22,826 WARNING [train.py:1060] (2/4) Exclude cut with ID 1914-133440-0031-94921-0_sp0.9 from training. Duration: 22.7444375
2022-12-24 02:44:33,105 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 2.381e+02 3.520e+02 3.982e+02 4.755e+02 1.366e+03, threshold=7.964e+02, percent-clipped=2.0
2022-12-24 02:44:51,894 INFO [zipformer.py:660] (2/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=95260.0, num_to_drop=0, layers_to_drop=set()
2022-12-24 02:45:00,342 INFO [train.py:894] (2/4) Epoch 28, batch 600, loss[loss=0.1485, simple_loss=0.2323, pruned_loss=0.03241, over 18695.00 frames. ], tot_loss[loss=0.1687, simple_loss=0.2584, pruned_loss=0.03951, over 3531225.02 frames. ], batch size: 46, lr: 4.10e-03, grad_scale: 16.0
2022-12-24 02:45:03,653 INFO [zipformer.py:660] (2/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=95268.0, num_to_drop=0, layers_to_drop=set()
2022-12-24 02:45:06,653 WARNING [train.py:1060] (2/4) Exclude cut with ID 4133-6541-0027-40495-0_sp1.1 from training. Duration: 0.9681875
2022-12-24 02:45:10,322 WARNING [train.py:1060] (2/4) Exclude cut with ID 6330-62851-0022-91297-0_sp0.9 from training. Duration: 22.3166875
2022-12-24 02:45:15,954 WARNING [train.py:1060] (2/4) Exclude cut with ID 543-133212-0015-59917-0_sp0.9 from training. Duration: 21.8166875
2022-12-24 02:45:38,854 INFO [scaling.py:679] (2/4) Whitening: num_groups=1, num_channels=384, metric=4.83 vs. limit=5.0
2022-12-24 02:46:15,547 INFO [train.py:894] (2/4) Epoch 28, batch 650, loss[loss=0.1669, simple_loss=0.255, pruned_loss=0.03939, over 18679.00 frames. ], tot_loss[loss=0.1686, simple_loss=0.2581, pruned_loss=0.03951, over 3571814.67 frames. ], batch size: 48, lr: 4.10e-03, grad_scale: 16.0
2022-12-24 02:46:23,520 INFO [zipformer.py:660] (2/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=95321.0, num_to_drop=0, layers_to_drop=set()
2022-12-24 02:46:29,986 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.6606, 1.9966, 1.7652, 2.3176, 2.8765, 1.7249, 1.8076, 1.4545],
       device='cuda:2'), covar=tensor([0.1987, 0.1904, 0.1626, 0.1090, 0.1077, 0.1132, 0.2037, 0.1608],
       device='cuda:2'), in_proj_covar=tensor([0.0251, 0.0232, 0.0223, 0.0204, 0.0265, 0.0200, 0.0231, 0.0205],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-24 02:46:35,543 INFO [zipformer.py:660] (2/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=95329.0, num_to_drop=0, layers_to_drop=set()
2022-12-24 02:46:49,745 INFO [zipformer.py:660] (2/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=95339.0, num_to_drop=0, layers_to_drop=set()
2022-12-24 02:46:58,405 WARNING [train.py:1060] (2/4) Exclude cut with ID 4957-30119-0041-23990-0_sp0.9 from training. Duration: 20.22775
2022-12-24 02:47:03,201 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 1.842e+02 3.223e+02 3.713e+02 4.578e+02 8.119e+02, threshold=7.427e+02, percent-clipped=2.0
2022-12-24 02:47:29,692 INFO [train.py:894] (2/4) Epoch 28, batch 700, loss[loss=0.1427, simple_loss=0.2274, pruned_loss=0.02902, over 18612.00 frames. ], tot_loss[loss=0.1681, simple_loss=0.2578, pruned_loss=0.03926, over 3602471.46 frames. ], batch size: 41, lr: 4.10e-03, grad_scale: 16.0
2022-12-24 02:47:33,491 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([2.0389, 1.8820, 1.6891, 1.0266, 2.2535, 2.1067, 1.9495, 1.5243],
       device='cuda:2'), covar=tensor([0.0388, 0.0474, 0.0480, 0.0786, 0.0352, 0.0393, 0.0408, 0.0927],
       device='cuda:2'), in_proj_covar=tensor([0.0125, 0.0132, 0.0130, 0.0118, 0.0104, 0.0128, 0.0134, 0.0162],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0001, 0.0001, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-24 02:47:42,385 WARNING [train.py:1060] (2/4) Exclude cut with ID 5239-32139-0047-9341-0_sp1.1 from training. Duration: 24.67275
2022-12-24 02:48:10,441 WARNING [train.py:1060] (2/4) Exclude cut with ID 3082-165428-0081-50734-0_sp0.9 from training. Duration: 21.8055625
2022-12-24 02:48:21,862 INFO [zipformer.py:660] (2/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=95400.0, num_to_drop=0, layers_to_drop=set()
2022-12-24 02:48:44,506 INFO [train.py:894] (2/4) Epoch 28, batch 750, loss[loss=0.1689, simple_loss=0.2686, pruned_loss=0.03454, over 18544.00 frames. ], tot_loss[loss=0.1681, simple_loss=0.2575, pruned_loss=0.0394, over 3627492.20 frames. ], batch size: 55, lr: 4.10e-03, grad_scale: 16.0
2022-12-24 02:48:47,740 WARNING [train.py:1060] (2/4) Exclude cut with ID 3340-169293-0054-76830-0_sp0.9 from training. Duration: 22.6666875
2022-12-24 02:49:32,038 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 2.240e+02 3.241e+02 3.902e+02 5.298e+02 1.359e+03, threshold=7.804e+02, percent-clipped=7.0
2022-12-24 02:49:32,376 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.6054, 1.5298, 1.5552, 1.4842, 1.2794, 3.6200, 1.4919, 1.9644],
       device='cuda:2'), covar=tensor([0.3267, 0.2194, 0.2186, 0.2276, 0.1594, 0.0215, 0.1746, 0.0956],
       device='cuda:2'), in_proj_covar=tensor([0.0132, 0.0118, 0.0125, 0.0123, 0.0107, 0.0097, 0.0090, 0.0090],
       device='cuda:2'), out_proj_covar=tensor([0.0005, 0.0005, 0.0005, 0.0005, 0.0004, 0.0003, 0.0004, 0.0004],
       device='cuda:2')
2022-12-24 02:49:43,143 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([2.2245, 2.1818, 2.9880, 1.4133, 2.8076, 2.6307, 1.8366, 2.7831],
       device='cuda:2'), covar=tensor([0.1635, 0.2121, 0.1274, 0.2657, 0.0948, 0.1562, 0.2475, 0.0981],
       device='cuda:2'), in_proj_covar=tensor([0.0198, 0.0214, 0.0206, 0.0194, 0.0171, 0.0215, 0.0215, 0.0193],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-24 02:49:50,294 WARNING [train.py:1060] (2/4) Exclude cut with ID 2411-132532-0017-82279-0_sp1.1 from training. Duration: 0.9681875
2022-12-24 02:49:51,890 INFO [zipformer.py:660] (2/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=95461.0, num_to_drop=1, layers_to_drop={3}
2022-12-24 02:50:00,125 INFO [train.py:894] (2/4) Epoch 28, batch 800, loss[loss=0.2081, simple_loss=0.2922, pruned_loss=0.06197, over 18520.00 frames. ], tot_loss[loss=0.1682, simple_loss=0.2575, pruned_loss=0.03943, over 3646804.54 frames. ], batch size: 58, lr: 4.10e-03, grad_scale: 16.0
2022-12-24 02:50:17,024 WARNING [train.py:1060] (2/4) Exclude cut with ID 6330-62850-0007-91323-0 from training. Duration: 22.485
2022-12-24 02:50:53,595 WARNING [train.py:1060] (2/4) Exclude cut with ID 3972-170212-0014-23379-0_sp1.1 from training. Duration: 23.82275
2022-12-24 02:51:05,169 WARNING [train.py:1060] (2/4) Exclude cut with ID 4860-13185-0032-76709-0 from training. Duration: 20.77
2022-12-24 02:51:13,290 INFO [train.py:894] (2/4) Epoch 28, batch 850, loss[loss=0.1571, simple_loss=0.2386, pruned_loss=0.03782, over 18624.00 frames. ], tot_loss[loss=0.1691, simple_loss=0.2586, pruned_loss=0.03981, over 3661433.83 frames. ], batch size: 41, lr: 4.10e-03, grad_scale: 16.0
2022-12-24 02:51:14,691 WARNING [train.py:1060] (2/4) Exclude cut with ID 6426-64292-0017-15984-0_sp0.9 from training. Duration: 24.088875
2022-12-24 02:51:42,226 WARNING [train.py:1060] (2/4) Exclude cut with ID 6330-62850-0007-91323-0_sp1.1 from training. Duration: 20.4409375
2022-12-24 02:51:59,546 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 2.124e+02 3.261e+02 3.961e+02 5.080e+02 1.056e+03, threshold=7.922e+02, percent-clipped=6.0
2022-12-24 02:52:27,514 INFO [train.py:894] (2/4) Epoch 28, batch 900, loss[loss=0.1631, simple_loss=0.2613, pruned_loss=0.03249, over 18645.00 frames. ], tot_loss[loss=0.1689, simple_loss=0.2586, pruned_loss=0.03959, over 3673257.89 frames. ], batch size: 69, lr: 4.10e-03, grad_scale: 16.0
2022-12-24 02:52:34,832 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.9237, 2.3349, 1.8474, 2.6965, 3.3541, 1.9262, 2.3382, 1.5671],
       device='cuda:2'), covar=tensor([0.1836, 0.1723, 0.1506, 0.1018, 0.1115, 0.1010, 0.1622, 0.1489],
       device='cuda:2'), in_proj_covar=tensor([0.0252, 0.0233, 0.0224, 0.0205, 0.0266, 0.0200, 0.0231, 0.0205],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-24 02:52:52,731 INFO [scaling.py:679] (2/4) Whitening: num_groups=8, num_channels=192, metric=1.90 vs. limit=2.0
2022-12-24 02:53:00,687 WARNING [train.py:1060] (2/4) Exclude cut with ID 2929-85685-0062-62366-0_sp0.9 from training. Duration: 22.511125
2022-12-24 02:53:00,709 WARNING [train.py:1060] (2/4) Exclude cut with ID 7395-89880-0031-39906-0 from training. Duration: 20.675
2022-12-24 02:53:43,112 INFO [train.py:894] (2/4) Epoch 28, batch 950, loss[loss=0.1495, simple_loss=0.2475, pruned_loss=0.02578, over 18610.00 frames. ], tot_loss[loss=0.1674, simple_loss=0.2572, pruned_loss=0.03885, over 3681978.95 frames. ], batch size: 77, lr: 4.10e-03, grad_scale: 16.0
2022-12-24 02:53:43,298 INFO [zipformer.py:660] (2/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=95616.0, num_to_drop=0, layers_to_drop=set()
2022-12-24 02:53:44,999 INFO [zipformer.py:660] (2/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=95617.0, num_to_drop=0, layers_to_drop=set()
2022-12-24 02:53:55,335 INFO [zipformer.py:660] (2/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=95624.0, num_to_drop=0, layers_to_drop=set()
2022-12-24 02:54:30,164 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 2.028e+02 3.050e+02 3.792e+02 4.585e+02 1.055e+03, threshold=7.584e+02, percent-clipped=2.0
2022-12-24 02:54:36,445 WARNING [train.py:1060] (2/4) Exclude cut with ID 6330-62850-0007-91323-0_sp0.9 from training. Duration: 24.9833125
2022-12-24 02:54:57,703 INFO [train.py:894] (2/4) Epoch 28, batch 1000, loss[loss=0.1468, simple_loss=0.2358, pruned_loss=0.02893, over 18530.00 frames. ], tot_loss[loss=0.1686, simple_loss=0.2586, pruned_loss=0.03929, over 3689775.34 frames. ], batch size: 47, lr: 4.10e-03, grad_scale: 16.0
2022-12-24 02:55:08,401 WARNING [train.py:1060] (2/4) Exclude cut with ID 5239-32139-0047-9341-0 from training. Duration: 27.14
2022-12-24 02:55:15,696 INFO [zipformer.py:660] (2/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=95678.0, num_to_drop=0, layers_to_drop=set()
2022-12-24 02:55:24,954 WARNING [train.py:1060] (2/4) Exclude cut with ID 2929-85685-0044-62348-0 from training. Duration: 22.44
2022-12-24 02:55:34,651 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.9614, 1.8591, 1.6837, 0.9313, 2.2922, 2.0700, 1.8386, 1.5242],
       device='cuda:2'), covar=tensor([0.0455, 0.0551, 0.0516, 0.0875, 0.0342, 0.0427, 0.0471, 0.0971],
       device='cuda:2'), in_proj_covar=tensor([0.0126, 0.0132, 0.0130, 0.0119, 0.0104, 0.0128, 0.0134, 0.0162],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0001, 0.0001, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-24 02:55:40,528 INFO [zipformer.py:660] (2/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=95695.0, num_to_drop=0, layers_to_drop=set()
2022-12-24 02:55:44,107 INFO [scaling.py:679] (2/4) Whitening: num_groups=8, num_channels=192, metric=1.85 vs. limit=2.0
2022-12-24 02:56:11,205 INFO [train.py:894] (2/4) Epoch 28, batch 1050, loss[loss=0.1721, simple_loss=0.2571, pruned_loss=0.0436, over 18535.00 frames. ], tot_loss[loss=0.1692, simple_loss=0.2591, pruned_loss=0.03964, over 3696065.09 frames. ], batch size: 47, lr: 4.09e-03, grad_scale: 16.0
2022-12-24 02:56:34,240 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.9919, 1.5590, 1.0423, 1.4724, 2.4845, 1.5308, 1.7943, 1.8270],
       device='cuda:2'), covar=tensor([0.1458, 0.1954, 0.2172, 0.1451, 0.1469, 0.1747, 0.1394, 0.1632],
       device='cuda:2'), in_proj_covar=tensor([0.0093, 0.0096, 0.0115, 0.0096, 0.0119, 0.0092, 0.0098, 0.0094],
       device='cuda:2'), out_proj_covar=tensor([0.0003, 0.0003, 0.0004, 0.0003, 0.0004, 0.0003, 0.0003, 0.0003],
       device='cuda:2')
2022-12-24 02:56:38,392 WARNING [train.py:1060] (2/4) Exclude cut with ID 2929-85685-0060-62364-0_sp0.9 from training. Duration: 21.361125
2022-12-24 02:56:45,083 WARNING [train.py:1060] (2/4) Exclude cut with ID 2929-85685-0079-62383-0_sp1.1 from training. Duration: 27.0318125
2022-12-24 02:56:55,935 WARNING [train.py:1060] (2/4) Exclude cut with ID 5622-44585-0006-90525-0_sp0.9 from training. Duration: 28.638875
2022-12-24 02:56:58,768 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 1.978e+02 3.368e+02 4.220e+02 5.256e+02 1.036e+03, threshold=8.440e+02, percent-clipped=6.0
2022-12-24 02:57:10,927 WARNING [train.py:1060] (2/4) Exclude cut with ID 3340-169293-0054-76830-0 from training. Duration: 20.4
2022-12-24 02:57:18,726 INFO [zipformer.py:660] (2/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=95761.0, num_to_drop=1, layers_to_drop={2}
2022-12-24 02:57:25,877 INFO [train.py:894] (2/4) Epoch 28, batch 1100, loss[loss=0.1559, simple_loss=0.244, pruned_loss=0.03391, over 18684.00 frames. ], tot_loss[loss=0.1682, simple_loss=0.2579, pruned_loss=0.03927, over 3700525.49 frames. ], batch size: 48, lr: 4.09e-03, grad_scale: 16.0
2022-12-24 02:57:43,357 WARNING [train.py:1060] (2/4) Exclude cut with ID 2929-85685-0071-62375-0 from training. Duration: 20.025
2022-12-24 02:57:43,370 WARNING [train.py:1060] (2/4) Exclude cut with ID 2364-131735-0112-64612-0_sp0.9 from training. Duration: 20.488875
2022-12-24 02:57:49,183 WARNING [train.py:1060] (2/4) Exclude cut with ID 2929-85685-0079-62383-0 from training. Duration: 29.735
2022-12-24 02:58:30,418 INFO [zipformer.py:660] (2/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=95809.0, num_to_drop=1, layers_to_drop={1}
2022-12-24 02:58:34,526 INFO [zipformer.py:660] (2/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=95812.0, num_to_drop=1, layers_to_drop={0}
2022-12-24 02:58:40,335 INFO [train.py:894] (2/4) Epoch 28, batch 1150, loss[loss=0.1683, simple_loss=0.255, pruned_loss=0.04079, over 18437.00 frames. ], tot_loss[loss=0.1681, simple_loss=0.2581, pruned_loss=0.03907, over 3703266.33 frames. ], batch size: 48, lr: 4.09e-03, grad_scale: 16.0
2022-12-24 02:59:07,932 WARNING [train.py:1060] (2/4) Exclude cut with ID 7276-92427-0014-12983-0_sp0.9 from training. Duration: 21.3055625
2022-12-24 02:59:09,736 WARNING [train.py:1060] (2/4) Exclude cut with ID 1025-75365-0008-79168-0_sp0.9 from training. Duration: 22.0666875
2022-12-24 02:59:25,947 INFO [zipformer.py:660] (2/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=95847.0, num_to_drop=0, layers_to_drop=set()
2022-12-24 02:59:26,972 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 2.213e+02 3.423e+02 3.966e+02 5.306e+02 8.843e+02, threshold=7.932e+02, percent-clipped=1.0
2022-12-24 02:59:54,523 INFO [train.py:894] (2/4) Epoch 28, batch 1200, loss[loss=0.1769, simple_loss=0.2796, pruned_loss=0.0371, over 18704.00 frames. ], tot_loss[loss=0.169, simple_loss=0.2591, pruned_loss=0.03944, over 3706233.11 frames. ], batch size: 60, lr: 4.09e-03, grad_scale: 16.0
2022-12-24 03:00:04,906 INFO [zipformer.py:660] (2/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=95873.0, num_to_drop=1, layers_to_drop={1}
2022-12-24 03:00:55,596 WARNING [train.py:1060] (2/4) Exclude cut with ID 2929-85685-0062-62366-0 from training. Duration: 20.26
2022-12-24 03:00:57,563 INFO [zipformer.py:660] (2/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=95908.0, num_to_drop=1, layers_to_drop={1}
2022-12-24 03:01:02,000 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.9829, 2.0920, 2.2840, 1.3350, 2.3657, 2.5105, 1.7183, 2.7789],
       device='cuda:2'), covar=tensor([0.1278, 0.1899, 0.1282, 0.2089, 0.0753, 0.1068, 0.2441, 0.0572],
       device='cuda:2'), in_proj_covar=tensor([0.0195, 0.0211, 0.0205, 0.0192, 0.0170, 0.0213, 0.0214, 0.0191],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-24 03:01:08,839 INFO [train.py:894] (2/4) Epoch 28, batch 1250, loss[loss=0.1642, simple_loss=0.2669, pruned_loss=0.0307, over 18479.00 frames. ], tot_loss[loss=0.1689, simple_loss=0.2591, pruned_loss=0.03937, over 3707450.73 frames. ], batch size: 54, lr: 4.09e-03, grad_scale: 16.0
2022-12-24 03:01:09,054 INFO [zipformer.py:660] (2/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=95916.0, num_to_drop=0, layers_to_drop=set()
2022-12-24 03:01:10,260 WARNING [train.py:1060] (2/4) Exclude cut with ID 5239-32139-0030-9324-0_sp0.9 from training. Duration: 21.3444375
2022-12-24 03:01:20,548 INFO [zipformer.py:660] (2/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=95924.0, num_to_drop=0, layers_to_drop=set()
2022-12-24 03:01:55,407 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 1.867e+02 3.056e+02 3.610e+02 4.347e+02 1.155e+03, threshold=7.221e+02, percent-clipped=3.0
2022-12-24 03:02:09,529 WARNING [train.py:1060] (2/4) Exclude cut with ID 497-129325-0061-62254-0_sp1.1 from training. Duration: 0.97725
2022-12-24 03:02:19,707 INFO [zipformer.py:660] (2/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=95964.0, num_to_drop=0, layers_to_drop=set()
2022-12-24 03:02:22,335 INFO [train.py:894] (2/4) Epoch 28, batch 1300, loss[loss=0.1702, simple_loss=0.2615, pruned_loss=0.03947, over 18533.00 frames. ], tot_loss[loss=0.1677, simple_loss=0.2576, pruned_loss=0.03889, over 3708814.40 frames. ], batch size: 98, lr: 4.09e-03, grad_scale: 16.0
2022-12-24 03:02:30,993 INFO [zipformer.py:660] (2/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=95972.0, num_to_drop=0, layers_to_drop=set()
2022-12-24 03:02:32,621 INFO [zipformer.py:660] (2/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=95973.0, num_to_drop=0, layers_to_drop=set()
2022-12-24 03:02:42,576 INFO [zipformer.py:660] (2/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=95980.0, num_to_drop=0, layers_to_drop=set()
2022-12-24 03:02:53,517 WARNING [train.py:1060] (2/4) Exclude cut with ID 7395-89880-0031-39906-0_sp0.9 from training. Duration: 22.97225
2022-12-24 03:03:04,932 INFO [zipformer.py:660] (2/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=95995.0, num_to_drop=0, layers_to_drop=set()
2022-12-24 03:03:29,263 WARNING [train.py:1060] (2/4) Exclude cut with ID 7395-89880-0047-39922-0_sp0.9 from training. Duration: 21.97775
2022-12-24 03:03:36,748 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.4578, 1.8492, 1.5326, 2.1525, 2.1760, 1.6464, 1.3657, 1.3005],
       device='cuda:2'), covar=tensor([0.2189, 0.1984, 0.1787, 0.1165, 0.1445, 0.1197, 0.2458, 0.1734],
       device='cuda:2'), in_proj_covar=tensor([0.0250, 0.0232, 0.0222, 0.0204, 0.0264, 0.0199, 0.0230, 0.0204],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-24 03:03:39,139 INFO [train.py:894] (2/4) Epoch 28, batch 1350, loss[loss=0.1782, simple_loss=0.2715, pruned_loss=0.04239, over 18537.00 frames. ], tot_loss[loss=0.1677, simple_loss=0.2579, pruned_loss=0.03873, over 3709820.76 frames. ], batch size: 55, lr: 4.09e-03, grad_scale: 16.0
2022-12-24 03:03:42,532 WARNING [train.py:1060] (2/4) Exclude cut with ID 1112-1043-0006-89194-0_sp0.9 from training. Duration: 21.8333125
2022-12-24 03:03:52,526 WARNING [train.py:1060] (2/4) Exclude cut with ID 1914-133440-0031-94921-0 from training. Duration: 20.47
2022-12-24 03:04:17,343 INFO [zipformer.py:660] (2/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=96041.0, num_to_drop=0, layers_to_drop=set()
2022-12-24 03:04:20,237 INFO [zipformer.py:660] (2/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=96043.0, num_to_drop=0, layers_to_drop=set()
2022-12-24 03:04:27,235 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 2.260e+02 3.035e+02 3.982e+02 4.869e+02 1.123e+03, threshold=7.963e+02, percent-clipped=2.0
2022-12-24 03:04:54,709 INFO [train.py:894] (2/4) Epoch 28, batch 1400, loss[loss=0.1587, simple_loss=0.2447, pruned_loss=0.03632, over 18701.00 frames. ], tot_loss[loss=0.1669, simple_loss=0.2573, pruned_loss=0.03822, over 3711349.23 frames. ], batch size: 50, lr: 4.09e-03, grad_scale: 16.0
2022-12-24 03:04:57,590 WARNING [train.py:1060] (2/4) Exclude cut with ID 7395-89880-0037-39912-0_sp0.9 from training. Duration: 20.67225
2022-12-24 03:05:11,282 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([3.9471, 3.4639, 3.4195, 3.8798, 3.6432, 3.4417, 4.0545, 1.2843],
       device='cuda:2'), covar=tensor([0.0716, 0.0739, 0.0708, 0.0778, 0.1220, 0.1193, 0.0625, 0.4911],
       device='cuda:2'), in_proj_covar=tensor([0.0357, 0.0235, 0.0244, 0.0281, 0.0336, 0.0273, 0.0301, 0.0293],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0001, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-24 03:05:16,023 WARNING [train.py:1060] (2/4) Exclude cut with ID 1914-133440-0024-94914-0_sp0.9 from training. Duration: 25.2444375
2022-12-24 03:05:16,382 INFO [zipformer.py:660] (2/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=96080.0, num_to_drop=0, layers_to_drop=set()
2022-12-24 03:05:37,410 WARNING [train.py:1060] (2/4) Exclude cut with ID 3340-169293-0021-76797-0_sp0.9 from training. Duration: 21.1445
2022-12-24 03:05:48,991 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([2.1813, 2.0730, 1.8579, 1.1521, 2.4852, 2.3169, 1.9581, 1.6100],
       device='cuda:2'), covar=tensor([0.0449, 0.0499, 0.0518, 0.0842, 0.0343, 0.0430, 0.0520, 0.0998],
       device='cuda:2'), in_proj_covar=tensor([0.0126, 0.0132, 0.0130, 0.0119, 0.0104, 0.0129, 0.0135, 0.0162],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0001, 0.0001, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-24 03:06:11,357 INFO [train.py:894] (2/4) Epoch 28, batch 1450, loss[loss=0.1545, simple_loss=0.2498, pruned_loss=0.02957, over 18473.00 frames. ], tot_loss[loss=0.1671, simple_loss=0.2573, pruned_loss=0.03845, over 3710384.35 frames. ], batch size: 54, lr: 4.09e-03, grad_scale: 16.0
2022-12-24 03:06:47,609 WARNING [train.py:1060] (2/4) Exclude cut with ID 2929-85685-0079-62383-0_sp0.9 from training. Duration: 33.038875
2022-12-24 03:06:47,976 INFO [zipformer.py:660] (2/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=96141.0, num_to_drop=0, layers_to_drop=set()
2022-12-24 03:06:58,572 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 2.188e+02 3.324e+02 3.803e+02 4.702e+02 9.088e+02, threshold=7.606e+02, percent-clipped=2.0
2022-12-24 03:07:22,820 INFO [zipformer.py:660] (2/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=96165.0, num_to_drop=0, layers_to_drop=set()
2022-12-24 03:07:23,683 INFO [train.py:894] (2/4) Epoch 28, batch 1500, loss[loss=0.1849, simple_loss=0.2747, pruned_loss=0.04759, over 18518.00 frames. ], tot_loss[loss=0.1675, simple_loss=0.2578, pruned_loss=0.03863, over 3711524.25 frames. ], batch size: 64, lr: 4.09e-03, grad_scale: 16.0
2022-12-24 03:07:26,484 WARNING [train.py:1060] (2/4) Exclude cut with ID 6426-64291-0000-16059-0_sp0.9 from training. Duration: 20.0944375
2022-12-24 03:07:26,629 INFO [zipformer.py:660] (2/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=96168.0, num_to_drop=1, layers_to_drop={1}
2022-12-24 03:07:41,935 WARNING [train.py:1060] (2/4) Exclude cut with ID 2929-85685-0044-62348-0_sp1.1 from training. Duration: 20.4
2022-12-24 03:07:49,174 WARNING [train.py:1060] (2/4) Exclude cut with ID 6330-62851-0022-91297-0 from training. Duration: 20.085
2022-12-24 03:07:53,415 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([2.6714, 2.3476, 1.7916, 0.8189, 1.8363, 2.2123, 1.7702, 2.0410],
       device='cuda:2'), covar=tensor([0.0708, 0.0723, 0.1464, 0.1936, 0.1440, 0.1581, 0.1950, 0.0966],
       device='cuda:2'), in_proj_covar=tensor([0.0176, 0.0190, 0.0209, 0.0189, 0.0210, 0.0205, 0.0217, 0.0204],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-24 03:07:59,110 WARNING [train.py:1060] (2/4) Exclude cut with ID 4860-13185-0032-76709-0_sp0.9 from training. Duration: 23.07775
2022-12-24 03:08:19,078 INFO [zipformer.py:660] (2/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=96203.0, num_to_drop=1, layers_to_drop={3}
2022-12-24 03:08:37,045 INFO [train.py:894] (2/4) Epoch 28, batch 1550, loss[loss=0.1854, simple_loss=0.264, pruned_loss=0.05343, over 18564.00 frames. ], tot_loss[loss=0.1674, simple_loss=0.2577, pruned_loss=0.03857, over 3712276.45 frames. ], batch size: 49, lr: 4.08e-03, grad_scale: 16.0
2022-12-24 03:08:38,994 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([2.3727, 2.2878, 1.6992, 2.6754, 2.5020, 2.2086, 3.0712, 2.3373],
       device='cuda:2'), covar=tensor([0.0899, 0.1973, 0.2973, 0.1807, 0.1898, 0.0926, 0.1000, 0.1351],
       device='cuda:2'), in_proj_covar=tensor([0.0185, 0.0220, 0.0260, 0.0294, 0.0246, 0.0198, 0.0210, 0.0212],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-24 03:08:42,562 WARNING [train.py:1060] (2/4) Exclude cut with ID 2929-85685-0044-62348-0_sp0.9 from training. Duration: 24.9333125
2022-12-24 03:08:52,688 INFO [zipformer.py:660] (2/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=96226.0, num_to_drop=0, layers_to_drop=set()
2022-12-24 03:09:01,599 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([2.2289, 2.2043, 1.6413, 2.6908, 2.4992, 2.1649, 3.0436, 2.3234],
       device='cuda:2'), covar=tensor([0.0845, 0.1999, 0.2901, 0.1656, 0.1786, 0.0889, 0.0933, 0.1240],
       device='cuda:2'), in_proj_covar=tensor([0.0184, 0.0219, 0.0259, 0.0293, 0.0245, 0.0197, 0.0210, 0.0212],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-24 03:09:24,986 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 2.260e+02 3.285e+02 3.978e+02 4.636e+02 1.099e+03, threshold=7.955e+02, percent-clipped=3.0
2022-12-24 03:09:26,621 WARNING [train.py:1060] (2/4) Exclude cut with ID 5118-111612-0016-124680-0_sp0.9 from training. Duration: 20.388875
2022-12-24 03:09:33,752 WARNING [train.py:1060] (2/4) Exclude cut with ID 432-122774-0017-62487-0_sp1.1 from training. Duration: 20.3590625
2022-12-24 03:09:50,894 INFO [train.py:894] (2/4) Epoch 28, batch 1600, loss[loss=0.191, simple_loss=0.276, pruned_loss=0.05295, over 18683.00 frames. ], tot_loss[loss=0.1679, simple_loss=0.2581, pruned_loss=0.03888, over 3712848.74 frames. ], batch size: 78, lr: 4.08e-03, grad_scale: 16.0
2022-12-24 03:10:02,752 INFO [zipformer.py:660] (2/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=96273.0, num_to_drop=0, layers_to_drop=set()
2022-12-24 03:10:40,915 WARNING [train.py:1060] (2/4) Exclude cut with ID 3557-8342-0013-54691-0_sp1.1 from training. Duration: 0.836375
2022-12-24 03:10:56,995 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([5.7210, 4.9109, 4.9918, 5.6993, 5.2871, 5.0757, 5.8257, 1.7127],
       device='cuda:2'), covar=tensor([0.0622, 0.0735, 0.0594, 0.0750, 0.1234, 0.1181, 0.0452, 0.5634],
       device='cuda:2'), in_proj_covar=tensor([0.0360, 0.0238, 0.0247, 0.0285, 0.0339, 0.0277, 0.0305, 0.0297],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0001, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-24 03:11:00,468 INFO [scaling.py:679] (2/4) Whitening: num_groups=1, num_channels=384, metric=2.79 vs. limit=5.0
2022-12-24 03:11:06,336 INFO [train.py:894] (2/4) Epoch 28, batch 1650, loss[loss=0.1839, simple_loss=0.2725, pruned_loss=0.04766, over 18463.00 frames. ], tot_loss[loss=0.1677, simple_loss=0.2576, pruned_loss=0.03889, over 3711852.37 frames. ], batch size: 64, lr: 4.08e-03, grad_scale: 32.0
2022-12-24 03:11:10,369 INFO [scaling.py:679] (2/4) Whitening: num_groups=8, num_channels=96, metric=1.17 vs. limit=2.0
2022-12-24 03:11:14,262 INFO [zipformer.py:660] (2/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=96321.0, num_to_drop=0, layers_to_drop=set()
2022-12-24 03:11:18,624 INFO [zipformer.py:660] (2/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=96324.0, num_to_drop=0, layers_to_drop=set()
2022-12-24 03:11:24,053 WARNING [train.py:1060] (2/4) Exclude cut with ID 8565-290391-0049-67394-0_sp0.9 from training. Duration: 21.3166875
2022-12-24 03:11:35,930 INFO [zipformer.py:660] (2/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=96336.0, num_to_drop=0, layers_to_drop=set()
2022-12-24 03:11:54,046 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 1.974e+02 3.462e+02 4.232e+02 5.237e+02 1.043e+03, threshold=8.464e+02, percent-clipped=3.0
2022-12-24 03:11:55,491 WARNING [train.py:1060] (2/4) Exclude cut with ID 6533-399-0029-104863-0_sp0.9 from training. Duration: 22.1055625
2022-12-24 03:12:06,254 WARNING [train.py:1060] (2/4) Exclude cut with ID 7699-105389-0094-26379-0_sp1.1 from training. Duration: 21.77725
2022-12-24 03:12:19,864 INFO [train.py:894] (2/4) Epoch 28, batch 1700, loss[loss=0.1534, simple_loss=0.2428, pruned_loss=0.03202, over 18578.00 frames. ], tot_loss[loss=0.1701, simple_loss=0.2592, pruned_loss=0.04052, over 3712333.89 frames. ], batch size: 49, lr: 4.08e-03, grad_scale: 32.0
2022-12-24 03:12:25,925 WARNING [train.py:1060] (2/4) Exclude cut with ID 7255-291500-0005-134304-0_sp0.9 from training. Duration: 27.8166875
2022-12-24 03:12:46,915 INFO [zipformer.py:660] (2/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=96385.0, num_to_drop=0, layers_to_drop=set()
2022-12-24 03:12:48,205 WARNING [train.py:1060] (2/4) Exclude cut with ID 7357-94126-0021-15852-0_sp1.1 from training. Duration: 22.5090625
2022-12-24 03:12:55,541 INFO [zipformer.py:660] (2/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=96390.0, num_to_drop=0, layers_to_drop=set()
2022-12-24 03:12:56,700 WARNING [train.py:1060] (2/4) Exclude cut with ID 7255-291500-0005-134304-0 from training. Duration: 25.035
2022-12-24 03:13:14,384 WARNING [train.py:1060] (2/4) Exclude cut with ID 774-127930-0014-10412-0_sp1.1 from training. Duration: 0.95
2022-12-24 03:13:32,273 WARNING [train.py:1060] (2/4) Exclude cut with ID 3033-130750-0096-55598-0_sp0.9 from training. Duration: 0.92225
2022-12-24 03:13:34,216 INFO [train.py:894] (2/4) Epoch 28, batch 1750, loss[loss=0.1847, simple_loss=0.2622, pruned_loss=0.05358, over 18399.00 frames. ], tot_loss[loss=0.1717, simple_loss=0.2601, pruned_loss=0.04169, over 3713912.76 frames. ], batch size: 46, lr: 4.08e-03, grad_scale: 32.0
2022-12-24 03:13:57,566 WARNING [train.py:1060] (2/4) Exclude cut with ID 4511-76322-0006-80011-0 from training. Duration: 21.97
2022-12-24 03:14:04,036 INFO [zipformer.py:660] (2/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=96436.0, num_to_drop=0, layers_to_drop=set()
2022-12-24 03:14:16,153 WARNING [train.py:1060] (2/4) Exclude cut with ID 7492-105653-0055-62765-0_sp0.9 from training. Duration: 21.97225
2022-12-24 03:14:17,447 WARNING [train.py:1060] (2/4) Exclude cut with ID 453-131332-0000-47844-0_sp0.9 from training. Duration: 25.3333125
2022-12-24 03:14:21,834 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 2.827e+02 4.069e+02 4.851e+02 6.169e+02 1.123e+03, threshold=9.703e+02, percent-clipped=7.0
2022-12-24 03:14:26,372 INFO [zipformer.py:660] (2/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=96451.0, num_to_drop=0, layers_to_drop=set()
2022-12-24 03:14:27,433 WARNING [train.py:1060] (2/4) Exclude cut with ID 5172-29468-0015-19128-0_sp0.9 from training. Duration: 21.5055625
2022-12-24 03:14:30,515 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([3.8176, 3.3445, 3.3062, 3.7428, 3.4633, 3.3494, 3.9341, 1.4197],
       device='cuda:2'), covar=tensor([0.0845, 0.0836, 0.0797, 0.0902, 0.1550, 0.1370, 0.0778, 0.4898],
       device='cuda:2'), in_proj_covar=tensor([0.0360, 0.0238, 0.0247, 0.0286, 0.0340, 0.0277, 0.0305, 0.0297],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0001, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-24 03:14:35,980 WARNING [train.py:1060] (2/4) Exclude cut with ID 453-131332-0000-47844-0_sp1.1 from training. Duration: 20.72725
2022-12-24 03:14:36,339 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([3.2711, 3.1935, 2.4104, 1.7777, 3.6247, 3.7608, 3.0473, 2.6721],
       device='cuda:2'), covar=tensor([0.0402, 0.0424, 0.0529, 0.0722, 0.0281, 0.0387, 0.0463, 0.0755],
       device='cuda:2'), in_proj_covar=tensor([0.0126, 0.0132, 0.0129, 0.0118, 0.0104, 0.0128, 0.0134, 0.0162],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0001, 0.0001, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-24 03:14:48,930 INFO [train.py:894] (2/4) Epoch 28, batch 1800, loss[loss=0.1762, simple_loss=0.2672, pruned_loss=0.04262, over 18528.00 frames. ], tot_loss[loss=0.1741, simple_loss=0.2614, pruned_loss=0.04344, over 3714730.31 frames. ], batch size: 55, lr: 4.08e-03, grad_scale: 32.0
