End of training
Browse files- README.md +116 -0
- all_results.json +12 -0
- config.json +52 -0
- eval_results.json +8 -0
- model.safetensors +3 -0
- preprocessor_config.json +22 -0
- runs/Dec01_18-39-04_DESKTOP-SKBE9FB/events.out.tfevents.1733099946.DESKTOP-SKBE9FB.5492.0 +3 -0
- runs/Dec01_18-39-04_DESKTOP-SKBE9FB/events.out.tfevents.1733100519.DESKTOP-SKBE9FB.5492.1 +3 -0
- train_results.json +7 -0
- trainer_state.json +1350 -0
- training_args.bin +3 -0
README.md
ADDED
|
@@ -0,0 +1,116 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
---
|
| 2 |
+
base_model: MBZUAI/swiftformer-xs
|
| 3 |
+
tags:
|
| 4 |
+
- generated_from_trainer
|
| 5 |
+
datasets:
|
| 6 |
+
- imagefolder
|
| 7 |
+
metrics:
|
| 8 |
+
- accuracy
|
| 9 |
+
model-index:
|
| 10 |
+
- name: swiftformer-xs-RD
|
| 11 |
+
results:
|
| 12 |
+
- task:
|
| 13 |
+
name: Image Classification
|
| 14 |
+
type: image-classification
|
| 15 |
+
dataset:
|
| 16 |
+
name: imagefolder
|
| 17 |
+
type: imagefolder
|
| 18 |
+
config: default
|
| 19 |
+
split: validation
|
| 20 |
+
args: default
|
| 21 |
+
metrics:
|
| 22 |
+
- name: Accuracy
|
| 23 |
+
type: accuracy
|
| 24 |
+
value: 0.8618181818181818
|
| 25 |
+
---
|
| 26 |
+
|
| 27 |
+
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
|
| 28 |
+
should probably proofread and complete it, then remove this comment. -->
|
| 29 |
+
|
| 30 |
+
# swiftformer-xs-RD
|
| 31 |
+
|
| 32 |
+
This model is a fine-tuned version of [MBZUAI/swiftformer-xs](https://huggingface.co/MBZUAI/swiftformer-xs) on the imagefolder dataset.
|
| 33 |
+
It achieves the following results on the evaluation set:
|
| 34 |
+
- Loss: 0.4005
|
| 35 |
+
- Accuracy: 0.8618
|
| 36 |
+
|
| 37 |
+
## Model description
|
| 38 |
+
|
| 39 |
+
More information needed
|
| 40 |
+
|
| 41 |
+
## Intended uses & limitations
|
| 42 |
+
|
| 43 |
+
More information needed
|
| 44 |
+
|
| 45 |
+
## Training and evaluation data
|
| 46 |
+
|
| 47 |
+
More information needed
|
| 48 |
+
|
| 49 |
+
## Training procedure
|
| 50 |
+
|
| 51 |
+
### Training hyperparameters
|
| 52 |
+
|
| 53 |
+
The following hyperparameters were used during training:
|
| 54 |
+
- learning_rate: 0.0003
|
| 55 |
+
- train_batch_size: 16
|
| 56 |
+
- eval_batch_size: 16
|
| 57 |
+
- seed: 42
|
| 58 |
+
- gradient_accumulation_steps: 4
|
| 59 |
+
- total_train_batch_size: 64
|
| 60 |
+
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
|
| 61 |
+
- lr_scheduler_type: linear
|
| 62 |
+
- lr_scheduler_warmup_ratio: 0.1
|
| 63 |
+
- num_epochs: 40
|
| 64 |
+
|
| 65 |
+
### Training results
|
| 66 |
+
|
| 67 |
+