2022-12-24 03:14:52,253 INFO [zipformer.py:660] (2/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=96468.0, num_to_drop=1, layers_to_drop={2}
2022-12-24 03:15:08,520 WARNING [train.py:1060] (2/4) Exclude cut with ID 8631-249866-0030-130156-0_sp0.9 from training. Duration: 26.32775
2022-12-24 03:15:12,350 INFO [scaling.py:679] (2/4) Whitening: num_groups=8, num_channels=192, metric=1.51 vs. limit=2.0
2022-12-24 03:15:40,595 WARNING [train.py:1060] (2/4) Exclude cut with ID 3867-173237-0077-144769-0 from training. Duration: 20.025
2022-12-24 03:15:45,180 INFO [zipformer.py:660] (2/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=96503.0, num_to_drop=0, layers_to_drop=set()
2022-12-24 03:15:46,601 WARNING [train.py:1060] (2/4) Exclude cut with ID 6709-74022-0004-86860-0_sp1.1 from training. Duration: 0.9409375
2022-12-24 03:15:46,615 WARNING [train.py:1060] (2/4) Exclude cut with ID 4757-1811-0023-62229-0_sp0.9 from training. Duration: 21.37775
2022-12-24 03:16:05,110 INFO [train.py:894] (2/4) Epoch 28, batch 1850, loss[loss=0.147, simple_loss=0.2268, pruned_loss=0.03356, over 18602.00 frames. ], tot_loss[loss=0.1754, simple_loss=0.262, pruned_loss=0.04439, over 3715721.16 frames. ], batch size: 41, lr: 4.08e-03, grad_scale: 32.0
2022-12-24 03:16:05,277 INFO [zipformer.py:660] (2/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=96516.0, num_to_drop=1, layers_to_drop={0}
2022-12-24 03:16:09,477 WARNING [train.py:1060] (2/4) Exclude cut with ID 1250-135782-0004-25974-0_sp0.9 from training. Duration: 21.17225
2022-12-24 03:16:09,488 WARNING [train.py:1060] (2/4) Exclude cut with ID 7357-94126-0021-15852-0_sp0.9 from training. Duration: 27.511125
2022-12-24 03:16:12,333 INFO [zipformer.py:660] (2/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=96521.0, num_to_drop=0, layers_to_drop=set()
2022-12-24 03:16:42,499 WARNING [train.py:1060] (2/4) Exclude cut with ID 453-131332-0000-47844-0 from training. Duration: 22.8
2022-12-24 03:16:46,767 WARNING [train.py:1060] (2/4) Exclude cut with ID 4964-30587-0040-44509-0 from training. Duration: 22.585
2022-12-24 03:16:52,435 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 2.179e+02 4.172e+02 5.277e+02 6.658e+02 9.431e+02, threshold=1.055e+03, percent-clipped=0.0
2022-12-24 03:16:56,802 INFO [zipformer.py:660] (2/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=96551.0, num_to_drop=0, layers_to_drop=set()
2022-12-24 03:16:57,147 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([2.3252, 2.2766, 1.6917, 2.6663, 2.4442, 2.1611, 3.0803, 2.3361],
       device='cuda:2'), covar=tensor([0.0834, 0.1726, 0.2813, 0.1687, 0.1816, 0.0878, 0.0929, 0.1224],
       device='cuda:2'), in_proj_covar=tensor([0.0185, 0.0219, 0.0260, 0.0293, 0.0245, 0.0197, 0.0209, 0.0211],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-24 03:17:10,039 INFO [scaling.py:679] (2/4) Whitening: num_groups=1, num_channels=384, metric=3.73 vs. limit=5.0
2022-12-24 03:17:10,915 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([2.2327, 2.7595, 2.7258, 1.0792, 2.9390, 2.0905, 0.4458, 1.7707],
       device='cuda:2'), covar=tensor([0.2194, 0.1651, 0.1768, 0.3772, 0.1286, 0.1136, 0.4844, 0.1725],
       device='cuda:2'), in_proj_covar=tensor([0.0152, 0.0151, 0.0163, 0.0126, 0.0153, 0.0118, 0.0148, 0.0118],
       device='cuda:2'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0002, 0.0003, 0.0002, 0.0003, 0.0002],
       device='cuda:2')
2022-12-24 03:17:17,023 WARNING [train.py:1060] (2/4) Exclude cut with ID 6978-92210-0001-146967-0_sp0.9 from training. Duration: 22.0166875
2022-12-24 03:17:18,925 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.8441, 1.3565, 1.0133, 1.3865, 2.1390, 1.2256, 1.5428, 1.7240],
       device='cuda:2'), covar=tensor([0.1628, 0.2055, 0.2063, 0.1549, 0.1711, 0.1800, 0.1475, 0.1707],
       device='cuda:2'), in_proj_covar=tensor([0.0094, 0.0097, 0.0116, 0.0096, 0.0120, 0.0092, 0.0099, 0.0094],
       device='cuda:2'), out_proj_covar=tensor([0.0003, 0.0003, 0.0004, 0.0003, 0.0004, 0.0003, 0.0003, 0.0003],
       device='cuda:2')
2022-12-24 03:17:20,005 INFO [train.py:894] (2/4) Epoch 28, batch 1900, loss[loss=0.194, simple_loss=0.2732, pruned_loss=0.05743, over 18494.00 frames. ], tot_loss[loss=0.1753, simple_loss=0.2615, pruned_loss=0.04456, over 3715900.10 frames. ], batch size: 64, lr: 4.08e-03, grad_scale: 32.0
2022-12-24 03:17:32,244 WARNING [train.py:1060] (2/4) Exclude cut with ID 7255-291500-0003-134302-0_sp1.1 from training. Duration: 24.395375
2022-12-24 03:17:39,806 WARNING [train.py:1060] (2/4) Exclude cut with ID 3488-85273-0017-41203-0_sp0.9 from training. Duration: 27.47775
2022-12-24 03:17:44,625 WARNING [train.py:1060] (2/4) Exclude cut with ID 432-122774-0017-62487-0_sp0.9 from training. Duration: 24.8833125
2022-12-24 03:17:47,640 WARNING [train.py:1060] (2/4) Exclude cut with ID 6758-72288-0033-108368-0 from training. Duration: 23.39
2022-12-24 03:17:53,367 WARNING [train.py:1060] (2/4) Exclude cut with ID 8291-282929-0007-12994-0_sp0.9 from training. Duration: 28.72225
2022-12-24 03:18:03,503 WARNING [train.py:1060] (2/4) Exclude cut with ID 585-294811-0110-133686-0_sp0.9 from training. Duration: 20.8944375
2022-12-24 03:18:18,248 WARNING [train.py:1060] (2/4) Exclude cut with ID 5796-66357-0007-116447-0_sp0.9 from training. Duration: 23.8444375
2022-12-24 03:18:36,187 INFO [train.py:894] (2/4) Epoch 28, batch 1950, loss[loss=0.1803, simple_loss=0.276, pruned_loss=0.04231, over 18584.00 frames. ], tot_loss[loss=0.1753, simple_loss=0.2612, pruned_loss=0.04468, over 3715578.72 frames. ], batch size: 69, lr: 4.08e-03, grad_scale: 32.0
2022-12-24 03:18:41,777 WARNING [train.py:1060] (2/4) Exclude cut with ID 8291-282929-0007-12994-0 from training. Duration: 25.85
2022-12-24 03:18:41,787 WARNING [train.py:1060] (2/4) Exclude cut with ID 8291-282929-0023-13010-0 from training. Duration: 21.39
2022-12-24 03:18:53,587 WARNING [train.py:1060] (2/4) Exclude cut with ID 7357-94126-0014-15845-0 from training. Duration: 27.92
2022-12-24 03:19:06,085 INFO [zipformer.py:660] (2/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=96636.0, num_to_drop=0, layers_to_drop=set()
2022-12-24 03:19:19,759 WARNING [train.py:1060] (2/4) Exclude cut with ID 8631-249866-0039-130165-0_sp0.9 from training. Duration: 20.661125
2022-12-24 03:19:23,237 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 2.734e+02 3.791e+02 4.616e+02 5.491e+02 1.283e+03, threshold=9.233e+02, percent-clipped=2.0
2022-12-24 03:19:44,187 WARNING [train.py:1060] (2/4) Exclude cut with ID 7357-94126-0043-15874-0_sp0.9 from training. Duration: 20.07225
2022-12-24 03:19:50,085 INFO [train.py:894] (2/4) Epoch 28, batch 2000, loss[loss=0.164, simple_loss=0.2425, pruned_loss=0.04275, over 18602.00 frames. ], tot_loss[loss=0.1751, simple_loss=0.2608, pruned_loss=0.0447, over 3715191.17 frames. ], batch size: 45, lr: 4.07e-03, grad_scale: 32.0
2022-12-24 03:19:51,524 WARNING [train.py:1060] (2/4) Exclude cut with ID 1085-156170-0017-128270-0 from training. Duration: 21.01
2022-12-24 03:20:11,534 INFO [zipformer.py:660] (2/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=96680.0, num_to_drop=0, layers_to_drop=set()
2022-12-24 03:20:17,231 INFO [zipformer.py:660] (2/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=96684.0, num_to_drop=0, layers_to_drop=set()
2022-12-24 03:20:59,278 WARNING [train.py:1060] (2/4) Exclude cut with ID 2195-150901-0045-59933-0 from training. Duration: 20.65
2022-12-24 03:21:05,004 INFO [train.py:894] (2/4) Epoch 28, batch 2050, loss[loss=0.2019, simple_loss=0.2813, pruned_loss=0.06128, over 18686.00 frames. ], tot_loss[loss=0.1754, simple_loss=0.2607, pruned_loss=0.04507, over 3715801.27 frames. ], batch size: 60, lr: 4.07e-03, grad_scale: 32.0
2022-12-24 03:21:06,672 WARNING [train.py:1060] (2/4) Exclude cut with ID 5796-66357-0007-116447-0 from training. Duration: 21.46
2022-12-24 03:21:35,246 INFO [zipformer.py:660] (2/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=96736.0, num_to_drop=0, layers_to_drop=set()
2022-12-24 03:21:50,876 INFO [zipformer.py:660] (2/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=96746.0, num_to_drop=0, layers_to_drop=set()
2022-12-24 03:21:52,030 WARNING [train.py:1060] (2/4) Exclude cut with ID 3557-8342-0013-54691-0 from training. Duration: 0.92
2022-12-24 03:21:53,231 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 2.630e+02 4.069e+02 4.932e+02 5.906e+02 1.093e+03, threshold=9.865e+02, percent-clipped=4.0
2022-12-24 03:21:58,972 WARNING [train.py:1060] (2/4) Exclude cut with ID 8291-282929-0023-13010-0_sp0.9 from training. Duration: 23.7666875
2022-12-24 03:22:19,498 INFO [train.py:894] (2/4) Epoch 28, batch 2100, loss[loss=0.1786, simple_loss=0.2671, pruned_loss=0.04505, over 18724.00 frames. ], tot_loss[loss=0.1755, simple_loss=0.2604, pruned_loss=0.04526, over 3715650.50 frames. ], batch size: 65, lr: 4.07e-03, grad_scale: 32.0
2022-12-24 03:22:34,967 INFO [zipformer.py:660] (2/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=96776.0, num_to_drop=0, layers_to_drop=set()
2022-12-24 03:22:36,130 WARNING [train.py:1060] (2/4) Exclude cut with ID 8544-281189-0060-101339-0_sp0.9 from training. Duration: 20.861125
2022-12-24 03:22:44,698 WARNING [train.py:1060] (2/4) Exclude cut with ID 3488-65654-0031-41259-0_sp0.9 from training. Duration: 22.711125
2022-12-24 03:22:46,442 INFO [zipformer.py:660] (2/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=96784.0, num_to_drop=0, layers_to_drop=set()
2022-12-24 03:23:26,175 WARNING [train.py:1060] (2/4) Exclude cut with ID 6951-79737-0043-132310-0_sp1.1 from training. Duration: 22.986375
2022-12-24 03:23:28,540 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.9689, 1.6356, 1.8118, 2.1801, 1.9121, 3.3179, 1.5633, 1.6849],
       device='cuda:2'), covar=tensor([0.0772, 0.1725, 0.1102, 0.0812, 0.1359, 0.0277, 0.1408, 0.1488],
       device='cuda:2'), in_proj_covar=tensor([0.0072, 0.0082, 0.0072, 0.0075, 0.0091, 0.0076, 0.0084, 0.0077],
       device='cuda:2'), out_proj_covar=tensor([0.0004, 0.0004, 0.0004, 0.0004, 0.0004, 0.0004, 0.0004, 0.0004],
       device='cuda:2')
2022-12-24 03:23:34,598 INFO [train.py:894] (2/4) Epoch 28, batch 2150, loss[loss=0.1842, simple_loss=0.2675, pruned_loss=0.05042, over 18530.00 frames. ], tot_loss[loss=0.1758, simple_loss=0.2602, pruned_loss=0.0457, over 3714739.15 frames. ], batch size: 55, lr: 4.07e-03, grad_scale: 16.0
2022-12-24 03:23:42,468 INFO [zipformer.py:660] (2/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=96821.0, num_to_drop=0, layers_to_drop=set()
2022-12-24 03:23:43,485 WARNING [train.py:1060] (2/4) Exclude cut with ID 8040-260924-0003-80960-0_sp0.9 from training. Duration: 22.07225
2022-12-24 03:23:47,742 WARNING [train.py:1060] (2/4) Exclude cut with ID 7699-105389-0045-26330-0_sp0.9 from training. Duration: 20.3055625
2022-12-24 03:23:49,238 WARNING [train.py:1060] (2/4) Exclude cut with ID 6356-271890-0060-94317-0_sp0.9 from training. Duration: 20.72225
2022-12-24 03:24:05,643 INFO [zipformer.py:660] (2/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=96837.0, num_to_drop=0, layers_to_drop=set()
2022-12-24 03:24:08,680 WARNING [train.py:1060] (2/4) Exclude cut with ID 3488-85273-0017-41203-0_sp1.1 from training. Duration: 22.4818125
2022-12-24 03:24:24,111 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 2.321e+02 3.798e+02 4.750e+02 6.031e+02 1.281e+03, threshold=9.499e+02, percent-clipped=3.0
2022-12-24 03:24:32,433 WARNING [train.py:1060] (2/4) Exclude cut with ID 4964-30587-0040-44509-0_sp0.9 from training. Duration: 25.0944375
2022-12-24 03:24:36,159 WARNING [train.py:1060] (2/4) Exclude cut with ID 6533-399-0047-104881-0 from training. Duration: 21.515
2022-12-24 03:24:43,850 WARNING [train.py:1060] (2/4) Exclude cut with ID 7357-94126-0009-15840-0_sp0.9 from training. Duration: 27.02225
2022-12-24 03:24:49,647 INFO [train.py:894] (2/4) Epoch 28, batch 2200, loss[loss=0.1766, simple_loss=0.268, pruned_loss=0.04265, over 18530.00 frames. ], tot_loss[loss=0.1757, simple_loss=0.2602, pruned_loss=0.04562, over 3713959.54 frames. ], batch size: 58, lr: 4.07e-03, grad_scale: 16.0
2022-12-24 03:24:49,648 WARNING [train.py:1060] (2/4) Exclude cut with ID 432-122774-0010-62480-0_sp0.9 from training. Duration: 22.22225
2022-12-24 03:24:54,266 INFO [zipformer.py:660] (2/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=96869.0, num_to_drop=0, layers_to_drop=set()
2022-12-24 03:24:55,789 WARNING [train.py:1060] (2/4) Exclude cut with ID 4964-30587-0085-44554-0_sp0.9 from training. Duration: 20.85
2022-12-24 03:25:30,330 WARNING [train.py:1060] (2/4) Exclude cut with ID 4295-39940-0007-92567-0 from training. Duration: 21.54
2022-12-24 03:25:34,623 WARNING [train.py:1060] (2/4) Exclude cut with ID 4964-30587-0040-44509-0_sp1.1 from training. Duration: 20.5318125
2022-12-24 03:25:44,790 WARNING [train.py:1060] (2/4) Exclude cut with ID 7255-291500-0012-134311-0_sp0.9 from training. Duration: 21.9333125
2022-12-24 03:26:04,707 INFO [train.py:894] (2/4) Epoch 28, batch 2250, loss[loss=0.1589, simple_loss=0.2463, pruned_loss=0.0357, over 18571.00 frames. ], tot_loss[loss=0.1757, simple_loss=0.26, pruned_loss=0.04567, over 3714230.32 frames. ], batch size: 49, lr: 4.07e-03, grad_scale: 16.0
2022-12-24 03:26:27,558 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.4288, 1.7053, 1.3826, 1.9762, 2.0938, 1.5264, 1.2266, 1.2367],
       device='cuda:2'), covar=tensor([0.2021, 0.1854, 0.1785, 0.1157, 0.1296, 0.1144, 0.2372, 0.1608],
       device='cuda:2'), in_proj_covar=tensor([0.0251, 0.0233, 0.0223, 0.0205, 0.0265, 0.0199, 0.0231, 0.0205],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-24 03:26:30,468 WARNING [train.py:1060] (2/4) Exclude cut with ID 8631-249866-0025-130151-0_sp0.9 from training. Duration: 21.7944375
2022-12-24 03:26:38,922 INFO [scaling.py:679] (2/4) Whitening: num_groups=8, num_channels=192, metric=1.67 vs. limit=2.0
2022-12-24 03:26:43,805 WARNING [train.py:1060] (2/4) Exclude cut with ID 8291-282929-0002-12989-0_sp0.9 from training. Duration: 22.4666875
2022-12-24 03:26:49,619 WARNING [train.py:1060] (2/4) Exclude cut with ID 6121-9014-0076-24124-0 from training. Duration: 21.635
2022-12-24 03:26:54,246 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 2.577e+02 4.074e+02 4.693e+02 5.920e+02 1.379e+03, threshold=9.385e+02, percent-clipped=3.0
2022-12-24 03:26:56,002 WARNING [train.py:1060] (2/4) Exclude cut with ID 6121-9014-0076-24124-0_sp0.9 from training. Duration: 24.038875
2022-12-24 03:27:19,895 INFO [train.py:894] (2/4) Epoch 28, batch 2300, loss[loss=0.2234, simple_loss=0.301, pruned_loss=0.07291, over 18721.00 frames. ], tot_loss[loss=0.1766, simple_loss=0.2609, pruned_loss=0.04614, over 3714558.20 frames. ], batch size: 65, lr: 4.07e-03, grad_scale: 16.0
2022-12-24 03:27:39,442 WARNING [train.py:1060] (2/4) Exclude cut with ID 7255-291500-0009-134308-0_sp1.1 from training. Duration: 21.786375
2022-12-24 03:27:41,160 INFO [zipformer.py:660] (2/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=96980.0, num_to_drop=0, layers_to_drop=set()
2022-12-24 03:27:52,315 WARNING [train.py:1060] (2/4) Exclude cut with ID 8291-282929-0002-12989-0 from training. Duration: 20.22
2022-12-24 03:28:10,518 INFO [zipformer.py:660] (2/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=97000.0, num_to_drop=0, layers_to_drop=set()
2022-12-24 03:28:33,343 INFO [train.py:894] (2/4) Epoch 28, batch 2350, loss[loss=0.1872, simple_loss=0.2773, pruned_loss=0.04854, over 18608.00 frames. ], tot_loss[loss=0.1769, simple_loss=0.261, pruned_loss=0.04639, over 3714504.57 frames. ], batch size: 78, lr: 4.07e-03, grad_scale: 16.0
2022-12-24 03:28:52,053 INFO [zipformer.py:660] (2/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=97028.0, num_to_drop=0, layers_to_drop=set()
2022-12-24 03:29:19,476 INFO [zipformer.py:660] (2/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=97046.0, num_to_drop=0, layers_to_drop=set()
2022-12-24 03:29:23,492 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 2.576e+02 4.083e+02 5.105e+02 6.686e+02 1.127e+03, threshold=1.021e+03, percent-clipped=9.0
2022-12-24 03:29:42,415 INFO [zipformer.py:660] (2/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=97061.0, num_to_drop=0, layers_to_drop=set()
2022-12-24 03:29:47,967 WARNING [train.py:1060] (2/4) Exclude cut with ID 6951-79737-0043-132310-0 from training. Duration: 25.285
2022-12-24 03:29:49,350 INFO [train.py:894] (2/4) Epoch 28, batch 2400, loss[loss=0.22, simple_loss=0.3012, pruned_loss=0.06941, over 18717.00 frames. ], tot_loss[loss=0.1774, simple_loss=0.2614, pruned_loss=0.04671, over 3714375.65 frames. ], batch size: 54, lr: 4.07e-03, grad_scale: 16.0
2022-12-24 03:30:21,545 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.9209, 1.8996, 2.1634, 1.2384, 2.1383, 2.2357, 1.6874, 2.6658],
       device='cuda:2'), covar=tensor([0.1171, 0.1855, 0.1201, 0.1867, 0.0753, 0.1097, 0.2204, 0.0535],
       device='cuda:2'), in_proj_covar=tensor([0.0198, 0.0215, 0.0208, 0.0195, 0.0171, 0.0218, 0.0217, 0.0194],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-24 03:30:29,867 INFO [zipformer.py:660] (2/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=97094.0, num_to_drop=0, layers_to_drop=set()
2022-12-24 03:30:35,785 INFO [scaling.py:679] (2/4) Whitening: num_groups=8, num_channels=192, metric=1.57 vs. limit=2.0
2022-12-24 03:30:52,959 WARNING [train.py:1060] (2/4) Exclude cut with ID 811-130148-0001-63453-0_sp0.9 from training. Duration: 20.861125
2022-12-24 03:31:03,779 INFO [train.py:894] (2/4) Epoch 28, batch 2450, loss[loss=0.161, simple_loss=0.2387, pruned_loss=0.04165, over 18590.00 frames. ], tot_loss[loss=0.1765, simple_loss=0.2604, pruned_loss=0.04635, over 3714553.27 frames. ], batch size: 45, lr: 4.07e-03, grad_scale: 16.0
2022-12-24 03:31:15,340 WARNING [train.py:1060] (2/4) Exclude cut with ID 6010-56788-0055-90261-0 from training. Duration: 20.88
2022-12-24 03:31:27,037 INFO [zipformer.py:660] (2/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=97132.0, num_to_drop=0, layers_to_drop=set()
2022-12-24 03:31:46,226 WARNING [train.py:1060] (2/4) Exclude cut with ID 7357-94126-0045-15876-0_sp0.9 from training. Duration: 23.4166875
2022-12-24 03:31:53,014 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 2.282e+02 3.799e+02 4.558e+02 5.930e+02 1.895e+03, threshold=9.116e+02, percent-clipped=2.0
2022-12-24 03:32:18,242 INFO [train.py:894] (2/4) Epoch 28, batch 2500, loss[loss=0.165, simple_loss=0.2567, pruned_loss=0.03667, over 18678.00 frames. ], tot_loss[loss=0.1767, simple_loss=0.2606, pruned_loss=0.04642, over 3714205.72 frames. ], batch size: 60, lr: 4.06e-03, grad_scale: 16.0
2022-12-24 03:33:01,416 WARNING [train.py:1060] (2/4) Exclude cut with ID 7255-291500-0006-134305-0 from training. Duration: 21.24
2022-12-24 03:33:03,022 WARNING [train.py:1060] (2/4) Exclude cut with ID 6533-399-0047-104881-0_sp0.9 from training. Duration: 23.9055625
2022-12-24 03:33:31,820 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([2.0146, 1.8212, 1.9316, 2.0110, 1.5763, 4.9550, 1.9667, 2.6013],
       device='cuda:2'), covar=tensor([0.3010, 0.2015, 0.1929, 0.1990, 0.1371, 0.0111, 0.1526, 0.0841],
       device='cuda:2'), in_proj_covar=tensor([0.0132, 0.0118, 0.0125, 0.0123, 0.0108, 0.0097, 0.0091, 0.0090],
       device='cuda:2'), out_proj_covar=tensor([0.0005, 0.0005, 0.0005, 0.0005, 0.0004, 0.0003, 0.0004, 0.0004],
       device='cuda:2')
2022-12-24 03:33:32,878 INFO [train.py:894] (2/4) Epoch 28, batch 2550, loss[loss=0.1616, simple_loss=0.255, pruned_loss=0.03404, over 18584.00 frames. ], tot_loss[loss=0.176, simple_loss=0.2601, pruned_loss=0.04593, over 3713672.40 frames. ], batch size: 56, lr: 4.06e-03, grad_scale: 16.0
2022-12-24 03:33:34,306 WARNING [train.py:1060] (2/4) Exclude cut with ID 6758-72288-0033-108368-0_sp0.9 from training. Duration: 25.988875
2022-12-24 03:33:43,035 WARNING [train.py:1060] (2/4) Exclude cut with ID 7255-291500-0001-134300-0_sp0.9 from training. Duration: 20.67225
2022-12-24 03:34:21,828 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 2.183e+02 3.973e+02 4.624e+02 5.734e+02 1.222e+03, threshold=9.247e+02, percent-clipped=1.0
2022-12-24 03:34:27,749 WARNING [train.py:1060] (2/4) Exclude cut with ID 3488-85273-0038-41224-0 from training. Duration: 20.34
2022-12-24 03:34:39,370 INFO [zipformer.py:660] (2/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=97260.0, num_to_drop=0, layers_to_drop=set()
2022-12-24 03:34:48,189 INFO [train.py:894] (2/4) Epoch 28, batch 2600, loss[loss=0.1499, simple_loss=0.2282, pruned_loss=0.03573, over 18586.00 frames. ], tot_loss[loss=0.1755, simple_loss=0.2596, pruned_loss=0.04575, over 3713487.76 frames. ], batch size: 41, lr: 4.06e-03, grad_scale: 16.0
2022-12-24 03:34:54,431 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.3532, 2.6790, 2.9563, 1.1282, 2.6418, 3.3673, 2.6037, 2.5357],
       device='cuda:2'), covar=tensor([0.0797, 0.0423, 0.0334, 0.0567, 0.0346, 0.0360, 0.0362, 0.0748],
       device='cuda:2'), in_proj_covar=tensor([0.0151, 0.0175, 0.0132, 0.0142, 0.0149, 0.0145, 0.0168, 0.0181],
       device='cuda:2'), out_proj_covar=tensor([1.1373e-04, 1.3127e-04, 9.6700e-05, 1.0390e-04, 1.0906e-04, 1.0907e-04,
        1.2711e-04, 1.3569e-04], device='cuda:2')
2022-12-24 03:35:43,797 INFO [scaling.py:679] (2/4) Whitening: num_groups=8, num_channels=96, metric=1.15 vs. limit=2.0
2022-12-24 03:35:44,221 WARNING [train.py:1060] (2/4) Exclude cut with ID 7357-94126-0026-15857-0_sp0.9 from training. Duration: 25.061125
2022-12-24 03:35:54,989 WARNING [train.py:1060] (2/4) Exclude cut with ID 3033-130750-0096-55598-0 from training. Duration: 0.83
2022-12-24 03:36:04,430 INFO [train.py:894] (2/4) Epoch 28, batch 2650, loss[loss=0.2166, simple_loss=0.2977, pruned_loss=0.06775, over 18459.00 frames. ], tot_loss[loss=0.1757, simple_loss=0.2596, pruned_loss=0.04589, over 3713120.29 frames. ], batch size: 64, lr: 4.06e-03, grad_scale: 16.0
2022-12-24 03:36:12,297 INFO [zipformer.py:660] (2/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=97321.0, num_to_drop=0, layers_to_drop=set()
2022-12-24 03:36:21,074 WARNING [train.py:1060] (2/4) Exclude cut with ID 3488-85273-0017-41203-0 from training. Duration: 24.73
2022-12-24 03:36:31,788 WARNING [train.py:1060] (2/4) Exclude cut with ID 7255-291500-0009-134308-0 from training. Duration: 23.965
2022-12-24 03:36:40,898 WARNING [train.py:1060] (2/4) Exclude cut with ID 6978-92210-0030-146996-0_sp0.9 from training. Duration: 22.088875
2022-12-24 03:36:44,027 INFO [zipformer.py:660] (2/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=97342.0, num_to_drop=0, layers_to_drop=set()
2022-12-24 03:36:51,791 INFO [zipformer.py:660] (2/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=97347.0, num_to_drop=0, layers_to_drop=set()
2022-12-24 03:36:54,148 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 2.790e+02 3.885e+02 4.802e+02 6.141e+02 1.138e+03, threshold=9.603e+02, percent-clipped=2.0
2022-12-24 03:36:57,322 WARNING [train.py:1060] (2/4) Exclude cut with ID 7255-291500-0006-134305-0_sp0.9 from training. Duration: 23.6
2022-12-24 03:37:05,650 INFO [zipformer.py:660] (2/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=97356.0, num_to_drop=0, layers_to_drop=set()
2022-12-24 03:37:20,347 INFO [train.py:894] (2/4) Epoch 28, batch 2700, loss[loss=0.1764, simple_loss=0.2691, pruned_loss=0.04186, over 18648.00 frames. ], tot_loss[loss=0.175, simple_loss=0.2593, pruned_loss=0.04538, over 3713914.34 frames. ], batch size: 60, lr: 4.06e-03, grad_scale: 16.0
2022-12-24 03:37:32,563 INFO [zipformer.py:660] (2/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=97374.0, num_to_drop=0, layers_to_drop=set()
2022-12-24 03:38:17,601 INFO [zipformer.py:660] (2/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=97403.0, num_to_drop=1, layers_to_drop={0}
2022-12-24 03:38:24,729 INFO [zipformer.py:660] (2/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=97408.0, num_to_drop=0, layers_to_drop=set()
2022-12-24 03:38:36,575 INFO [train.py:894] (2/4) Epoch 28, batch 2750, loss[loss=0.1412, simple_loss=0.2239, pruned_loss=0.0293, over 18669.00 frames. ], tot_loss[loss=0.1745, simple_loss=0.2589, pruned_loss=0.04508, over 3714301.41 frames. ], batch size: 48, lr: 4.06e-03, grad_scale: 16.0
2022-12-24 03:38:39,328 WARNING [train.py:1060] (2/4) Exclude cut with ID 8291-282929-0024-13011-0 from training. Duration: 23.795
2022-12-24 03:38:57,723 WARNING [train.py:1060] (2/4) Exclude cut with ID 8631-249866-0030-130156-0_sp1.1 from training. Duration: 21.5409375
2022-12-24 03:38:59,062 WARNING [train.py:1060] (2/4) Exclude cut with ID 6978-92210-0019-146985-0_sp0.9 from training. Duration: 24.97775
2022-12-24 03:39:00,744 INFO [zipformer.py:660] (2/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=97432.0, num_to_drop=0, layers_to_drop=set()
2022-12-24 03:39:05,190 INFO [zipformer.py:660] (2/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=97435.0, num_to_drop=0, layers_to_drop=set()
2022-12-24 03:39:10,607 WARNING [train.py:1060] (2/4) Exclude cut with ID 1085-156170-0017-128270-0_sp0.9 from training. Duration: 23.3444375
2022-12-24 03:39:26,353 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 2.529e+02 3.785e+02 4.498e+02 5.381e+02 9.968e+02, threshold=8.997e+02, percent-clipped=1.0
2022-12-24 03:39:36,851 WARNING [train.py:1060] (2/4) Exclude cut with ID 6010-56788-0055-90261-0_sp0.9 from training. Duration: 23.2
2022-12-24 03:39:43,022 WARNING [train.py:1060] (2/4) Exclude cut with ID 5653-46179-0060-117930-0_sp0.9 from training. Duration: 21.17225
2022-12-24 03:39:52,088 INFO [train.py:894] (2/4) Epoch 28, batch 2800, loss[loss=0.1923, simple_loss=0.2816, pruned_loss=0.05154, over 18654.00 frames. ], tot_loss[loss=0.1745, simple_loss=0.2589, pruned_loss=0.04509, over 3714289.88 frames. ], batch size: 62, lr: 4.06e-03, grad_scale: 16.0
2022-12-24 03:40:03,036 WARNING [train.py:1060] (2/4) Exclude cut with ID 7255-291500-0008-134307-0_sp0.9 from training. Duration: 24.6555625
2022-12-24 03:40:13,275 INFO [zipformer.py:660] (2/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=97480.0, num_to_drop=0, layers_to_drop=set()
2022-12-24 03:40:30,820 INFO [scaling.py:679] (2/4) Whitening: num_groups=8, num_channels=192, metric=1.49 vs. limit=2.0
2022-12-24 03:40:58,311 WARNING [train.py:1060] (2/4) Exclude cut with ID 3488-65654-0031-41259-0 from training. Duration: 20.44
2022-12-24 03:41:06,756 INFO [train.py:894] (2/4) Epoch 28, batch 2850, loss[loss=0.1871, simple_loss=0.2594, pruned_loss=0.05741, over 18671.00 frames. ], tot_loss[loss=0.1746, simple_loss=0.2589, pruned_loss=0.04513, over 3715392.09 frames. ], batch size: 46, lr: 4.06e-03, grad_scale: 16.0
2022-12-24 03:41:11,491 WARNING [train.py:1060] (2/4) Exclude cut with ID 6951-79737-0018-132285-0_sp0.9 from training. Duration: 23.45
2022-12-24 03:41:17,407 INFO [scaling.py:679] (2/4) Whitening: num_groups=1, num_channels=384, metric=4.13 vs. limit=5.0
2022-12-24 03:41:42,343 WARNING [train.py:1060] (2/4) Exclude cut with ID 6945-60535-0076-12784-0_sp0.9 from training. Duration: 20.52225
2022-12-24 03:41:48,207 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.9325, 1.4314, 1.0189, 1.5456, 2.1619, 1.4051, 1.5774, 1.7895],
       device='cuda:2'), covar=tensor([0.1488, 0.2014, 0.2012, 0.1367, 0.1772, 0.1693, 0.1418, 0.1613],
       device='cuda:2'), in_proj_covar=tensor([0.0095, 0.0097, 0.0117, 0.0096, 0.0121, 0.0093, 0.0099, 0.0095],
       device='cuda:2'), out_proj_covar=tensor([0.0003, 0.0003, 0.0004, 0.0003, 0.0004, 0.0003, 0.0003, 0.0003],
       device='cuda:2')
2022-12-24 03:41:49,354 WARNING [train.py:1060] (2/4) Exclude cut with ID 7255-291500-0008-134307-0 from training. Duration: 22.19
2022-12-24 03:41:56,695 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 2.835e+02 4.191e+02 5.018e+02 6.319e+02 1.560e+03, threshold=1.004e+03, percent-clipped=3.0
2022-12-24 03:41:59,605 WARNING [train.py:1060] (2/4) Exclude cut with ID 7357-94126-0014-15845-0_sp1.1 from training. Duration: 25.3818125
2022-12-24 03:42:06,189 INFO [zipformer.py:660] (2/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=97555.0, num_to_drop=0, layers_to_drop=set()
2022-12-24 03:42:13,031 WARNING [train.py:1060] (2/4) Exclude cut with ID 6951-79737-0043-132310-0_sp0.9 from training. Duration: 28.0944375
2022-12-24 03:42:16,509 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([2.5403, 2.3673, 2.1458, 1.4255, 2.8574, 2.6858, 2.3841, 1.9136],
       device='cuda:2'), covar=tensor([0.0382, 0.0438, 0.0472, 0.0726, 0.0284, 0.0380, 0.0421, 0.0875],
       device='cuda:2'), in_proj_covar=tensor([0.0124, 0.0131, 0.0128, 0.0116, 0.0103, 0.0126, 0.0133, 0.0161],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0001, 0.0001, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-24 03:42:17,807 WARNING [train.py:1060] (2/4) Exclude cut with ID 2195-150901-0045-59933-0_sp0.9 from training. Duration: 22.9444375
2022-12-24 03:42:22,873 INFO [train.py:894] (2/4) Epoch 28, batch 2900, loss[loss=0.207, simple_loss=0.2882, pruned_loss=0.0629, over 18641.00 frames. ], tot_loss[loss=0.1749, simple_loss=0.2588, pruned_loss=0.04552, over 3716320.55 frames. ], batch size: 175, lr: 4.06e-03, grad_scale: 16.0
2022-12-24 03:42:27,045 WARNING [train.py:1060] (2/4) Exclude cut with ID 8291-282929-0024-13011-0_sp1.1 from training. Duration: 21.6318125
2022-12-24 03:42:46,866 WARNING [train.py:1060] (2/4) Exclude cut with ID 8631-249866-0030-130156-0 from training. Duration: 23.695
2022-12-24 03:43:12,554 WARNING [train.py:1060] (2/4) Exclude cut with ID 7699-105389-0094-26379-0 from training. Duration: 23.955
2022-12-24 03:43:20,872 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([2.1564, 2.1671, 1.6142, 2.6012, 2.3149, 2.0372, 2.8828, 2.1863],
       device='cuda:2'), covar=tensor([0.0991, 0.1857, 0.2930, 0.1772, 0.1937, 0.0984, 0.0960, 0.1373],
       device='cuda:2'), in_proj_covar=tensor([0.0185, 0.0220, 0.0260, 0.0294, 0.0245, 0.0198, 0.0209, 0.0211],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-24 03:43:37,272 INFO [train.py:894] (2/4) Epoch 28, batch 2950, loss[loss=0.1653, simple_loss=0.26, pruned_loss=0.03535, over 18519.00 frames. ], tot_loss[loss=0.175, simple_loss=0.2589, pruned_loss=0.04554, over 3716293.61 frames. ], batch size: 52, lr: 4.05e-03, grad_scale: 16.0
2022-12-24 03:43:37,448 INFO [zipformer.py:660] (2/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=97616.0, num_to_drop=0, layers_to_drop=set()
2022-12-24 03:43:37,664 INFO [zipformer.py:660] (2/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=97616.0, num_to_drop=0, layers_to_drop=set()
2022-12-24 03:43:46,377 WARNING [train.py:1060] (2/4) Exclude cut with ID 8291-282929-0024-13011-0_sp0.9 from training. Duration: 26.438875
2022-12-24 03:44:26,831 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 2.787e+02 3.883e+02 4.695e+02 5.764e+02 1.290e+03, threshold=9.391e+02, percent-clipped=3.0
2022-12-24 03:44:31,217 WARNING [train.py:1060] (2/4) Exclude cut with ID 7699-105389-0021-26306-0_sp0.9 from training. Duration: 21.2444375
2022-12-24 03:44:31,242 WARNING [train.py:1060] (2/4) Exclude cut with ID 7357-94126-0014-15845-0_sp0.9 from training. Duration: 31.02225
2022-12-24 03:44:37,372 INFO [zipformer.py:660] (2/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=97656.0, num_to_drop=0, layers_to_drop=set()
2022-12-24 03:44:42,147 WARNING [train.py:1060] (2/4) Exclude cut with ID 432-122774-0017-62487-0 from training. Duration: 22.395
2022-12-24 03:44:52,229 INFO [train.py:894] (2/4) Epoch 28, batch 3000, loss[loss=0.1465, simple_loss=0.2281, pruned_loss=0.03243, over 18434.00 frames. ], tot_loss[loss=0.1749, simple_loss=0.2593, pruned_loss=0.04529, over 3715680.81 frames. ], batch size: 42, lr: 4.05e-03, grad_scale: 16.0
2022-12-24 03:44:52,229 INFO [train.py:919] (2/4) Computing validation loss
2022-12-24 03:45:02,930 INFO [train.py:928] (2/4) Epoch 28, validation: loss=0.1624, simple_loss=0.2586, pruned_loss=0.03309, over 944034.00 frames. 
2022-12-24 03:45:02,931 INFO [train.py:929] (2/4) Maximum memory allocated so far is 24680MB
2022-12-24 03:45:10,257 WARNING [train.py:1060] (2/4) Exclude cut with ID 7357-94126-0045-15876-0 from training. Duration: 21.075
2022-12-24 03:45:14,664 WARNING [train.py:1060] (2/4) Exclude cut with ID 6482-98857-0025-147532-0_sp0.9 from training. Duration: 20.0055625
2022-12-24 03:45:14,671 WARNING [train.py:1060] (2/4) Exclude cut with ID 6951-79737-0037-132304-0_sp0.9 from training. Duration: 22.05
2022-12-24 03:45:14,681 WARNING [train.py:1060] (2/4) Exclude cut with ID 7255-291500-0003-134302-0 from training. Duration: 26.8349375
2022-12-24 03:45:15,069 INFO [zipformer.py:660] (2/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=97674.0, num_to_drop=1, layers_to_drop={0}
2022-12-24 03:45:18,854 WARNING [train.py:1060] (2/4) Exclude cut with ID 7357-94126-0009-15840-0_sp1.1 from training. Duration: 22.1090625
2022-12-24 03:45:26,382 WARNING [train.py:1060] (2/4) Exclude cut with ID 7699-105389-0094-26379-0_sp0.9 from training. Duration: 26.6166875
2022-12-24 03:45:44,618 WARNING [train.py:1060] (2/4) Exclude cut with ID 2046-178027-0000-53705-0_sp0.9 from training. Duration: 20.3055625
2022-12-24 03:45:50,670 INFO [zipformer.py:660] (2/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=97698.0, num_to_drop=1, layers_to_drop={1}
2022-12-24 03:45:58,315 INFO [zipformer.py:660] (2/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=97703.0, num_to_drop=0, layers_to_drop=set()
2022-12-24 03:45:59,741 INFO [zipformer.py:660] (2/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=97704.0, num_to_drop=0, layers_to_drop=set()
2022-12-24 03:46:07,131 WARNING [train.py:1060] (2/4) Exclude cut with ID 7205-50138-0008-5373-0_sp0.9 from training. Duration: 20.7
2022-12-24 03:46:12,120 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.2905, 1.4865, 1.3186, 1.7044, 1.7050, 1.2977, 1.2377, 1.1690],
       device='cuda:2'), covar=tensor([0.1587, 0.1429, 0.1396, 0.0906, 0.1178, 0.0952, 0.2273, 0.1295],
       device='cuda:2'), in_proj_covar=tensor([0.0252, 0.0231, 0.0223, 0.0205, 0.0265, 0.0199, 0.0231, 0.0205],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-24 03:46:17,322 INFO [train.py:894] (2/4) Epoch 28, batch 3050, loss[loss=0.1933, simple_loss=0.2743, pruned_loss=0.05613, over 18624.00 frames. ], tot_loss[loss=0.1751, simple_loss=0.2592, pruned_loss=0.04549, over 3715186.44 frames. ], batch size: 98, lr: 4.05e-03, grad_scale: 16.0
2022-12-24 03:46:31,025 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([3.9223, 3.4441, 3.3768, 3.8354, 3.6256, 3.3822, 4.0274, 1.2525],
       device='cuda:2'), covar=tensor([0.0802, 0.0765, 0.0792, 0.0891, 0.1390, 0.1331, 0.0721, 0.5057],
       device='cuda:2'), in_proj_covar=tensor([0.0367, 0.0240, 0.0251, 0.0291, 0.0345, 0.0282, 0.0309, 0.0300],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0001, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-24 03:46:38,865 INFO [zipformer.py:660] (2/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=97730.0, num_to_drop=0, layers_to_drop=set()
2022-12-24 03:46:46,693 INFO [zipformer.py:660] (2/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=97735.0, num_to_drop=1, layers_to_drop={2}
2022-12-24 03:46:47,700 WARNING [train.py:1060] (2/4) Exclude cut with ID 6978-92210-0019-146985-0 from training. Duration: 22.48
2022-12-24 03:47:04,931 WARNING [train.py:1060] (2/4) Exclude cut with ID 7255-291500-0003-134302-0_sp0.9 from training. Duration: 29.816625
2022-12-24 03:47:08,411 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 2.276e+02 3.752e+02 4.841e+02 5.897e+02 1.196e+03, threshold=9.682e+02, percent-clipped=5.0
2022-12-24 03:47:11,834 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([2.7567, 2.4681, 2.0126, 0.9737, 1.8687, 2.3377, 1.9750, 2.0938],
       device='cuda:2'), covar=tensor([0.0600, 0.0577, 0.1176, 0.1619, 0.1199, 0.1260, 0.1506, 0.0827],
       device='cuda:2'), in_proj_covar=tensor([0.0176, 0.0189, 0.0209, 0.0189, 0.0210, 0.0205, 0.0217, 0.0204],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-24 03:47:24,794 WARNING [train.py:1060] (2/4) Exclude cut with ID 7255-291500-0005-134304-0_sp1.1 from training. Duration: 22.7590625
2022-12-24 03:47:30,699 WARNING [train.py:1060] (2/4) Exclude cut with ID 7357-94126-0026-15857-0 from training. Duration: 22.555
2022-12-24 03:47:33,713 INFO [train.py:894] (2/4) Epoch 28, batch 3100, loss[loss=0.223, simple_loss=0.2944, pruned_loss=0.0758, over 18597.00 frames. ], tot_loss[loss=0.175, simple_loss=0.259, pruned_loss=0.04547, over 3715039.36 frames. ], batch size: 172, lr: 4.05e-03, grad_scale: 16.0
2022-12-24 03:47:51,757 WARNING [train.py:1060] (2/4) Exclude cut with ID 1250-135782-0005-25975-0_sp0.9 from training. Duration: 21.688875
2022-12-24 03:47:58,429 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.9412, 1.5674, 1.8508, 1.6313, 1.9235, 1.9258, 1.7486, 1.8121],
       device='cuda:2'), covar=tensor([0.1918, 0.2794, 0.2068, 0.2807, 0.1984, 0.0956, 0.3074, 0.1185],
       device='cuda:2'), in_proj_covar=tensor([0.0274, 0.0304, 0.0290, 0.0330, 0.0323, 0.0261, 0.0358, 0.0252],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-24 03:48:09,256 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([2.5704, 2.1442, 1.7818, 2.3583, 2.0014, 2.2331, 2.1204, 2.5128],
       device='cuda:2'), covar=tensor([0.2357, 0.3478, 0.2268, 0.2845, 0.3920, 0.1218, 0.3176, 0.1107],
       device='cuda:2'), in_proj_covar=tensor([0.0302, 0.0304, 0.0258, 0.0352, 0.0284, 0.0236, 0.0299, 0.0224],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0003, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-24 03:48:24,201 WARNING [train.py:1060] (2/4) Exclude cut with ID 3488-85273-0038-41224-0_sp0.9 from training. Duration: 22.6
2022-12-24 03:48:50,863 INFO [train.py:894] (2/4) Epoch 28, batch 3150, loss[loss=0.1812, simple_loss=0.2658, pruned_loss=0.04828, over 18645.00 frames. ], tot_loss[loss=0.1745, simple_loss=0.2586, pruned_loss=0.0452, over 3715568.08 frames. ], batch size: 53, lr: 4.05e-03, grad_scale: 16.0
2022-12-24 03:49:01,559 WARNING [train.py:1060] (2/4) Exclude cut with ID 7357-94126-0009-15840-0 from training. Duration: 24.32
2022-12-24 03:49:41,081 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 2.574e+02 3.893e+02 4.688e+02 5.671e+02 9.844e+02, threshold=9.376e+02, percent-clipped=1.0
2022-12-24 03:49:58,917 WARNING [train.py:1060] (2/4) Exclude cut with ID 8291-276745-0093-13116-0_sp0.9 from training. Duration: 21.061125
2022-12-24 03:50:06,571 INFO [train.py:894] (2/4) Epoch 28, batch 3200, loss[loss=0.1645, simple_loss=0.2572, pruned_loss=0.03596, over 18675.00 frames. ], tot_loss[loss=0.1745, simple_loss=0.2584, pruned_loss=0.04528, over 3715454.97 frames. ], batch size: 60, lr: 4.05e-03, grad_scale: 16.0
2022-12-24 03:50:14,017 WARNING [train.py:1060] (2/4) Exclude cut with ID 7357-94126-0024-15855-0_sp0.9 from training. Duration: 20.32225
2022-12-24 03:50:26,266 WARNING [train.py:1060] (2/4) Exclude cut with ID 3033-130750-0096-55598-0_sp1.1 from training. Duration: 0.7545625
2022-12-24 03:50:34,280 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.9859, 1.8685, 2.2056, 1.3270, 2.1098, 2.2057, 1.5951, 2.5051],
       device='cuda:2'), covar=tensor([0.1123, 0.1833, 0.1137, 0.1780, 0.0733, 0.1065, 0.2339, 0.0558],
       device='cuda:2'), in_proj_covar=tensor([0.0200, 0.0217, 0.0210, 0.0198, 0.0174, 0.0220, 0.0218, 0.0195],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-24 03:50:39,870 WARNING [train.py:1060] (2/4) Exclude cut with ID 4295-39940-0007-92567-0_sp0.9 from training. Duration: 23.9333125
2022-12-24 03:50:40,213 INFO [zipformer.py:660] (2/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=97888.0, num_to_drop=0, layers_to_drop=set()
2022-12-24 03:50:58,451 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.3946, 0.9932, 0.7625, 1.1177, 1.8486, 0.7471, 1.0628, 1.2445],
       device='cuda:2'), covar=tensor([0.2419, 0.3187, 0.2411, 0.2137, 0.2464, 0.2476, 0.2262, 0.2625],
       device='cuda:2'), in_proj_covar=tensor([0.0095, 0.0098, 0.0118, 0.0097, 0.0122, 0.0093, 0.0100, 0.0095],
       device='cuda:2'), out_proj_covar=tensor([0.0003, 0.0003, 0.0004, 0.0003, 0.0004, 0.0003, 0.0003, 0.0003],
       device='cuda:2')
2022-12-24 03:51:09,529 INFO [scaling.py:679] (2/4) Whitening: num_groups=8, num_channels=192, metric=1.91 vs. limit=2.0
2022-12-24 03:51:11,336 WARNING [train.py:1060] (2/4) Exclude cut with ID 7255-291500-0008-134307-0_sp1.1 from training. Duration: 20.17275
2022-12-24 03:51:14,918 INFO [zipformer.py:660] (2/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=97911.0, num_to_drop=0, layers_to_drop=set()
2022-12-24 03:51:17,513 WARNING [train.py:1060] (2/4) Exclude cut with ID 6978-92210-0019-146985-0_sp1.1 from training. Duration: 20.436375
2022-12-24 03:51:22,072 INFO [train.py:894] (2/4) Epoch 28, batch 3250, loss[loss=0.1888, simple_loss=0.2618, pruned_loss=0.0579, over 18602.00 frames. ], tot_loss[loss=0.1758, simple_loss=0.2595, pruned_loss=0.04602, over 3715793.83 frames. ], batch size: 45, lr: 4.05e-03, grad_scale: 16.0
2022-12-24 03:51:22,351 INFO [zipformer.py:660] (2/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=97916.0, num_to_drop=0, layers_to_drop=set()
2022-12-24 03:52:13,181 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 3.219e+02 4.264e+02 4.990e+02 6.052e+02 1.194e+03, threshold=9.979e+02, percent-clipped=2.0
2022-12-24 03:52:13,634 INFO [zipformer.py:660] (2/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=97949.0, num_to_drop=0, layers_to_drop=set()
2022-12-24 03:52:33,185 WARNING [train.py:1060] (2/4) Exclude cut with ID 4234-40345-0022-142709-0_sp0.9 from training. Duration: 23.1055625
2022-12-24 03:52:35,094 WARNING [train.py:1060] (2/4) Exclude cut with ID 8291-282929-0007-12994-0_sp1.1 from training. Duration: 23.5
2022-12-24 03:52:36,936 INFO [zipformer.py:660] (2/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=97964.0, num_to_drop=0, layers_to_drop=set()
2022-12-24 03:52:39,767 INFO [train.py:894] (2/4) Epoch 28, batch 3300, loss[loss=0.1825, simple_loss=0.2651, pruned_loss=0.04993, over 18573.00 frames. ], tot_loss[loss=0.1742, simple_loss=0.2582, pruned_loss=0.04512, over 3714399.99 frames. ], batch size: 69, lr: 4.05e-03, grad_scale: 16.0
2022-12-24 03:52:44,332 WARNING [train.py:1060] (2/4) Exclude cut with ID 7255-291500-0009-134308-0_sp0.9 from training. Duration: 26.62775
2022-12-24 03:52:59,122 WARNING [train.py:1060] (2/4) Exclude cut with ID 6951-79737-0018-132285-0 from training. Duration: 21.105
2022-12-24 03:53:02,171 WARNING [train.py:1060] (2/4) Exclude cut with ID 4511-76322-0006-80011-0_sp0.9 from training. Duration: 24.411125
2022-12-24 03:53:28,586 INFO [zipformer.py:660] (2/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=97998.0, num_to_drop=0, layers_to_drop=set()
2022-12-24 03:53:29,873 WARNING [train.py:1060] (2/4) Exclude cut with ID 6758-72288-0033-108368-0_sp1.1 from training. Duration: 21.263625
2022-12-24 03:53:39,597 INFO [zipformer.py:660] (2/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=98003.0, num_to_drop=0, layers_to_drop=set()
2022-12-24 03:54:00,163 INFO [train.py:894] (2/4) Epoch 28, batch 3350, loss[loss=0.1718, simple_loss=0.2604, pruned_loss=0.04156, over 18456.00 frames. ], tot_loss[loss=0.1739, simple_loss=0.2579, pruned_loss=0.04494, over 3714089.45 frames. ], batch size: 50, lr: 4.05e-03, grad_scale: 16.0
2022-12-24 03:54:07,632 WARNING [train.py:1060] (2/4) Exclude cut with ID 4234-40345-0022-142709-0 from training. Duration: 20.795
2022-12-24 03:54:09,572 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.9917, 1.9084, 1.5197, 1.5379, 1.6768, 1.8469, 1.7052, 1.8196],
       device='cuda:2'), covar=tensor([0.2319, 0.3043, 0.2121, 0.2695, 0.3582, 0.1169, 0.2690, 0.1154],
       device='cuda:2'), in_proj_covar=tensor([0.0300, 0.0302, 0.0256, 0.0350, 0.0283, 0.0235, 0.0298, 0.0224],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0003, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-24 03:54:16,489 WARNING [train.py:1060] (2/4) Exclude cut with ID 7357-94126-0021-15852-0 from training. Duration: 24.76
2022-12-24 03:54:16,503 WARNING [train.py:1060] (2/4) Exclude cut with ID 3867-173237-0077-144769-0_sp0.9 from training. Duration: 22.25
2022-12-24 03:54:21,845 INFO [zipformer.py:660] (2/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=98030.0, num_to_drop=1, layers_to_drop={2}
2022-12-24 03:54:22,015 INFO [zipformer.py:660] (2/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=98030.0, num_to_drop=0, layers_to_drop=set()
2022-12-24 03:54:42,836 WARNING [train.py:1060] (2/4) Exclude cut with ID 7357-94126-0026-15857-0_sp1.1 from training. Duration: 20.5045625
2022-12-24 03:54:46,008 INFO [zipformer.py:660] (2/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=98046.0, num_to_drop=0, layers_to_drop=set()
2022-12-24 03:54:49,991 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 2.669e+02 3.930e+02 4.880e+02 5.841e+02 1.288e+03, threshold=9.759e+02, percent-clipped=4.0
2022-12-24 03:54:53,416 INFO [zipformer.py:660] (2/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=98051.0, num_to_drop=0, layers_to_drop=set()
2022-12-24 03:55:15,433 INFO [train.py:894] (2/4) Epoch 28, batch 3400, loss[loss=0.195, simple_loss=0.2776, pruned_loss=0.05626, over 18618.00 frames. ], tot_loss[loss=0.1739, simple_loss=0.2581, pruned_loss=0.04483, over 3714664.64 frames. ], batch size: 97, lr: 4.05e-03, grad_scale: 16.0
2022-12-24 03:55:33,225 INFO [zipformer.py:660] (2/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=98078.0, num_to_drop=0, layers_to_drop=set()
2022-12-24 03:55:33,716 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([2.5075, 2.1712, 1.9455, 2.4512, 2.0467, 2.1212, 2.0770, 2.5549],
       device='cuda:2'), covar=tensor([0.1779, 0.2713, 0.1594, 0.2178, 0.3049, 0.0961, 0.2653, 0.0850],
       device='cuda:2'), in_proj_covar=tensor([0.0299, 0.0302, 0.0256, 0.0350, 0.0282, 0.0235, 0.0298, 0.0223],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0003, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-24 03:56:14,251 INFO [zipformer.py:660] (2/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=98106.0, num_to_drop=0, layers_to_drop=set()
2022-12-24 03:56:28,563 INFO [train.py:894] (2/4) Epoch 28, batch 3450, loss[loss=0.1559, simple_loss=0.2309, pruned_loss=0.04045, over 18425.00 frames. ], tot_loss[loss=0.1733, simple_loss=0.2576, pruned_loss=0.04451, over 3713537.41 frames. ], batch size: 42, lr: 4.04e-03, grad_scale: 16.0
2022-12-24 03:57:15,592 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 2.690e+02 3.760e+02 4.523e+02 5.539e+02 7.784e+02, threshold=9.046e+02, percent-clipped=1.0
2022-12-24 03:57:40,698 INFO [train.py:894] (2/4) Epoch 28, batch 3500, loss[loss=0.1782, simple_loss=0.2624, pruned_loss=0.04702, over 18653.00 frames. ], tot_loss[loss=0.1741, simple_loss=0.2581, pruned_loss=0.04507, over 3712489.21 frames. ], batch size: 177, lr: 4.04e-03, grad_scale: 16.0
2022-12-24 03:57:42,528 INFO [zipformer.py:660] (2/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=98167.0, num_to_drop=0, layers_to_drop=set()
2022-12-24 03:58:00,536 WARNING [train.py:1060] (2/4) Exclude cut with ID 7859-102521-0017-7548-0_sp1.1 from training. Duration: 22.2954375
2022-12-24 03:58:08,570 INFO [train.py:894] (2/4) Epoch 29, batch 0, loss[loss=0.1743, simple_loss=0.2663, pruned_loss=0.04108, over 18457.00 frames. ], tot_loss[loss=0.1743, simple_loss=0.2663, pruned_loss=0.04108, over 18457.00 frames. ], batch size: 64, lr: 3.97e-03, grad_scale: 16.0
2022-12-24 03:58:08,570 INFO [train.py:919] (2/4) Computing validation loss
2022-12-24 03:58:19,243 INFO [train.py:928] (2/4) Epoch 29, validation: loss=0.163, simple_loss=0.259, pruned_loss=0.03348, over 944034.00 frames. 
2022-12-24 03:58:19,244 INFO [train.py:929] (2/4) Maximum memory allocated so far is 24680MB
2022-12-24 03:59:12,525 WARNING [train.py:1060] (2/4) Exclude cut with ID 298-126791-0067-24026-0_sp0.9 from training. Duration: 21.438875
2022-12-24 03:59:17,021 WARNING [train.py:1060] (2/4) Exclude cut with ID 5652-39938-0025-23684-0_sp0.9 from training. Duration: 22.2055625
2022-12-24 03:59:18,617 INFO [zipformer.py:660] (2/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=98211.0, num_to_drop=0, layers_to_drop=set()
2022-12-24 03:59:34,737 INFO [train.py:894] (2/4) Epoch 29, batch 50, loss[loss=0.1866, simple_loss=0.2774, pruned_loss=0.04791, over 18498.00 frames. ], tot_loss[loss=0.1661, simple_loss=0.2563, pruned_loss=0.03795, over 837841.98 frames. ], batch size: 58, lr: 3.97e-03, grad_scale: 16.0
2022-12-24 03:59:36,660 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.4824, 1.3816, 1.3631, 1.3639, 1.7101, 1.5954, 1.5490, 1.1808],
       device='cuda:2'), covar=tensor([0.0314, 0.0236, 0.0493, 0.0224, 0.0200, 0.0391, 0.0272, 0.0318],
       device='cuda:2'), in_proj_covar=tensor([0.0098, 0.0129, 0.0154, 0.0122, 0.0118, 0.0123, 0.0101, 0.0129],
       device='cuda:2'), out_proj_covar=tensor([7.7267e-05, 1.0209e-04, 1.2621e-04, 9.6400e-05, 9.4068e-05, 9.4295e-05,
        7.8484e-05, 1.0157e-04], device='cuda:2')
2022-12-24 03:59:52,719 INFO [scaling.py:679] (2/4) Whitening: num_groups=8, num_channels=96, metric=1.23 vs. limit=2.0
2022-12-24 04:00:08,513 INFO [zipformer.py:660] (2/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=98244.0, num_to_drop=0, layers_to_drop=set()
2022-12-24 04:00:15,073 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 2.323e+02 3.149e+02 4.026e+02 5.359e+02 1.218e+03, threshold=8.052e+02, percent-clipped=4.0
2022-12-24 04:00:29,804 INFO [zipformer.py:660] (2/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=98259.0, num_to_drop=0, layers_to_drop=set()
2022-12-24 04:00:38,983 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.3732, 2.2185, 2.7174, 0.8400, 2.5261, 3.4036, 2.1564, 2.4955],
       device='cuda:2'), covar=tensor([0.1062, 0.0802, 0.0483, 0.0772, 0.0488, 0.0441, 0.0664, 0.0809],
       device='cuda:2'), in_proj_covar=tensor([0.0151, 0.0174, 0.0131, 0.0143, 0.0150, 0.0144, 0.0168, 0.0180],
       device='cuda:2'), out_proj_covar=tensor([1.1340e-04, 1.3037e-04, 9.6099e-05, 1.0424e-04, 1.0951e-04, 1.0815e-04,
        1.2659e-04, 1.3515e-04], device='cuda:2')
2022-12-24 04:00:49,565 INFO [train.py:894] (2/4) Epoch 29, batch 100, loss[loss=0.1464, simple_loss=0.2359, pruned_loss=0.02846, over 18431.00 frames. ], tot_loss[loss=0.1655, simple_loss=0.2561, pruned_loss=0.03746, over 1475369.91 frames. ], batch size: 48, lr: 3.97e-03, grad_scale: 16.0
2022-12-24 04:01:37,771 INFO [scaling.py:679] (2/4) Whitening: num_groups=8, num_channels=192, metric=1.89 vs. limit=2.0
2022-12-24 04:02:06,025 INFO [train.py:894] (2/4) Epoch 29, batch 150, loss[loss=0.1767, simple_loss=0.2704, pruned_loss=0.04156, over 18660.00 frames. ], tot_loss[loss=0.1665, simple_loss=0.2569, pruned_loss=0.03806, over 1972217.31 frames. ], batch size: 69, lr: 3.97e-03, grad_scale: 16.0
2022-12-24 04:02:18,107 WARNING [train.py:1060] (2/4) Exclude cut with ID 7859-102521-0017-7548-0 from training. Duration: 24.525
2022-12-24 04:02:18,334 INFO [zipformer.py:660] (2/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=98330.0, num_to_drop=1, layers_to_drop={2}
2022-12-24 04:02:47,186 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 1.956e+02 3.151e+02 3.628e+02 4.247e+02 8.996e+02, threshold=7.256e+02, percent-clipped=1.0
2022-12-24 04:02:49,012 WARNING [train.py:1060] (2/4) Exclude cut with ID 3699-47246-0007-3408-0_sp0.9 from training. Duration: 20.26675
2022-12-24 04:03:01,265 WARNING [train.py:1060] (2/4) Exclude cut with ID 7859-102521-0017-7548-0_sp0.9 from training. Duration: 27.25
2022-12-24 04:03:23,391 INFO [train.py:894] (2/4) Epoch 29, batch 200, loss[loss=0.1766, simple_loss=0.2695, pruned_loss=0.04187, over 18497.00 frames. ], tot_loss[loss=0.1659, simple_loss=0.2558, pruned_loss=0.03802, over 2357561.32 frames. ], batch size: 64, lr: 3.97e-03, grad_scale: 16.0
2022-12-24 04:03:32,497 INFO [zipformer.py:660] (2/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=98378.0, num_to_drop=1, layers_to_drop={1}
2022-12-24 04:04:18,036 WARNING [train.py:1060] (2/4) Exclude cut with ID 6426-64292-0017-15984-0 from training. Duration: 21.68
2022-12-24 04:04:19,584 INFO [zipformer.py:660] (2/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=98410.0, num_to_drop=0, layers_to_drop=set()
2022-12-24 04:04:29,377 WARNING [train.py:1060] (2/4) Exclude cut with ID 4278-13270-0007-59342-0 from training. Duration: 21.6300625
2022-12-24 04:04:37,742 INFO [train.py:894] (2/4) Epoch 29, batch 250, loss[loss=0.1497, simple_loss=0.2291, pruned_loss=0.03518, over 18420.00 frames. ], tot_loss[loss=0.165, simple_loss=0.2547, pruned_loss=0.03762, over 2658416.18 frames. ], batch size: 42, lr: 3.97e-03, grad_scale: 16.0
2022-12-24 04:04:45,505 INFO [scaling.py:679] (2/4) Whitening: num_groups=8, num_channels=96, metric=1.22 vs. limit=2.0
2022-12-24 04:04:54,922 WARNING [train.py:1060] (2/4) Exclude cut with ID 4278-13270-0007-59342-0_sp0.9 from training. Duration: 24.033375
2022-12-24 04:05:17,892 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 2.102e+02 2.945e+02 3.476e+02 4.234e+02 1.493e+03, threshold=6.951e+02, percent-clipped=2.0
2022-12-24 04:05:37,607 INFO [zipformer.py:660] (2/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=98462.0, num_to_drop=0, layers_to_drop=set()
2022-12-24 04:05:50,790 WARNING [train.py:1060] (2/4) Exclude cut with ID 4278-13270-0009-59344-0 from training. Duration: 22.905
2022-12-24 04:05:52,431 WARNING [train.py:1060] (2/4) Exclude cut with ID 5622-44585-0006-90525-0_sp1.1 from training. Duration: 23.4318125
2022-12-24 04:05:52,775 INFO [zipformer.py:660] (2/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=98471.0, num_to_drop=0, layers_to_drop=set()
2022-12-24 04:05:53,821 INFO [train.py:894] (2/4) Epoch 29, batch 300, loss[loss=0.1628, simple_loss=0.2574, pruned_loss=0.03413, over 18688.00 frames. ], tot_loss[loss=0.1646, simple_loss=0.2544, pruned_loss=0.03742, over 2893177.63 frames. ], batch size: 62, lr: 3.97e-03, grad_scale: 16.0
2022-12-24 04:06:54,611 INFO [scaling.py:679] (2/4) Whitening: num_groups=1, num_channels=384, metric=2.80 vs. limit=5.0
2022-12-24 04:07:07,860 INFO [train.py:894] (2/4) Epoch 29, batch 350, loss[loss=0.1784, simple_loss=0.2706, pruned_loss=0.04316, over 18533.00 frames. ], tot_loss[loss=0.164, simple_loss=0.2541, pruned_loss=0.03693, over 3075614.05 frames. ], batch size: 96, lr: 3.96e-03, grad_scale: 16.0
2022-12-24 04:07:39,676 INFO [zipformer.py:660] (2/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=98543.0, num_to_drop=0, layers_to_drop=set()
2022-12-24 04:07:40,889 INFO [zipformer.py:660] (2/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=98544.0, num_to_drop=0, layers_to_drop=set()
2022-12-24 04:07:47,848 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 1.942e+02 3.244e+02 3.661e+02 4.280e+02 7.325e+02, threshold=7.323e+02, percent-clipped=2.0
2022-12-24 04:07:52,175 WARNING [train.py:1060] (2/4) Exclude cut with ID 4278-13270-0009-59344-0_sp1.1 from training. Duration: 20.82275
2022-12-24 04:07:52,228 WARNING [train.py:1060] (2/4) Exclude cut with ID 4278-13270-0009-59344-0_sp0.9 from training. Duration: 25.45
2022-12-24 04:08:13,840 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([0.6798, 2.1033, 2.2904, 1.1392, 1.7829, 2.5346, 2.3144, 1.9220],
       device='cuda:2'), covar=tensor([0.0891, 0.0392, 0.0343, 0.0509, 0.0358, 0.0451, 0.0253, 0.0790],
       device='cuda:2'), in_proj_covar=tensor([0.0151, 0.0174, 0.0131, 0.0142, 0.0149, 0.0144, 0.0167, 0.0180],
       device='cuda:2'), out_proj_covar=tensor([1.1346e-04, 1.3035e-04, 9.6089e-05, 1.0397e-04, 1.0859e-04, 1.0777e-04,
        1.2590e-04, 1.3503e-04], device='cuda:2')
2022-12-24 04:08:22,543 INFO [train.py:894] (2/4) Epoch 29, batch 400, loss[loss=0.1611, simple_loss=0.2589, pruned_loss=0.03164, over 18379.00 frames. ], tot_loss[loss=0.1641, simple_loss=0.2544, pruned_loss=0.03689, over 3217087.54 frames. ], batch size: 53, lr: 3.96e-03, grad_scale: 16.0
2022-12-24 04:08:52,256 INFO [zipformer.py:660] (2/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=98592.0, num_to_drop=0, layers_to_drop=set()
2022-12-24 04:08:53,681 WARNING [train.py:1060] (2/4) Exclude cut with ID 5622-44585-0006-90525-0 from training. Duration: 25.775
2022-12-24 04:09:10,223 INFO [zipformer.py:660] (2/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=98604.0, num_to_drop=0, layers_to_drop=set()
2022-12-24 04:09:16,350 WARNING [train.py:1060] (2/4) Exclude cut with ID 2929-85685-0071-62375-0_sp0.9 from training. Duration: 22.25
2022-12-24 04:09:25,362 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([2.7901, 2.3174, 1.9133, 0.8317, 1.9231, 2.1310, 1.6802, 2.2193],
       device='cuda:2'), covar=tensor([0.0817, 0.0914, 0.2101, 0.2417, 0.1629, 0.1917, 0.2441, 0.1193],
       device='cuda:2'), in_proj_covar=tensor([0.0178, 0.0190, 0.0211, 0.0191, 0.0212, 0.0207, 0.0220, 0.0206],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-24 04:09:37,589 INFO [train.py:894] (2/4) Epoch 29, batch 450, loss[loss=0.1539, simple_loss=0.2436, pruned_loss=0.0321, over 18586.00 frames. ], tot_loss[loss=0.164, simple_loss=0.2545, pruned_loss=0.0368, over 3327165.84 frames. ], batch size: 49, lr: 3.96e-03, grad_scale: 16.0
2022-12-24 04:09:43,379 WARNING [train.py:1060] (2/4) Exclude cut with ID 3972-170212-0014-23379-0 from training. Duration: 26.205
2022-12-24 04:10:02,934 WARNING [train.py:1060] (2/4) Exclude cut with ID 5239-32139-0047-9341-0_sp0.9 from training. Duration: 30.1555625
2022-12-24 04:10:08,679 WARNING [train.py:1060] (2/4) Exclude cut with ID 1265-135635-0050-6781-0_sp0.9 from training. Duration: 21.8333125
2022-12-24 04:10:16,402 WARNING [train.py:1060] (2/4) Exclude cut with ID 1914-133440-0024-94914-0_sp1.1 from training. Duration: 20.6545625
2022-12-24 04:10:17,783 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 1.919e+02 3.115e+02 3.804e+02 4.610e+02 9.757e+02, threshold=7.608e+02, percent-clipped=2.0
2022-12-24 04:10:53,225 INFO [train.py:894] (2/4) Epoch 29, batch 500, loss[loss=0.1863, simple_loss=0.2815, pruned_loss=0.04559, over 18454.00 frames. ], tot_loss[loss=0.1653, simple_loss=0.2555, pruned_loss=0.03752, over 3413122.34 frames. ], batch size: 54, lr: 3.96e-03, grad_scale: 16.0
2022-12-24 04:10:57,543 WARNING [train.py:1060] (2/4) Exclude cut with ID 7395-89880-0045-39920-0_sp0.9 from training. Duration: 20.52225
2022-12-24 04:11:18,707 WARNING [train.py:1060] (2/4) Exclude cut with ID 3972-170212-0014-23379-0_sp0.9 from training. Duration: 29.1166875
2022-12-24 04:12:08,560 INFO [train.py:894] (2/4) Epoch 29, batch 550, loss[loss=0.1929, simple_loss=0.2764, pruned_loss=0.05463, over 18610.00 frames. ], tot_loss[loss=0.1661, simple_loss=0.2562, pruned_loss=0.03801, over 3480532.01 frames. ], batch size: 169, lr: 3.96e-03, grad_scale: 16.0
2022-12-24 04:12:17,580 WARNING [train.py:1060] (2/4) Exclude cut with ID 543-133211-0007-59831-0_sp0.9 from training. Duration: 21.388875
2022-12-24 04:12:49,075 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 2.526e+02 3.198e+02 3.726e+02 4.559e+02 9.672e+02, threshold=7.451e+02, percent-clipped=1.0
2022-12-24 04:12:52,788 WARNING [train.py:1060] (2/4) Exclude cut with ID 1914-133440-0024-94914-0 from training. Duration: 22.72
2022-12-24 04:12:53,930 WARNING [train.py:1060] (2/4) Exclude cut with ID 1914-133440-0031-94921-0_sp0.9 from training. Duration: 22.7444375
2022-12-24 04:12:58,736 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.7430, 1.7982, 1.5573, 1.8158, 2.0281, 1.9590, 1.9532, 1.4081],
       device='cuda:2'), covar=tensor([0.0414, 0.0265, 0.0524, 0.0212, 0.0216, 0.0400, 0.0289, 0.0352],
       device='cuda:2'), in_proj_covar=tensor([0.0099, 0.0131, 0.0158, 0.0124, 0.0120, 0.0126, 0.0104, 0.0132],
       device='cuda:2'), out_proj_covar=tensor([7.8457e-05, 1.0344e-04, 1.2876e-04, 9.7975e-05, 9.6018e-05, 9.6553e-05,
        8.0253e-05, 1.0354e-04], device='cuda:2')
2022-12-24 04:13:10,025 INFO [zipformer.py:660] (2/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=98762.0, num_to_drop=0, layers_to_drop=set()
2022-12-24 04:13:15,396 INFO [zipformer.py:660] (2/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=98766.0, num_to_drop=0, layers_to_drop=set()
2022-12-24 04:13:24,498 INFO [train.py:894] (2/4) Epoch 29, batch 600, loss[loss=0.1671, simple_loss=0.2636, pruned_loss=0.03531, over 18600.00 frames. ], tot_loss[loss=0.1658, simple_loss=0.2557, pruned_loss=0.03797, over 3532831.45 frames. ], batch size: 57, lr: 3.96e-03, grad_scale: 16.0
2022-12-24 04:13:33,258 WARNING [train.py:1060] (2/4) Exclude cut with ID 4133-6541-0027-40495-0_sp1.1 from training. Duration: 0.9681875
2022-12-24 04:13:37,737 WARNING [train.py:1060] (2/4) Exclude cut with ID 6330-62851-0022-91297-0_sp0.9 from training. Duration: 22.3166875
2022-12-24 04:13:43,996 WARNING [train.py:1060] (2/4) Exclude cut with ID 543-133212-0015-59917-0_sp0.9 from training. Duration: 21.8166875
2022-12-24 04:14:14,192 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([2.2327, 2.1805, 1.7986, 1.9552, 2.5264, 2.9772, 2.6912, 1.9636],
       device='cuda:2'), covar=tensor([0.0433, 0.0362, 0.0520, 0.0275, 0.0255, 0.0317, 0.0315, 0.0376],
       device='cuda:2'), in_proj_covar=tensor([0.0099, 0.0130, 0.0157, 0.0124, 0.0120, 0.0126, 0.0103, 0.0131],
       device='cuda:2'), out_proj_covar=tensor([7.8186e-05, 1.0287e-04, 1.2865e-04, 9.7677e-05, 9.5575e-05, 9.6021e-05,
        7.9842e-05, 1.0290e-04], device='cuda:2')
2022-12-24 04:14:23,240 INFO [zipformer.py:660] (2/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=98810.0, num_to_drop=0, layers_to_drop=set()
2022-12-24 04:14:40,756 INFO [train.py:894] (2/4) Epoch 29, batch 650, loss[loss=0.1594, simple_loss=0.2446, pruned_loss=0.03713, over 18422.00 frames. ], tot_loss[loss=0.1661, simple_loss=0.2559, pruned_loss=0.03809, over 3572789.16 frames. ], batch size: 48, lr: 3.96e-03, grad_scale: 32.0
2022-12-24 04:15:00,709 INFO [scaling.py:679] (2/4) Whitening: num_groups=8, num_channels=192, metric=1.50 vs. limit=2.0
2022-12-24 04:15:16,374 INFO [scaling.py:679] (2/4) Whitening: num_groups=8, num_channels=192, metric=1.80 vs. limit=2.0
2022-12-24 04:15:21,392 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 1.842e+02 3.342e+02 3.913e+02 4.819e+02 8.981e+02, threshold=7.826e+02, percent-clipped=4.0
2022-12-24 04:15:25,678 WARNING [train.py:1060] (2/4) Exclude cut with ID 4957-30119-0041-23990-0_sp0.9 from training. Duration: 20.22775
2022-12-24 04:15:55,862 INFO [train.py:894] (2/4) Epoch 29, batch 700, loss[loss=0.1719, simple_loss=0.2665, pruned_loss=0.03863, over 18387.00 frames. ], tot_loss[loss=0.166, simple_loss=0.256, pruned_loss=0.03803, over 3604150.91 frames. ], batch size: 53, lr: 3.96e-03, grad_scale: 16.0
2022-12-24 04:15:58,137 INFO [scaling.py:679] (2/4) Whitening: num_groups=8, num_channels=96, metric=1.17 vs. limit=2.0
2022-12-24 04:16:09,346 WARNING [train.py:1060] (2/4) Exclude cut with ID 5239-32139-0047-9341-0_sp1.1 from training. Duration: 24.67275
2022-12-24 04:16:22,661 INFO [zipformer.py:660] (2/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=98890.0, num_to_drop=0, layers_to_drop=set()
2022-12-24 04:16:30,534 INFO [zipformer.py:660] (2/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=98895.0, num_to_drop=0, layers_to_drop=set()
2022-12-24 04:16:36,255 WARNING [train.py:1060] (2/4) Exclude cut with ID 3082-165428-0081-50734-0_sp0.9 from training. Duration: 21.8055625
2022-12-24 04:16:36,355 INFO [zipformer.py:660] (2/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=98899.0, num_to_drop=0, layers_to_drop=set()
2022-12-24 04:16:50,758 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([2.5208, 3.4213, 3.1829, 1.4372, 3.5197, 2.6789, 0.6868, 2.1057],
       device='cuda:2'), covar=tensor([0.2127, 0.1134, 0.1539, 0.3509, 0.0962, 0.0890, 0.4740, 0.1680],
       device='cuda:2'), in_proj_covar=tensor([0.0151, 0.0149, 0.0161, 0.0125, 0.0152, 0.0116, 0.0145, 0.0118],
       device='cuda:2'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0002, 0.0003, 0.0002, 0.0003, 0.0002],
       device='cuda:2')
2022-12-24 04:17:10,932 INFO [train.py:894] (2/4) Epoch 29, batch 750, loss[loss=0.1562, simple_loss=0.2366, pruned_loss=0.03785, over 18552.00 frames. ], tot_loss[loss=0.1664, simple_loss=0.2564, pruned_loss=0.03821, over 3628543.93 frames. ], batch size: 41, lr: 3.96e-03, grad_scale: 16.0
2022-12-24 04:17:12,316 WARNING [train.py:1060] (2/4) Exclude cut with ID 3340-169293-0054-76830-0_sp0.9 from training. Duration: 22.6666875