| Training Loss | Epoch | Step | Validation Loss | Accuracy |
|
| 68 |
+
|:-------------:|:-----:|:----:|:---------------:|:--------:|
|
| 69 |
+
| 1.5571 | 0.99 | 40 | 1.5191 | 0.5182 |
|
| 70 |
+
| 0.979 | 1.99 | 80 | 0.8953 | 0.7091 |
|
| 71 |
+
| 0.8343 | 2.98 | 120 | 0.7520 | 0.7273 |
|
| 72 |
+
| 0.7755 | 4.0 | 161 | 0.7223 | 0.7527 |
|
| 73 |
+
| 0.7064 | 4.99 | 201 | 0.6760 | 0.74 |
|
| 74 |
+
| 0.6296 | 5.99 | 241 | 0.5624 | 0.8018 |
|
| 75 |
+
| 0.6387 | 6.98 | 281 | 0.5362 | 0.8073 |
|
| 76 |
+
| 0.5793 | 8.0 | 322 | 0.5242 | 0.7964 |
|
| 77 |
+
| 0.5723 | 8.99 | 362 | 0.4544 | 0.8309 |
|
| 78 |
+
| 0.5897 | 9.99 | 402 | 0.4510 | 0.8273 |
|
| 79 |
+
| 0.5443 | 10.98 | 442 | 0.4685 | 0.8055 |
|
| 80 |
+
| 0.4959 | 12.0 | 483 | 0.4248 | 0.8255 |
|
| 81 |
+
| 0.4743 | 12.99 | 523 | 0.4308 | 0.8309 |
|
| 82 |
+
| 0.4679 | 13.99 | 563 | 0.3999 | 0.84 |
|
| 83 |
+
| 0.4997 | 14.98 | 603 | 0.4151 | 0.8382 |
|
| 84 |
+
| 0.4602 | 16.0 | 644 | 0.4112 | 0.8218 |
|
| 85 |
+
| 0.4459 | 16.99 | 684 | 0.4196 | 0.8345 |
|
| 86 |
+
| 0.4668 | 17.99 | 724 | 0.4042 | 0.8291 |
|
| 87 |
+
| 0.41 | 18.98 | 764 | 0.4112 | 0.8436 |
|
| 88 |
+
| 0.4349 | 20.0 | 805 | 0.4120 | 0.8418 |
|
| 89 |
+
| 0.4558 | 20.99 | 845 | 0.4554 | 0.8218 |
|
| 90 |
+
| 0.3952 | 21.99 | 885 | 0.3775 | 0.8509 |
|
| 91 |
+
| 0.3395 | 22.98 | 925 | 0.4027 | 0.84 |
|
| 92 |
+
| 0.3525 | 24.0 | 966 | 0.3957 | 0.8509 |
|
| 93 |
+
| 0.3497 | 24.99 | 1006 | 0.4020 | 0.8527 |
|
| 94 |
+
| 0.3625 | 25.99 | 1046 | 0.3910 | 0.8564 |
|
| 95 |
+
| 0.3008 | 26.98 | 1086 | 0.3790 | 0.8582 |
|
| 96 |
+
| 0.2907 | 28.0 | 1127 | 0.3831 | 0.8564 |
|
| 97 |
+
| 0.3165 | 28.99 | 1167 | 0.4005 | 0.8618 |
|
| 98 |
+
| 0.3231 | 29.99 | 1207 | 0.4512 | 0.8327 |
|
| 99 |
+
| 0.2819 | 30.98 | 1247 | 0.4228 | 0.8455 |
|
| 100 |
+
| 0.2704 | 32.0 | 1288 | 0.4074 | 0.86 |
|
| 101 |
+
| 0.2429 | 32.99 | 1328 | 0.4405 | 0.8545 |
|
| 102 |
+
| 0.2421 | 33.99 | 1368 | 0.4337 | 0.8527 |
|
| 103 |
+
| 0.3039 | 34.98 | 1408 | 0.4628 | 0.8473 |
|
| 104 |
+
| 0.2677 | 36.0 | 1449 | 0.4411 | 0.8545 |
|
| 105 |
+
| 0.2171 | 36.99 | 1489 | 0.4755 | 0.8618 |
|
| 106 |
+
| 0.2268 | 37.99 | 1529 | 0.4506 | 0.86 |
|
| 107 |
+
| 0.2378 | 38.98 | 1569 | 0.4633 | 0.8527 |
|
| 108 |
+
| 0.2021 | 39.75 | 1600 | 0.4492 | 0.8564 |
|
| 109 |
+
|
| 110 |
+
|
| 111 |
+
### Framework versions
|
| 112 |
+
|
| 113 |
+
- Transformers 4.36.2
|
| 114 |
+
- Pytorch 2.1.2+cu118
|
| 115 |
+
- Datasets 2.16.1
|
| 116 |
+
- Tokenizers 0.15.0
|
all_results.json
ADDED
|
@@ -0,0 +1,12 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 39.75,
|
| 3 |