2022-12-24 04:17:28,086 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.9392, 2.1289, 1.6701, 2.2441, 3.0478, 1.7093, 2.0807, 1.5047],
       device='cuda:2'), covar=tensor([0.1989, 0.1929, 0.1866, 0.1250, 0.1364, 0.1340, 0.2007, 0.1758],
       device='cuda:2'), in_proj_covar=tensor([0.0247, 0.0230, 0.0222, 0.0203, 0.0262, 0.0198, 0.0228, 0.0203],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-24 04:17:53,677 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 2.092e+02 3.312e+02 3.981e+02 4.974e+02 8.508e+02, threshold=7.962e+02, percent-clipped=2.0
2022-12-24 04:17:55,617 INFO [zipformer.py:660] (2/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=98951.0, num_to_drop=0, layers_to_drop=set()
2022-12-24 04:18:03,419 INFO [zipformer.py:660] (2/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=98956.0, num_to_drop=0, layers_to_drop=set()
2022-12-24 04:18:10,212 INFO [scaling.py:679] (2/4) Whitening: num_groups=8, num_channels=96, metric=1.44 vs. limit=2.0
2022-12-24 04:18:14,980 WARNING [train.py:1060] (2/4) Exclude cut with ID 2411-132532-0017-82279-0_sp1.1 from training. Duration: 0.9681875
2022-12-24 04:18:25,278 INFO [zipformer.py:660] (2/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=98971.0, num_to_drop=0, layers_to_drop=set()
2022-12-24 04:18:26,282 INFO [train.py:894] (2/4) Epoch 29, batch 800, loss[loss=0.1779, simple_loss=0.2676, pruned_loss=0.04406, over 18371.00 frames. ], tot_loss[loss=0.1671, simple_loss=0.2573, pruned_loss=0.03848, over 3646859.39 frames. ], batch size: 51, lr: 3.96e-03, grad_scale: 16.0
2022-12-24 04:18:41,373 WARNING [train.py:1060] (2/4) Exclude cut with ID 6330-62850-0007-91323-0 from training. Duration: 22.485
2022-12-24 04:19:19,651 WARNING [train.py:1060] (2/4) Exclude cut with ID 3972-170212-0014-23379-0_sp1.1 from training. Duration: 23.82275
2022-12-24 04:19:34,198 WARNING [train.py:1060] (2/4) Exclude cut with ID 4860-13185-0032-76709-0 from training. Duration: 20.77
2022-12-24 04:19:40,151 WARNING [train.py:1060] (2/4) Exclude cut with ID 6426-64292-0017-15984-0_sp0.9 from training. Duration: 24.088875
2022-12-24 04:19:41,613 INFO [train.py:894] (2/4) Epoch 29, batch 850, loss[loss=0.1834, simple_loss=0.2742, pruned_loss=0.0463, over 18528.00 frames. ], tot_loss[loss=0.1668, simple_loss=0.257, pruned_loss=0.03836, over 3661297.17 frames. ], batch size: 58, lr: 3.95e-03, grad_scale: 16.0
2022-12-24 04:19:57,549 INFO [zipformer.py:660] (2/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=99032.0, num_to_drop=0, layers_to_drop=set()
2022-12-24 04:20:12,114 WARNING [train.py:1060] (2/4) Exclude cut with ID 6330-62850-0007-91323-0_sp1.1 from training. Duration: 20.4409375
2022-12-24 04:20:24,862 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 2.082e+02 3.230e+02 3.807e+02 4.729e+02 8.839e+02, threshold=7.614e+02, percent-clipped=1.0
2022-12-24 04:20:26,564 INFO [zipformer.py:660] (2/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=99051.0, num_to_drop=0, layers_to_drop=set()
2022-12-24 04:20:47,171 INFO [zipformer.py:660] (2/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=99066.0, num_to_drop=0, layers_to_drop=set()
2022-12-24 04:20:55,347 INFO [train.py:894] (2/4) Epoch 29, batch 900, loss[loss=0.1645, simple_loss=0.2494, pruned_loss=0.0398, over 18486.00 frames. ], tot_loss[loss=0.1674, simple_loss=0.2579, pruned_loss=0.03845, over 3673026.74 frames. ], batch size: 43, lr: 3.95e-03, grad_scale: 16.0
2022-12-24 04:21:28,303 WARNING [train.py:1060] (2/4) Exclude cut with ID 2929-85685-0062-62366-0_sp0.9 from training. Duration: 22.511125
2022-12-24 04:21:28,331 WARNING [train.py:1060] (2/4) Exclude cut with ID 7395-89880-0031-39906-0 from training. Duration: 20.675
2022-12-24 04:21:55,145 INFO [zipformer.py:660] (2/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=99112.0, num_to_drop=0, layers_to_drop=set()
2022-12-24 04:21:57,799 INFO [zipformer.py:660] (2/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=99114.0, num_to_drop=0, layers_to_drop=set()
2022-12-24 04:22:09,356 INFO [train.py:894] (2/4) Epoch 29, batch 950, loss[loss=0.1393, simple_loss=0.2254, pruned_loss=0.0266, over 18527.00 frames. ], tot_loss[loss=0.168, simple_loss=0.2584, pruned_loss=0.03886, over 3682048.91 frames. ], batch size: 44, lr: 3.95e-03, grad_scale: 16.0
2022-12-24 04:22:18,954 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([3.9664, 3.4456, 3.5381, 3.8820, 3.6792, 3.4791, 4.0576, 2.1206],
       device='cuda:2'), covar=tensor([0.0698, 0.0722, 0.0623, 0.0814, 0.1177, 0.1110, 0.0868, 0.3654],
       device='cuda:2'), in_proj_covar=tensor([0.0361, 0.0234, 0.0245, 0.0282, 0.0335, 0.0275, 0.0302, 0.0292],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0001, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-24 04:22:51,782 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 2.071e+02 3.012e+02 3.640e+02 4.456e+02 1.368e+03, threshold=7.280e+02, percent-clipped=4.0
2022-12-24 04:22:58,243 INFO [scaling.py:679] (2/4) Whitening: num_groups=8, num_channels=96, metric=1.63 vs. limit=2.0
2022-12-24 04:23:04,825 WARNING [train.py:1060] (2/4) Exclude cut with ID 6330-62850-0007-91323-0_sp0.9 from training. Duration: 24.9833125
2022-12-24 04:23:23,378 INFO [train.py:894] (2/4) Epoch 29, batch 1000, loss[loss=0.1504, simple_loss=0.2329, pruned_loss=0.03397, over 18577.00 frames. ], tot_loss[loss=0.1677, simple_loss=0.258, pruned_loss=0.03872, over 3689379.47 frames. ], batch size: 41, lr: 3.95e-03, grad_scale: 16.0
2022-12-24 04:23:31,556 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.5925, 1.7900, 1.5792, 2.1192, 2.2459, 1.6867, 1.4312, 1.4614],
       device='cuda:2'), covar=tensor([0.1806, 0.1763, 0.1566, 0.1017, 0.1099, 0.1044, 0.2200, 0.1466],
       device='cuda:2'), in_proj_covar=tensor([0.0247, 0.0230, 0.0222, 0.0202, 0.0261, 0.0197, 0.0227, 0.0203],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-24 04:23:36,988 WARNING [train.py:1060] (2/4) Exclude cut with ID 5239-32139-0047-9341-0 from training. Duration: 27.14
2022-12-24 04:23:53,416 WARNING [train.py:1060] (2/4) Exclude cut with ID 2929-85685-0044-62348-0 from training. Duration: 22.44
2022-12-24 04:24:04,830 INFO [zipformer.py:660] (2/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=99199.0, num_to_drop=0, layers_to_drop=set()
2022-12-24 04:24:37,472 INFO [train.py:894] (2/4) Epoch 29, batch 1050, loss[loss=0.1526, simple_loss=0.2437, pruned_loss=0.03073, over 18677.00 frames. ], tot_loss[loss=0.1673, simple_loss=0.2576, pruned_loss=0.03853, over 3695004.61 frames. ], batch size: 48, lr: 3.95e-03, grad_scale: 16.0
2022-12-24 04:24:41,492 INFO [scaling.py:679] (2/4) Whitening: num_groups=8, num_channels=192, metric=1.53 vs. limit=2.0
2022-12-24 04:24:48,875 INFO [zipformer.py:660] (2/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=99229.0, num_to_drop=0, layers_to_drop=set()
2022-12-24 04:25:11,841 WARNING [train.py:1060] (2/4) Exclude cut with ID 2929-85685-0060-62364-0_sp0.9 from training. Duration: 21.361125
2022-12-24 04:25:15,380 INFO [zipformer.py:660] (2/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=99246.0, num_to_drop=0, layers_to_drop=set()
2022-12-24 04:25:16,057 INFO [scaling.py:679] (2/4) Whitening: num_groups=8, num_channels=96, metric=1.16 vs. limit=2.0
2022-12-24 04:25:16,893 INFO [zipformer.py:660] (2/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=99247.0, num_to_drop=0, layers_to_drop=set()
2022-12-24 04:25:18,204 WARNING [train.py:1060] (2/4) Exclude cut with ID 2929-85685-0079-62383-0_sp1.1 from training. Duration: 27.0318125
2022-12-24 04:25:20,995 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 2.030e+02 3.056e+02 3.707e+02 4.620e+02 9.577e+02, threshold=7.414e+02, percent-clipped=2.0
2022-12-24 04:25:22,737 INFO [zipformer.py:660] (2/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=99251.0, num_to_drop=0, layers_to_drop=set()
2022-12-24 04:25:27,997 WARNING [train.py:1060] (2/4) Exclude cut with ID 5622-44585-0006-90525-0_sp0.9 from training. Duration: 28.638875
2022-12-24 04:25:42,335 WARNING [train.py:1060] (2/4) Exclude cut with ID 3340-169293-0054-76830-0 from training. Duration: 20.4
2022-12-24 04:25:46,957 INFO [zipformer.py:660] (2/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=99268.0, num_to_drop=0, layers_to_drop=set()
2022-12-24 04:25:52,966 INFO [train.py:894] (2/4) Epoch 29, batch 1100, loss[loss=0.1545, simple_loss=0.2438, pruned_loss=0.03261, over 18543.00 frames. ], tot_loss[loss=0.1671, simple_loss=0.2572, pruned_loss=0.03852, over 3699022.93 frames. ], batch size: 47, lr: 3.95e-03, grad_scale: 16.0
2022-12-24 04:26:17,105 WARNING [train.py:1060] (2/4) Exclude cut with ID 2929-85685-0071-62375-0 from training. Duration: 20.025
2022-12-24 04:26:17,118 WARNING [train.py:1060] (2/4) Exclude cut with ID 2364-131735-0112-64612-0_sp0.9 from training. Duration: 20.488875
2022-12-24 04:26:20,179 WARNING [train.py:1060] (2/4) Exclude cut with ID 2929-85685-0079-62383-0 from training. Duration: 29.735
2022-12-24 04:26:21,143 INFO [zipformer.py:660] (2/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=99290.0, num_to_drop=0, layers_to_drop=set()
2022-12-24 04:27:08,761 INFO [train.py:894] (2/4) Epoch 29, batch 1150, loss[loss=0.1611, simple_loss=0.2569, pruned_loss=0.03265, over 18575.00 frames. ], tot_loss[loss=0.1666, simple_loss=0.2567, pruned_loss=0.03827, over 3703111.71 frames. ], batch size: 56, lr: 3.95e-03, grad_scale: 16.0
2022-12-24 04:27:12,604 INFO [zipformer.py:660] (2/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=99324.0, num_to_drop=0, layers_to_drop=set()
2022-12-24 04:27:16,530 INFO [zipformer.py:660] (2/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=99327.0, num_to_drop=0, layers_to_drop=set()
2022-12-24 04:27:19,600 INFO [zipformer.py:660] (2/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=99329.0, num_to_drop=0, layers_to_drop=set()
2022-12-24 04:27:48,217 WARNING [train.py:1060] (2/4) Exclude cut with ID 7276-92427-0014-12983-0_sp0.9 from training. Duration: 21.3055625
2022-12-24 04:27:49,624 WARNING [train.py:1060] (2/4) Exclude cut with ID 1025-75365-0008-79168-0_sp0.9 from training. Duration: 22.0666875
2022-12-24 04:27:51,033 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 1.723e+02 3.197e+02 3.795e+02 4.539e+02 9.903e+02, threshold=7.590e+02, percent-clipped=2.0
2022-12-24 04:28:23,583 INFO [train.py:894] (2/4) Epoch 29, batch 1200, loss[loss=0.1835, simple_loss=0.2775, pruned_loss=0.04478, over 18728.00 frames. ], tot_loss[loss=0.1669, simple_loss=0.257, pruned_loss=0.03837, over 3705493.44 frames. ], batch size: 52, lr: 3.95e-03, grad_scale: 16.0
2022-12-24 04:28:43,312 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([2.2741, 2.7235, 3.0282, 1.4973, 2.8449, 2.8994, 2.0959, 3.2225],
       device='cuda:2'), covar=tensor([0.1290, 0.1566, 0.1382, 0.2260, 0.0722, 0.1192, 0.2085, 0.0540],
       device='cuda:2'), in_proj_covar=tensor([0.0197, 0.0214, 0.0208, 0.0194, 0.0170, 0.0217, 0.0215, 0.0193],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-24 04:28:43,327 INFO [zipformer.py:660] (2/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=99385.0, num_to_drop=0, layers_to_drop=set()
2022-12-24 04:29:07,429 INFO [scaling.py:679] (2/4) Whitening: num_groups=8, num_channels=96, metric=1.15 vs. limit=2.0
2022-12-24 04:29:14,357 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.6270, 1.4977, 1.4218, 0.7234, 1.6797, 1.5567, 1.4967, 1.3308],
       device='cuda:2'), covar=tensor([0.0462, 0.0620, 0.0537, 0.0900, 0.0484, 0.0481, 0.0513, 0.1041],
       device='cuda:2'), in_proj_covar=tensor([0.0126, 0.0131, 0.0129, 0.0117, 0.0105, 0.0129, 0.0135, 0.0162],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0001, 0.0001, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-24 04:29:15,561 INFO [zipformer.py:660] (2/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=99407.0, num_to_drop=0, layers_to_drop=set()
2022-12-24 04:29:32,793 WARNING [train.py:1060] (2/4) Exclude cut with ID 2929-85685-0062-62366-0 from training. Duration: 20.26
2022-12-24 04:29:35,384 INFO [zipformer.py:660] (2/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=99420.0, num_to_drop=0, layers_to_drop=set()
2022-12-24 04:29:37,888 INFO [train.py:894] (2/4) Epoch 29, batch 1250, loss[loss=0.1589, simple_loss=0.245, pruned_loss=0.03643, over 18571.00 frames. ], tot_loss[loss=0.1668, simple_loss=0.2569, pruned_loss=0.03835, over 3707399.56 frames. ], batch size: 49, lr: 3.95e-03, grad_scale: 16.0
2022-12-24 04:29:48,586 WARNING [train.py:1060] (2/4) Exclude cut with ID 5239-32139-0030-9324-0_sp0.9 from training. Duration: 21.3444375
2022-12-24 04:30:19,587 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 1.792e+02 3.036e+02 3.747e+02 4.299e+02 8.271e+02, threshold=7.494e+02, percent-clipped=1.0
2022-12-24 04:30:44,499 WARNING [train.py:1060] (2/4) Exclude cut with ID 497-129325-0061-62254-0_sp1.1 from training. Duration: 0.97725
2022-12-24 04:30:47,256 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([2.2746, 2.0228, 1.5445, 0.5140, 1.4653, 1.9758, 1.7325, 1.8021],
       device='cuda:2'), covar=tensor([0.0634, 0.0544, 0.1235, 0.1807, 0.1276, 0.1700, 0.1797, 0.0816],
       device='cuda:2'), in_proj_covar=tensor([0.0176, 0.0189, 0.0209, 0.0189, 0.0212, 0.0206, 0.0218, 0.0205],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-24 04:30:47,258 INFO [zipformer.py:660] (2/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=99468.0, num_to_drop=1, layers_to_drop={1}
2022-12-24 04:30:47,531 INFO [scaling.py:679] (2/4) Whitening: num_groups=1, num_channels=384, metric=2.84 vs. limit=5.0
2022-12-24 04:30:48,804 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.4746, 1.4134, 1.4377, 1.3502, 0.8487, 2.4065, 0.8256, 1.3532],
       device='cuda:2'), covar=tensor([0.3130, 0.2142, 0.2085, 0.2196, 0.1602, 0.0310, 0.1784, 0.0904],
       device='cuda:2'), in_proj_covar=tensor([0.0132, 0.0119, 0.0126, 0.0124, 0.0107, 0.0096, 0.0091, 0.0090],
       device='cuda:2'), out_proj_covar=tensor([0.0005, 0.0005, 0.0005, 0.0005, 0.0004, 0.0003, 0.0004, 0.0004],
       device='cuda:2')
2022-12-24 04:30:52,825 INFO [train.py:894] (2/4) Epoch 29, batch 1300, loss[loss=0.1791, simple_loss=0.2789, pruned_loss=0.03962, over 18629.00 frames. ], tot_loss[loss=0.166, simple_loss=0.256, pruned_loss=0.03804, over 3709037.99 frames. ], batch size: 69, lr: 3.95e-03, grad_scale: 16.0
2022-12-24 04:31:06,882 INFO [zipformer.py:660] (2/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=99481.0, num_to_drop=0, layers_to_drop=set()
2022-12-24 04:31:15,651 INFO [scaling.py:679] (2/4) Whitening: num_groups=1, num_channels=384, metric=2.89 vs. limit=5.0
2022-12-24 04:31:26,335 WARNING [train.py:1060] (2/4) Exclude cut with ID 7395-89880-0031-39906-0_sp0.9 from training. Duration: 22.97225
2022-12-24 04:31:51,667 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.6009, 3.0000, 3.3098, 1.4424, 2.8932, 3.6553, 2.9859, 2.7785],
       device='cuda:2'), covar=tensor([0.0844, 0.0405, 0.0322, 0.0567, 0.0371, 0.0560, 0.0324, 0.0726],
       device='cuda:2'), in_proj_covar=tensor([0.0152, 0.0174, 0.0132, 0.0142, 0.0149, 0.0144, 0.0168, 0.0182],
       device='cuda:2'), out_proj_covar=tensor([1.1403e-04, 1.3087e-04, 9.6836e-05, 1.0404e-04, 1.0888e-04, 1.0823e-04,
        1.2675e-04, 1.3635e-04], device='cuda:2')
2022-12-24 04:31:55,878 WARNING [train.py:1060] (2/4) Exclude cut with ID 7395-89880-0047-39922-0_sp0.9 from training. Duration: 21.97775
2022-12-24 04:32:08,087 INFO [train.py:894] (2/4) Epoch 29, batch 1350, loss[loss=0.162, simple_loss=0.258, pruned_loss=0.03295, over 18579.00 frames. ], tot_loss[loss=0.1665, simple_loss=0.2566, pruned_loss=0.03819, over 3710106.02 frames. ], batch size: 56, lr: 3.94e-03, grad_scale: 16.0
2022-12-24 04:32:10,251 WARNING [train.py:1060] (2/4) Exclude cut with ID 1112-1043-0006-89194-0_sp0.9 from training. Duration: 21.8333125
2022-12-24 04:32:19,379 INFO [zipformer.py:660] (2/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=99529.0, num_to_drop=1, layers_to_drop={0}
2022-12-24 04:32:20,324 WARNING [train.py:1060] (2/4) Exclude cut with ID 1914-133440-0031-94921-0 from training. Duration: 20.47
2022-12-24 04:32:44,446 INFO [zipformer.py:660] (2/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=99546.0, num_to_drop=0, layers_to_drop=set()
2022-12-24 04:32:48,457 INFO [scaling.py:679] (2/4) Whitening: num_groups=8, num_channels=96, metric=1.59 vs. limit=2.0
2022-12-24 04:32:50,068 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 1.781e+02 3.126e+02 3.882e+02 4.637e+02 1.252e+03, threshold=7.765e+02, percent-clipped=4.0
2022-12-24 04:32:51,755 INFO [zipformer.py:660] (2/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=99551.0, num_to_drop=0, layers_to_drop=set()
2022-12-24 04:33:16,928 INFO [zipformer.py:660] (2/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=99568.0, num_to_drop=0, layers_to_drop=set()
2022-12-24 04:33:22,788 INFO [train.py:894] (2/4) Epoch 29, batch 1400, loss[loss=0.1713, simple_loss=0.2583, pruned_loss=0.04216, over 18527.00 frames. ], tot_loss[loss=0.1665, simple_loss=0.2566, pruned_loss=0.03816, over 3710524.17 frames. ], batch size: 58, lr: 3.94e-03, grad_scale: 16.0
2022-12-24 04:33:24,445 WARNING [train.py:1060] (2/4) Exclude cut with ID 7395-89880-0037-39912-0_sp0.9 from training. Duration: 20.67225
2022-12-24 04:33:38,980 INFO [scaling.py:679] (2/4) Whitening: num_groups=8, num_channels=96, metric=1.18 vs. limit=2.0
2022-12-24 04:33:43,060 INFO [zipformer.py:660] (2/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=99585.0, num_to_drop=0, layers_to_drop=set()
2022-12-24 04:33:44,519 WARNING [train.py:1060] (2/4) Exclude cut with ID 1914-133440-0024-94914-0_sp0.9 from training. Duration: 25.2444375
2022-12-24 04:33:49,136 INFO [zipformer.py:660] (2/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=99589.0, num_to_drop=0, layers_to_drop=set()
2022-12-24 04:33:55,840 INFO [zipformer.py:660] (2/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=99594.0, num_to_drop=0, layers_to_drop=set()
2022-12-24 04:34:02,717 INFO [zipformer.py:660] (2/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=99599.0, num_to_drop=0, layers_to_drop=set()
2022-12-24 04:34:05,949 WARNING [train.py:1060] (2/4) Exclude cut with ID 3340-169293-0021-76797-0_sp0.9 from training. Duration: 21.1445
2022-12-24 04:34:38,032 INFO [train.py:894] (2/4) Epoch 29, batch 1450, loss[loss=0.1781, simple_loss=0.2779, pruned_loss=0.0392, over 18622.00 frames. ], tot_loss[loss=0.1662, simple_loss=0.2564, pruned_loss=0.03794, over 3711033.01 frames. ], batch size: 62, lr: 3.94e-03, grad_scale: 16.0
2022-12-24 04:34:41,295 INFO [zipformer.py:660] (2/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=99624.0, num_to_drop=0, layers_to_drop=set()
2022-12-24 04:34:45,481 INFO [zipformer.py:660] (2/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=99627.0, num_to_drop=0, layers_to_drop=set()
2022-12-24 04:34:49,917 INFO [zipformer.py:660] (2/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=99629.0, num_to_drop=0, layers_to_drop=set()
2022-12-24 04:35:19,634 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 1.877e+02 3.031e+02 3.905e+02 4.844e+02 1.106e+03, threshold=7.811e+02, percent-clipped=1.0
2022-12-24 04:35:20,033 INFO [zipformer.py:660] (2/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=99650.0, num_to_drop=0, layers_to_drop=set()
2022-12-24 04:35:21,110 WARNING [train.py:1060] (2/4) Exclude cut with ID 2929-85685-0079-62383-0_sp0.9 from training. Duration: 33.038875
2022-12-24 04:35:53,197 INFO [train.py:894] (2/4) Epoch 29, batch 1500, loss[loss=0.16, simple_loss=0.258, pruned_loss=0.03106, over 18634.00 frames. ], tot_loss[loss=0.1653, simple_loss=0.2558, pruned_loss=0.03746, over 3712510.18 frames. ], batch size: 53, lr: 3.94e-03, grad_scale: 16.0
2022-12-24 04:35:54,969 WARNING [train.py:1060] (2/4) Exclude cut with ID 6426-64291-0000-16059-0_sp0.9 from training. Duration: 20.0944375
2022-12-24 04:35:56,297 INFO [scaling.py:679] (2/4) Whitening: num_groups=1, num_channels=384, metric=2.46 vs. limit=5.0
2022-12-24 04:35:58,685 INFO [zipformer.py:660] (2/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=99675.0, num_to_drop=0, layers_to_drop=set()
2022-12-24 04:36:00,552 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([2.2626, 2.1689, 1.6223, 2.5394, 2.4551, 2.1422, 2.9214, 2.2791],
       device='cuda:2'), covar=tensor([0.0834, 0.1796, 0.2824, 0.1707, 0.1735, 0.0881, 0.0940, 0.1247],
       device='cuda:2'), in_proj_covar=tensor([0.0185, 0.0220, 0.0260, 0.0294, 0.0245, 0.0197, 0.0207, 0.0212],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-24 04:36:06,127 INFO [zipformer.py:660] (2/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=99680.0, num_to_drop=0, layers_to_drop=set()
2022-12-24 04:36:06,296 INFO [zipformer.py:660] (2/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=99680.0, num_to_drop=0, layers_to_drop=set()
2022-12-24 04:36:10,105 WARNING [train.py:1060] (2/4) Exclude cut with ID 2929-85685-0044-62348-0_sp1.1 from training. Duration: 20.4
2022-12-24 04:36:18,564 WARNING [train.py:1060] (2/4) Exclude cut with ID 6330-62851-0022-91297-0 from training. Duration: 20.085
2022-12-24 04:36:24,459 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.6710, 1.5761, 1.6318, 1.5273, 1.0898, 3.6601, 1.4658, 1.8914],
       device='cuda:2'), covar=tensor([0.3084, 0.2170, 0.2023, 0.2140, 0.1602, 0.0173, 0.1618, 0.0925],
       device='cuda:2'), in_proj_covar=tensor([0.0132, 0.0119, 0.0125, 0.0124, 0.0107, 0.0096, 0.0091, 0.0090],
       device='cuda:2'), out_proj_covar=tensor([0.0005, 0.0005, 0.0005, 0.0005, 0.0004, 0.0003, 0.0004, 0.0004],
       device='cuda:2')
2022-12-24 04:36:26,955 WARNING [train.py:1060] (2/4) Exclude cut with ID 4860-13185-0032-76709-0_sp0.9 from training. Duration: 23.07775
2022-12-24 04:36:44,892 INFO [zipformer.py:660] (2/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=99707.0, num_to_drop=0, layers_to_drop=set()
2022-12-24 04:36:57,534 INFO [scaling.py:679] (2/4) Whitening: num_groups=8, num_channels=96, metric=1.15 vs. limit=2.0
2022-12-24 04:37:07,842 INFO [train.py:894] (2/4) Epoch 29, batch 1550, loss[loss=0.1677, simple_loss=0.2553, pruned_loss=0.04008, over 18694.00 frames. ], tot_loss[loss=0.1647, simple_loss=0.2554, pruned_loss=0.03699, over 3712996.10 frames. ], batch size: 50, lr: 3.94e-03, grad_scale: 16.0
2022-12-24 04:37:12,968 WARNING [train.py:1060] (2/4) Exclude cut with ID 2929-85685-0044-62348-0_sp0.9 from training. Duration: 24.9333125
2022-12-24 04:37:16,177 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([2.4424, 3.6640, 3.6001, 1.2699, 3.8342, 2.7568, 1.1998, 2.3404],
       device='cuda:2'), covar=tensor([0.2290, 0.1283, 0.1439, 0.3942, 0.0822, 0.0946, 0.4114, 0.1593],
       device='cuda:2'), in_proj_covar=tensor([0.0152, 0.0150, 0.0161, 0.0126, 0.0153, 0.0117, 0.0146, 0.0118],
       device='cuda:2'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0002, 0.0003, 0.0002, 0.0003, 0.0002],
       device='cuda:2')
2022-12-24 04:37:36,329 INFO [zipformer.py:660] (2/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=99741.0, num_to_drop=0, layers_to_drop=set()
2022-12-24 04:37:49,146 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 2.330e+02 3.376e+02 3.911e+02 4.789e+02 1.084e+03, threshold=7.821e+02, percent-clipped=4.0
2022-12-24 04:37:53,699 WARNING [train.py:1060] (2/4) Exclude cut with ID 5118-111612-0016-124680-0_sp0.9 from training. Duration: 20.388875
2022-12-24 04:37:56,713 INFO [zipformer.py:660] (2/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=99755.0, num_to_drop=0, layers_to_drop=set()
2022-12-24 04:37:59,956 WARNING [train.py:1060] (2/4) Exclude cut with ID 432-122774-0017-62487-0_sp1.1 from training. Duration: 20.3590625
2022-12-24 04:38:11,416 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.6427, 1.2690, 1.4882, 1.7864, 1.5380, 3.3264, 1.3882, 1.4624],
       device='cuda:2'), covar=tensor([0.0973, 0.2285, 0.1273, 0.1128, 0.1787, 0.0300, 0.1754, 0.2003],
       device='cuda:2'), in_proj_covar=tensor([0.0074, 0.0083, 0.0073, 0.0076, 0.0092, 0.0077, 0.0085, 0.0079],
       device='cuda:2'), out_proj_covar=tensor([0.0004, 0.0004, 0.0004, 0.0004, 0.0004, 0.0004, 0.0004, 0.0004],
       device='cuda:2')
2022-12-24 04:38:23,631 INFO [train.py:894] (2/4) Epoch 29, batch 1600, loss[loss=0.1633, simple_loss=0.2589, pruned_loss=0.03383, over 18729.00 frames. ], tot_loss[loss=0.1655, simple_loss=0.2563, pruned_loss=0.03729, over 3713890.41 frames. ], batch size: 52, lr: 3.94e-03, grad_scale: 16.0
2022-12-24 04:38:29,388 INFO [zipformer.py:660] (2/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=99776.0, num_to_drop=0, layers_to_drop=set()
2022-12-24 04:39:10,507 WARNING [train.py:1060] (2/4) Exclude cut with ID 3557-8342-0013-54691-0_sp1.1 from training. Duration: 0.836375
2022-12-24 04:39:12,415 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([0.6080, 2.0182, 2.2808, 1.1881, 1.4813, 2.4467, 2.2786, 1.8360],
       device='cuda:2'), covar=tensor([0.0844, 0.0366, 0.0295, 0.0463, 0.0383, 0.0428, 0.0232, 0.0721],
       device='cuda:2'), in_proj_covar=tensor([0.0151, 0.0173, 0.0132, 0.0142, 0.0149, 0.0143, 0.0167, 0.0180],
       device='cuda:2'), out_proj_covar=tensor([1.1358e-04, 1.2994e-04, 9.6681e-05, 1.0339e-04, 1.0849e-04, 1.0725e-04,
        1.2582e-04, 1.3517e-04], device='cuda:2')
2022-12-24 04:39:37,377 INFO [train.py:894] (2/4) Epoch 29, batch 1650, loss[loss=0.1732, simple_loss=0.2682, pruned_loss=0.03914, over 18690.00 frames. ], tot_loss[loss=0.1663, simple_loss=0.2565, pruned_loss=0.038, over 3714399.81 frames. ], batch size: 62, lr: 3.94e-03, grad_scale: 16.0
2022-12-24 04:39:40,327 INFO [zipformer.py:660] (2/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=99824.0, num_to_drop=1, layers_to_drop={1}
2022-12-24 04:39:46,815 INFO [scaling.py:679] (2/4) Whitening: num_groups=8, num_channels=96, metric=1.18 vs. limit=2.0
2022-12-24 04:39:52,162 WARNING [train.py:1060] (2/4) Exclude cut with ID 8565-290391-0049-67394-0_sp0.9 from training. Duration: 21.3166875
2022-12-24 04:40:17,780 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 2.398e+02 3.479e+02 4.265e+02 5.308e+02 1.026e+03, threshold=8.530e+02, percent-clipped=3.0
2022-12-24 04:40:22,166 WARNING [train.py:1060] (2/4) Exclude cut with ID 6533-399-0029-104863-0_sp0.9 from training. Duration: 22.1055625
2022-12-24 04:40:33,452 WARNING [train.py:1060] (2/4) Exclude cut with ID 7699-105389-0094-26379-0_sp1.1 from training. Duration: 21.77725
2022-12-24 04:40:52,339 INFO [train.py:894] (2/4) Epoch 29, batch 1700, loss[loss=0.1754, simple_loss=0.2693, pruned_loss=0.0408, over 18724.00 frames. ], tot_loss[loss=0.1678, simple_loss=0.257, pruned_loss=0.0393, over 3714938.07 frames. ], batch size: 52, lr: 3.94e-03, grad_scale: 16.0
2022-12-24 04:40:54,109 WARNING [train.py:1060] (2/4) Exclude cut with ID 7255-291500-0005-134304-0_sp0.9 from training. Duration: 27.8166875
2022-12-24 04:41:11,367 INFO [zipformer.py:660] (2/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=99885.0, num_to_drop=0, layers_to_drop=set()
2022-12-24 04:41:18,352 WARNING [train.py:1060] (2/4) Exclude cut with ID 7357-94126-0021-15852-0_sp1.1 from training. Duration: 22.5090625
2022-12-24 04:41:23,697 WARNING [train.py:1060] (2/4) Exclude cut with ID 7255-291500-0005-134304-0 from training. Duration: 25.035
2022-12-24 04:41:41,814 WARNING [train.py:1060] (2/4) Exclude cut with ID 774-127930-0014-10412-0_sp1.1 from training. Duration: 0.95
2022-12-24 04:42:00,794 WARNING [train.py:1060] (2/4) Exclude cut with ID 3033-130750-0096-55598-0_sp0.9 from training. Duration: 0.92225
2022-12-24 04:42:06,483 INFO [train.py:894] (2/4) Epoch 29, batch 1750, loss[loss=0.1505, simple_loss=0.2432, pruned_loss=0.02893, over 18569.00 frames. ], tot_loss[loss=0.1688, simple_loss=0.2572, pruned_loss=0.04025, over 3714211.51 frames. ], batch size: 49, lr: 3.94e-03, grad_scale: 16.0
2022-12-24 04:42:09,574 INFO [zipformer.py:660] (2/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=99924.0, num_to_drop=0, layers_to_drop=set()
2022-12-24 04:42:09,669 INFO [zipformer.py:660] (2/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=99924.0, num_to_drop=0, layers_to_drop=set()
2022-12-24 04:42:22,319 INFO [zipformer.py:660] (2/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=99933.0, num_to_drop=0, layers_to_drop=set()
2022-12-24 04:42:23,683 WARNING [train.py:1060] (2/4) Exclude cut with ID 4511-76322-0006-80011-0 from training. Duration: 21.97
2022-12-24 04:42:40,160 INFO [zipformer.py:660] (2/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=99945.0, num_to_drop=0, layers_to_drop=set()
2022-12-24 04:42:44,120 WARNING [train.py:1060] (2/4) Exclude cut with ID 7492-105653-0055-62765-0_sp0.9 from training. Duration: 21.97225
2022-12-24 04:42:44,151 WARNING [train.py:1060] (2/4) Exclude cut with ID 453-131332-0000-47844-0_sp0.9 from training. Duration: 25.3333125
2022-12-24 04:42:44,476 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.8654, 1.2281, 1.0313, 1.4528, 2.2850, 1.2933, 1.5121, 1.6814],
       device='cuda:2'), covar=tensor([0.1571, 0.2175, 0.2070, 0.1463, 0.1636, 0.1821, 0.1456, 0.1722],
       device='cuda:2'), in_proj_covar=tensor([0.0095, 0.0098, 0.0117, 0.0097, 0.0121, 0.0093, 0.0098, 0.0095],
       device='cuda:2'), out_proj_covar=tensor([0.0003, 0.0003, 0.0004, 0.0003, 0.0004, 0.0003, 0.0003, 0.0003],
       device='cuda:2')
2022-12-24 04:42:47,625 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 2.605e+02 3.896e+02 4.661e+02 6.128e+02 1.064e+03, threshold=9.321e+02, percent-clipped=8.0
2022-12-24 04:42:56,453 WARNING [train.py:1060] (2/4) Exclude cut with ID 5172-29468-0015-19128-0_sp0.9 from training. Duration: 21.5055625
2022-12-24 04:43:07,237 WARNING [train.py:1060] (2/4) Exclude cut with ID 453-131332-0000-47844-0_sp1.1 from training. Duration: 20.72725
2022-12-24 04:43:21,508 INFO [train.py:894] (2/4) Epoch 29, batch 1800, loss[loss=0.1771, simple_loss=0.2593, pruned_loss=0.04751, over 18478.00 frames. ], tot_loss[loss=0.1705, simple_loss=0.2582, pruned_loss=0.04141, over 3714828.07 frames. ], batch size: 50, lr: 3.94e-03, grad_scale: 16.0
2022-12-24 04:43:21,705 INFO [zipformer.py:660] (2/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=99972.0, num_to_drop=0, layers_to_drop=set()
2022-12-24 04:43:33,064 INFO [zipformer.py:660] (2/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=99980.0, num_to_drop=0, layers_to_drop=set()
2022-12-24 04:43:40,215 WARNING [train.py:1060] (2/4) Exclude cut with ID 8631-249866-0030-130156-0_sp0.9 from training. Duration: 26.32775
2022-12-24 04:44:16,711 WARNING [train.py:1060] (2/4) Exclude cut with ID 3867-173237-0077-144769-0 from training. Duration: 20.025
2022-12-24 04:44:17,043 INFO [zipformer.py:660] (2/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=100007.0, num_to_drop=1, layers_to_drop={0}
2022-12-24 04:44:21,029 WARNING [train.py:1060] (2/4) Exclude cut with ID 6709-74022-0004-86860-0_sp1.1 from training. Duration: 0.9409375
2022-12-24 04:44:21,035 WARNING [train.py:1060] (2/4) Exclude cut with ID 4757-1811-0023-62229-0_sp0.9 from training. Duration: 21.37775
2022-12-24 04:44:39,152 INFO [train.py:894] (2/4) Epoch 29, batch 1850, loss[loss=0.1401, simple_loss=0.2196, pruned_loss=0.03033, over 18478.00 frames. ], tot_loss[loss=0.1712, simple_loss=0.2578, pruned_loss=0.04234, over 3714977.38 frames. ], batch size: 43, lr: 3.93e-03, grad_scale: 16.0
2022-12-24 04:44:42,160 WARNING [train.py:1060] (2/4) Exclude cut with ID 1250-135782-0004-25974-0_sp0.9 from training. Duration: 21.17225
2022-12-24 04:44:42,172 WARNING [train.py:1060] (2/4) Exclude cut with ID 7357-94126-0021-15852-0_sp0.9 from training. Duration: 27.511125
2022-12-24 04:44:48,209 INFO [zipformer.py:660] (2/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=100028.0, num_to_drop=0, layers_to_drop=set()
2022-12-24 04:44:59,251 INFO [zipformer.py:660] (2/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=100036.0, num_to_drop=0, layers_to_drop=set()
2022-12-24 04:45:12,934 WARNING [train.py:1060] (2/4) Exclude cut with ID 453-131332-0000-47844-0 from training. Duration: 22.8
2022-12-24 04:45:17,144 WARNING [train.py:1060] (2/4) Exclude cut with ID 4964-30587-0040-44509-0 from training. Duration: 22.585
2022-12-24 04:45:20,321 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 2.870e+02 3.993e+02 4.913e+02 5.637e+02 1.150e+03, threshold=9.826e+02, percent-clipped=4.0
2022-12-24 04:45:48,570 INFO [zipformer.py:660] (2/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=100068.0, num_to_drop=1, layers_to_drop={2}
2022-12-24 04:45:49,486 WARNING [train.py:1060] (2/4) Exclude cut with ID 6978-92210-0001-146967-0_sp0.9 from training. Duration: 22.0166875
2022-12-24 04:45:53,702 INFO [train.py:894] (2/4) Epoch 29, batch 1900, loss[loss=0.2048, simple_loss=0.2914, pruned_loss=0.05909, over 18712.00 frames. ], tot_loss[loss=0.1726, simple_loss=0.2586, pruned_loss=0.04332, over 3714455.69 frames. ], batch size: 60, lr: 3.93e-03, grad_scale: 16.0
2022-12-24 04:45:59,795 INFO [zipformer.py:660] (2/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=100076.0, num_to_drop=0, layers_to_drop=set()
2022-12-24 04:46:06,087 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([2.2878, 1.6210, 1.9909, 1.9177, 2.2879, 2.3334, 2.0829, 1.9832],
       device='cuda:2'), covar=tensor([0.2384, 0.3422, 0.2767, 0.3119, 0.2198, 0.1031, 0.3559, 0.1421],
       device='cuda:2'), in_proj_covar=tensor([0.0273, 0.0301, 0.0289, 0.0328, 0.0319, 0.0260, 0.0357, 0.0250],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-24 04:46:06,998 WARNING [train.py:1060] (2/4) Exclude cut with ID 7255-291500-0003-134302-0_sp1.1 from training. Duration: 24.395375
2022-12-24 04:46:14,261 WARNING [train.py:1060] (2/4) Exclude cut with ID 3488-85273-0017-41203-0_sp0.9 from training. Duration: 27.47775
2022-12-24 04:46:19,642 WARNING [train.py:1060] (2/4) Exclude cut with ID 432-122774-0017-62487-0_sp0.9 from training. Duration: 24.8833125
2022-12-24 04:46:22,578 WARNING [train.py:1060] (2/4) Exclude cut with ID 6758-72288-0033-108368-0 from training. Duration: 23.39
2022-12-24 04:46:27,436 WARNING [train.py:1060] (2/4) Exclude cut with ID 8291-282929-0007-12994-0_sp0.9 from training. Duration: 28.72225
2022-12-24 04:46:38,415 WARNING [train.py:1060] (2/4) Exclude cut with ID 585-294811-0110-133686-0_sp0.9 from training. Duration: 20.8944375
2022-12-24 04:46:53,124 WARNING [train.py:1060] (2/4) Exclude cut with ID 5796-66357-0007-116447-0_sp0.9 from training. Duration: 23.8444375
2022-12-24 04:47:06,743 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.9633, 1.3752, 0.9033, 1.5748, 2.2688, 1.3426, 1.6103, 1.7889],
       device='cuda:2'), covar=tensor([0.1518, 0.2059, 0.2114, 0.1379, 0.1640, 0.1868, 0.1368, 0.1660],
       device='cuda:2'), in_proj_covar=tensor([0.0095, 0.0098, 0.0117, 0.0097, 0.0122, 0.0093, 0.0098, 0.0095],
       device='cuda:2'), out_proj_covar=tensor([0.0003, 0.0003, 0.0004, 0.0003, 0.0004, 0.0003, 0.0003, 0.0003],
       device='cuda:2')
2022-12-24 04:47:07,844 INFO [train.py:894] (2/4) Epoch 29, batch 1950, loss[loss=0.1806, simple_loss=0.2693, pruned_loss=0.04595, over 18717.00 frames. ], tot_loss[loss=0.1727, simple_loss=0.2584, pruned_loss=0.04348, over 3713403.91 frames. ], batch size: 52, lr: 3.93e-03, grad_scale: 8.0
2022-12-24 04:47:11,089 INFO [zipformer.py:660] (2/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=100124.0, num_to_drop=0, layers_to_drop=set()
2022-12-24 04:47:11,208 INFO [zipformer.py:660] (2/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=100124.0, num_to_drop=1, layers_to_drop={0}
2022-12-24 04:47:18,149 WARNING [train.py:1060] (2/4) Exclude cut with ID 8291-282929-0007-12994-0 from training. Duration: 25.85
2022-12-24 04:47:18,159 WARNING [train.py:1060] (2/4) Exclude cut with ID 8291-282929-0023-13010-0 from training. Duration: 21.39
2022-12-24 04:47:29,848 WARNING [train.py:1060] (2/4) Exclude cut with ID 7357-94126-0014-15845-0 from training. Duration: 27.92
2022-12-24 04:47:50,707 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 2.524e+02 3.680e+02 4.686e+02 5.699e+02 9.097e+02, threshold=9.372e+02, percent-clipped=0.0
2022-12-24 04:47:51,642 INFO [scaling.py:679] (2/4) Whitening: num_groups=1, num_channels=384, metric=4.29 vs. limit=5.0
2022-12-24 04:47:57,131 WARNING [train.py:1060] (2/4) Exclude cut with ID 8631-249866-0039-130165-0_sp0.9 from training. Duration: 20.661125
2022-12-24 04:48:22,914 INFO [train.py:894] (2/4) Epoch 29, batch 2000, loss[loss=0.1719, simple_loss=0.2626, pruned_loss=0.04063, over 18642.00 frames. ], tot_loss[loss=0.1727, simple_loss=0.2583, pruned_loss=0.04354, over 3714111.58 frames. ], batch size: 60, lr: 3.93e-03, grad_scale: 8.0
2022-12-24 04:48:22,987 WARNING [train.py:1060] (2/4) Exclude cut with ID 7357-94126-0043-15874-0_sp0.9 from training. Duration: 20.07225
2022-12-24 04:48:23,057 INFO [zipformer.py:660] (2/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=100172.0, num_to_drop=1, layers_to_drop={0}
2022-12-24 04:48:29,664 WARNING [train.py:1060] (2/4) Exclude cut with ID 1085-156170-0017-128270-0 from training. Duration: 21.01
2022-12-24 04:49:34,540 WARNING [train.py:1060] (2/4) Exclude cut with ID 2195-150901-0045-59933-0 from training. Duration: 20.65
2022-12-24 04:49:37,252 INFO [train.py:894] (2/4) Epoch 29, batch 2050, loss[loss=0.1892, simple_loss=0.2774, pruned_loss=0.05052, over 18531.00 frames. ], tot_loss[loss=0.1728, simple_loss=0.2579, pruned_loss=0.0439, over 3714581.62 frames. ], batch size: 58, lr: 3.93e-03, grad_scale: 8.0
2022-12-24 04:49:40,815 INFO [zipformer.py:660] (2/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=100224.0, num_to_drop=0, layers_to_drop=set()
2022-12-24 04:49:43,275 WARNING [train.py:1060] (2/4) Exclude cut with ID 5796-66357-0007-116447-0 from training. Duration: 21.46
2022-12-24 04:49:47,864 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.6447, 1.5714, 1.7078, 1.5092, 1.1157, 2.9830, 1.2436, 1.7761],
       device='cuda:2'), covar=tensor([0.3204, 0.2074, 0.1909, 0.2204, 0.1597, 0.0256, 0.1805, 0.0877],
       device='cuda:2'), in_proj_covar=tensor([0.0132, 0.0118, 0.0125, 0.0123, 0.0107, 0.0096, 0.0091, 0.0090],
       device='cuda:2'), out_proj_covar=tensor([0.0005, 0.0005, 0.0005, 0.0005, 0.0004, 0.0003, 0.0004, 0.0004],
       device='cuda:2')
2022-12-24 04:49:52,349 INFO [scaling.py:679] (2/4) Whitening: num_groups=8, num_channels=96, metric=1.48 vs. limit=2.0
2022-12-24 04:50:00,974 INFO [zipformer.py:660] (2/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=100238.0, num_to_drop=0, layers_to_drop=set()
2022-12-24 04:50:11,265 INFO [zipformer.py:660] (2/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=100245.0, num_to_drop=0, layers_to_drop=set()
2022-12-24 04:50:20,512 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 2.560e+02 3.985e+02 4.677e+02 5.805e+02 1.123e+03, threshold=9.354e+02, percent-clipped=1.0
2022-12-24 04:50:26,249 WARNING [train.py:1060] (2/4) Exclude cut with ID 3557-8342-0013-54691-0 from training. Duration: 0.92
2022-12-24 04:50:33,880 WARNING [train.py:1060] (2/4) Exclude cut with ID 8291-282929-0023-13010-0_sp0.9 from training. Duration: 23.7666875
2022-12-24 04:50:42,659 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([2.5162, 2.1340, 1.6840, 2.1959, 1.9011, 2.1154, 1.9912, 2.3152],
       device='cuda:2'), covar=tensor([0.2195, 0.3170, 0.2189, 0.2854, 0.3740, 0.1112, 0.2999, 0.1104],
       device='cuda:2'), in_proj_covar=tensor([0.0302, 0.0301, 0.0256, 0.0351, 0.0283, 0.0236, 0.0299, 0.0224],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0003, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-24 04:50:52,854 INFO [train.py:894] (2/4) Epoch 29, batch 2100, loss[loss=0.1488, simple_loss=0.2378, pruned_loss=0.02991, over 18453.00 frames. ], tot_loss[loss=0.1735, simple_loss=0.2583, pruned_loss=0.04435, over 3715522.55 frames. ], batch size: 50, lr: 3.93e-03, grad_scale: 8.0
2022-12-24 04:50:52,977 INFO [zipformer.py:660] (2/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=100272.0, num_to_drop=0, layers_to_drop=set()
2022-12-24 04:51:09,544 WARNING [train.py:1060] (2/4) Exclude cut with ID 8544-281189-0060-101339-0_sp0.9 from training. Duration: 20.861125
2022-12-24 04:51:19,081 WARNING [train.py:1060] (2/4) Exclude cut with ID 3488-65654-0031-41259-0_sp0.9 from training. Duration: 22.711125
2022-12-24 04:51:26,306 INFO [zipformer.py:660] (2/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=100293.0, num_to_drop=0, layers_to_drop=set()
2022-12-24 04:51:35,018 INFO [zipformer.py:660] (2/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=100299.0, num_to_drop=0, layers_to_drop=set()
2022-12-24 04:52:00,248 WARNING [train.py:1060] (2/4) Exclude cut with ID 6951-79737-0043-132310-0_sp1.1 from training. Duration: 22.986375
2022-12-24 04:52:09,438 INFO [train.py:894] (2/4) Epoch 29, batch 2150, loss[loss=0.1742, simple_loss=0.265, pruned_loss=0.04166, over 18396.00 frames. ], tot_loss[loss=0.1747, simple_loss=0.2591, pruned_loss=0.04515, over 3715269.50 frames. ], batch size: 53, lr: 3.93e-03, grad_scale: 8.0
2022-12-24 04:52:15,334 WARNING [train.py:1060] (2/4) Exclude cut with ID 8040-260924-0003-80960-0_sp0.9 from training. Duration: 22.07225
2022-12-24 04:52:21,781 WARNING [train.py:1060] (2/4) Exclude cut with ID 7699-105389-0045-26330-0_sp0.9 from training. Duration: 20.3055625
2022-12-24 04:52:23,443 WARNING [train.py:1060] (2/4) Exclude cut with ID 6356-271890-0060-94317-0_sp0.9 from training. Duration: 20.72225
2022-12-24 04:52:32,861 INFO [zipformer.py:660] (2/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=100336.0, num_to_drop=0, layers_to_drop=set()
2022-12-24 04:52:43,356 WARNING [train.py:1060] (2/4) Exclude cut with ID 3488-85273-0017-41203-0_sp1.1 from training. Duration: 22.4818125
2022-12-24 04:52:54,788 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 2.550e+02 4.110e+02 5.035e+02 6.020e+02 1.639e+03, threshold=1.007e+03, percent-clipped=1.0
2022-12-24 04:53:08,328 WARNING [train.py:1060] (2/4) Exclude cut with ID 4964-30587-0040-44509-0_sp0.9 from training. Duration: 25.0944375
2022-12-24 04:53:11,316 WARNING [train.py:1060] (2/4) Exclude cut with ID 6533-399-0047-104881-0 from training. Duration: 21.515
2022-12-24 04:53:12,846 INFO [zipformer.py:660] (2/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=100363.0, num_to_drop=1, layers_to_drop={2}
2022-12-24 04:53:18,627 WARNING [train.py:1060] (2/4) Exclude cut with ID 7357-94126-0009-15840-0_sp0.9 from training. Duration: 27.02225
2022-12-24 04:53:22,921 WARNING [train.py:1060] (2/4) Exclude cut with ID 432-122774-0010-62480-0_sp0.9 from training. Duration: 22.22225
2022-12-24 04:53:25,696 INFO [train.py:894] (2/4) Epoch 29, batch 2200, loss[loss=0.1724, simple_loss=0.2625, pruned_loss=0.04109, over 18574.00 frames. ], tot_loss[loss=0.1742, simple_loss=0.2584, pruned_loss=0.045, over 3715795.17 frames. ], batch size: 51, lr: 3.93e-03, grad_scale: 8.0
2022-12-24 04:53:29,252 WARNING [train.py:1060] (2/4) Exclude cut with ID 4964-30587-0085-44554-0_sp0.9 from training. Duration: 20.85
2022-12-24 04:53:45,584 INFO [zipformer.py:660] (2/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=100384.0, num_to_drop=0, layers_to_drop=set()
2022-12-24 04:53:54,162 INFO [scaling.py:679] (2/4) Whitening: num_groups=1, num_channels=384, metric=4.73 vs. limit=5.0
2022-12-24 04:54:01,936 WARNING [train.py:1060] (2/4) Exclude cut with ID 4295-39940-0007-92567-0 from training. Duration: 21.54
2022-12-24 04:54:05,075 WARNING [train.py:1060] (2/4) Exclude cut with ID 4964-30587-0040-44509-0_sp1.1 from training. Duration: 20.5318125
2022-12-24 04:54:13,198 INFO [zipformer.py:660] (2/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=100402.0, num_to_drop=0, layers_to_drop=set()
2022-12-24 04:54:15,895 WARNING [train.py:1060] (2/4) Exclude cut with ID 7255-291500-0012-134311-0_sp0.9 from training. Duration: 21.9333125
2022-12-24 04:54:16,478 INFO [scaling.py:679] (2/4) Whitening: num_groups=1, num_channels=384, metric=3.27 vs. limit=5.0
2022-12-24 04:54:22,319 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([2.6293, 2.2246, 1.8111, 2.2622, 2.0091, 2.1898, 2.1076, 2.4492],
       device='cuda:2'), covar=tensor([0.2265, 0.3453, 0.2264, 0.3116, 0.3920, 0.1239, 0.3137, 0.1169],
       device='cuda:2'), in_proj_covar=tensor([0.0302, 0.0302, 0.0256, 0.0351, 0.0283, 0.0237, 0.0299, 0.0225],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0003, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-24 04:54:43,502 INFO [train.py:894] (2/4) Epoch 29, batch 2250, loss[loss=0.2189, simple_loss=0.2919, pruned_loss=0.07301, over 18599.00 frames. ], tot_loss[loss=0.1747, simple_loss=0.2585, pruned_loss=0.04541, over 3714268.12 frames. ], batch size: 184, lr: 3.93e-03, grad_scale: 8.0
2022-12-24 04:54:53,829 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([2.5141, 2.0082, 2.1500, 2.2180, 2.4872, 2.4725, 2.3202, 2.0554],
       device='cuda:2'), covar=tensor([0.2378, 0.3509, 0.2828, 0.3266, 0.2147, 0.1055, 0.3873, 0.1478],
       device='cuda:2'), in_proj_covar=tensor([0.0273, 0.0301, 0.0289, 0.0328, 0.0320, 0.0260, 0.0357, 0.0250],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-24 04:55:05,000 WARNING [train.py:1060] (2/4) Exclude cut with ID 8631-249866-0025-130151-0_sp0.9 from training. Duration: 21.7944375
2022-12-24 04:55:16,408 WARNING [train.py:1060] (2/4) Exclude cut with ID 8291-282929-0002-12989-0_sp0.9 from training. Duration: 22.4666875
2022-12-24 04:55:23,991 WARNING [train.py:1060] (2/4) Exclude cut with ID 6121-9014-0076-24124-0 from training. Duration: 21.635
2022-12-24 04:55:26,873 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 2.445e+02 4.059e+02 4.926e+02 5.682e+02 1.179e+03, threshold=9.852e+02, percent-clipped=4.0
2022-12-24 04:55:28,598 WARNING [train.py:1060] (2/4) Exclude cut with ID 6121-9014-0076-24124-0_sp0.9 from training. Duration: 24.038875
2022-12-24 04:55:45,374 INFO [zipformer.py:660] (2/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=100463.0, num_to_drop=1, layers_to_drop={0}
2022-12-24 04:56:00,249 INFO [train.py:894] (2/4) Epoch 29, batch 2300, loss[loss=0.1641, simple_loss=0.2538, pruned_loss=0.03722, over 18676.00 frames. ], tot_loss[loss=0.1748, simple_loss=0.259, pruned_loss=0.0453, over 3715223.44 frames. ], batch size: 48, lr: 3.93e-03, grad_scale: 8.0
2022-12-24 04:56:11,136 WARNING [train.py:1060] (2/4) Exclude cut with ID 7255-291500-0009-134308-0_sp1.1 from training. Duration: 21.786375
2022-12-24 04:56:22,798 WARNING [train.py:1060] (2/4) Exclude cut with ID 8291-282929-0002-12989-0 from training. Duration: 20.22
2022-12-24 04:57:15,046 INFO [train.py:894] (2/4) Epoch 29, batch 2350, loss[loss=0.1901, simple_loss=0.2781, pruned_loss=0.05103, over 18500.00 frames. ], tot_loss[loss=0.1749, simple_loss=0.2593, pruned_loss=0.0453, over 3715249.07 frames. ], batch size: 58, lr: 3.92e-03, grad_scale: 8.0
2022-12-24 04:57:28,998 INFO [scaling.py:679] (2/4) Whitening: num_groups=1, num_channels=384, metric=3.64 vs. limit=5.0
2022-12-24 04:57:30,520 INFO [scaling.py:679] (2/4) Whitening: num_groups=8, num_channels=96, metric=1.25 vs. limit=2.0
2022-12-24 04:57:56,842 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 2.255e+02 3.961e+02 4.890e+02 6.164e+02 1.267e+03, threshold=9.780e+02, percent-clipped=4.0
2022-12-24 04:58:25,805 WARNING [train.py:1060] (2/4) Exclude cut with ID 6951-79737-0043-132310-0 from training. Duration: 25.285
2022-12-24 04:58:29,158 INFO [train.py:894] (2/4) Epoch 29, batch 2400, loss[loss=0.1698, simple_loss=0.2588, pruned_loss=0.04046, over 18573.00 frames. ], tot_loss[loss=0.1744, simple_loss=0.2589, pruned_loss=0.04497, over 3714750.31 frames. ], batch size: 56, lr: 3.92e-03, grad_scale: 8.0
2022-12-24 04:59:01,586 INFO [zipformer.py:660] (2/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=100594.0, num_to_drop=0, layers_to_drop=set()
2022-12-24 04:59:09,421 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.6860, 1.8109, 1.5867, 2.2080, 2.6507, 1.6687, 1.7544, 1.3571],
       device='cuda:2'), covar=tensor([0.2119, 0.2109, 0.1896, 0.1209, 0.1275, 0.1356, 0.2143, 0.1812],
       device='cuda:2'), in_proj_covar=tensor([0.0252, 0.0234, 0.0224, 0.0205, 0.0265, 0.0200, 0.0230, 0.0206],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-24 04:59:29,541 WARNING [train.py:1060] (2/4) Exclude cut with ID 811-130148-0001-63453-0_sp0.9 from training. Duration: 20.861125
2022-12-24 04:59:44,334 INFO [train.py:894] (2/4) Epoch 29, batch 2450, loss[loss=0.1595, simple_loss=0.2347, pruned_loss=0.04217, over 18528.00 frames. ], tot_loss[loss=0.1746, simple_loss=0.2587, pruned_loss=0.04521, over 3714186.12 frames. ], batch size: 44, lr: 3.92e-03, grad_scale: 8.0
2022-12-24 04:59:48,917 WARNING [train.py:1060] (2/4) Exclude cut with ID 6010-56788-0055-90261-0 from training. Duration: 20.88
2022-12-24 05:00:21,516 WARNING [train.py:1060] (2/4) Exclude cut with ID 7357-94126-0045-15876-0_sp0.9 from training. Duration: 23.4166875
2022-12-24 05:00:27,740 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 2.767e+02 3.997e+02 4.879e+02 5.983e+02 1.573e+03, threshold=9.759e+02, percent-clipped=4.0
2022-12-24 05:00:46,124 INFO [zipformer.py:660] (2/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=100663.0, num_to_drop=1, layers_to_drop={0}
2022-12-24 05:00:59,032 INFO [train.py:894] (2/4) Epoch 29, batch 2500, loss[loss=0.1822, simple_loss=0.2686, pruned_loss=0.04789, over 18514.00 frames. ], tot_loss[loss=0.174, simple_loss=0.2583, pruned_loss=0.04489, over 3714273.57 frames. ], batch size: 58, lr: 3.92e-03, grad_scale: 8.0
2022-12-24 05:01:32,373 INFO [scaling.py:679] (2/4) Whitening: num_groups=8, num_channels=192, metric=1.84 vs. limit=2.0
2022-12-24 05:01:40,065 WARNING [train.py:1060] (2/4) Exclude cut with ID 7255-291500-0006-134305-0 from training. Duration: 21.24
2022-12-24 05:01:40,079 WARNING [train.py:1060] (2/4) Exclude cut with ID 6533-399-0047-104881-0_sp0.9 from training. Duration: 23.9055625
2022-12-24 05:01:50,602 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.5250, 1.0610, 0.7600, 1.1777, 1.9139, 0.6679, 1.2008, 1.3672],
       device='cuda:2'), covar=tensor([0.1630, 0.2182, 0.1848, 0.1501, 0.1841, 0.1907, 0.1485, 0.1728],
       device='cuda:2'), in_proj_covar=tensor([0.0094, 0.0097, 0.0116, 0.0097, 0.0120, 0.0092, 0.0098, 0.0094],
       device='cuda:2'), out_proj_covar=tensor([0.0003, 0.0003, 0.0004, 0.0003, 0.0004, 0.0003, 0.0003, 0.0003],
       device='cuda:2')
2022-12-24 05:01:57,799 INFO [zipformer.py:660] (2/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=100711.0, num_to_drop=1, layers_to_drop={0}
2022-12-24 05:02:13,671 INFO [train.py:894] (2/4) Epoch 29, batch 2550, loss[loss=0.1687, simple_loss=0.2558, pruned_loss=0.04074, over 18658.00 frames. ], tot_loss[loss=0.1749, simple_loss=0.2587, pruned_loss=0.04555, over 3714890.60 frames. ], batch size: 98, lr: 3.92e-03, grad_scale: 8.0
2022-12-24 05:02:13,746 WARNING [train.py:1060] (2/4) Exclude cut with ID 6758-72288-0033-108368-0_sp0.9 from training. Duration: 25.988875
2022-12-24 05:02:24,008 WARNING [train.py:1060] (2/4) Exclude cut with ID 7255-291500-0001-134300-0_sp0.9 from training. Duration: 20.67225
2022-12-24 05:02:25,878 INFO [zipformer.py:660] (2/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=100730.0, num_to_drop=1, layers_to_drop={1}
2022-12-24 05:02:42,991 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.6182, 2.7555, 3.1338, 1.1566, 2.8867, 3.5854, 2.8249, 2.6219],
       device='cuda:2'), covar=tensor([0.0713, 0.0419, 0.0375, 0.0536, 0.0349, 0.0341, 0.0359, 0.0682],
       device='cuda:2'), in_proj_covar=tensor([0.0153, 0.0175, 0.0133, 0.0144, 0.0149, 0.0146, 0.0170, 0.0183],
       device='cuda:2'), out_proj_covar=tensor([1.1481e-04, 1.3142e-04, 9.8047e-05, 1.0488e-04, 1.0868e-04, 1.0926e-04,
        1.2795e-04, 1.3771e-04], device='cuda:2')
2022-12-24 05:02:55,944 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 2.707e+02 4.004e+02 4.626e+02 5.678e+02 1.065e+03, threshold=9.253e+02, percent-clipped=1.0
2022-12-24 05:03:01,286 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([2.1245, 1.9996, 2.4019, 1.5197, 2.3347, 2.4435, 1.6234, 2.8033],
       device='cuda:2'), covar=tensor([0.1146, 0.1843, 0.1227, 0.1829, 0.0715, 0.1100, 0.2467, 0.0508],
       device='cuda:2'), in_proj_covar=tensor([0.0198, 0.0214, 0.0209, 0.0194, 0.0170, 0.0218, 0.0216, 0.0194],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-24 05:03:07,219 INFO [zipformer.py:660] (2/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=100758.0, num_to_drop=1, layers_to_drop={1}
2022-12-24 05:03:12,205 WARNING [train.py:1060] (2/4) Exclude cut with ID 3488-85273-0038-41224-0 from training. Duration: 20.34
2022-12-24 05:03:28,308 INFO [train.py:894] (2/4) Epoch 29, batch 2600, loss[loss=0.143, simple_loss=0.2243, pruned_loss=0.03088, over 18394.00 frames. ], tot_loss[loss=0.1742, simple_loss=0.2581, pruned_loss=0.04518, over 3714531.63 frames. ], batch size: 42, lr: 3.92e-03, grad_scale: 8.0
2022-12-24 05:03:56,134 INFO [zipformer.py:660] (2/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=100791.0, num_to_drop=1, layers_to_drop={3}
2022-12-24 05:04:24,504 WARNING [train.py:1060] (2/4) Exclude cut with ID 7357-94126-0026-15857-0_sp0.9 from training. Duration: 25.061125
2022-12-24 05:04:35,159 WARNING [train.py:1060] (2/4) Exclude cut with ID 3033-130750-0096-55598-0 from training. Duration: 0.83
2022-12-24 05:04:43,744 INFO [train.py:894] (2/4) Epoch 29, batch 2650, loss[loss=0.1994, simple_loss=0.2805, pruned_loss=0.0591, over 18669.00 frames. ], tot_loss[loss=0.1751, simple_loss=0.2589, pruned_loss=0.04561, over 3714240.94 frames. ], batch size: 62, lr: 3.92e-03, grad_scale: 8.0
2022-12-24 05:04:57,880 WARNING [train.py:1060] (2/4) Exclude cut with ID 3488-85273-0017-41203-0 from training. Duration: 24.73
2022-12-24 05:05:03,963 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.9626, 1.4717, 2.6958, 4.2662, 3.1738, 2.7804, 1.1020, 3.1211],
       device='cuda:2'), covar=tensor([0.1911, 0.1741, 0.1378, 0.0538, 0.0900, 0.1280, 0.2054, 0.0902],
       device='cuda:2'), in_proj_covar=tensor([0.0106, 0.0121, 0.0141, 0.0161, 0.0110, 0.0148, 0.0131, 0.0118],
       device='cuda:2'), out_proj_covar=tensor([0.0003, 0.0003, 0.0004, 0.0004, 0.0003, 0.0004, 0.0004, 0.0003],
       device='cuda:2')
2022-12-24 05:05:11,185 WARNING [train.py:1060] (2/4) Exclude cut with ID 7255-291500-0009-134308-0 from training. Duration: 23.965
2022-12-24 05:05:18,469 WARNING [train.py:1060] (2/4) Exclude cut with ID 6978-92210-0030-146996-0_sp0.9 from training. Duration: 22.088875
2022-12-24 05:05:27,245 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 2.255e+02 4.318e+02 5.283e+02 6.216e+02 2.785e+03, threshold=1.057e+03, percent-clipped=2.0
2022-12-24 05:05:38,002 WARNING [train.py:1060] (2/4) Exclude cut with ID 7255-291500-0006-134305-0_sp0.9 from training. Duration: 23.6
2022-12-24 05:05:57,954 INFO [train.py:894] (2/4) Epoch 29, batch 2700, loss[loss=0.1525, simple_loss=0.2419, pruned_loss=0.0315, over 18701.00 frames. ], tot_loss[loss=0.1743, simple_loss=0.2583, pruned_loss=0.04515, over 3713645.40 frames. ], batch size: 50, lr: 3.92e-03, grad_scale: 8.0
2022-12-24 05:06:06,329 INFO [scaling.py:679] (2/4) Whitening: num_groups=1, num_channels=384, metric=3.19 vs. limit=5.0
2022-12-24 05:06:31,439 INFO [zipformer.py:660] (2/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=100894.0, num_to_drop=0, layers_to_drop=set()
2022-12-24 05:07:12,431 INFO [train.py:894] (2/4) Epoch 29, batch 2750, loss[loss=0.1863, simple_loss=0.2702, pruned_loss=0.05116, over 18655.00 frames. ], tot_loss[loss=0.1736, simple_loss=0.2576, pruned_loss=0.04479, over 3714874.43 frames. ], batch size: 78, lr: 3.92e-03, grad_scale: 8.0
2022-12-24 05:07:14,022 WARNING [train.py:1060] (2/4) Exclude cut with ID 8291-282929-0024-13011-0 from training. Duration: 23.795
2022-12-24 05:07:30,376 WARNING [train.py:1060] (2/4) Exclude cut with ID 8631-249866-0030-130156-0_sp1.1 from training. Duration: 21.5409375
2022-12-24 05:07:33,498 WARNING [train.py:1060] (2/4) Exclude cut with ID 6978-92210-0019-146985-0_sp0.9 from training. Duration: 24.97775
2022-12-24 05:07:33,888 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([2.3008, 1.9778, 1.5108, 1.9877, 1.8982, 1.8903, 1.8031, 2.2204],
       device='cuda:2'), covar=tensor([0.2494, 0.3241, 0.2580, 0.2859, 0.3683, 0.1439, 0.3531, 0.1106],
       device='cuda:2'), in_proj_covar=tensor([0.0303, 0.0303, 0.0257, 0.0352, 0.0285, 0.0237, 0.0301, 0.0225],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0003, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-24 05:07:43,261 INFO [zipformer.py:660] (2/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=100942.0, num_to_drop=0, layers_to_drop=set()
2022-12-24 05:07:44,793 WARNING [train.py:1060] (2/4) Exclude cut with ID 1085-156170-0017-128270-0_sp0.9 from training. Duration: 23.3444375
2022-12-24 05:07:57,584 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 2.603e+02 4.059e+02 4.995e+02 5.667e+02 1.400e+03, threshold=9.990e+02, percent-clipped=2.0
2022-12-24 05:08:13,953 WARNING [train.py:1060] (2/4) Exclude cut with ID 6010-56788-0055-90261-0_sp0.9 from training. Duration: 23.2
2022-12-24 05:08:20,088 WARNING [train.py:1060] (2/4) Exclude cut with ID 5653-46179-0060-117930-0_sp0.9 from training. Duration: 21.17225
2022-12-24 05:08:28,844 INFO [train.py:894] (2/4) Epoch 29, batch 2800, loss[loss=0.1933, simple_loss=0.2782, pruned_loss=0.05415, over 18666.00 frames. ], tot_loss[loss=0.1735, simple_loss=0.2577, pruned_loss=0.04462, over 3715840.23 frames. ], batch size: 69, lr: 3.92e-03, grad_scale: 8.0
2022-12-24 05:08:38,901 WARNING [train.py:1060] (2/4) Exclude cut with ID 7255-291500-0008-134307-0_sp0.9 from training. Duration: 24.6555625
2022-12-24 05:09:13,646 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.6654, 1.3783, 2.0462, 3.3013, 2.5155, 2.6224, 0.7738, 2.4451],
       device='cuda:2'), covar=tensor([0.1916, 0.1663, 0.1473, 0.0681, 0.1019, 0.1246, 0.2234, 0.0964],
       device='cuda:2'), in_proj_covar=tensor([0.0105, 0.0120, 0.0140, 0.0160, 0.0109, 0.0146, 0.0130, 0.0117],
       device='cuda:2'), out_proj_covar=tensor([0.0003, 0.0003, 0.0004, 0.0004, 0.0003, 0.0004, 0.0004, 0.0003],
       device='cuda:2')
2022-12-24 05:09:34,866 WARNING [train.py:1060] (2/4) Exclude cut with ID 3488-65654-0031-41259-0 from training. Duration: 20.44
2022-12-24 05:09:45,294 INFO [train.py:894] (2/4) Epoch 29, batch 2850, loss[loss=0.1659, simple_loss=0.2583, pruned_loss=0.03675, over 18714.00 frames. ], tot_loss[loss=0.1735, simple_loss=0.2579, pruned_loss=0.04458, over 3715837.48 frames. ], batch size: 54, lr: 3.92e-03, grad_scale: 8.0
2022-12-24 05:09:46,369 INFO [scaling.py:679] (2/4) Whitening: num_groups=1, num_channels=384, metric=4.72 vs. limit=5.0
2022-12-24 05:09:49,968 WARNING [train.py:1060] (2/4) Exclude cut with ID 6951-79737-0018-132285-0_sp0.9 from training. Duration: 23.45
2022-12-24 05:10:15,297 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.5887, 1.9881, 1.6561, 2.2617, 2.5780, 1.7028, 1.5281, 1.3789],
       device='cuda:2'), covar=tensor([0.1956, 0.1728, 0.1611, 0.1041, 0.1173, 0.1059, 0.2175, 0.1589],
       device='cuda:2'), in_proj_covar=tensor([0.0251, 0.0234, 0.0223, 0.0205, 0.0265, 0.0199, 0.0229, 0.0205],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-24 05:10:19,355 WARNING [train.py:1060] (2/4) Exclude cut with ID 6945-60535-0076-12784-0_sp0.9 from training. Duration: 20.52225
2022-12-24 05:10:25,193 WARNING [train.py:1060] (2/4) Exclude cut with ID 7255-291500-0008-134307-0 from training. Duration: 22.19
2022-12-24 05:10:31,126 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 2.944e+02 4.277e+02 5.213e+02 6.121e+02 9.835e+02, threshold=1.043e+03, percent-clipped=0.0
2022-12-24 05:10:35,722 WARNING [train.py:1060] (2/4) Exclude cut with ID 7357-94126-0014-15845-0_sp1.1 from training. Duration: 25.3818125
2022-12-24 05:10:41,805 INFO [zipformer.py:660] (2/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=101058.0, num_to_drop=0, layers_to_drop=set()
2022-12-24 05:10:52,140 WARNING [train.py:1060] (2/4) Exclude cut with ID 6951-79737-0043-132310-0_sp0.9 from training. Duration: 28.0944375
2022-12-24 05:10:59,575 WARNING [train.py:1060] (2/4) Exclude cut with ID 2195-150901-0045-59933-0_sp0.9 from training. Duration: 22.9444375
2022-12-24 05:11:02,463 INFO [train.py:894] (2/4) Epoch 29, batch 2900, loss[loss=0.1564, simple_loss=0.2503, pruned_loss=0.03122, over 18454.00 frames. ], tot_loss[loss=0.1728, simple_loss=0.2573, pruned_loss=0.04409, over 3715158.64 frames. ], batch size: 54, lr: 3.91e-03, grad_scale: 8.0
2022-12-24 05:11:03,171 INFO [scaling.py:679] (2/4) Whitening: num_groups=8, num_channels=96, metric=1.14 vs. limit=2.0
2022-12-24 05:11:09,284 WARNING [train.py:1060] (2/4) Exclude cut with ID 8291-282929-0024-13011-0_sp1.1 from training. Duration: 21.6318125
2022-12-24 05:11:20,915 INFO [scaling.py:679] (2/4) Whitening: num_groups=8, num_channels=96, metric=1.18 vs. limit=2.0
2022-12-24 05:11:24,492 INFO [zipformer.py:660] (2/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=101086.0, num_to_drop=1, layers_to_drop={3}
2022-12-24 05:11:27,349 WARNING [train.py:1060] (2/4) Exclude cut with ID 8631-249866-0030-130156-0 from training. Duration: 23.695
2022-12-24 05:11:49,567 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.3171, 1.5344, 1.2588, 1.7812, 1.7597, 1.4446, 1.0333, 1.2598],
       device='cuda:2'), covar=tensor([0.2039, 0.1885, 0.1816, 0.1223, 0.1209, 0.1159, 0.2339, 0.1661],
       device='cuda:2'), in_proj_covar=tensor([0.0252, 0.0235, 0.0225, 0.0206, 0.0266, 0.0200, 0.0230, 0.0206],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-24 05:11:50,546 WARNING [train.py:1060] (2/4) Exclude cut with ID 7699-105389-0094-26379-0 from training. Duration: 23.955
2022-12-24 05:11:53,538 INFO [zipformer.py:660] (2/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=101106.0, num_to_drop=0, layers_to_drop=set()
2022-12-24 05:11:55,269 INFO [zipformer.py:660] (2/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=101107.0, num_to_drop=0, layers_to_drop=set()
2022-12-24 05:12:17,126 INFO [train.py:894] (2/4) Epoch 29, batch 2950, loss[loss=0.1573, simple_loss=0.2422, pruned_loss=0.03618, over 18672.00 frames. ], tot_loss[loss=0.1724, simple_loss=0.257, pruned_loss=0.04388, over 3715094.36 frames. ], batch size: 48, lr: 3.91e-03, grad_scale: 8.0
2022-12-24 05:12:24,843 WARNING [train.py:1060] (2/4) Exclude cut with ID 8291-282929-0024-13011-0_sp0.9 from training. Duration: 26.438875
2022-12-24 05:12:26,372 INFO [scaling.py:679] (2/4) Whitening: num_groups=8, num_channels=96, metric=1.38 vs. limit=2.0
2022-12-24 05:12:51,427 INFO [scaling.py:679] (2/4) Whitening: num_groups=8, num_channels=96, metric=1.22 vs. limit=2.0
2022-12-24 05:13:00,819 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 2.294e+02 3.879e+02 5.080e+02 6.075e+02 1.162e+03, threshold=1.016e+03, percent-clipped=1.0
2022-12-24 05:13:03,834 WARNING [train.py:1060] (2/4) Exclude cut with ID 7699-105389-0021-26306-0_sp0.9 from training. Duration: 21.2444375
2022-12-24 05:13:05,414 WARNING [train.py:1060] (2/4) Exclude cut with ID 7357-94126-0014-15845-0_sp0.9 from training. Duration: 31.02225
2022-12-24 05:13:14,393 WARNING [train.py:1060] (2/4) Exclude cut with ID 432-122774-0017-62487-0 from training. Duration: 22.395
2022-12-24 05:13:19,975 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.8390, 1.4112, 1.7877, 1.9631, 1.7191, 3.5824, 1.6166, 1.6088],
       device='cuda:2'), covar=tensor([0.0839, 0.1960, 0.0983, 0.0925, 0.1502, 0.0251, 0.1425, 0.1617],
       device='cuda:2'), in_proj_covar=tensor([0.0074, 0.0083, 0.0073, 0.0075, 0.0092, 0.0077, 0.0086, 0.0079],
       device='cuda:2'), out_proj_covar=tensor([0.0004, 0.0004, 0.0004, 0.0004, 0.0005, 0.0004, 0.0004, 0.0004],
       device='cuda:2')
2022-12-24 05:13:28,297 INFO [zipformer.py:660] (2/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=101168.0, num_to_drop=0, layers_to_drop=set()
2022-12-24 05:13:30,943 WARNING [train.py:1060] (2/4) Exclude cut with ID 7357-94126-0045-15876-0 from training. Duration: 21.075
2022-12-24 05:13:34,504 INFO [train.py:894] (2/4) Epoch 29, batch 3000, loss[loss=0.1775, simple_loss=0.2692, pruned_loss=0.04294, over 18502.00 frames. ], tot_loss[loss=0.172, simple_loss=0.2566, pruned_loss=0.04373, over 3714761.44 frames. ], batch size: 52, lr: 3.91e-03, grad_scale: 8.0
2022-12-24 05:13:34,505 INFO [train.py:919] (2/4) Computing validation loss
2022-12-24 05:13:40,914 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([0.7413, 1.9713, 2.0770, 1.2351, 1.5743, 2.1879, 2.1403, 1.7507],
       device='cuda:2'), covar=tensor([0.0880, 0.0360, 0.0366, 0.0450, 0.0416, 0.0523, 0.0272, 0.0769],
       device='cuda:2'), in_proj_covar=tensor([0.0150, 0.0173, 0.0132, 0.0142, 0.0148, 0.0144, 0.0167, 0.0182],
       device='cuda:2'), out_proj_covar=tensor([1.1285e-04, 1.2939e-04, 9.7004e-05, 1.0368e-04, 1.0766e-04, 1.0801e-04,
        1.2597e-04, 1.3655e-04], device='cuda:2')
2022-12-24 05:13:45,185 INFO [train.py:928] (2/4) Epoch 29, validation: loss=0.1637, simple_loss=0.2594, pruned_loss=0.034, over 944034.00 frames. 