+
"eval_accuracy": 0.8618181818181818,
|
| 4 |
+
"eval_loss": 0.4005330204963684,
|
| 5 |
+
"eval_runtime": 1.3983,
|
| 6 |
+
"eval_samples_per_second": 393.325,
|
| 7 |
+
"eval_steps_per_second": 25.03,
|
| 8 |
+
"train_loss": 0.4774747739732266,
|
| 9 |
+
"train_runtime": 572.0649,
|
| 10 |
+
"train_samples_per_second": 179.21,
|
| 11 |
+
"train_steps_per_second": 2.797
|
| 12 |
+
}
|
config.json
ADDED
|
@@ -0,0 +1,52 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_name_or_path": "MBZUAI/swiftformer-xs",
|
| 3 |
+
"architectures": [
|
| 4 |
+
"SwiftFormerForImageClassification"
|
| 5 |
+
],
|
| 6 |
+
"batch_norm_eps": 1e-05,
|
| 7 |
+
"depths": [
|
| 8 |
+
3,
|
| 9 |
+
3,
|
| 10 |
+
6,
|
| 11 |
+
4
|
| 12 |
+
],
|
| 13 |
+
"down_pad": 1,
|
| 14 |
+
"down_patch_size": 3,
|
| 15 |
+
"down_stride": 2,
|
| 16 |
+
"downsamples": [
|
| 17 |
+
true,
|
| 18 |
+
true,
|
| 19 |
+
true,
|
| 20 |
+
true
|
| 21 |
+
],
|
| 22 |
+
"drop_path_rate": 0.0,
|
| 23 |
+
"embed_dims": [
|
| 24 |
+
48,
|
| 25 |
+
56,
|
| 26 |
+
112,
|
| 27 |
+
220
|
| 28 |
+
],
|
| 29 |
+
"hidden_act": "gelu",
|
| 30 |
+
"id2label": {
|
| 31 |
+
"0": "Mild",
|
| 32 |
+
"1": "Moderate",
|
| 33 |
+
"2": "No_DR",
|
| 34 |
+
"3": "Proliferate_DR",
|
| 35 |
+
"4": "Severe"
|
| 36 |
+
},
|
| 37 |
+
"label2id": {
|
| 38 |
+
"Mild": 0,
|
| 39 |
+
"Moderate": 1,
|
| 40 |
+
"No_DR": 2,
|
| 41 |
+
"Proliferate_DR": 3,
|
| 42 |
+
"Severe": 4
|
| 43 |
+
},
|
| 44 |
+
"layer_scale_init_value": 1e-05,
|
| 45 |
+
"mlp_ratio": 4,
|
| 46 |
+
"model_type": "swiftformer",
|
| 47 |
+
"num_channels": 3,
|
| 48 |
+
"problem_type": "single_label_classification",
|
| 49 |
+
"torch_dtype": "float32",
|
| 50 |
+
"transformers_version": "4.36.2",
|
| 51 |
+
"use_layer_scale": true
|
| 52 |
+
}
|
eval_results.json
ADDED
|
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 39.75,
|
| 3 |
+
"eval_accuracy": 0.8618181818181818,
|
| 4 |
+
"eval_loss": 0.4005330204963684,
|
| 5 |
+
"eval_runtime": 1.3983,
|
| 6 |
+
"eval_samples_per_second": 393.325,
|
| 7 |
+
"eval_steps_per_second": 25.03
|
| 8 |
+
}
|
model.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:df752bc48fa2c9186ad5adc73347b9ce8b0043698128e0370cc83326177670af
|
| 3 |
+
size 12205416
|
preprocessor_config.json
ADDED
|
@@ -0,0 +1,22 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"do_normalize": true,
|
| 3 |
+
"do_rescale": true,
|
| 4 |
+
"do_resize": true,
|
| 5 |
+
"image_mean": [
|
| 6 |
+
0.485,
|
| 7 |
+
0.456,
|
| 8 |
+
0.406
|
| 9 |
+
],
|
| 10 |
+
"image_processor_type": "ViTImageProcessor",
|
| 11 |
+
"image_std": [
|
| 12 |