2022-12-24 05:13:45,186 INFO [train.py:929] (2/4) Maximum memory allocated so far is 24680MB
2022-12-24 05:13:49,695 WARNING [train.py:1060] (2/4) Exclude cut with ID 6482-98857-0025-147532-0_sp0.9 from training. Duration: 20.0055625
2022-12-24 05:13:49,706 WARNING [train.py:1060] (2/4) Exclude cut with ID 6951-79737-0037-132304-0_sp0.9 from training. Duration: 22.05
2022-12-24 05:13:49,716 WARNING [train.py:1060] (2/4) Exclude cut with ID 7255-291500-0003-134302-0 from training. Duration: 26.8349375
2022-12-24 05:13:52,717 WARNING [train.py:1060] (2/4) Exclude cut with ID 7357-94126-0009-15840-0_sp1.1 from training. Duration: 22.1090625
2022-12-24 05:13:59,701 WARNING [train.py:1060] (2/4) Exclude cut with ID 7699-105389-0094-26379-0_sp0.9 from training. Duration: 26.6166875
2022-12-24 05:14:16,734 WARNING [train.py:1060] (2/4) Exclude cut with ID 2046-178027-0000-53705-0_sp0.9 from training. Duration: 20.3055625
2022-12-24 05:14:40,820 WARNING [train.py:1060] (2/4) Exclude cut with ID 7205-50138-0008-5373-0_sp0.9 from training. Duration: 20.7
2022-12-24 05:14:59,864 INFO [train.py:894] (2/4) Epoch 29, batch 3050, loss[loss=0.1728, simple_loss=0.2647, pruned_loss=0.0405, over 18583.00 frames. ], tot_loss[loss=0.1733, simple_loss=0.2578, pruned_loss=0.04443, over 3714793.78 frames. ], batch size: 69, lr: 3.91e-03, grad_scale: 8.0
2022-12-24 05:15:22,327 WARNING [train.py:1060] (2/4) Exclude cut with ID 6978-92210-0019-146985-0 from training. Duration: 22.48
2022-12-24 05:15:40,329 WARNING [train.py:1060] (2/4) Exclude cut with ID 7255-291500-0003-134302-0_sp0.9 from training. Duration: 29.816625
2022-12-24 05:15:43,406 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 2.815e+02 4.122e+02 4.831e+02 6.036e+02 1.084e+03, threshold=9.661e+02, percent-clipped=3.0
2022-12-24 05:15:45,331 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([2.5293, 1.9617, 2.1096, 2.2275, 2.4209, 2.4767, 2.2491, 2.0822],
       device='cuda:2'), covar=tensor([0.2233, 0.3376, 0.2676, 0.3090, 0.2134, 0.0979, 0.3501, 0.1372],
       device='cuda:2'), in_proj_covar=tensor([0.0274, 0.0303, 0.0290, 0.0331, 0.0323, 0.0262, 0.0358, 0.0252],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-24 05:16:01,393 WARNING [train.py:1060] (2/4) Exclude cut with ID 7255-291500-0005-134304-0_sp1.1 from training. Duration: 22.7590625
2022-12-24 05:16:07,363 WARNING [train.py:1060] (2/4) Exclude cut with ID 7357-94126-0026-15857-0 from training. Duration: 22.555
2022-12-24 05:16:14,397 INFO [train.py:894] (2/4) Epoch 29, batch 3100, loss[loss=0.1617, simple_loss=0.2404, pruned_loss=0.04154, over 18601.00 frames. ], tot_loss[loss=0.1725, simple_loss=0.2568, pruned_loss=0.04411, over 3714338.52 frames. ], batch size: 45, lr: 3.91e-03, grad_scale: 8.0
2022-12-24 05:16:27,282 WARNING [train.py:1060] (2/4) Exclude cut with ID 1250-135782-0005-25975-0_sp0.9 from training. Duration: 21.688875
2022-12-24 05:16:27,927 INFO [scaling.py:679] (2/4) Whitening: num_groups=1, num_channels=384, metric=3.75 vs. limit=5.0
2022-12-24 05:17:00,286 WARNING [train.py:1060] (2/4) Exclude cut with ID 3488-85273-0038-41224-0_sp0.9 from training. Duration: 22.6
2022-12-24 05:17:29,624 INFO [train.py:894] (2/4) Epoch 29, batch 3150, loss[loss=0.1748, simple_loss=0.2619, pruned_loss=0.04388, over 18623.00 frames. ], tot_loss[loss=0.1722, simple_loss=0.2564, pruned_loss=0.04396, over 3714378.82 frames. ], batch size: 53, lr: 3.91e-03, grad_scale: 8.0
2022-12-24 05:17:37,274 WARNING [train.py:1060] (2/4) Exclude cut with ID 7357-94126-0009-15840-0 from training. Duration: 24.32
2022-12-24 05:18:13,841 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 2.460e+02 3.834e+02 4.753e+02 5.855e+02 1.523e+03, threshold=9.507e+02, percent-clipped=2.0
2022-12-24 05:18:38,949 WARNING [train.py:1060] (2/4) Exclude cut with ID 8291-276745-0093-13116-0_sp0.9 from training. Duration: 21.061125
2022-12-24 05:18:44,383 INFO [train.py:894] (2/4) Epoch 29, batch 3200, loss[loss=0.1423, simple_loss=0.2257, pruned_loss=0.02951, over 18390.00 frames. ], tot_loss[loss=0.1725, simple_loss=0.257, pruned_loss=0.04399, over 3714619.52 frames. ], batch size: 46, lr: 3.91e-03, grad_scale: 8.0
2022-12-24 05:18:51,435 WARNING [train.py:1060] (2/4) Exclude cut with ID 7357-94126-0024-15855-0_sp0.9 from training. Duration: 20.32225
2022-12-24 05:19:01,704 WARNING [train.py:1060] (2/4) Exclude cut with ID 3033-130750-0096-55598-0_sp1.1 from training. Duration: 0.7545625
2022-12-24 05:19:04,760 INFO [zipformer.py:660] (2/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=101386.0, num_to_drop=1, layers_to_drop={0}
2022-12-24 05:19:18,691 WARNING [train.py:1060] (2/4) Exclude cut with ID 4295-39940-0007-92567-0_sp0.9 from training. Duration: 23.9333125
2022-12-24 05:19:33,175 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([2.1227, 2.0465, 1.4557, 2.3141, 2.2734, 1.7926, 2.9849, 2.1498],
       device='cuda:2'), covar=tensor([0.1017, 0.1809, 0.3191, 0.1839, 0.1949, 0.1158, 0.0885, 0.1487],
       device='cuda:2'), in_proj_covar=tensor([0.0187, 0.0222, 0.0263, 0.0296, 0.0247, 0.0199, 0.0211, 0.0214],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-24 05:19:45,108 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.9106, 2.3485, 1.7849, 2.5953, 2.9781, 1.7505, 1.9996, 1.4851],
       device='cuda:2'), covar=tensor([0.1937, 0.1682, 0.1651, 0.1005, 0.1378, 0.1153, 0.1997, 0.1622],
       device='cuda:2'), in_proj_covar=tensor([0.0252, 0.0235, 0.0225, 0.0206, 0.0267, 0.0199, 0.0231, 0.0206],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-24 05:19:49,093 WARNING [train.py:1060] (2/4) Exclude cut with ID 7255-291500-0008-134307-0_sp1.1 from training. Duration: 20.17275
2022-12-24 05:19:55,368 WARNING [train.py:1060] (2/4) Exclude cut with ID 6978-92210-0019-146985-0_sp1.1 from training. Duration: 20.436375
2022-12-24 05:19:58,836 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([2.3171, 2.3261, 1.7991, 2.6979, 2.5481, 2.1235, 3.0464, 2.3785],
       device='cuda:2'), covar=tensor([0.0870, 0.1729, 0.2669, 0.1707, 0.1767, 0.0916, 0.0932, 0.1188],
       device='cuda:2'), in_proj_covar=tensor([0.0187, 0.0222, 0.0263, 0.0296, 0.0247, 0.0199, 0.0211, 0.0214],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-24 05:19:59,784 INFO [train.py:894] (2/4) Epoch 29, batch 3250, loss[loss=0.1648, simple_loss=0.2498, pruned_loss=0.03986, over 18448.00 frames. ], tot_loss[loss=0.1721, simple_loss=0.2567, pruned_loss=0.04372, over 3713530.48 frames. ], batch size: 54, lr: 3.91e-03, grad_scale: 8.0
2022-12-24 05:20:19,095 INFO [zipformer.py:660] (2/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=101434.0, num_to_drop=1, layers_to_drop={1}
2022-12-24 05:20:42,416 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.5511, 1.4604, 1.4611, 0.8884, 1.6365, 1.5177, 1.4607, 1.3453],
       device='cuda:2'), covar=tensor([0.0456, 0.0619, 0.0532, 0.0771, 0.0492, 0.0493, 0.0482, 0.1096],
       device='cuda:2'), in_proj_covar=tensor([0.0127, 0.0133, 0.0131, 0.0119, 0.0107, 0.0130, 0.0136, 0.0164],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0001, 0.0001, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-24 05:20:43,368 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 2.471e+02 3.720e+02 4.479e+02 6.030e+02 1.410e+03, threshold=8.958e+02, percent-clipped=2.0
2022-12-24 05:21:01,320 INFO [zipformer.py:660] (2/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=101463.0, num_to_drop=0, layers_to_drop=set()
2022-12-24 05:21:12,861 WARNING [train.py:1060] (2/4) Exclude cut with ID 4234-40345-0022-142709-0_sp0.9 from training. Duration: 23.1055625
2022-12-24 05:21:14,202 INFO [train.py:894] (2/4) Epoch 29, batch 3300, loss[loss=0.1551, simple_loss=0.2427, pruned_loss=0.03371, over 18596.00 frames. ], tot_loss[loss=0.1716, simple_loss=0.2559, pruned_loss=0.0437, over 3713948.59 frames. ], batch size: 45, lr: 3.91e-03, grad_scale: 8.0
2022-12-24 05:21:16,152 WARNING [train.py:1060] (2/4) Exclude cut with ID 8291-282929-0007-12994-0_sp1.1 from training. Duration: 23.5
2022-12-24 05:21:28,345 WARNING [train.py:1060] (2/4) Exclude cut with ID 7255-291500-0009-134308-0_sp0.9 from training. Duration: 26.62775
2022-12-24 05:21:40,954 WARNING [train.py:1060] (2/4) Exclude cut with ID 6951-79737-0018-132285-0 from training. Duration: 21.105
2022-12-24 05:21:44,994 WARNING [train.py:1060] (2/4) Exclude cut with ID 4511-76322-0006-80011-0_sp0.9 from training. Duration: 24.411125
2022-12-24 05:22:12,526 WARNING [train.py:1060] (2/4) Exclude cut with ID 6758-72288-0033-108368-0_sp1.1 from training. Duration: 21.263625
2022-12-24 05:22:25,904 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([2.8213, 2.3162, 1.8870, 2.5278, 2.0903, 2.2633, 2.2051, 2.7214],
       device='cuda:2'), covar=tensor([0.2076, 0.3449, 0.2108, 0.2939, 0.4113, 0.1142, 0.3278, 0.1004],
       device='cuda:2'), in_proj_covar=tensor([0.0304, 0.0303, 0.0258, 0.0353, 0.0286, 0.0239, 0.0302, 0.0227],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0003, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-24 05:22:29,736 INFO [train.py:894] (2/4) Epoch 29, batch 3350, loss[loss=0.1857, simple_loss=0.2712, pruned_loss=0.05017, over 18685.00 frames. ], tot_loss[loss=0.1711, simple_loss=0.2556, pruned_loss=0.04324, over 3713966.15 frames. ], batch size: 62, lr: 3.91e-03, grad_scale: 8.0
2022-12-24 05:22:30,238 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([2.1326, 1.4175, 1.8679, 1.7539, 2.1418, 2.1742, 1.9399, 1.9121],
       device='cuda:2'), covar=tensor([0.2590, 0.3760, 0.3066, 0.3397, 0.2407, 0.1213, 0.3916, 0.1547],
       device='cuda:2'), in_proj_covar=tensor([0.0275, 0.0304, 0.0291, 0.0332, 0.0324, 0.0262, 0.0360, 0.0253],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-24 05:22:43,789 WARNING [train.py:1060] (2/4) Exclude cut with ID 4234-40345-0022-142709-0 from training. Duration: 20.795
2022-12-24 05:22:55,860 WARNING [train.py:1060] (2/4) Exclude cut with ID 7357-94126-0021-15852-0 from training. Duration: 24.76
2022-12-24 05:22:55,878 WARNING [train.py:1060] (2/4) Exclude cut with ID 3867-173237-0077-144769-0_sp0.9 from training. Duration: 22.25
2022-12-24 05:22:57,891 INFO [scaling.py:679] (2/4) Whitening: num_groups=8, num_channels=96, metric=1.31 vs. limit=2.0
2022-12-24 05:23:11,984 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([3.2215, 2.4452, 1.6622, 2.7985, 2.2421, 2.2568, 2.3850, 3.2666],
       device='cuda:2'), covar=tensor([0.2076, 0.3513, 0.2423, 0.3286, 0.4240, 0.1246, 0.3517, 0.0901],
       device='cuda:2'), in_proj_covar=tensor([0.0303, 0.0303, 0.0258, 0.0353, 0.0285, 0.0238, 0.0301, 0.0227],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0003, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-24 05:23:12,765 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 2.686e+02 3.942e+02 4.707e+02 5.878e+02 1.188e+03, threshold=9.415e+02, percent-clipped=4.0
2022-12-24 05:23:20,033 WARNING [train.py:1060] (2/4) Exclude cut with ID 7357-94126-0026-15857-0_sp1.1 from training. Duration: 20.5045625
2022-12-24 05:23:43,806 INFO [train.py:894] (2/4) Epoch 29, batch 3400, loss[loss=0.1607, simple_loss=0.2445, pruned_loss=0.03849, over 18561.00 frames. ], tot_loss[loss=0.1718, simple_loss=0.2562, pruned_loss=0.04368, over 3713337.84 frames. ], batch size: 49, lr: 3.90e-03, grad_scale: 8.0
2022-12-24 05:24:57,356 INFO [train.py:894] (2/4) Epoch 29, batch 3450, loss[loss=0.1739, simple_loss=0.2662, pruned_loss=0.04078, over 18479.00 frames. ], tot_loss[loss=0.1711, simple_loss=0.2554, pruned_loss=0.04341, over 3713487.45 frames. ], batch size: 54, lr: 3.90e-03, grad_scale: 8.0
2022-12-24 05:25:38,891 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 2.398e+02 3.927e+02 4.688e+02 5.596e+02 9.547e+02, threshold=9.377e+02, percent-clipped=1.0
2022-12-24 05:26:10,012 INFO [train.py:894] (2/4) Epoch 29, batch 3500, loss[loss=0.1883, simple_loss=0.2755, pruned_loss=0.05055, over 18653.00 frames. ], tot_loss[loss=0.1718, simple_loss=0.2562, pruned_loss=0.04373, over 3714324.61 frames. ], batch size: 180, lr: 3.90e-03, grad_scale: 8.0
2022-12-24 05:26:30,293 WARNING [train.py:1060] (2/4) Exclude cut with ID 7859-102521-0017-7548-0_sp1.1 from training. Duration: 22.2954375
2022-12-24 05:26:36,017 INFO [train.py:894] (2/4) Epoch 30, batch 0, loss[loss=0.1724, simple_loss=0.2635, pruned_loss=0.04067, over 18462.00 frames. ], tot_loss[loss=0.1724, simple_loss=0.2635, pruned_loss=0.04067, over 18462.00 frames. ], batch size: 54, lr: 3.84e-03, grad_scale: 8.0
2022-12-24 05:26:36,017 INFO [train.py:919] (2/4) Computing validation loss
2022-12-24 05:26:44,340 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([2.0492, 1.6357, 1.9467, 2.3419, 2.1030, 3.7145, 1.8413, 1.7659],
       device='cuda:2'), covar=tensor([0.0853, 0.1967, 0.1054, 0.0890, 0.1438, 0.0298, 0.1422, 0.1621],
       device='cuda:2'), in_proj_covar=tensor([0.0074, 0.0083, 0.0073, 0.0075, 0.0092, 0.0077, 0.0086, 0.0078],
       device='cuda:2'), out_proj_covar=tensor([0.0004, 0.0004, 0.0004, 0.0004, 0.0005, 0.0004, 0.0004, 0.0004],
       device='cuda:2')
2022-12-24 05:26:46,582 INFO [train.py:928] (2/4) Epoch 30, validation: loss=0.1625, simple_loss=0.2589, pruned_loss=0.03306, over 944034.00 frames. 
2022-12-24 05:26:46,583 INFO [train.py:929] (2/4) Maximum memory allocated so far is 24680MB
2022-12-24 05:26:55,648 INFO [zipformer.py:660] (2/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=101684.0, num_to_drop=0, layers_to_drop=set()
2022-12-24 05:27:35,076 WARNING [train.py:1060] (2/4) Exclude cut with ID 298-126791-0067-24026-0_sp0.9 from training. Duration: 21.438875
2022-12-24 05:27:39,834 WARNING [train.py:1060] (2/4) Exclude cut with ID 5652-39938-0025-23684-0_sp0.9 from training. Duration: 22.2055625
2022-12-24 05:27:54,900 INFO [scaling.py:679] (2/4) Whitening: num_groups=8, num_channels=96, metric=1.29 vs. limit=2.0
2022-12-24 05:28:01,815 INFO [train.py:894] (2/4) Epoch 30, batch 50, loss[loss=0.1726, simple_loss=0.2705, pruned_loss=0.0374, over 18505.00 frames. ], tot_loss[loss=0.1625, simple_loss=0.2528, pruned_loss=0.03609, over 837686.02 frames. ], batch size: 52, lr: 3.83e-03, grad_scale: 8.0
2022-12-24 05:28:23,945 INFO [scaling.py:679] (2/4) Whitening: num_groups=1, num_channels=384, metric=3.32 vs. limit=5.0
2022-12-24 05:28:26,572 INFO [zipformer.py:660] (2/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=101745.0, num_to_drop=0, layers_to_drop=set()
2022-12-24 05:28:29,544 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([2.5895, 2.7762, 3.1209, 1.7270, 3.1705, 3.4043, 2.0993, 3.4356],
       device='cuda:2'), covar=tensor([0.1333, 0.1755, 0.1539, 0.2382, 0.0736, 0.1003, 0.2375, 0.0623],
       device='cuda:2'), in_proj_covar=tensor([0.0199, 0.0217, 0.0211, 0.0197, 0.0172, 0.0220, 0.0218, 0.0197],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-24 05:28:35,183 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 1.974e+02 3.070e+02 3.838e+02 4.860e+02 9.164e+02, threshold=7.675e+02, percent-clipped=0.0
2022-12-24 05:28:52,112 INFO [zipformer.py:660] (2/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=101763.0, num_to_drop=0, layers_to_drop=set()
2022-12-24 05:29:14,269 INFO [train.py:894] (2/4) Epoch 30, batch 100, loss[loss=0.1509, simple_loss=0.2349, pruned_loss=0.03349, over 18600.00 frames. ], tot_loss[loss=0.1643, simple_loss=0.2539, pruned_loss=0.03738, over 1475127.48 frames. ], batch size: 45, lr: 3.83e-03, grad_scale: 8.0
2022-12-24 05:29:26,622 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.7633, 1.7022, 1.8655, 1.6880, 1.3872, 3.8165, 1.6680, 2.1382],
       device='cuda:2'), covar=tensor([0.3211, 0.2084, 0.1894, 0.2091, 0.1450, 0.0168, 0.1622, 0.0882],
       device='cuda:2'), in_proj_covar=tensor([0.0133, 0.0118, 0.0125, 0.0123, 0.0107, 0.0097, 0.0091, 0.0090],
       device='cuda:2'), out_proj_covar=tensor([0.0005, 0.0005, 0.0005, 0.0005, 0.0004, 0.0003, 0.0004, 0.0004],
       device='cuda:2')
2022-12-24 05:30:03,213 INFO [zipformer.py:660] (2/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=101811.0, num_to_drop=0, layers_to_drop=set()
2022-12-24 05:30:30,035 INFO [train.py:894] (2/4) Epoch 30, batch 150, loss[loss=0.1416, simple_loss=0.2269, pruned_loss=0.02813, over 18591.00 frames. ], tot_loss[loss=0.1634, simple_loss=0.253, pruned_loss=0.03686, over 1972510.01 frames. ], batch size: 41, lr: 3.83e-03, grad_scale: 8.0
2022-12-24 05:30:38,817 WARNING [train.py:1060] (2/4) Exclude cut with ID 7859-102521-0017-7548-0 from training. Duration: 24.525
2022-12-24 05:31:03,396 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 2.089e+02 3.245e+02 3.691e+02 4.719e+02 9.898e+02, threshold=7.382e+02, percent-clipped=3.0
2022-12-24 05:31:09,267 WARNING [train.py:1060] (2/4) Exclude cut with ID 3699-47246-0007-3408-0_sp0.9 from training. Duration: 20.26675
2022-12-24 05:31:23,876 WARNING [train.py:1060] (2/4) Exclude cut with ID 7859-102521-0017-7548-0_sp0.9 from training. Duration: 27.25
2022-12-24 05:31:45,372 INFO [train.py:894] (2/4) Epoch 30, batch 200, loss[loss=0.1484, simple_loss=0.2261, pruned_loss=0.03532, over 18627.00 frames. ], tot_loss[loss=0.1627, simple_loss=0.2521, pruned_loss=0.03662, over 2359403.46 frames. ], batch size: 41, lr: 3.83e-03, grad_scale: 8.0
2022-12-24 05:32:21,989 INFO [scaling.py:679] (2/4) Whitening: num_groups=8, num_channels=96, metric=1.30 vs. limit=2.0
2022-12-24 05:32:35,390 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.3748, 1.5411, 1.3861, 1.8544, 1.7368, 1.4986, 0.9665, 1.3347],
       device='cuda:2'), covar=tensor([0.1944, 0.1898, 0.1674, 0.1080, 0.1185, 0.1123, 0.2405, 0.1531],
       device='cuda:2'), in_proj_covar=tensor([0.0250, 0.0234, 0.0224, 0.0205, 0.0265, 0.0199, 0.0230, 0.0205],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-24 05:32:37,775 WARNING [train.py:1060] (2/4) Exclude cut with ID 6426-64292-0017-15984-0 from training. Duration: 21.68
2022-12-24 05:32:47,023 INFO [zipformer.py:660] (2/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=101920.0, num_to_drop=1, layers_to_drop={1}
2022-12-24 05:32:48,285 WARNING [train.py:1060] (2/4) Exclude cut with ID 4278-13270-0007-59342-0 from training. Duration: 21.6300625
2022-12-24 05:32:58,775 INFO [train.py:894] (2/4) Epoch 30, batch 250, loss[loss=0.1596, simple_loss=0.2566, pruned_loss=0.03132, over 18631.00 frames. ], tot_loss[loss=0.1621, simple_loss=0.2518, pruned_loss=0.03618, over 2659239.30 frames. ], batch size: 53, lr: 3.83e-03, grad_scale: 8.0
2022-12-24 05:33:12,766 WARNING [train.py:1060] (2/4) Exclude cut with ID 4278-13270-0007-59342-0_sp0.9 from training. Duration: 24.033375
2022-12-24 05:33:31,317 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 1.846e+02 3.129e+02 3.860e+02 4.670e+02 7.401e+02, threshold=7.720e+02, percent-clipped=1.0
2022-12-24 05:33:34,978 INFO [scaling.py:679] (2/4) Whitening: num_groups=1, num_channels=384, metric=2.91 vs. limit=5.0
2022-12-24 05:33:41,915 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.5163, 1.3778, 1.4302, 1.3441, 0.8091, 2.3778, 0.7941, 1.3491],
       device='cuda:2'), covar=tensor([0.3259, 0.2262, 0.2106, 0.2308, 0.1786, 0.0312, 0.1942, 0.0975],
       device='cuda:2'), in_proj_covar=tensor([0.0132, 0.0118, 0.0125, 0.0123, 0.0107, 0.0097, 0.0091, 0.0090],
       device='cuda:2'), out_proj_covar=tensor([0.0005, 0.0005, 0.0005, 0.0005, 0.0004, 0.0003, 0.0004, 0.0004],
       device='cuda:2')
2022-12-24 05:34:07,662 WARNING [train.py:1060] (2/4) Exclude cut with ID 4278-13270-0009-59344-0 from training. Duration: 22.905
2022-12-24 05:34:09,910 WARNING [train.py:1060] (2/4) Exclude cut with ID 5622-44585-0006-90525-0_sp1.1 from training. Duration: 23.4318125
2022-12-24 05:34:13,089 INFO [train.py:894] (2/4) Epoch 30, batch 300, loss[loss=0.139, simple_loss=0.2257, pruned_loss=0.02617, over 18397.00 frames. ], tot_loss[loss=0.1616, simple_loss=0.2514, pruned_loss=0.03587, over 2892255.65 frames. ], batch size: 42, lr: 3.83e-03, grad_scale: 8.0
2022-12-24 05:34:17,763 INFO [zipformer.py:660] (2/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=101981.0, num_to_drop=1, layers_to_drop={0}
2022-12-24 05:34:26,722 INFO [zipformer.py:660] (2/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=101987.0, num_to_drop=0, layers_to_drop=set()
2022-12-24 05:34:50,757 INFO [scaling.py:679] (2/4) Whitening: num_groups=8, num_channels=96, metric=1.17 vs. limit=2.0
2022-12-24 05:35:32,120 INFO [train.py:894] (2/4) Epoch 30, batch 350, loss[loss=0.1645, simple_loss=0.2468, pruned_loss=0.04108, over 18657.00 frames. ], tot_loss[loss=0.1619, simple_loss=0.2517, pruned_loss=0.03603, over 3074619.80 frames. ], batch size: 48, lr: 3.83e-03, grad_scale: 8.0
2022-12-24 05:35:49,613 INFO [zipformer.py:660] (2/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=102040.0, num_to_drop=0, layers_to_drop=set()
2022-12-24 05:36:01,102 INFO [zipformer.py:660] (2/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=102048.0, num_to_drop=0, layers_to_drop=set()
2022-12-24 05:36:05,011 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 1.772e+02 3.223e+02 3.909e+02 4.619e+02 7.719e+02, threshold=7.817e+02, percent-clipped=0.0
2022-12-24 05:36:09,508 WARNING [train.py:1060] (2/4) Exclude cut with ID 4278-13270-0009-59344-0_sp1.1 from training. Duration: 20.82275
2022-12-24 05:36:11,153 WARNING [train.py:1060] (2/4) Exclude cut with ID 4278-13270-0009-59344-0_sp0.9 from training. Duration: 25.45
2022-12-24 05:36:33,025 INFO [scaling.py:679] (2/4) Whitening: num_groups=8, num_channels=96, metric=1.15 vs. limit=2.0
2022-12-24 05:36:45,861 INFO [train.py:894] (2/4) Epoch 30, batch 400, loss[loss=0.1717, simple_loss=0.2657, pruned_loss=0.03891, over 18464.00 frames. ], tot_loss[loss=0.1626, simple_loss=0.2524, pruned_loss=0.0364, over 3215795.54 frames. ], batch size: 54, lr: 3.83e-03, grad_scale: 8.0
2022-12-24 05:37:06,319 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([4.5973, 3.8622, 3.9942, 4.5285, 4.2125, 4.0140, 4.7499, 1.3296],
       device='cuda:2'), covar=tensor([0.0675, 0.0781, 0.0649, 0.0765, 0.1233, 0.1110, 0.0547, 0.5287],
       device='cuda:2'), in_proj_covar=tensor([0.0364, 0.0238, 0.0251, 0.0286, 0.0341, 0.0278, 0.0305, 0.0296],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0001, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-24 05:37:13,382 WARNING [train.py:1060] (2/4) Exclude cut with ID 5622-44585-0006-90525-0 from training. Duration: 25.775
2022-12-24 05:37:26,642 INFO [scaling.py:679] (2/4) Whitening: num_groups=8, num_channels=96, metric=1.33 vs. limit=2.0
2022-12-24 05:37:34,907 WARNING [train.py:1060] (2/4) Exclude cut with ID 2929-85685-0071-62375-0_sp0.9 from training. Duration: 22.25
2022-12-24 05:38:01,026 INFO [train.py:894] (2/4) Epoch 30, batch 450, loss[loss=0.1578, simple_loss=0.2609, pruned_loss=0.02735, over 18712.00 frames. ], tot_loss[loss=0.1629, simple_loss=0.2531, pruned_loss=0.03636, over 3325784.39 frames. ], batch size: 54, lr: 3.83e-03, grad_scale: 16.0
2022-12-24 05:38:01,101 WARNING [train.py:1060] (2/4) Exclude cut with ID 3972-170212-0014-23379-0 from training. Duration: 26.205
2022-12-24 05:38:18,720 WARNING [train.py:1060] (2/4) Exclude cut with ID 5239-32139-0047-9341-0_sp0.9 from training. Duration: 30.1555625
2022-12-24 05:38:20,574 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([2.3780, 2.7296, 3.0126, 1.5947, 2.9208, 2.9695, 2.2219, 3.2029],
       device='cuda:2'), covar=tensor([0.1332, 0.1675, 0.1480, 0.2322, 0.0727, 0.1257, 0.2156, 0.0612],
       device='cuda:2'), in_proj_covar=tensor([0.0197, 0.0216, 0.0209, 0.0195, 0.0171, 0.0219, 0.0217, 0.0195],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-24 05:38:24,318 WARNING [train.py:1060] (2/4) Exclude cut with ID 1265-135635-0050-6781-0_sp0.9 from training. Duration: 21.8333125
2022-12-24 05:38:34,419 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 2.124e+02 3.155e+02 3.798e+02 4.778e+02 1.868e+03, threshold=7.596e+02, percent-clipped=6.0
2022-12-24 05:38:34,467 WARNING [train.py:1060] (2/4) Exclude cut with ID 1914-133440-0024-94914-0_sp1.1 from training. Duration: 20.6545625
2022-12-24 05:39:15,304 INFO [train.py:894] (2/4) Epoch 30, batch 500, loss[loss=0.1623, simple_loss=0.2538, pruned_loss=0.03543, over 18714.00 frames. ], tot_loss[loss=0.1635, simple_loss=0.2536, pruned_loss=0.0367, over 3412487.52 frames. ], batch size: 60, lr: 3.83e-03, grad_scale: 16.0
2022-12-24 05:39:17,017 WARNING [train.py:1060] (2/4) Exclude cut with ID 7395-89880-0045-39920-0_sp0.9 from training. Duration: 20.52225
2022-12-24 05:39:35,488 WARNING [train.py:1060] (2/4) Exclude cut with ID 3972-170212-0014-23379-0_sp0.9 from training. Duration: 29.1166875
2022-12-24 05:40:31,532 INFO [train.py:894] (2/4) Epoch 30, batch 550, loss[loss=0.1748, simple_loss=0.269, pruned_loss=0.04032, over 18624.00 frames. ], tot_loss[loss=0.1644, simple_loss=0.2546, pruned_loss=0.03707, over 3478939.09 frames. ], batch size: 53, lr: 3.83e-03, grad_scale: 16.0
2022-12-24 05:40:34,941 WARNING [train.py:1060] (2/4) Exclude cut with ID 543-133211-0007-59831-0_sp0.9 from training. Duration: 21.388875
2022-12-24 05:41:05,076 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 2.220e+02 3.027e+02 3.739e+02 4.714e+02 8.555e+02, threshold=7.479e+02, percent-clipped=1.0
2022-12-24 05:41:12,789 WARNING [train.py:1060] (2/4) Exclude cut with ID 1914-133440-0024-94914-0 from training. Duration: 22.72
2022-12-24 05:41:14,763 WARNING [train.py:1060] (2/4) Exclude cut with ID 1914-133440-0031-94921-0_sp0.9 from training. Duration: 22.7444375
2022-12-24 05:41:39,509 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.5272, 1.3813, 1.4355, 1.3699, 0.9343, 2.3333, 0.8721, 1.3384],
       device='cuda:2'), covar=tensor([0.3217, 0.2311, 0.2103, 0.2246, 0.1610, 0.0326, 0.1840, 0.0988],
       device='cuda:2'), in_proj_covar=tensor([0.0133, 0.0119, 0.0125, 0.0123, 0.0108, 0.0097, 0.0091, 0.0090],
       device='cuda:2'), out_proj_covar=tensor([0.0005, 0.0005, 0.0005, 0.0005, 0.0004, 0.0003, 0.0004, 0.0004],
       device='cuda:2')
2022-12-24 05:41:43,490 INFO [zipformer.py:660] (2/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=102276.0, num_to_drop=1, layers_to_drop={3}
2022-12-24 05:41:46,132 INFO [train.py:894] (2/4) Epoch 30, batch 600, loss[loss=0.1615, simple_loss=0.2477, pruned_loss=0.03766, over 18397.00 frames. ], tot_loss[loss=0.1653, simple_loss=0.2554, pruned_loss=0.03756, over 3530432.96 frames. ], batch size: 46, lr: 3.82e-03, grad_scale: 16.0
2022-12-24 05:41:56,095 INFO [scaling.py:679] (2/4) Whitening: num_groups=8, num_channels=192, metric=1.66 vs. limit=2.0
2022-12-24 05:41:58,155 WARNING [train.py:1060] (2/4) Exclude cut with ID 4133-6541-0027-40495-0_sp1.1 from training. Duration: 0.9681875
2022-12-24 05:42:00,990 WARNING [train.py:1060] (2/4) Exclude cut with ID 6330-62851-0022-91297-0_sp0.9 from training. Duration: 22.3166875
2022-12-24 05:42:06,728 WARNING [train.py:1060] (2/4) Exclude cut with ID 543-133212-0015-59917-0_sp0.9 from training. Duration: 21.8166875
2022-12-24 05:43:02,698 INFO [train.py:894] (2/4) Epoch 30, batch 650, loss[loss=0.1655, simple_loss=0.2653, pruned_loss=0.03283, over 18477.00 frames. ], tot_loss[loss=0.1644, simple_loss=0.2546, pruned_loss=0.03712, over 3571278.77 frames. ], batch size: 54, lr: 3.82e-03, grad_scale: 8.0
2022-12-24 05:43:20,994 INFO [zipformer.py:660] (2/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=102340.0, num_to_drop=0, layers_to_drop=set()
2022-12-24 05:43:25,231 INFO [zipformer.py:660] (2/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=102343.0, num_to_drop=0, layers_to_drop=set()
2022-12-24 05:43:38,572 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 2.067e+02 2.967e+02 3.701e+02 4.563e+02 8.252e+02, threshold=7.403e+02, percent-clipped=2.0
2022-12-24 05:43:48,916 WARNING [train.py:1060] (2/4) Exclude cut with ID 4957-30119-0041-23990-0_sp0.9 from training. Duration: 20.22775
2022-12-24 05:44:16,267 INFO [train.py:894] (2/4) Epoch 30, batch 700, loss[loss=0.1881, simple_loss=0.2919, pruned_loss=0.04209, over 18705.00 frames. ], tot_loss[loss=0.1656, simple_loss=0.256, pruned_loss=0.03766, over 3602776.39 frames. ], batch size: 62, lr: 3.82e-03, grad_scale: 8.0
2022-12-24 05:44:31,282 INFO [zipformer.py:660] (2/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=102388.0, num_to_drop=0, layers_to_drop=set()
2022-12-24 05:44:32,503 WARNING [train.py:1060] (2/4) Exclude cut with ID 5239-32139-0047-9341-0_sp1.1 from training. Duration: 24.67275
2022-12-24 05:45:01,249 WARNING [train.py:1060] (2/4) Exclude cut with ID 3082-165428-0081-50734-0_sp0.9 from training. Duration: 21.8055625
2022-12-24 05:45:32,303 INFO [train.py:894] (2/4) Epoch 30, batch 750, loss[loss=0.1515, simple_loss=0.2418, pruned_loss=0.03065, over 18688.00 frames. ], tot_loss[loss=0.1659, simple_loss=0.2563, pruned_loss=0.03779, over 3627724.06 frames. ], batch size: 50, lr: 3.82e-03, grad_scale: 8.0
2022-12-24 05:45:36,660 WARNING [train.py:1060] (2/4) Exclude cut with ID 3340-169293-0054-76830-0_sp0.9 from training. Duration: 22.6666875
2022-12-24 05:46:07,509 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 2.493e+02 3.194e+02 3.640e+02 4.304e+02 1.220e+03, threshold=7.280e+02, percent-clipped=2.0
2022-12-24 05:46:38,451 WARNING [train.py:1060] (2/4) Exclude cut with ID 2411-132532-0017-82279-0_sp1.1 from training. Duration: 0.9681875
2022-12-24 05:46:47,413 INFO [train.py:894] (2/4) Epoch 30, batch 800, loss[loss=0.1959, simple_loss=0.2861, pruned_loss=0.05288, over 18471.00 frames. ], tot_loss[loss=0.1666, simple_loss=0.2571, pruned_loss=0.03805, over 3646952.99 frames. ], batch size: 64, lr: 3.82e-03, grad_scale: 8.0
2022-12-24 05:47:04,464 WARNING [train.py:1060] (2/4) Exclude cut with ID 6330-62850-0007-91323-0 from training. Duration: 22.485
2022-12-24 05:47:20,152 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.8158, 1.6839, 1.8126, 1.7597, 1.2795, 3.7256, 1.6295, 2.1595],
       device='cuda:2'), covar=tensor([0.3083, 0.2092, 0.1906, 0.2038, 0.1514, 0.0171, 0.1581, 0.0782],
       device='cuda:2'), in_proj_covar=tensor([0.0132, 0.0118, 0.0124, 0.0123, 0.0107, 0.0096, 0.0091, 0.0090],
       device='cuda:2'), out_proj_covar=tensor([0.0005, 0.0005, 0.0005, 0.0005, 0.0004, 0.0003, 0.0004, 0.0004],
       device='cuda:2')
2022-12-24 05:47:23,457 INFO [scaling.py:679] (2/4) Whitening: num_groups=8, num_channels=192, metric=1.91 vs. limit=2.0
2022-12-24 05:47:35,085 INFO [scaling.py:679] (2/4) Whitening: num_groups=8, num_channels=96, metric=1.17 vs. limit=2.0
2022-12-24 05:47:44,898 WARNING [train.py:1060] (2/4) Exclude cut with ID 3972-170212-0014-23379-0_sp1.1 from training. Duration: 23.82275
2022-12-24 05:47:58,234 WARNING [train.py:1060] (2/4) Exclude cut with ID 4860-13185-0032-76709-0 from training. Duration: 20.77
2022-12-24 05:48:00,507 INFO [train.py:894] (2/4) Epoch 30, batch 850, loss[loss=0.1568, simple_loss=0.2446, pruned_loss=0.03447, over 18718.00 frames. ], tot_loss[loss=0.166, simple_loss=0.2562, pruned_loss=0.03788, over 3662010.43 frames. ], batch size: 50, lr: 3.82e-03, grad_scale: 8.0
2022-12-24 05:48:05,874 WARNING [train.py:1060] (2/4) Exclude cut with ID 6426-64292-0017-15984-0_sp0.9 from training. Duration: 24.088875
2022-12-24 05:48:34,400 WARNING [train.py:1060] (2/4) Exclude cut with ID 6330-62850-0007-91323-0_sp1.1 from training. Duration: 20.4409375
2022-12-24 05:48:35,719 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 1.988e+02 3.397e+02 3.948e+02 4.845e+02 7.696e+02, threshold=7.896e+02, percent-clipped=1.0
2022-12-24 05:49:11,983 INFO [zipformer.py:660] (2/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=102576.0, num_to_drop=1, layers_to_drop={2}
2022-12-24 05:49:14,488 INFO [train.py:894] (2/4) Epoch 30, batch 900, loss[loss=0.1591, simple_loss=0.2552, pruned_loss=0.03148, over 18648.00 frames. ], tot_loss[loss=0.1651, simple_loss=0.2551, pruned_loss=0.03752, over 3673609.89 frames. ], batch size: 78, lr: 3.82e-03, grad_scale: 8.0
2022-12-24 05:49:37,351 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.3784, 2.8653, 3.1765, 1.2318, 2.7203, 3.5731, 2.7230, 2.6973],
       device='cuda:2'), covar=tensor([0.0851, 0.0417, 0.0366, 0.0597, 0.0441, 0.0407, 0.0402, 0.0747],
       device='cuda:2'), in_proj_covar=tensor([0.0151, 0.0174, 0.0132, 0.0143, 0.0149, 0.0145, 0.0168, 0.0183],
       device='cuda:2'), out_proj_covar=tensor([1.1377e-04, 1.3024e-04, 9.7106e-05, 1.0442e-04, 1.0842e-04, 1.0865e-04,
        1.2646e-04, 1.3717e-04], device='cuda:2')
2022-12-24 05:49:48,898 WARNING [train.py:1060] (2/4) Exclude cut with ID 2929-85685-0062-62366-0_sp0.9 from training. Duration: 22.511125
2022-12-24 05:49:48,912 WARNING [train.py:1060] (2/4) Exclude cut with ID 7395-89880-0031-39906-0 from training. Duration: 20.675
2022-12-24 05:50:23,373 INFO [zipformer.py:660] (2/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=102624.0, num_to_drop=1, layers_to_drop={0}
2022-12-24 05:50:28,617 INFO [train.py:894] (2/4) Epoch 30, batch 950, loss[loss=0.1684, simple_loss=0.2596, pruned_loss=0.03863, over 18604.00 frames. ], tot_loss[loss=0.166, simple_loss=0.2559, pruned_loss=0.03803, over 3682777.96 frames. ], batch size: 51, lr: 3.82e-03, grad_scale: 8.0
2022-12-24 05:50:51,146 INFO [zipformer.py:660] (2/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=102643.0, num_to_drop=0, layers_to_drop=set()
2022-12-24 05:51:03,633 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 1.806e+02 3.148e+02 4.122e+02 4.963e+02 1.598e+03, threshold=8.245e+02, percent-clipped=4.0
2022-12-24 05:51:27,754 WARNING [train.py:1060] (2/4) Exclude cut with ID 6330-62850-0007-91323-0_sp0.9 from training. Duration: 24.9833125
2022-12-24 05:51:42,750 INFO [train.py:894] (2/4) Epoch 30, batch 1000, loss[loss=0.1691, simple_loss=0.268, pruned_loss=0.03505, over 18588.00 frames. ], tot_loss[loss=0.1663, simple_loss=0.2564, pruned_loss=0.03812, over 3689949.48 frames. ], batch size: 78, lr: 3.82e-03, grad_scale: 8.0
2022-12-24 05:51:59,571 WARNING [train.py:1060] (2/4) Exclude cut with ID 5239-32139-0047-9341-0 from training. Duration: 27.14
2022-12-24 05:52:01,635 INFO [zipformer.py:660] (2/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=102691.0, num_to_drop=0, layers_to_drop=set()
2022-12-24 05:52:13,683 WARNING [train.py:1060] (2/4) Exclude cut with ID 2929-85685-0044-62348-0 from training. Duration: 22.44
2022-12-24 05:52:26,976 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([2.8188, 4.1972, 3.9931, 1.8818, 4.3041, 3.3498, 0.8469, 2.7375],
       device='cuda:2'), covar=tensor([0.1933, 0.1030, 0.1282, 0.3404, 0.0693, 0.0807, 0.4947, 0.1443],