+
0.229,
|
| 13 |
+
0.224,
|
| 14 |
+
0.225
|
| 15 |
+
],
|
| 16 |
+
"resample": 2,
|
| 17 |
+
"rescale_factor": 0.00392156862745098,
|
| 18 |
+
"size": {
|
| 19 |
+
"height": 224,
|
| 20 |
+
"width": 224
|
| 21 |
+
}
|
| 22 |
+
}
|
runs/Dec01_18-39-04_DESKTOP-SKBE9FB/events.out.tfevents.1733099946.DESKTOP-SKBE9FB.5492.0
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:2db7ee74c6a19cba67e2ec1b72d6e19f54794ed477ce81b1c5686ce6bd357b54
|
| 3 |
+
size 42921
|
runs/Dec01_18-39-04_DESKTOP-SKBE9FB/events.out.tfevents.1733100519.DESKTOP-SKBE9FB.5492.1
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:0c2e5cfcf4ca68ee721df4d23426044936546fd5f055c6c1aff81387461621d4
|
| 3 |
+
size 411
|
train_results.json
ADDED
|
@@ -0,0 +1,7 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 39.75,
|
| 3 |
+
"train_loss": 0.4774747739732266,
|
| 4 |
+
"train_runtime": 572.0649,
|
| 5 |
+
"train_samples_per_second": 179.21,
|
| 6 |
+
"train_steps_per_second": 2.797
|
| 7 |
+
}
|
trainer_state.json
ADDED
|
@@ -0,0 +1,1350 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": 0.8618181818181818,
|
| 3 |
+
"best_model_checkpoint": "swiftformer-xs-RD\\checkpoint-1167",
|
| 4 |
+
"epoch": 39.75155279503105,
|
| 5 |
+
"eval_steps": 500,
|
| 6 |
+
"global_step": 1600,
|
| 7 |
+
"is_hyper_param_search": false,
|
| 8 |
+
"is_local_process_zero": true,
|
| 9 |
+
"is_world_process_zero": true,
|
| 10 |
+
"log_history": [
|
| 11 |
+
{
|
| 12 |
+
"epoch": 0.25,
|
| 13 |
+
"learning_rate": 1.875e-05,
|
| 14 |
+
"loss": 1.6091,
|
| 15 |
+
"step": 10
|
| 16 |
+
},
|
| 17 |
+
{
|
| 18 |
+
"epoch": 0.5,
|
| 19 |
+
"learning_rate": 3.75e-05,
|
| 20 |
+
"loss": 1.6066,
|
| 21 |
+
"step": 20
|
| 22 |
+
},
|
| 23 |
+
{
|
| 24 |
+
"epoch": 0.75,
|
| 25 |
+
"learning_rate": 5.625e-05,
|
| 26 |
+
"loss": 1.5954,
|
| 27 |
+
"step": 30
|
| 28 |
+
},
|
| 29 |
+
{
|
| 30 |
+
"epoch": 0.99,
|
| 31 |
+
"learning_rate": 7.5e-05,
|
| 32 |
+
"loss": 1.5571,
|
| 33 |
+
"step": 40
|
| 34 |
+
},
|
| 35 |
+
{
|
| 36 |
+
"epoch": 0.99,
|
| 37 |
+
"eval_accuracy": 0.5181818181818182,
|
| 38 |
+
"eval_loss": 1.519059658050537,
|
| 39 |
+
"eval_runtime": 2.8787,
|
| 40 |
+
"eval_samples_per_second": 191.057,
|
| 41 |
+
"eval_steps_per_second": 12.158,
|
| 42 |
+
"step": 40
|
| 43 |
+
},
|
| 44 |
+
{
|
| 45 |
+
"epoch": 1.24,
|
| 46 |
+
"learning_rate": 9.374999999999999e-05,
|
| 47 |
+
"loss": 1.4441,
|
| 48 |
+
"step": 50
|
| 49 |
+
},
|
| 50 |
+
{
|
| 51 |
+
"epoch": 1.49,
|
| 52 |
+
"learning_rate": 0.0001125,
|
| 53 |
+
"loss": 1.255,
|
| 54 |
+
"step": 60
|
| 55 |
+
},
|
| 56 |
+
{
|
| 57 |
+
"epoch": 1.74,
|
| 58 |
+
"learning_rate": 0.00013125,
|
| 59 |
+
"loss": 1.0283,
|
| 60 |
+
"step": 70
|
| 61 |
+
},
|
| 62 |
+
{
|
| 63 |
+
"epoch": 1.99,
|
| 64 |
+
"learning_rate": 0.00015,
|
| 65 |
+
"loss": 0.979,
|
| 66 |
+
"step": 80
|
| 67 |
+
},
|
| 68 |
+
{
|
| 69 |
+
"epoch": 1.99,
|
| 70 |
+
"eval_accuracy": 0.7090909090909091,
|
| 71 |
+
"eval_loss": 0.8952864408493042,
|
| 72 |
+
"eval_runtime": 1.7304,
|
| 73 |
+
"eval_samples_per_second": 317.84,
|
| 74 |
+
"eval_steps_per_second": 20.226,
|
| 75 |
+
"step": 80
|
| 76 |
+
},
|
| 77 |
+
{
|
| 78 |
+
"epoch": 2.24,
|
| 79 |
+
"learning_rate": 0.00016874999999999998,
|
| 80 |
+
"loss": 0.864,
|
| 81 |
+
"step": 90
|
| 82 |
+
},
|
| 83 |
+
{
|
| 84 |
+
"epoch": 2.48,
|
| 85 |
+
"learning_rate": 0.00018749999999999998,
|
| 86 |
+
"loss": 0.859,
|
| 87 |
+
"step": 100
|
| 88 |
+
},
|
| 89 |
+
{
|
| 90 |
+
"epoch": 2.73,
|
| 91 |
+
"learning_rate": 0.00020624999999999997,
|
| 92 |
+
"loss": 0.7978,
|
| 93 |
+
"step": 110
|
| 94 |
+
},
|
| 95 |
+
{
|
| 96 |
+
"epoch": 2.98,
|
| 97 |
+
"learning_rate": 0.000225,
|
| 98 |
+
"loss": 0.8343,
|
| 99 |
+
"step": 120
|
| 100 |
+
},
|
| 101 |
+
{
|
| 102 |
+
"epoch": 2.98,
|
| 103 |
+
"eval_accuracy": 0.7272727272727273,
|
| 104 |
+
"eval_loss": 0.7519798874855042,
|
| 105 |
+
"eval_runtime": 1.3968,
|
| 106 |
+
"eval_samples_per_second": 393.746,
|
| 107 |
+
"eval_steps_per_second": 25.057,
|
| 108 |
+
"step": 120
|
| 109 |
+
},
|
| 110 |
+
{
|
| 111 |
+
"epoch": 3.23,
|
| 112 |
+
"learning_rate": 0.00024375,
|
| 113 |
+
"loss": 0.7879,
|
| 114 |
+
"step": 130
|
| 115 |
+
},
|
| 116 |
+
{
|
| 117 |
+
"epoch": 3.48,
|
| 118 |
+
"learning_rate": 0.0002625,
|
| 119 |
+
"loss": 0.7814,
|
| 120 |
+
"step": 140
|
| 121 |
+
},
|
| 122 |
+
{
|
| 123 |
+
"epoch": 3.73,
|
| 124 |
+
"learning_rate": 0.00028125,
|
| 125 |
+
"loss": 0.7806,
|
| 126 |
+
"step": 150
|
| 127 |
+
},
|
| 128 |
+
{
|
| 129 |
+
"epoch": 3.98,
|
| 130 |
+
"learning_rate": 0.0003,
|
| 131 |
+
"loss": 0.7755,
|
| 132 |
+
"step": 160
|
| 133 |
+
},
|
| 134 |
+
{
|
| 135 |
+
"epoch": 4.0,
|
| 136 |
+
"eval_accuracy": 0.7527272727272727,
|
| 137 |
+
"eval_loss": 0.7222614288330078,
|
| 138 |
+
"eval_runtime": 1.3418,
|
| 139 |
+
"eval_samples_per_second": 409.89,
|
| 140 |
+
"eval_steps_per_second": 26.084,
|
| 141 |
+
"step": 161
|
| 142 |
+
},
|
| 143 |
+
{
|
| 144 |
+
"epoch": 4.22,
|
| 145 |
+
"learning_rate": 0.00029791666666666665,
|
| 146 |
+
"loss": 0.7536,
|
| 147 |
+
"step": 170
|
| 148 |
+
},
|
| 149 |
+
{
|
| 150 |
+
"epoch": 4.47,
|
| 151 |
+
"learning_rate": 0.00029583333333333333,