       device='cuda:2'), in_proj_covar=tensor([0.0153, 0.0151, 0.0162, 0.0126, 0.0153, 0.0116, 0.0146, 0.0118],
       device='cuda:2'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0002, 0.0003, 0.0002, 0.0003, 0.0002],
       device='cuda:2')
2022-12-24 05:52:53,001 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.7594, 1.6112, 1.7158, 1.9946, 1.8637, 3.6206, 1.5085, 1.6440],
       device='cuda:2'), covar=tensor([0.0820, 0.1755, 0.1013, 0.0902, 0.1402, 0.0220, 0.1444, 0.1596],
       device='cuda:2'), in_proj_covar=tensor([0.0074, 0.0083, 0.0073, 0.0075, 0.0092, 0.0078, 0.0086, 0.0078],
       device='cuda:2'), out_proj_covar=tensor([0.0004, 0.0004, 0.0004, 0.0004, 0.0005, 0.0004, 0.0004, 0.0004],
       device='cuda:2')
2022-12-24 05:52:57,216 INFO [train.py:894] (2/4) Epoch 30, batch 1050, loss[loss=0.1591, simple_loss=0.2545, pruned_loss=0.03189, over 18575.00 frames. ], tot_loss[loss=0.1659, simple_loss=0.256, pruned_loss=0.03784, over 3694067.03 frames. ], batch size: 56, lr: 3.82e-03, grad_scale: 8.0
2022-12-24 05:53:03,793 INFO [scaling.py:679] (2/4) Whitening: num_groups=1, num_channels=384, metric=3.05 vs. limit=5.0
2022-12-24 05:53:32,766 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 1.729e+02 3.103e+02 3.715e+02 4.329e+02 7.586e+02, threshold=7.430e+02, percent-clipped=0.0
2022-12-24 05:53:35,403 WARNING [train.py:1060] (2/4) Exclude cut with ID 2929-85685-0060-62364-0_sp0.9 from training. Duration: 21.361125
2022-12-24 05:53:41,376 WARNING [train.py:1060] (2/4) Exclude cut with ID 2929-85685-0079-62383-0_sp1.1 from training. Duration: 27.0318125
2022-12-24 05:53:50,810 WARNING [train.py:1060] (2/4) Exclude cut with ID 5622-44585-0006-90525-0_sp0.9 from training. Duration: 28.638875
2022-12-24 05:54:05,459 WARNING [train.py:1060] (2/4) Exclude cut with ID 3340-169293-0054-76830-0 from training. Duration: 20.4
2022-12-24 05:54:11,562 INFO [train.py:894] (2/4) Epoch 30, batch 1100, loss[loss=0.1432, simple_loss=0.231, pruned_loss=0.02771, over 18611.00 frames. ], tot_loss[loss=0.1657, simple_loss=0.2559, pruned_loss=0.03777, over 3697597.23 frames. ], batch size: 45, lr: 3.82e-03, grad_scale: 8.0
2022-12-24 05:54:38,002 WARNING [train.py:1060] (2/4) Exclude cut with ID 2929-85685-0071-62375-0 from training. Duration: 20.025
2022-12-24 05:54:38,014 WARNING [train.py:1060] (2/4) Exclude cut with ID 2364-131735-0112-64612-0_sp0.9 from training. Duration: 20.488875
2022-12-24 05:54:38,467 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([0.7326, 2.1757, 2.3688, 1.3734, 1.6422, 2.5227, 2.1971, 1.9813],
       device='cuda:2'), covar=tensor([0.0855, 0.0361, 0.0311, 0.0436, 0.0406, 0.0446, 0.0303, 0.0740],
       device='cuda:2'), in_proj_covar=tensor([0.0152, 0.0175, 0.0133, 0.0144, 0.0149, 0.0145, 0.0168, 0.0183],
       device='cuda:2'), out_proj_covar=tensor([1.1400e-04, 1.3077e-04, 9.7484e-05, 1.0494e-04, 1.0842e-04, 1.0887e-04,
        1.2672e-04, 1.3760e-04], device='cuda:2')
2022-12-24 05:54:41,760 INFO [scaling.py:679] (2/4) Whitening: num_groups=8, num_channels=96, metric=1.14 vs. limit=2.0
2022-12-24 05:54:43,784 WARNING [train.py:1060] (2/4) Exclude cut with ID 2929-85685-0079-62383-0 from training. Duration: 29.735
2022-12-24 05:55:05,322 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.5753, 1.2979, 1.9928, 2.9256, 2.1295, 2.4760, 0.8082, 2.1144],
       device='cuda:2'), covar=tensor([0.1861, 0.1695, 0.1346, 0.0637, 0.1012, 0.1192, 0.2260, 0.1014],
       device='cuda:2'), in_proj_covar=tensor([0.0103, 0.0118, 0.0137, 0.0157, 0.0107, 0.0144, 0.0129, 0.0115],
       device='cuda:2'), out_proj_covar=tensor([0.0003, 0.0003, 0.0004, 0.0004, 0.0003, 0.0004, 0.0004, 0.0003],
       device='cuda:2')
2022-12-24 05:55:26,093 INFO [train.py:894] (2/4) Epoch 30, batch 1150, loss[loss=0.178, simple_loss=0.2673, pruned_loss=0.04431, over 18489.00 frames. ], tot_loss[loss=0.1648, simple_loss=0.2549, pruned_loss=0.03741, over 3700474.54 frames. ], batch size: 64, lr: 3.81e-03, grad_scale: 8.0
2022-12-24 05:55:59,390 WARNING [train.py:1060] (2/4) Exclude cut with ID 7276-92427-0014-12983-0_sp0.9 from training. Duration: 21.3055625
2022-12-24 05:56:01,148 WARNING [train.py:1060] (2/4) Exclude cut with ID 1025-75365-0008-79168-0_sp0.9 from training. Duration: 22.0666875
2022-12-24 05:56:01,480 INFO [zipformer.py:660] (2/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=102851.0, num_to_drop=0, layers_to_drop=set()
2022-12-24 05:56:02,557 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 1.993e+02 3.311e+02 3.980e+02 4.684e+02 1.228e+03, threshold=7.959e+02, percent-clipped=4.0
2022-12-24 05:56:06,521 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([2.8087, 2.4006, 1.9249, 0.9455, 2.0373, 2.0414, 1.6717, 2.0627],
       device='cuda:2'), covar=tensor([0.0809, 0.0900, 0.2014, 0.2373, 0.1724, 0.2017, 0.2515, 0.1292],
       device='cuda:2'), in_proj_covar=tensor([0.0176, 0.0189, 0.0210, 0.0190, 0.0212, 0.0206, 0.0219, 0.0205],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-24 05:56:41,192 INFO [train.py:894] (2/4) Epoch 30, batch 1200, loss[loss=0.1554, simple_loss=0.2487, pruned_loss=0.03101, over 18463.00 frames. ], tot_loss[loss=0.1654, simple_loss=0.2554, pruned_loss=0.03768, over 3703108.82 frames. ], batch size: 50, lr: 3.81e-03, grad_scale: 8.0
2022-12-24 05:56:55,347 INFO [zipformer.py:660] (2/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=102887.0, num_to_drop=0, layers_to_drop=set()
2022-12-24 05:57:28,457 INFO [scaling.py:679] (2/4) Whitening: num_groups=8, num_channels=96, metric=1.20 vs. limit=2.0
2022-12-24 05:57:33,142 INFO [zipformer.py:660] (2/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=102912.0, num_to_drop=0, layers_to_drop=set()
2022-12-24 05:57:47,501 WARNING [train.py:1060] (2/4) Exclude cut with ID 2929-85685-0062-62366-0 from training. Duration: 20.26
2022-12-24 05:57:51,454 INFO [scaling.py:679] (2/4) Whitening: num_groups=8, num_channels=96, metric=1.25 vs. limit=2.0
2022-12-24 05:57:55,701 INFO [train.py:894] (2/4) Epoch 30, batch 1250, loss[loss=0.1573, simple_loss=0.2525, pruned_loss=0.03106, over 18394.00 frames. ], tot_loss[loss=0.1651, simple_loss=0.2552, pruned_loss=0.03753, over 3705120.68 frames. ], batch size: 53, lr: 3.81e-03, grad_scale: 8.0
2022-12-24 05:57:56,111 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([2.4998, 2.1400, 2.3747, 1.4217, 2.7488, 2.6140, 2.3510, 1.8216],
       device='cuda:2'), covar=tensor([0.0396, 0.0567, 0.0406, 0.0850, 0.0312, 0.0385, 0.0443, 0.1035],
       device='cuda:2'), in_proj_covar=tensor([0.0126, 0.0133, 0.0129, 0.0118, 0.0106, 0.0128, 0.0134, 0.0162],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0001, 0.0001, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-24 05:58:00,494 WARNING [train.py:1060] (2/4) Exclude cut with ID 5239-32139-0030-9324-0_sp0.9 from training. Duration: 21.3444375
2022-12-24 05:58:26,599 INFO [zipformer.py:660] (2/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=102948.0, num_to_drop=0, layers_to_drop=set()
2022-12-24 05:58:32,823 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 2.010e+02 3.139e+02 3.911e+02 4.609e+02 1.084e+03, threshold=7.821e+02, percent-clipped=2.0
2022-12-24 05:58:55,602 WARNING [train.py:1060] (2/4) Exclude cut with ID 497-129325-0061-62254-0_sp1.1 from training. Duration: 0.97725
2022-12-24 05:59:12,449 INFO [train.py:894] (2/4) Epoch 30, batch 1300, loss[loss=0.1519, simple_loss=0.2523, pruned_loss=0.0258, over 18724.00 frames. ], tot_loss[loss=0.1645, simple_loss=0.2548, pruned_loss=0.03711, over 3706642.45 frames. ], batch size: 54, lr: 3.81e-03, grad_scale: 8.0
2022-12-24 05:59:20,022 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.7236, 2.1803, 1.7346, 2.4160, 3.1155, 1.7650, 1.9018, 1.3629],
       device='cuda:2'), covar=tensor([0.2195, 0.1955, 0.1809, 0.1210, 0.1309, 0.1251, 0.2145, 0.1783],
       device='cuda:2'), in_proj_covar=tensor([0.0252, 0.0234, 0.0225, 0.0206, 0.0265, 0.0201, 0.0230, 0.0206],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-24 05:59:37,126 WARNING [train.py:1060] (2/4) Exclude cut with ID 7395-89880-0031-39906-0_sp0.9 from training. Duration: 22.97225
2022-12-24 06:00:08,733 WARNING [train.py:1060] (2/4) Exclude cut with ID 7395-89880-0047-39922-0_sp0.9 from training. Duration: 21.97775
2022-12-24 06:00:18,130 INFO [scaling.py:679] (2/4) Whitening: num_groups=8, num_channels=192, metric=1.68 vs. limit=2.0
2022-12-24 06:00:24,044 WARNING [train.py:1060] (2/4) Exclude cut with ID 1112-1043-0006-89194-0_sp0.9 from training. Duration: 21.8333125
2022-12-24 06:00:27,021 INFO [train.py:894] (2/4) Epoch 30, batch 1350, loss[loss=0.1577, simple_loss=0.241, pruned_loss=0.03718, over 18591.00 frames. ], tot_loss[loss=0.1644, simple_loss=0.2549, pruned_loss=0.0369, over 3709414.50 frames. ], batch size: 41, lr: 3.81e-03, grad_scale: 8.0
2022-12-24 06:00:34,246 WARNING [train.py:1060] (2/4) Exclude cut with ID 1914-133440-0031-94921-0 from training. Duration: 20.47
2022-12-24 06:01:01,599 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 2.043e+02 3.138e+02 3.681e+02 4.641e+02 8.476e+02, threshold=7.361e+02, percent-clipped=1.0
2022-12-24 06:01:37,459 WARNING [train.py:1060] (2/4) Exclude cut with ID 7395-89880-0037-39912-0_sp0.9 from training. Duration: 20.67225
2022-12-24 06:01:41,954 INFO [train.py:894] (2/4) Epoch 30, batch 1400, loss[loss=0.1511, simple_loss=0.2321, pruned_loss=0.03505, over 18609.00 frames. ], tot_loss[loss=0.1649, simple_loss=0.2555, pruned_loss=0.03715, over 3709676.79 frames. ], batch size: 45, lr: 3.81e-03, grad_scale: 8.0
2022-12-24 06:01:55,129 WARNING [train.py:1060] (2/4) Exclude cut with ID 1914-133440-0024-94914-0_sp0.9 from training. Duration: 25.2444375
2022-12-24 06:02:11,575 INFO [zipformer.py:660] (2/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=103098.0, num_to_drop=0, layers_to_drop=set()
2022-12-24 06:02:20,390 WARNING [train.py:1060] (2/4) Exclude cut with ID 3340-169293-0021-76797-0_sp0.9 from training. Duration: 21.1445
2022-12-24 06:02:55,825 INFO [train.py:894] (2/4) Epoch 30, batch 1450, loss[loss=0.15, simple_loss=0.2353, pruned_loss=0.0324, over 18532.00 frames. ], tot_loss[loss=0.1655, simple_loss=0.2563, pruned_loss=0.03732, over 3709901.00 frames. ], batch size: 47, lr: 3.81e-03, grad_scale: 8.0
2022-12-24 06:03:01,874 INFO [zipformer.py:660] (2/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=103132.0, num_to_drop=0, layers_to_drop=set()
2022-12-24 06:03:31,248 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 1.686e+02 3.036e+02 3.905e+02 4.908e+02 8.550e+02, threshold=7.811e+02, percent-clipped=3.0
2022-12-24 06:03:34,109 WARNING [train.py:1060] (2/4) Exclude cut with ID 2929-85685-0079-62383-0_sp0.9 from training. Duration: 33.038875
2022-12-24 06:03:41,529 INFO [zipformer.py:660] (2/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=103159.0, num_to_drop=0, layers_to_drop=set()
2022-12-24 06:04:09,807 INFO [train.py:894] (2/4) Epoch 30, batch 1500, loss[loss=0.1514, simple_loss=0.2377, pruned_loss=0.0325, over 18574.00 frames. ], tot_loss[loss=0.1653, simple_loss=0.2558, pruned_loss=0.03739, over 3710276.15 frames. ], batch size: 45, lr: 3.81e-03, grad_scale: 8.0
2022-12-24 06:04:14,020 WARNING [train.py:1060] (2/4) Exclude cut with ID 6426-64291-0000-16059-0_sp0.9 from training. Duration: 20.0944375
2022-12-24 06:04:21,730 INFO [scaling.py:679] (2/4) Whitening: num_groups=8, num_channels=96, metric=1.20 vs. limit=2.0
2022-12-24 06:04:26,986 WARNING [train.py:1060] (2/4) Exclude cut with ID 2929-85685-0044-62348-0_sp1.1 from training. Duration: 20.4
2022-12-24 06:04:31,996 INFO [zipformer.py:660] (2/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=103193.0, num_to_drop=0, layers_to_drop=set()
2022-12-24 06:04:34,223 WARNING [train.py:1060] (2/4) Exclude cut with ID 6330-62851-0022-91297-0 from training. Duration: 20.085
2022-12-24 06:04:47,563 WARNING [train.py:1060] (2/4) Exclude cut with ID 4860-13185-0032-76709-0_sp0.9 from training. Duration: 23.07775
2022-12-24 06:04:53,224 INFO [zipformer.py:660] (2/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=103207.0, num_to_drop=0, layers_to_drop=set()
2022-12-24 06:04:54,912 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([2.3841, 2.1231, 1.7817, 1.9885, 1.9142, 2.1664, 1.9390, 2.2155],
       device='cuda:2'), covar=tensor([0.2287, 0.3569, 0.2125, 0.2773, 0.3771, 0.1136, 0.3064, 0.1133],
       device='cuda:2'), in_proj_covar=tensor([0.0303, 0.0305, 0.0258, 0.0352, 0.0284, 0.0238, 0.0301, 0.0227],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0003, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-24 06:04:57,873 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.9073, 0.7494, 1.7704, 1.4877, 1.9598, 1.9998, 1.5488, 1.7892],
       device='cuda:2'), covar=tensor([0.2495, 0.3712, 0.2922, 0.3118, 0.2463, 0.1084, 0.3860, 0.1542],
       device='cuda:2'), in_proj_covar=tensor([0.0273, 0.0301, 0.0290, 0.0330, 0.0323, 0.0261, 0.0359, 0.0252],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-24 06:05:10,190 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([2.6594, 2.4204, 2.2937, 1.4608, 2.8173, 2.8246, 2.4844, 2.1175],
       device='cuda:2'), covar=tensor([0.0373, 0.0452, 0.0441, 0.0797, 0.0314, 0.0358, 0.0410, 0.0764],
       device='cuda:2'), in_proj_covar=tensor([0.0127, 0.0133, 0.0130, 0.0119, 0.0106, 0.0129, 0.0135, 0.0163],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0001, 0.0001, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-24 06:05:24,587 INFO [train.py:894] (2/4) Epoch 30, batch 1550, loss[loss=0.1507, simple_loss=0.2369, pruned_loss=0.03223, over 18396.00 frames. ], tot_loss[loss=0.1654, simple_loss=0.2562, pruned_loss=0.03734, over 3711063.04 frames. ], batch size: 46, lr: 3.81e-03, grad_scale: 8.0
2022-12-24 06:05:31,863 WARNING [train.py:1060] (2/4) Exclude cut with ID 2929-85685-0044-62348-0_sp0.9 from training. Duration: 24.9333125
2022-12-24 06:05:44,233 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([2.0171, 1.1732, 1.7673, 1.6161, 2.0366, 2.1099, 1.7623, 1.8345],
       device='cuda:2'), covar=tensor([0.2505, 0.3682, 0.3009, 0.3199, 0.2531, 0.1152, 0.3866, 0.1561],
       device='cuda:2'), in_proj_covar=tensor([0.0273, 0.0300, 0.0289, 0.0329, 0.0322, 0.0260, 0.0358, 0.0251],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-24 06:05:47,092 INFO [zipformer.py:660] (2/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=103243.0, num_to_drop=0, layers_to_drop=set()
2022-12-24 06:06:00,765 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 1.963e+02 3.194e+02 3.778e+02 4.720e+02 8.597e+02, threshold=7.556e+02, percent-clipped=1.0
2022-12-24 06:06:14,013 WARNING [train.py:1060] (2/4) Exclude cut with ID 5118-111612-0016-124680-0_sp0.9 from training. Duration: 20.388875
2022-12-24 06:06:17,948 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([2.6359, 2.3223, 1.8223, 2.3445, 2.1158, 2.3034, 2.1951, 2.6216],
       device='cuda:2'), covar=tensor([0.2188, 0.3328, 0.1999, 0.3067, 0.3742, 0.1041, 0.3197, 0.1003],
       device='cuda:2'), in_proj_covar=tensor([0.0303, 0.0305, 0.0258, 0.0352, 0.0284, 0.0239, 0.0301, 0.0227],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0003, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-24 06:06:20,290 WARNING [train.py:1060] (2/4) Exclude cut with ID 432-122774-0017-62487-0_sp1.1 from training. Duration: 20.3590625
2022-12-24 06:06:39,750 INFO [train.py:894] (2/4) Epoch 30, batch 1600, loss[loss=0.1976, simple_loss=0.2815, pruned_loss=0.0568, over 18664.00 frames. ], tot_loss[loss=0.1661, simple_loss=0.2569, pruned_loss=0.03762, over 3712270.68 frames. ], batch size: 62, lr: 3.81e-03, grad_scale: 8.0
2022-12-24 06:07:26,567 WARNING [train.py:1060] (2/4) Exclude cut with ID 3557-8342-0013-54691-0_sp1.1 from training. Duration: 0.836375
2022-12-24 06:07:53,612 INFO [train.py:894] (2/4) Epoch 30, batch 1650, loss[loss=0.1545, simple_loss=0.2381, pruned_loss=0.03544, over 18556.00 frames. ], tot_loss[loss=0.166, simple_loss=0.2564, pruned_loss=0.03783, over 3712762.80 frames. ], batch size: 49, lr: 3.81e-03, grad_scale: 8.0
2022-12-24 06:08:07,506 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.4267, 1.1554, 1.5254, 2.2749, 1.6371, 2.4334, 0.8070, 1.8028],
       device='cuda:2'), covar=tensor([0.1725, 0.1609, 0.1268, 0.0745, 0.1062, 0.0757, 0.1835, 0.1060],
       device='cuda:2'), in_proj_covar=tensor([0.0103, 0.0118, 0.0137, 0.0157, 0.0107, 0.0144, 0.0129, 0.0115],
       device='cuda:2'), out_proj_covar=tensor([0.0003, 0.0003, 0.0004, 0.0004, 0.0003, 0.0004, 0.0004, 0.0003],
       device='cuda:2')
2022-12-24 06:08:08,714 WARNING [train.py:1060] (2/4) Exclude cut with ID 8565-290391-0049-67394-0_sp0.9 from training. Duration: 21.3166875
2022-12-24 06:08:21,187 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.9380, 1.4801, 0.7856, 1.3399, 2.2704, 1.1397, 1.7113, 1.7217],
       device='cuda:2'), covar=tensor([0.1535, 0.2023, 0.2286, 0.1587, 0.1709, 0.1888, 0.1387, 0.1745],
       device='cuda:2'), in_proj_covar=tensor([0.0094, 0.0098, 0.0116, 0.0098, 0.0120, 0.0092, 0.0099, 0.0095],
       device='cuda:2'), out_proj_covar=tensor([0.0003, 0.0003, 0.0004, 0.0003, 0.0004, 0.0003, 0.0003, 0.0003],
       device='cuda:2')
2022-12-24 06:08:29,720 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 2.022e+02 3.417e+02 4.120e+02 5.329e+02 9.621e+02, threshold=8.239e+02, percent-clipped=4.0
2022-12-24 06:08:39,702 WARNING [train.py:1060] (2/4) Exclude cut with ID 6533-399-0029-104863-0_sp0.9 from training. Duration: 22.1055625
2022-12-24 06:08:51,005 WARNING [train.py:1060] (2/4) Exclude cut with ID 7699-105389-0094-26379-0_sp1.1 from training. Duration: 21.77725
2022-12-24 06:08:51,335 INFO [zipformer.py:660] (2/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=103366.0, num_to_drop=1, layers_to_drop={0}
2022-12-24 06:09:08,252 INFO [train.py:894] (2/4) Epoch 30, batch 1700, loss[loss=0.1585, simple_loss=0.2511, pruned_loss=0.03297, over 18641.00 frames. ], tot_loss[loss=0.1664, simple_loss=0.2562, pruned_loss=0.03835, over 3712652.99 frames. ], batch size: 69, lr: 3.80e-03, grad_scale: 8.0
2022-12-24 06:09:11,471 WARNING [train.py:1060] (2/4) Exclude cut with ID 7255-291500-0005-134304-0_sp0.9 from training. Duration: 27.8166875
2022-12-24 06:09:35,192 WARNING [train.py:1060] (2/4) Exclude cut with ID 7357-94126-0021-15852-0_sp1.1 from training. Duration: 22.5090625
2022-12-24 06:09:41,396 WARNING [train.py:1060] (2/4) Exclude cut with ID 7255-291500-0005-134304-0 from training. Duration: 25.035
2022-12-24 06:09:59,344 WARNING [train.py:1060] (2/4) Exclude cut with ID 774-127930-0014-10412-0_sp1.1 from training. Duration: 0.95
2022-12-24 06:10:16,982 WARNING [train.py:1060] (2/4) Exclude cut with ID 3033-130750-0096-55598-0_sp0.9 from training. Duration: 0.92225
2022-12-24 06:10:23,395 INFO [zipformer.py:660] (2/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=103427.0, num_to_drop=1, layers_to_drop={0}
2022-12-24 06:10:24,380 INFO [train.py:894] (2/4) Epoch 30, batch 1750, loss[loss=0.1696, simple_loss=0.2483, pruned_loss=0.04549, over 18440.00 frames. ], tot_loss[loss=0.1686, simple_loss=0.2574, pruned_loss=0.03988, over 3712314.89 frames. ], batch size: 42, lr: 3.80e-03, grad_scale: 8.0
2022-12-24 06:10:26,148 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.4773, 1.1135, 1.8733, 2.6908, 2.0059, 2.3891, 0.7332, 1.9835],
       device='cuda:2'), covar=tensor([0.1892, 0.1715, 0.1315, 0.0695, 0.1028, 0.1099, 0.2117, 0.1154],
       device='cuda:2'), in_proj_covar=tensor([0.0104, 0.0120, 0.0139, 0.0159, 0.0108, 0.0145, 0.0131, 0.0117],
       device='cuda:2'), out_proj_covar=tensor([0.0003, 0.0003, 0.0004, 0.0004, 0.0003, 0.0004, 0.0004, 0.0003],
       device='cuda:2')
2022-12-24 06:10:44,837 WARNING [train.py:1060] (2/4) Exclude cut with ID 4511-76322-0006-80011-0 from training. Duration: 21.97
2022-12-24 06:11:00,832 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 2.344e+02 3.614e+02 4.053e+02 5.108e+02 1.085e+03, threshold=8.106e+02, percent-clipped=5.0
2022-12-24 06:11:02,337 WARNING [train.py:1060] (2/4) Exclude cut with ID 7492-105653-0055-62765-0_sp0.9 from training. Duration: 21.97225
2022-12-24 06:11:04,036 WARNING [train.py:1060] (2/4) Exclude cut with ID 453-131332-0000-47844-0_sp0.9 from training. Duration: 25.3333125
2022-12-24 06:11:04,137 INFO [zipformer.py:660] (2/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=103454.0, num_to_drop=0, layers_to_drop=set()
2022-12-24 06:11:15,073 WARNING [train.py:1060] (2/4) Exclude cut with ID 5172-29468-0015-19128-0_sp0.9 from training. Duration: 21.5055625
2022-12-24 06:11:25,442 WARNING [train.py:1060] (2/4) Exclude cut with ID 453-131332-0000-47844-0_sp1.1 from training. Duration: 20.72725
2022-12-24 06:11:40,431 INFO [train.py:894] (2/4) Epoch 30, batch 1800, loss[loss=0.2231, simple_loss=0.2986, pruned_loss=0.07375, over 18717.00 frames. ], tot_loss[loss=0.1698, simple_loss=0.2579, pruned_loss=0.04089, over 3713153.38 frames. ], batch size: 54, lr: 3.80e-03, grad_scale: 8.0
2022-12-24 06:11:55,350 INFO [zipformer.py:660] (2/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=103488.0, num_to_drop=0, layers_to_drop=set()
2022-12-24 06:11:59,452 WARNING [train.py:1060] (2/4) Exclude cut with ID 8631-249866-0030-130156-0_sp0.9 from training. Duration: 26.32775
2022-12-24 06:12:19,648 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([3.0696, 2.4744, 2.0049, 2.8971, 2.3457, 2.4952, 2.5208, 3.2709],
       device='cuda:2'), covar=tensor([0.1980, 0.3300, 0.1986, 0.2918, 0.3643, 0.1035, 0.3082, 0.0840],
       device='cuda:2'), in_proj_covar=tensor([0.0302, 0.0303, 0.0257, 0.0351, 0.0283, 0.0237, 0.0299, 0.0225],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0003, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-24 06:12:24,052 INFO [zipformer.py:660] (2/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=103507.0, num_to_drop=0, layers_to_drop=set()
2022-12-24 06:12:30,042 WARNING [train.py:1060] (2/4) Exclude cut with ID 3867-173237-0077-144769-0 from training. Duration: 20.025
2022-12-24 06:12:34,289 WARNING [train.py:1060] (2/4) Exclude cut with ID 6709-74022-0004-86860-0_sp1.1 from training. Duration: 0.9409375
2022-12-24 06:12:35,863 WARNING [train.py:1060] (2/4) Exclude cut with ID 4757-1811-0023-62229-0_sp0.9 from training. Duration: 21.37775
2022-12-24 06:12:55,189 INFO [train.py:894] (2/4) Epoch 30, batch 1850, loss[loss=0.1506, simple_loss=0.2416, pruned_loss=0.02979, over 18707.00 frames. ], tot_loss[loss=0.1707, simple_loss=0.2582, pruned_loss=0.04164, over 3713761.46 frames. ], batch size: 54, lr: 3.80e-03, grad_scale: 8.0
2022-12-24 06:12:55,227 WARNING [train.py:1060] (2/4) Exclude cut with ID 1250-135782-0004-25974-0_sp0.9 from training. Duration: 21.17225
2022-12-24 06:12:55,237 WARNING [train.py:1060] (2/4) Exclude cut with ID 7357-94126-0021-15852-0_sp0.9 from training. Duration: 27.511125
2022-12-24 06:13:14,539 INFO [scaling.py:679] (2/4) Whitening: num_groups=1, num_channels=384, metric=4.26 vs. limit=5.0
2022-12-24 06:13:18,052 INFO [zipformer.py:660] (2/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=103543.0, num_to_drop=0, layers_to_drop=set()
2022-12-24 06:13:31,004 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 2.198e+02 4.176e+02 5.067e+02 6.015e+02 1.138e+03, threshold=1.013e+03, percent-clipped=6.0
2022-12-24 06:13:31,052 WARNING [train.py:1060] (2/4) Exclude cut with ID 453-131332-0000-47844-0 from training. Duration: 22.8
2022-12-24 06:13:35,896 WARNING [train.py:1060] (2/4) Exclude cut with ID 4964-30587-0040-44509-0 from training. Duration: 22.585
2022-12-24 06:13:36,007 INFO [zipformer.py:660] (2/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=103555.0, num_to_drop=0, layers_to_drop=set()
2022-12-24 06:14:06,203 WARNING [train.py:1060] (2/4) Exclude cut with ID 6978-92210-0001-146967-0_sp0.9 from training. Duration: 22.0166875
2022-12-24 06:14:11,259 INFO [train.py:894] (2/4) Epoch 30, batch 1900, loss[loss=0.1663, simple_loss=0.245, pruned_loss=0.04375, over 18395.00 frames. ], tot_loss[loss=0.1718, simple_loss=0.2583, pruned_loss=0.04266, over 3714002.12 frames. ], batch size: 46, lr: 3.80e-03, grad_scale: 8.0
2022-12-24 06:14:22,816 WARNING [train.py:1060] (2/4) Exclude cut with ID 7255-291500-0003-134302-0_sp1.1 from training. Duration: 24.395375
2022-12-24 06:14:28,684 WARNING [train.py:1060] (2/4) Exclude cut with ID 3488-85273-0017-41203-0_sp0.9 from training. Duration: 27.47775
2022-12-24 06:14:30,246 INFO [zipformer.py:660] (2/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=103591.0, num_to_drop=0, layers_to_drop=set()
2022-12-24 06:14:33,068 WARNING [train.py:1060] (2/4) Exclude cut with ID 432-122774-0017-62487-0_sp0.9 from training. Duration: 24.8833125
2022-12-24 06:14:36,009 WARNING [train.py:1060] (2/4) Exclude cut with ID 6758-72288-0033-108368-0 from training. Duration: 23.39
2022-12-24 06:14:42,243 WARNING [train.py:1060] (2/4) Exclude cut with ID 8291-282929-0007-12994-0_sp0.9 from training. Duration: 28.72225
2022-12-24 06:14:51,589 WARNING [train.py:1060] (2/4) Exclude cut with ID 585-294811-0110-133686-0_sp0.9 from training. Duration: 20.8944375
2022-12-24 06:14:57,893 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.3730, 1.1406, 1.6296, 2.3209, 1.6559, 2.3125, 0.8528, 1.7752],
       device='cuda:2'), covar=tensor([0.1796, 0.1626, 0.1189, 0.0761, 0.1141, 0.0835, 0.1766, 0.1128],
       device='cuda:2'), in_proj_covar=tensor([0.0104, 0.0120, 0.0139, 0.0159, 0.0108, 0.0146, 0.0131, 0.0117],
       device='cuda:2'), out_proj_covar=tensor([0.0003, 0.0003, 0.0004, 0.0004, 0.0003, 0.0004, 0.0004, 0.0003],
       device='cuda:2')
2022-12-24 06:15:05,787 WARNING [train.py:1060] (2/4) Exclude cut with ID 5796-66357-0007-116447-0_sp0.9 from training. Duration: 23.8444375
2022-12-24 06:15:26,316 INFO [train.py:894] (2/4) Epoch 30, batch 1950, loss[loss=0.2148, simple_loss=0.2959, pruned_loss=0.06691, over 18645.00 frames. ], tot_loss[loss=0.1726, simple_loss=0.2585, pruned_loss=0.04333, over 3714421.37 frames. ], batch size: 178, lr: 3.80e-03, grad_scale: 8.0
2022-12-24 06:15:30,797 WARNING [train.py:1060] (2/4) Exclude cut with ID 8291-282929-0007-12994-0 from training. Duration: 25.85
2022-12-24 06:15:30,807 WARNING [train.py:1060] (2/4) Exclude cut with ID 8291-282929-0023-13010-0 from training. Duration: 21.39
2022-12-24 06:15:41,179 WARNING [train.py:1060] (2/4) Exclude cut with ID 7357-94126-0014-15845-0 from training. Duration: 27.92
2022-12-24 06:16:04,197 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 2.424e+02 3.831e+02 4.811e+02 6.471e+02 1.162e+03, threshold=9.622e+02, percent-clipped=1.0
2022-12-24 06:16:09,871 WARNING [train.py:1060] (2/4) Exclude cut with ID 8631-249866-0039-130165-0_sp0.9 from training. Duration: 20.661125
2022-12-24 06:16:37,234 WARNING [train.py:1060] (2/4) Exclude cut with ID 7357-94126-0043-15874-0_sp0.9 from training. Duration: 20.07225
2022-12-24 06:16:41,392 INFO [train.py:894] (2/4) Epoch 30, batch 2000, loss[loss=0.1676, simple_loss=0.2574, pruned_loss=0.03891, over 18596.00 frames. ], tot_loss[loss=0.1723, simple_loss=0.2578, pruned_loss=0.04337, over 3713440.39 frames. ], batch size: 56, lr: 3.80e-03, grad_scale: 8.0
2022-12-24 06:16:44,389 WARNING [train.py:1060] (2/4) Exclude cut with ID 1085-156170-0017-128270-0 from training. Duration: 21.01
2022-12-24 06:17:48,358 WARNING [train.py:1060] (2/4) Exclude cut with ID 2195-150901-0045-59933-0 from training. Duration: 20.65
2022-12-24 06:17:48,468 INFO [zipformer.py:660] (2/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=103722.0, num_to_drop=1, layers_to_drop={2}
2022-12-24 06:17:57,499 INFO [train.py:894] (2/4) Epoch 30, batch 2050, loss[loss=0.1867, simple_loss=0.2702, pruned_loss=0.05163, over 18560.00 frames. ], tot_loss[loss=0.1738, simple_loss=0.2588, pruned_loss=0.0444, over 3713055.39 frames. ], batch size: 49, lr: 3.80e-03, grad_scale: 8.0
2022-12-24 06:17:57,531 WARNING [train.py:1060] (2/4) Exclude cut with ID 5796-66357-0007-116447-0 from training. Duration: 21.46
2022-12-24 06:18:15,890 INFO [zipformer.py:660] (2/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=103740.0, num_to_drop=0, layers_to_drop=set()
2022-12-24 06:18:34,846 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 2.766e+02 4.041e+02 4.718e+02 5.719e+02 1.210e+03, threshold=9.436e+02, percent-clipped=5.0
2022-12-24 06:18:36,552 INFO [zipformer.py:660] (2/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=103754.0, num_to_drop=0, layers_to_drop=set()
2022-12-24 06:18:41,981 WARNING [train.py:1060] (2/4) Exclude cut with ID 3557-8342-0013-54691-0 from training. Duration: 0.92
2022-12-24 06:18:42,335 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([2.5064, 2.4705, 2.8658, 1.7345, 2.8599, 2.8337, 1.8514, 3.2521],
       device='cuda:2'), covar=tensor([0.1181, 0.1746, 0.1468, 0.2036, 0.0725, 0.1259, 0.2310, 0.0640],
       device='cuda:2'), in_proj_covar=tensor([0.0199, 0.0217, 0.0211, 0.0196, 0.0173, 0.0221, 0.0219, 0.0197],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-24 06:18:48,407 WARNING [train.py:1060] (2/4) Exclude cut with ID 8291-282929-0023-13010-0_sp0.9 from training. Duration: 23.7666875
2022-12-24 06:18:53,262 INFO [zipformer.py:660] (2/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=103765.0, num_to_drop=0, layers_to_drop=set()
2022-12-24 06:18:59,597 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([2.5005, 2.3944, 2.8855, 2.2061, 2.7686, 2.7163, 2.1966, 3.0217],
       device='cuda:2'), covar=tensor([0.0975, 0.1513, 0.1293, 0.1518, 0.0571, 0.0977, 0.1698, 0.0504],
       device='cuda:2'), in_proj_covar=tensor([0.0199, 0.0217, 0.0211, 0.0196, 0.0173, 0.0220, 0.0218, 0.0197],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-24 06:19:11,914 INFO [train.py:894] (2/4) Epoch 30, batch 2100, loss[loss=0.1776, simple_loss=0.2644, pruned_loss=0.04541, over 18596.00 frames. ], tot_loss[loss=0.1737, simple_loss=0.2582, pruned_loss=0.04455, over 3713219.66 frames. ], batch size: 51, lr: 3.80e-03, grad_scale: 8.0
2022-12-24 06:19:24,117 WARNING [train.py:1060] (2/4) Exclude cut with ID 8544-281189-0060-101339-0_sp0.9 from training. Duration: 20.861125
2022-12-24 06:19:28,369 INFO [zipformer.py:660] (2/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=103788.0, num_to_drop=0, layers_to_drop=set()
2022-12-24 06:19:35,246 WARNING [train.py:1060] (2/4) Exclude cut with ID 3488-65654-0031-41259-0_sp0.9 from training. Duration: 22.711125
2022-12-24 06:19:42,999 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.5000, 1.4048, 1.3330, 1.4061, 1.7238, 1.5475, 1.5590, 1.2552],
       device='cuda:2'), covar=tensor([0.0304, 0.0253, 0.0557, 0.0239, 0.0213, 0.0457, 0.0306, 0.0344],
       device='cuda:2'), in_proj_covar=tensor([0.0100, 0.0130, 0.0160, 0.0125, 0.0121, 0.0126, 0.0104, 0.0132],
       device='cuda:2'), out_proj_covar=tensor([7.8668e-05, 1.0290e-04, 1.3054e-04, 9.9300e-05, 9.6250e-05, 9.6230e-05,
        8.0519e-05, 1.0396e-04], device='cuda:2')
2022-12-24 06:19:47,464 INFO [zipformer.py:660] (2/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=103801.0, num_to_drop=0, layers_to_drop=set()
2022-12-24 06:19:48,649 INFO [zipformer.py:660] (2/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=103802.0, num_to_drop=0, layers_to_drop=set()
2022-12-24 06:20:16,699 WARNING [train.py:1060] (2/4) Exclude cut with ID 6951-79737-0043-132310-0_sp1.1 from training. Duration: 22.986375
2022-12-24 06:20:24,164 INFO [zipformer.py:660] (2/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=103826.0, num_to_drop=0, layers_to_drop=set()
2022-12-24 06:20:26,361 INFO [train.py:894] (2/4) Epoch 30, batch 2150, loss[loss=0.1762, simple_loss=0.2589, pruned_loss=0.04681, over 18701.00 frames. ], tot_loss[loss=0.1734, simple_loss=0.2579, pruned_loss=0.04446, over 3713024.75 frames. ], batch size: 50, lr: 3.80e-03, grad_scale: 8.0
2022-12-24 06:20:30,733 WARNING [train.py:1060] (2/4) Exclude cut with ID 8040-260924-0003-80960-0_sp0.9 from training. Duration: 22.07225
2022-12-24 06:20:38,299 WARNING [train.py:1060] (2/4) Exclude cut with ID 7699-105389-0045-26330-0_sp0.9 from training. Duration: 20.3055625
2022-12-24 06:20:39,795 WARNING [train.py:1060] (2/4) Exclude cut with ID 6356-271890-0060-94317-0_sp0.9 from training. Duration: 20.72225
2022-12-24 06:20:39,960 INFO [zipformer.py:660] (2/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=103836.0, num_to_drop=0, layers_to_drop=set()
2022-12-24 06:20:58,684 WARNING [train.py:1060] (2/4) Exclude cut with ID 3488-85273-0017-41203-0_sp1.1 from training. Duration: 22.4818125
2022-12-24 06:21:04,441 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 2.179e+02 3.826e+02 4.710e+02 5.981e+02 1.265e+03, threshold=9.420e+02, percent-clipped=2.0
2022-12-24 06:21:24,606 WARNING [train.py:1060] (2/4) Exclude cut with ID 4964-30587-0040-44509-0_sp0.9 from training. Duration: 25.0944375
2022-12-24 06:21:28,861 WARNING [train.py:1060] (2/4) Exclude cut with ID 6533-399-0047-104881-0 from training. Duration: 21.515
2022-12-24 06:21:34,276 INFO [scaling.py:679] (2/4) Whitening: num_groups=8, num_channels=96, metric=1.31 vs. limit=2.0
2022-12-24 06:21:36,247 WARNING [train.py:1060] (2/4) Exclude cut with ID 7357-94126-0009-15840-0_sp0.9 from training. Duration: 27.02225
2022-12-24 06:21:39,533 INFO [zipformer.py:660] (2/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=103876.0, num_to_drop=0, layers_to_drop=set()
2022-12-24 06:21:40,690 WARNING [train.py:1060] (2/4) Exclude cut with ID 432-122774-0010-62480-0_sp0.9 from training. Duration: 22.22225
2022-12-24 06:21:42,074 INFO [train.py:894] (2/4) Epoch 30, batch 2200, loss[loss=0.1651, simple_loss=0.2448, pruned_loss=0.04269, over 18565.00 frames. ], tot_loss[loss=0.1727, simple_loss=0.257, pruned_loss=0.04421, over 3712912.30 frames. ], batch size: 49, lr: 3.79e-03, grad_scale: 8.0
2022-12-24 06:21:48,400 WARNING [train.py:1060] (2/4) Exclude cut with ID 4964-30587-0085-44554-0_sp0.9 from training. Duration: 20.85
2022-12-24 06:22:01,005 INFO [scaling.py:679] (2/4) Whitening: num_groups=1, num_channels=384, metric=2.85 vs. limit=5.0
2022-12-24 06:22:04,713 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([4.5842, 3.9562, 3.9508, 4.5099, 4.2300, 4.0000, 4.7448, 1.5454],
       device='cuda:2'), covar=tensor([0.0650, 0.0724, 0.0709, 0.0838, 0.1212, 0.1236, 0.0553, 0.4866],
       device='cuda:2'), in_proj_covar=tensor([0.0367, 0.0240, 0.0253, 0.0290, 0.0343, 0.0280, 0.0305, 0.0299],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0001, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-24 06:22:22,456 WARNING [train.py:1060] (2/4) Exclude cut with ID 4295-39940-0007-92567-0 from training. Duration: 21.54
2022-12-24 06:22:28,051 WARNING [train.py:1060] (2/4) Exclude cut with ID 4964-30587-0040-44509-0_sp1.1 from training. Duration: 20.5318125
2022-12-24 06:22:36,706 WARNING [train.py:1060] (2/4) Exclude cut with ID 7255-291500-0012-134311-0_sp0.9 from training. Duration: 21.9333125
2022-12-24 06:22:39,127 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([3.0781, 2.5031, 1.8729, 2.9676, 2.3374, 2.4717, 2.5033, 3.2763],
       device='cuda:2'), covar=tensor([0.2196, 0.3297, 0.2148, 0.2999, 0.4266, 0.1041, 0.3358, 0.0894],
       device='cuda:2'), in_proj_covar=tensor([0.0305, 0.0305, 0.0259, 0.0354, 0.0286, 0.0238, 0.0301, 0.0227],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0003, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-24 06:22:56,281 INFO [train.py:894] (2/4) Epoch 30, batch 2250, loss[loss=0.1687, simple_loss=0.2446, pruned_loss=0.04645, over 18619.00 frames. ], tot_loss[loss=0.1737, simple_loss=0.2578, pruned_loss=0.04481, over 3713932.30 frames. ], batch size: 45, lr: 3.79e-03, grad_scale: 8.0
2022-12-24 06:23:11,379 INFO [zipformer.py:660] (2/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=103937.0, num_to_drop=1, layers_to_drop={2}
2022-12-24 06:23:21,641 WARNING [train.py:1060] (2/4) Exclude cut with ID 8631-249866-0025-130151-0_sp0.9 from training. Duration: 21.7944375
2022-12-24 06:23:34,017 WARNING [train.py:1060] (2/4) Exclude cut with ID 8291-282929-0002-12989-0_sp0.9 from training. Duration: 22.4666875
2022-12-24 06:23:35,554 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 2.692e+02 4.009e+02 4.873e+02 5.788e+02 1.098e+03, threshold=9.745e+02, percent-clipped=7.0
2022-12-24 06:23:41,364 WARNING [train.py:1060] (2/4) Exclude cut with ID 6121-9014-0076-24124-0 from training. Duration: 21.635
2022-12-24 06:23:44,130 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.4901, 1.0260, 0.7066, 1.1709, 2.0092, 0.6431, 1.2125, 1.4383],
       device='cuda:2'), covar=tensor([0.1722, 0.2285, 0.2016, 0.1556, 0.1764, 0.1883, 0.1568, 0.1719],
       device='cuda:2'), in_proj_covar=tensor([0.0096, 0.0099, 0.0118, 0.0098, 0.0122, 0.0093, 0.0100, 0.0096],
       device='cuda:2'), out_proj_covar=tensor([0.0003, 0.0003, 0.0004, 0.0003, 0.0004, 0.0003, 0.0003, 0.0003],
       device='cuda:2')
2022-12-24 06:23:47,206 WARNING [train.py:1060] (2/4) Exclude cut with ID 6121-9014-0076-24124-0_sp0.9 from training. Duration: 24.038875
2022-12-24 06:24:13,202 INFO [train.py:894] (2/4) Epoch 30, batch 2300, loss[loss=0.1934, simple_loss=0.2761, pruned_loss=0.05539, over 18539.00 frames. ], tot_loss[loss=0.174, simple_loss=0.2581, pruned_loss=0.04494, over 3713542.79 frames. ], batch size: 58, lr: 3.79e-03, grad_scale: 8.0
2022-12-24 06:24:30,965 INFO [zipformer.py:660] (2/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=103990.0, num_to_drop=0, layers_to_drop=set()
2022-12-24 06:24:31,960 WARNING [train.py:1060] (2/4) Exclude cut with ID 7255-291500-0009-134308-0_sp1.1 from training. Duration: 21.786375
2022-12-24 06:24:41,079 INFO [zipformer.py:660] (2/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=103997.0, num_to_drop=0, layers_to_drop=set()
2022-12-24 06:24:44,093 WARNING [train.py:1060] (2/4) Exclude cut with ID 8291-282929-0002-12989-0 from training. Duration: 20.22
2022-12-24 06:25:21,829 INFO [zipformer.py:660] (2/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=104022.0, num_to_drop=1, layers_to_drop={2}
2022-12-24 06:25:31,240 INFO [train.py:894] (2/4) Epoch 30, batch 2350, loss[loss=0.1464, simple_loss=0.2216, pruned_loss=0.03564, over 18545.00 frames. ], tot_loss[loss=0.1738, simple_loss=0.2577, pruned_loss=0.04495, over 3714000.67 frames. ], batch size: 41, lr: 3.79e-03, grad_scale: 8.0
2022-12-24 06:25:36,702 INFO [scaling.py:679] (2/4) Whitening: num_groups=8, num_channels=192, metric=1.89 vs. limit=2.0
2022-12-24 06:25:39,313 INFO [scaling.py:679] (2/4) Whitening: num_groups=8, num_channels=192, metric=1.77 vs. limit=2.0
2022-12-24 06:26:02,186 INFO [zipformer.py:660] (2/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=104049.0, num_to_drop=0, layers_to_drop=set()
2022-12-24 06:26:05,076 INFO [zipformer.py:660] (2/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=104051.0, num_to_drop=0, layers_to_drop=set()
2022-12-24 06:26:07,809 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 2.777e+02 3.890e+02 4.874e+02 6.166e+02 1.500e+03, threshold=9.748e+02, percent-clipped=4.0
2022-12-24 06:26:16,616 INFO [zipformer.py:660] (2/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=104058.0, num_to_drop=0, layers_to_drop=set()
2022-12-24 06:26:34,427 INFO [zipformer.py:660] (2/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=104070.0, num_to_drop=1, layers_to_drop={1}
2022-12-24 06:26:34,747 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([2.0091, 1.0805, 1.7804, 1.6451, 2.0484, 2.0570, 1.7524, 1.7969],
       device='cuda:2'), covar=tensor([0.2379, 0.3639, 0.2932, 0.2936, 0.2296, 0.1121, 0.3341, 0.1492],
       device='cuda:2'), in_proj_covar=tensor([0.0276, 0.0303, 0.0292, 0.0332, 0.0324, 0.0263, 0.0361, 0.0253],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-24 06:26:36,297 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.7298, 1.3264, 1.9622, 3.4452, 2.5238, 2.5619, 0.8799, 2.5271],
       device='cuda:2'), covar=tensor([0.1834, 0.1622, 0.1487, 0.0580, 0.0953, 0.1139, 0.2207, 0.0978],
       device='cuda:2'), in_proj_covar=tensor([0.0104, 0.0120, 0.0138, 0.0159, 0.0108, 0.0146, 0.0130, 0.0117],
       device='cuda:2'), out_proj_covar=tensor([0.0003, 0.0003, 0.0004, 0.0004, 0.0003, 0.0004, 0.0004, 0.0003],
       device='cuda:2')
2022-12-24 06:26:44,635 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([2.2924, 1.5530, 0.9162, 1.8036, 2.5085, 1.7469, 1.9109, 2.3760],
       device='cuda:2'), covar=tensor([0.1559, 0.2052, 0.2404, 0.1417, 0.1661, 0.1858, 0.1398, 0.1516],
       device='cuda:2'), in_proj_covar=tensor([0.0095, 0.0098, 0.0117, 0.0097, 0.0121, 0.0093, 0.0099, 0.0095],
       device='cuda:2'), out_proj_covar=tensor([0.0003, 0.0003, 0.0004, 0.0003, 0.0004, 0.0003, 0.0003, 0.0003],
       device='cuda:2')
2022-12-24 06:26:45,787 INFO [train.py:894] (2/4) Epoch 30, batch 2400, loss[loss=0.1834, simple_loss=0.277, pruned_loss=0.04487, over 18667.00 frames. ], tot_loss[loss=0.1745, simple_loss=0.2586, pruned_loss=0.04518, over 3714321.49 frames. ], batch size: 60, lr: 3.79e-03, grad_scale: 8.0
2022-12-24 06:26:47,285 WARNING [train.py:1060] (2/4) Exclude cut with ID 6951-79737-0043-132310-0 from training. Duration: 25.285
2022-12-24 06:27:02,719 INFO [scaling.py:679] (2/4) Whitening: num_groups=8, num_channels=96, metric=1.48 vs. limit=2.0
2022-12-24 06:27:12,780 INFO [zipformer.py:660] (2/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=104096.0, num_to_drop=0, layers_to_drop=set()
2022-12-24 06:27:29,030 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([2.6885, 2.4621, 2.0313, 0.8787, 1.9197, 2.0438, 1.7473, 2.2340],
       device='cuda:2'), covar=tensor([0.0658, 0.0617, 0.1312, 0.1871, 0.1403, 0.1622, 0.1822, 0.0935],
       device='cuda:2'), in_proj_covar=tensor([0.0177, 0.0191, 0.0211, 0.0191, 0.0213, 0.0207, 0.0219, 0.0207],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-24 06:27:33,327 INFO [zipformer.py:660] (2/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=104110.0, num_to_drop=0, layers_to_drop=set()
2022-12-24 06:27:47,413 INFO [scaling.py:679] (2/4) Whitening: num_groups=8, num_channels=192, metric=1.91 vs. limit=2.0
2022-12-24 06:27:47,809 WARNING [train.py:1060] (2/4) Exclude cut with ID 811-130148-0001-63453-0_sp0.9 from training. Duration: 20.861125
2022-12-24 06:27:50,974 INFO [zipformer.py:660] (2/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=104121.0, num_to_drop=0, layers_to_drop=set()
2022-12-24 06:28:01,169 INFO [train.py:894] (2/4) Epoch 30, batch 2450, loss[loss=0.159, simple_loss=0.2468, pruned_loss=0.0356, over 18472.00 frames. ], tot_loss[loss=0.1744, simple_loss=0.2584, pruned_loss=0.04517, over 3714378.14 frames. ], batch size: 50, lr: 3.79e-03, grad_scale: 8.0
2022-12-24 06:28:03,552 INFO [scaling.py:679] (2/4) Whitening: num_groups=8, num_channels=96, metric=1.18 vs. limit=2.0
2022-12-24 06:28:08,982 WARNING [train.py:1060] (2/4) Exclude cut with ID 6010-56788-0055-90261-0 from training. Duration: 20.88
2022-12-24 06:28:34,483 INFO [scaling.py:679] (2/4) Whitening: num_groups=8, num_channels=192, metric=1.92 vs. limit=2.0
2022-12-24 06:28:39,221 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 2.777e+02 4.495e+02 5.196e+02 6.656e+02 1.409e+03, threshold=1.039e+03, percent-clipped=6.0
2022-12-24 06:28:40,761 WARNING [train.py:1060] (2/4) Exclude cut with ID 7357-94126-0045-15876-0_sp0.9 from training. Duration: 23.4166875
2022-12-24 06:29:16,541 INFO [train.py:894] (2/4) Epoch 30, batch 2500, loss[loss=0.1616, simple_loss=0.2472, pruned_loss=0.03805, over 18709.00 frames. ], tot_loss[loss=0.1736, simple_loss=0.2581, pruned_loss=0.0446, over 3713474.88 frames. ], batch size: 50, lr: 3.79e-03, grad_scale: 8.0
2022-12-24 06:29:22,600 INFO [scaling.py:679] (2/4) Whitening: num_groups=8, num_channels=96, metric=1.15 vs. limit=2.0
2022-12-24 06:30:01,888 WARNING [train.py:1060] (2/4) Exclude cut with ID 7255-291500-0006-134305-0 from training. Duration: 21.24
2022-12-24 06:30:01,907 WARNING [train.py:1060] (2/4) Exclude cut with ID 6533-399-0047-104881-0_sp0.9 from training. Duration: 23.9055625
2022-12-24 06:30:31,028 INFO [train.py:894] (2/4) Epoch 30, batch 2550, loss[loss=0.1738, simple_loss=0.2625, pruned_loss=0.04254, over 18501.00 frames. ], tot_loss[loss=0.1722, simple_loss=0.2566, pruned_loss=0.04391, over 3713540.43 frames. ], batch size: 52, lr: 3.79e-03, grad_scale: 8.0
2022-12-24 06:30:32,360 WARNING [train.py:1060] (2/4) Exclude cut with ID 6758-72288-0033-108368-0_sp0.9 from training. Duration: 25.988875
2022-12-24 06:30:34,588 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([2.0918, 1.4774, 2.4839, 4.4050, 3.1423, 2.8951, 0.7568, 3.2541],
       device='cuda:2'), covar=tensor([0.1816, 0.1622, 0.1467, 0.0611, 0.0897, 0.1161, 0.2362, 0.0849],
       device='cuda:2'), in_proj_covar=tensor([0.0103, 0.0119, 0.0137, 0.0157, 0.0108, 0.0144, 0.0129, 0.0116],
       device='cuda:2'), out_proj_covar=tensor([0.0003, 0.0003, 0.0004, 0.0004, 0.0003, 0.0004, 0.0004, 0.0003],
       device='cuda:2')
2022-12-24 06:30:37,577 INFO [zipformer.py:660] (2/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=104232.0, num_to_drop=1, layers_to_drop={0}
2022-12-24 06:30:42,348 WARNING [train.py:1060] (2/4) Exclude cut with ID 7255-291500-0001-134300-0_sp0.9 from training. Duration: 20.67225
2022-12-24 06:30:55,659 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([2.5398, 2.3331, 1.8223, 0.7569, 1.7855, 2.1288, 1.8019, 2.0359],
       device='cuda:2'), covar=tensor([0.0668, 0.0550, 0.1244, 0.1732, 0.1227, 0.1510, 0.1715, 0.0844],
       device='cuda:2'), in_proj_covar=tensor([0.0177, 0.0191, 0.0212, 0.0192, 0.0214, 0.0207, 0.0220, 0.0207],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-24 06:31:04,407 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([2.0951, 1.9533, 1.6448, 1.6298, 1.8218, 1.9716, 1.7955, 1.8736],
       device='cuda:2'), covar=tensor([0.2335, 0.3120, 0.2092, 0.2552, 0.3508, 0.1104, 0.2951, 0.1103],
       device='cuda:2'), in_proj_covar=tensor([0.0302, 0.0301, 0.0257, 0.0350, 0.0283, 0.0236, 0.0298, 0.0225],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0003, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-24 06:31:10,010 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 2.524e+02 4.032e+02 4.876e+02 5.931e+02 1.170e+03, threshold=9.752e+02, percent-clipped=3.0
2022-12-24 06:31:29,723 WARNING [train.py:1060] (2/4) Exclude cut with ID 3488-85273-0038-41224-0 from training. Duration: 20.34
2022-12-24 06:31:46,936 INFO [train.py:894] (2/4) Epoch 30, batch 2600, loss[loss=0.1837, simple_loss=0.267, pruned_loss=0.05018, over 18651.00 frames. ], tot_loss[loss=0.1728, simple_loss=0.2571, pruned_loss=0.04423, over 3713350.26 frames. ], batch size: 78, lr: 3.79e-03, grad_scale: 8.0
2022-12-24 06:32:09,103 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.8055, 1.3158, 1.5986, 1.9374, 1.6554, 3.1627, 1.3158, 1.3687],
       device='cuda:2'), covar=tensor([0.1005, 0.2602, 0.1168, 0.1025, 0.1832, 0.0343, 0.2073, 0.2273],
       device='cuda:2'), in_proj_covar=tensor([0.0074, 0.0083, 0.0072, 0.0075, 0.0092, 0.0077, 0.0085, 0.0078],
       device='cuda:2'), out_proj_covar=tensor([0.0004, 0.0004, 0.0004, 0.0004, 0.0005, 0.0004, 0.0004, 0.0004],
       device='cuda:2')
2022-12-24 06:32:41,906 WARNING [train.py:1060] (2/4) Exclude cut with ID 7357-94126-0026-15857-0_sp0.9 from training. Duration: 25.061125
2022-12-24 06:32:54,329 WARNING [train.py:1060] (2/4) Exclude cut with ID 3033-130750-0096-55598-0 from training. Duration: 0.83
2022-12-24 06:33:01,508 INFO [train.py:894] (2/4) Epoch 30, batch 2650, loss[loss=0.2327, simple_loss=0.2998, pruned_loss=0.08282, over 18616.00 frames. ], tot_loss[loss=0.1724, simple_loss=0.2567, pruned_loss=0.04406, over 3713463.79 frames. ], batch size: 170, lr: 3.79e-03, grad_scale: 8.0
2022-12-24 06:33:19,781 WARNING [train.py:1060] (2/4) Exclude cut with ID 3488-85273-0017-41203-0 from training. Duration: 24.73
2022-12-24 06:33:22,034 INFO [scaling.py:679] (2/4) Whitening: num_groups=8, num_channels=192, metric=1.81 vs. limit=2.0
2022-12-24 06:33:29,475 INFO [zipformer.py:660] (2/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=104346.0, num_to_drop=0, layers_to_drop=set()
2022-12-24 06:33:30,809 WARNING [train.py:1060] (2/4) Exclude cut with ID 7255-291500-0009-134308-0 from training. Duration: 23.965
2022-12-24 06:33:39,922 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 2.438e+02 4.320e+02 5.325e+02 6.686e+02 2.435e+03, threshold=1.065e+03, percent-clipped=5.0
2022-12-24 06:33:39,960 WARNING [train.py:1060] (2/4) Exclude cut with ID 6978-92210-0030-146996-0_sp0.9 from training. Duration: 22.088875
2022-12-24 06:33:40,062 INFO [zipformer.py:660] (2/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=104353.0, num_to_drop=0, layers_to_drop=set()
2022-12-24 06:33:55,224 WARNING [train.py:1060] (2/4) Exclude cut with ID 7255-291500-0006-134305-0_sp0.9 from training. Duration: 23.6
2022-12-24 06:34:17,275 INFO [train.py:894] (2/4) Epoch 30, batch 2700, loss[loss=0.1865, simple_loss=0.2698, pruned_loss=0.05159, over 18522.00 frames. ], tot_loss[loss=0.1729, simple_loss=0.2573, pruned_loss=0.04421, over 3713187.11 frames. ], batch size: 64, lr: 3.79e-03, grad_scale: 8.0
2022-12-24 06:34:26,653 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([2.4822, 2.1198, 1.4504, 2.2385, 2.8049, 2.4515, 2.3252, 2.5290],
       device='cuda:2'), covar=tensor([0.1192, 0.1564, 0.1978, 0.1066, 0.1377, 0.1389, 0.1033, 0.1151],
       device='cuda:2'), in_proj_covar=tensor([0.0095, 0.0098, 0.0117, 0.0097, 0.0121, 0.0093, 0.0099, 0.0095],
       device='cuda:2'), out_proj_covar=tensor([0.0003, 0.0003, 0.0004, 0.0003, 0.0004, 0.0003, 0.0003, 0.0003],
       device='cuda:2')
2022-12-24 06:34:46,168 INFO [zipformer.py:660] (2/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=104396.0, num_to_drop=0, layers_to_drop=set()
2022-12-24 06:34:58,877 INFO [zipformer.py:660] (2/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=104405.0, num_to_drop=0, layers_to_drop=set()
2022-12-24 06:35:22,319 INFO [zipformer.py:660] (2/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=104421.0, num_to_drop=0, layers_to_drop=set()
2022-12-24 06:35:32,648 INFO [train.py:894] (2/4) Epoch 30, batch 2750, loss[loss=0.1853, simple_loss=0.2731, pruned_loss=0.0487, over 18509.00 frames. ], tot_loss[loss=0.1724, simple_loss=0.2569, pruned_loss=0.04397, over 3712894.84 frames. ], batch size: 58, lr: 3.78e-03, grad_scale: 8.0
2022-12-24 06:35:36,956 WARNING [train.py:1060] (2/4) Exclude cut with ID 8291-282929-0024-13011-0 from training. Duration: 23.795
2022-12-24 06:35:51,726 INFO [scaling.py:679] (2/4) Whitening: num_groups=8, num_channels=192, metric=1.91 vs. limit=2.0
2022-12-24 06:35:54,423 WARNING [train.py:1060] (2/4) Exclude cut with ID 8631-249866-0030-130156-0_sp1.1 from training. Duration: 21.5409375
2022-12-24 06:35:57,326 WARNING [train.py:1060] (2/4) Exclude cut with ID 6978-92210-0019-146985-0_sp0.9 from training. Duration: 24.97775
2022-12-24 06:35:57,446 INFO [zipformer.py:660] (2/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=104444.0, num_to_drop=0, layers_to_drop=set()
2022-12-24 06:36:06,959 WARNING [train.py:1060] (2/4) Exclude cut with ID 1085-156170-0017-128270-0_sp0.9 from training. Duration: 23.3444375
2022-12-24 06:36:09,546 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 2.603e+02 3.880e+02 4.528e+02 5.642e+02 1.210e+03, threshold=9.056e+02, percent-clipped=1.0
2022-12-24 06:36:32,891 WARNING [train.py:1060] (2/4) Exclude cut with ID 6010-56788-0055-90261-0_sp0.9 from training. Duration: 23.2
2022-12-24 06:36:34,532 INFO [zipformer.py:660] (2/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=104469.0, num_to_drop=0, layers_to_drop=set()
2022-12-24 06:36:40,446 WARNING [train.py:1060] (2/4) Exclude cut with ID 5653-46179-0060-117930-0_sp0.9 from training. Duration: 21.17225
2022-12-24 06:36:47,359 INFO [train.py:894] (2/4) Epoch 30, batch 2800, loss[loss=0.2515, simple_loss=0.3149, pruned_loss=0.09408, over 18636.00 frames. ], tot_loss[loss=0.1727, simple_loss=0.2572, pruned_loss=0.04416, over 3713102.21 frames. ], batch size: 184, lr: 3.78e-03, grad_scale: 8.0
2022-12-24 06:36:59,951 WARNING [train.py:1060] (2/4) Exclude cut with ID 7255-291500-0008-134307-0_sp0.9 from training. Duration: 24.6555625
2022-12-24 06:37:43,509 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([2.4651, 3.2755, 3.2947, 1.2637, 3.5265, 2.7470, 0.7493, 2.3024],
       device='cuda:2'), covar=tensor([0.2244, 0.1724, 0.1533, 0.3790, 0.0998, 0.0884, 0.4834, 0.1544],
       device='cuda:2'), in_proj_covar=tensor([0.0156, 0.0155, 0.0165, 0.0127, 0.0156, 0.0120, 0.0148, 0.0122],
       device='cuda:2'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0002, 0.0003, 0.0002, 0.0003, 0.0002],
       device='cuda:2')
2022-12-24 06:37:54,810 WARNING [train.py:1060] (2/4) Exclude cut with ID 3488-65654-0031-41259-0 from training. Duration: 20.44
2022-12-24 06:38:02,535 INFO [train.py:894] (2/4) Epoch 30, batch 2850, loss[loss=0.1517, simple_loss=0.2289, pruned_loss=0.0373, over 18536.00 frames. ], tot_loss[loss=0.1731, simple_loss=0.2573, pruned_loss=0.04444, over 3713978.86 frames. ], batch size: 44, lr: 3.78e-03, grad_scale: 8.0
2022-12-24 06:38:08,369 INFO [zipformer.py:660] (2/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=104532.0, num_to_drop=1, layers_to_drop={0}
2022-12-24 06:38:10,375 WARNING [train.py:1060] (2/4) Exclude cut with ID 6951-79737-0018-132285-0_sp0.9 from training. Duration: 23.45
2022-12-24 06:38:39,964 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 2.496e+02 3.637e+02 4.715e+02 6.506e+02 1.297e+03, threshold=9.431e+02, percent-clipped=5.0
2022-12-24 06:38:40,057 WARNING [train.py:1060] (2/4) Exclude cut with ID 6945-60535-0076-12784-0_sp0.9 from training. Duration: 20.52225
2022-12-24 06:38:47,311 WARNING [train.py:1060] (2/4) Exclude cut with ID 7255-291500-0008-134307-0 from training. Duration: 22.19
2022-12-24 06:38:52,429 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.9983, 1.2568, 1.7031, 1.6846, 1.9988, 2.0828, 1.7567, 1.7427],
       device='cuda:2'), covar=tensor([0.2925, 0.4028, 0.3399, 0.3356, 0.2693, 0.1346, 0.4010, 0.1791],
       device='cuda:2'), in_proj_covar=tensor([0.0276, 0.0305, 0.0294, 0.0333, 0.0325, 0.0264, 0.0362, 0.0253],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-24 06:38:57,775 WARNING [train.py:1060] (2/4) Exclude cut with ID 7357-94126-0014-15845-0_sp1.1 from training. Duration: 25.3818125
2022-12-24 06:39:12,888 WARNING [train.py:1060] (2/4) Exclude cut with ID 6951-79737-0043-132310-0_sp0.9 from training. Duration: 28.0944375
2022-12-24 06:39:16,974 INFO [train.py:894] (2/4) Epoch 30, batch 2900, loss[loss=0.1688, simple_loss=0.2407, pruned_loss=0.04846, over 18541.00 frames. ], tot_loss[loss=0.1728, simple_loss=0.2573, pruned_loss=0.04415, over 3714633.90 frames. ], batch size: 44, lr: 3.78e-03, grad_scale: 8.0
2022-12-24 06:39:20,646 WARNING [train.py:1060] (2/4) Exclude cut with ID 2195-150901-0045-59933-0_sp0.9 from training. Duration: 22.9444375
2022-12-24 06:39:20,770 INFO [zipformer.py:660] (2/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=104580.0, num_to_drop=0, layers_to_drop=set()
2022-12-24 06:39:27,885 WARNING [train.py:1060] (2/4) Exclude cut with ID 8291-282929-0024-13011-0_sp1.1 from training. Duration: 21.6318125
2022-12-24 06:39:29,657 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.8104, 1.8421, 1.9817, 1.2328, 2.0371, 2.0602, 1.5476, 2.4120],
       device='cuda:2'), covar=tensor([0.1124, 0.1811, 0.1243, 0.1795, 0.0719, 0.1120, 0.2420, 0.0572],
       device='cuda:2'), in_proj_covar=tensor([0.0197, 0.0216, 0.0209, 0.0194, 0.0173, 0.0219, 0.0218, 0.0195],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-24 06:39:44,904 WARNING [train.py:1060] (2/4) Exclude cut with ID 8631-249866-0030-130156-0 from training. Duration: 23.695
2022-12-24 06:40:10,366 WARNING [train.py:1060] (2/4) Exclude cut with ID 7699-105389-0094-26379-0 from training. Duration: 23.955
2022-12-24 06:40:14,349 INFO [scaling.py:679] (2/4) Whitening: num_groups=8, num_channels=96, metric=1.15 vs. limit=2.0
2022-12-24 06:40:32,904 INFO [train.py:894] (2/4) Epoch 30, batch 2950, loss[loss=0.1719, simple_loss=0.2479, pruned_loss=0.04789, over 18674.00 frames. ], tot_loss[loss=0.1727, simple_loss=0.2573, pruned_loss=0.04404, over 3713747.06 frames. ], batch size: 46, lr: 3.78e-03, grad_scale: 8.0
2022-12-24 06:40:43,202 WARNING [train.py:1060] (2/4) Exclude cut with ID 8291-282929-0024-13011-0_sp0.9 from training. Duration: 26.438875
2022-12-24 06:40:59,832 INFO [zipformer.py:660] (2/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=104646.0, num_to_drop=0, layers_to_drop=set()
2022-12-24 06:41:09,936 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 2.578e+02 3.847e+02 4.605e+02 6.019e+02 2.047e+03, threshold=9.211e+02, percent-clipped=4.0
2022-12-24 06:41:10,316 INFO [zipformer.py:660] (2/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=104653.0, num_to_drop=0, layers_to_drop=set()
2022-12-24 06:41:28,280 WARNING [train.py:1060] (2/4) Exclude cut with ID 7699-105389-0021-26306-0_sp0.9 from training. Duration: 21.2444375
2022-12-24 06:41:28,301 WARNING [train.py:1060] (2/4) Exclude cut with ID 7357-94126-0014-15845-0_sp0.9 from training. Duration: 31.02225
2022-12-24 06:41:38,734 WARNING [train.py:1060] (2/4) Exclude cut with ID 432-122774-0017-62487-0 from training. Duration: 22.395
2022-12-24 06:41:47,953 INFO [train.py:894] (2/4) Epoch 30, batch 3000, loss[loss=0.153, simple_loss=0.2361, pruned_loss=0.03501, over 18622.00 frames. ], tot_loss[loss=0.1727, simple_loss=0.2571, pruned_loss=0.04414, over 3713665.56 frames. ], batch size: 45, lr: 3.78e-03, grad_scale: 8.0
2022-12-24 06:41:47,953 INFO [train.py:919] (2/4) Computing validation loss
2022-12-24 06:41:58,660 INFO [train.py:928] (2/4) Epoch 30, validation: loss=0.163, simple_loss=0.2585, pruned_loss=0.0338, over 944034.00 frames. 
2022-12-24 06:41:58,661 INFO [train.py:929] (2/4) Maximum memory allocated so far is 24680MB
2022-12-24 06:42:06,008 WARNING [train.py:1060] (2/4) Exclude cut with ID 7357-94126-0045-15876-0 from training. Duration: 21.075
2022-12-24 06:42:11,957 WARNING [train.py:1060] (2/4) Exclude cut with ID 6482-98857-0025-147532-0_sp0.9 from training. Duration: 20.0055625
2022-12-24 06:42:11,968 WARNING [train.py:1060] (2/4) Exclude cut with ID 6951-79737-0037-132304-0_sp0.9 from training. Duration: 22.05
2022-12-24 06:42:11,979 WARNING [train.py:1060] (2/4) Exclude cut with ID 7255-291500-0003-134302-0 from training. Duration: 26.8349375
2022-12-24 06:42:15,057 WARNING [train.py:1060] (2/4) Exclude cut with ID 7357-94126-0009-15840-0_sp1.1 from training. Duration: 22.1090625
2022-12-24 06:42:15,341 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([2.4439, 1.7187, 0.5255, 1.9091, 2.6479, 1.7719, 2.1071, 2.3014],
       device='cuda:2'), covar=tensor([0.1492, 0.2071, 0.2692, 0.1453, 0.1729, 0.1917, 0.1434, 0.1619],
       device='cuda:2'), in_proj_covar=tensor([0.0096, 0.0098, 0.0117, 0.0098, 0.0122, 0.0093, 0.0100, 0.0095],
       device='cuda:2'), out_proj_covar=tensor([0.0003, 0.0003, 0.0004, 0.0003, 0.0004, 0.0003, 0.0003, 0.0003],
       device='cuda:2')
2022-12-24 06:42:22,338 WARNING [train.py:1060] (2/4) Exclude cut with ID 7699-105389-0094-26379-0_sp0.9 from training. Duration: 26.6166875
2022-12-24 06:42:22,500 INFO [zipformer.py:660] (2/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=104694.0, num_to_drop=0, layers_to_drop=set()
2022-12-24 06:42:32,976 INFO [zipformer.py:660] (2/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=104701.0, num_to_drop=0, layers_to_drop=set()
2022-12-24 06:42:39,407 INFO [zipformer.py:660] (2/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=104705.0, num_to_drop=0, layers_to_drop=set()
2022-12-24 06:42:40,725 WARNING [train.py:1060] (2/4) Exclude cut with ID 2046-178027-0000-53705-0_sp0.9 from training. Duration: 20.3055625
2022-12-24 06:43:01,689 WARNING [train.py:1060] (2/4) Exclude cut with ID 7205-50138-0008-5373-0_sp0.9 from training. Duration: 20.7
2022-12-24 06:43:13,899 INFO [train.py:894] (2/4) Epoch 30, batch 3050, loss[loss=0.1664, simple_loss=0.2617, pruned_loss=0.03554, over 18477.00 frames. ], tot_loss[loss=0.1723, simple_loss=0.257, pruned_loss=0.0438, over 3713190.02 frames. ], batch size: 54, lr: 3.78e-03, grad_scale: 8.0
2022-12-24 06:43:44,623 WARNING [train.py:1060] (2/4) Exclude cut with ID 6978-92210-0019-146985-0 from training. Duration: 22.48
2022-12-24 06:43:48,769 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([1.6258, 1.8944, 1.4592, 2.1542, 2.4287, 1.6451, 1.4193, 1.3493],
       device='cuda:2'), covar=tensor([0.1934, 0.1844, 0.1817, 0.1077, 0.1307, 0.1131, 0.2320, 0.1612],
       device='cuda:2'), in_proj_covar=tensor([0.0255, 0.0237, 0.0228, 0.0207, 0.0269, 0.0203, 0.0233, 0.0207],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0003, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-24 06:43:51,975 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 2.496e+02 3.921e+02 4.500e+02 5.468e+02 9.473e+02, threshold=9.001e+02, percent-clipped=1.0
2022-12-24 06:43:52,161 INFO [zipformer.py:660] (2/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=104753.0, num_to_drop=0, layers_to_drop=set()
2022-12-24 06:43:59,249 WARNING [train.py:1060] (2/4) Exclude cut with ID 7255-291500-0003-134302-0_sp0.9 from training. Duration: 29.816625
2022-12-24 06:44:09,486 INFO [scaling.py:679] (2/4) Whitening: num_groups=8, num_channels=192, metric=1.47 vs. limit=2.0
2022-12-24 06:44:19,597 WARNING [train.py:1060] (2/4) Exclude cut with ID 7255-291500-0005-134304-0_sp1.1 from training. Duration: 22.7590625
2022-12-24 06:44:25,513 WARNING [train.py:1060] (2/4) Exclude cut with ID 7357-94126-0026-15857-0 from training. Duration: 22.555
2022-12-24 06:44:29,640 INFO [train.py:894] (2/4) Epoch 30, batch 3100, loss[loss=0.1537, simple_loss=0.2383, pruned_loss=0.03458, over 18433.00 frames. ], tot_loss[loss=0.1718, simple_loss=0.2568, pruned_loss=0.04347, over 3714273.98 frames. ], batch size: 48, lr: 3.78e-03, grad_scale: 8.0
2022-12-24 06:44:45,408 WARNING [train.py:1060] (2/4) Exclude cut with ID 1250-135782-0005-25975-0_sp0.9 from training. Duration: 21.688875
2022-12-24 06:45:18,396 WARNING [train.py:1060] (2/4) Exclude cut with ID 3488-85273-0038-41224-0_sp0.9 from training. Duration: 22.6
2022-12-24 06:45:44,263 INFO [train.py:894] (2/4) Epoch 30, batch 3150, loss[loss=0.1687, simple_loss=0.2427, pruned_loss=0.04733, over 18424.00 frames. ], tot_loss[loss=0.1723, simple_loss=0.2568, pruned_loss=0.04388, over 3713556.80 frames. ], batch size: 42, lr: 3.78e-03, grad_scale: 8.0
2022-12-24 06:45:55,320 WARNING [train.py:1060] (2/4) Exclude cut with ID 7357-94126-0009-15840-0 from training. Duration: 24.32
2022-12-24 06:46:21,956 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 2.960e+02 3.977e+02 4.589e+02 5.752e+02 1.230e+03, threshold=9.178e+02, percent-clipped=2.0
2022-12-24 06:46:32,278 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([0.7172, 2.0592, 2.2656, 1.2279, 1.5981, 2.4387, 2.2117, 1.8359],
       device='cuda:2'), covar=tensor([0.0889, 0.0406, 0.0329, 0.0466, 0.0405, 0.0514, 0.0290, 0.0874],
       device='cuda:2'), in_proj_covar=tensor([0.0150, 0.0175, 0.0133, 0.0143, 0.0147, 0.0145, 0.0167, 0.0182],
       device='cuda:2'), out_proj_covar=tensor([1.1238e-04, 1.3085e-04, 9.7472e-05, 1.0448e-04, 1.0737e-04, 1.0826e-04,
        1.2586e-04, 1.3636e-04], device='cuda:2')
2022-12-24 06:46:53,114 WARNING [train.py:1060] (2/4) Exclude cut with ID 8291-276745-0093-13116-0_sp0.9 from training. Duration: 21.061125
2022-12-24 06:46:59,065 INFO [train.py:894] (2/4) Epoch 30, batch 3200, loss[loss=0.1932, simple_loss=0.28, pruned_loss=0.05321, over 18657.00 frames. ], tot_loss[loss=0.1726, simple_loss=0.2568, pruned_loss=0.04423, over 3713910.14 frames. ], batch size: 97, lr: 3.78e-03, grad_scale: 8.0
2022-12-24 06:47:07,000 WARNING [train.py:1060] (2/4) Exclude cut with ID 7357-94126-0024-15855-0_sp0.9 from training. Duration: 20.32225
2022-12-24 06:47:20,247 WARNING [train.py:1060] (2/4) Exclude cut with ID 3033-130750-0096-55598-0_sp1.1 from training. Duration: 0.7545625
2022-12-24 06:47:25,843 INFO [scaling.py:679] (2/4) Whitening: num_groups=8, num_channels=192, metric=1.53 vs. limit=2.0
2022-12-24 06:47:34,984 WARNING [train.py:1060] (2/4) Exclude cut with ID 4295-39940-0007-92567-0_sp0.9 from training. Duration: 23.9333125
2022-12-24 06:48:07,351 WARNING [train.py:1060] (2/4) Exclude cut with ID 7255-291500-0008-134307-0_sp1.1 from training. Duration: 20.17275
2022-12-24 06:48:12,145 WARNING [train.py:1060] (2/4) Exclude cut with ID 6978-92210-0019-146985-0_sp1.1 from training. Duration: 20.436375
2022-12-24 06:48:13,574 INFO [train.py:894] (2/4) Epoch 30, batch 3250, loss[loss=0.1954, simple_loss=0.2798, pruned_loss=0.05553, over 18618.00 frames. ], tot_loss[loss=0.1721, simple_loss=0.2564, pruned_loss=0.04388, over 3715062.61 frames. ], batch size: 100, lr: 3.78e-03, grad_scale: 8.0
2022-12-24 06:48:47,923 INFO [scaling.py:679] (2/4) Whitening: num_groups=8, num_channels=96, metric=1.48 vs. limit=2.0
2022-12-24 06:48:51,166 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 2.506e+02 3.857e+02 4.754e+02 5.995e+02 1.031e+03, threshold=9.507e+02, percent-clipped=5.0
2022-12-24 06:49:27,997 INFO [scaling.py:679] (2/4) Whitening: num_groups=8, num_channels=192, metric=1.57 vs. limit=2.0
2022-12-24 06:49:28,777 INFO [train.py:894] (2/4) Epoch 30, batch 3300, loss[loss=0.1456, simple_loss=0.227, pruned_loss=0.03213, over 18398.00 frames. ], tot_loss[loss=0.1726, simple_loss=0.2567, pruned_loss=0.04421, over 3714145.20 frames. ], batch size: 46, lr: 3.78e-03, grad_scale: 8.0
2022-12-24 06:49:33,952 WARNING [train.py:1060] (2/4) Exclude cut with ID 4234-40345-0022-142709-0_sp0.9 from training. Duration: 23.1055625
2022-12-24 06:49:35,361 WARNING [train.py:1060] (2/4) Exclude cut with ID 8291-282929-0007-12994-0_sp1.1 from training. Duration: 23.5
2022-12-24 06:49:47,677 WARNING [train.py:1060] (2/4) Exclude cut with ID 7255-291500-0009-134308-0_sp0.9 from training. Duration: 26.62775
2022-12-24 06:49:59,187 WARNING [train.py:1060] (2/4) Exclude cut with ID 6951-79737-0018-132285-0 from training. Duration: 21.105
2022-12-24 06:50:04,285 WARNING [train.py:1060] (2/4) Exclude cut with ID 4511-76322-0006-80011-0_sp0.9 from training. Duration: 24.411125
2022-12-24 06:50:30,953 WARNING [train.py:1060] (2/4) Exclude cut with ID 6758-72288-0033-108368-0_sp1.1 from training. Duration: 21.263625
2022-12-24 06:50:45,221 INFO [train.py:894] (2/4) Epoch 30, batch 3350, loss[loss=0.1657, simple_loss=0.2625, pruned_loss=0.03444, over 18501.00 frames. ], tot_loss[loss=0.1724, simple_loss=0.2568, pruned_loss=0.044, over 3713942.23 frames. ], batch size: 52, lr: 3.77e-03, grad_scale: 8.0
2022-12-24 06:51:06,200 WARNING [train.py:1060] (2/4) Exclude cut with ID 4234-40345-0022-142709-0 from training. Duration: 20.795
2022-12-24 06:51:17,017 WARNING [train.py:1060] (2/4) Exclude cut with ID 7357-94126-0021-15852-0 from training. Duration: 24.76
2022-12-24 06:51:17,031 WARNING [train.py:1060] (2/4) Exclude cut with ID 3867-173237-0077-144769-0_sp0.9 from training. Duration: 22.25
2022-12-24 06:51:23,334 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 2.673e+02 4.112e+02 5.188e+02 6.129e+02 1.207e+03, threshold=1.038e+03, percent-clipped=1.0
2022-12-24 06:51:39,579 WARNING [train.py:1060] (2/4) Exclude cut with ID 7357-94126-0026-15857-0_sp1.1 from training. Duration: 20.5045625
2022-12-24 06:52:01,705 INFO [train.py:894] (2/4) Epoch 30, batch 3400, loss[loss=0.1834, simple_loss=0.2716, pruned_loss=0.04755, over 18665.00 frames. ], tot_loss[loss=0.1715, simple_loss=0.2557, pruned_loss=0.04363, over 3713974.79 frames. ], batch size: 181, lr: 3.77e-03, grad_scale: 8.0
2022-12-24 06:53:00,983 INFO [zipformer.py:1480] (2/4) attn_weights_entropy = tensor([2.4963, 2.3074, 2.0044, 1.4458, 2.7428, 2.6078, 2.2796, 1.9136],
       device='cuda:2'), covar=tensor([0.0429, 0.0496, 0.0528, 0.0775, 0.0317, 0.0422, 0.0503, 0.0910],
       device='cuda:2'), in_proj_covar=tensor([0.0125, 0.0132, 0.0129, 0.0118, 0.0106, 0.0128, 0.0134, 0.0163],
       device='cuda:2'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0001, 0.0001, 0.0002, 0.0002, 0.0002],
       device='cuda:2')
2022-12-24 06:53:12,287 INFO [train.py:894] (2/4) Epoch 30, batch 3450, loss[loss=0.1768, simple_loss=0.2692, pruned_loss=0.04224, over 18693.00 frames. ], tot_loss[loss=0.1715, simple_loss=0.2559, pruned_loss=0.04357, over 3714455.66 frames. ], batch size: 62, lr: 3.77e-03, grad_scale: 8.0
2022-12-24 06:53:49,269 INFO [optim.py:369] (2/4) Clipping_scale=2.0, grad-norm quartiles 2.749e+02 3.937e+02 4.807e+02 6.235e+02 1.177e+03, threshold=9.613e+02, percent-clipped=2.0
2022-12-24 06:54:25,725 INFO [train.py:894] (2/4) Epoch 30, batch 3500, loss[loss=0.1783, simple_loss=0.2638, pruned_loss=0.04643, over 18611.00 frames. ], tot_loss[loss=0.172, simple_loss=0.2567, pruned_loss=0.04369, over 3715066.67 frames. ], batch size: 176, lr: 3.77e-03, grad_scale: 8.0
2022-12-24 06:54:36,585 INFO [train.py:1158] (2/4) Done!