|
| 152 |
+
"loss": 0.6664,
|
| 153 |
+
"step": 180
|
| 154 |
+
},
|
| 155 |
+
{
|
| 156 |
+
"epoch": 4.72,
|
| 157 |
+
"learning_rate": 0.00029374999999999996,
|
| 158 |
+
"loss": 0.6218,
|
| 159 |
+
"step": 190
|
| 160 |
+
},
|
| 161 |
+
{
|
| 162 |
+
"epoch": 4.97,
|
| 163 |
+
"learning_rate": 0.00029166666666666664,
|
| 164 |
+
"loss": 0.7064,
|
| 165 |
+
"step": 200
|
| 166 |
+
},
|
| 167 |
+
{
|
| 168 |
+
"epoch": 4.99,
|
| 169 |
+
"eval_accuracy": 0.74,
|
| 170 |
+
"eval_loss": 0.6760307550430298,
|
| 171 |
+
"eval_runtime": 1.3443,
|
| 172 |
+
"eval_samples_per_second": 409.125,
|
| 173 |
+
"eval_steps_per_second": 26.035,
|
| 174 |
+
"step": 201
|
| 175 |
+
},
|
| 176 |
+
{
|
| 177 |
+
"epoch": 5.22,
|
| 178 |
+
"learning_rate": 0.0002895833333333333,
|
| 179 |
+
"loss": 0.715,
|
| 180 |
+
"step": 210
|
| 181 |
+
},
|
| 182 |
+
{
|
| 183 |
+
"epoch": 5.47,
|
| 184 |
+
"learning_rate": 0.0002875,
|
| 185 |
+
"loss": 0.7128,
|
| 186 |
+
"step": 220
|
| 187 |
+
},
|
| 188 |
+
{
|
| 189 |
+
"epoch": 5.71,
|
| 190 |
+
"learning_rate": 0.0002854166666666666,
|
| 191 |
+
"loss": 0.6436,
|
| 192 |
+
"step": 230
|
| 193 |
+
},
|
| 194 |
+
{
|
| 195 |
+
"epoch": 5.96,
|
| 196 |
+
"learning_rate": 0.0002833333333333333,
|
| 197 |
+
"loss": 0.6296,
|
| 198 |
+
"step": 240
|
| 199 |
+
},
|
| 200 |
+
{
|
| 201 |
+
"epoch": 5.99,
|
| 202 |
+
"eval_accuracy": 0.8018181818181818,
|
| 203 |
+
"eval_loss": 0.5624179840087891,
|
| 204 |
+
"eval_runtime": 1.5624,
|
| 205 |
+
"eval_samples_per_second": 352.03,
|
| 206 |
+
"eval_steps_per_second": 22.402,
|
| 207 |
+
"step": 241
|
| 208 |
+
},
|
| 209 |
+
{
|
| 210 |
+
"epoch": 6.21,
|
| 211 |
+
"learning_rate": 0.00028125,
|
| 212 |
+
"loss": 0.6154,
|
| 213 |
+
"step": 250
|
| 214 |
+
},
|
| 215 |
+
{
|
| 216 |
+
"epoch": 6.46,
|
| 217 |
+
"learning_rate": 0.00027916666666666666,
|
| 218 |
+
"loss": 0.6749,
|
| 219 |
+
"step": 260
|
| 220 |
+
},
|
| 221 |
+
{
|
| 222 |
+
"epoch": 6.71,
|
| 223 |
+
"learning_rate": 0.00027708333333333334,
|
| 224 |
+
"loss": 0.6336,
|
| 225 |
+
"step": 270
|
| 226 |
+
},
|
| 227 |
+
{
|
| 228 |
+
"epoch": 6.96,
|
| 229 |
+
"learning_rate": 0.00027499999999999996,
|
| 230 |
+
"loss": 0.6387,
|
| 231 |
+
"step": 280
|
| 232 |
+
},
|
| 233 |
+
{
|
| 234 |
+
"epoch": 6.98,
|
| 235 |
+
"eval_accuracy": 0.8072727272727273,
|
| 236 |
+
"eval_loss": 0.5362208485603333,
|
| 237 |
+
"eval_runtime": 1.4093,
|
| 238 |
+
"eval_samples_per_second": 390.255,
|
| 239 |
+
"eval_steps_per_second": 24.834,
|
| 240 |
+
"step": 281
|
| 241 |
+
},
|
| 242 |
+
{
|
| 243 |
+
"epoch": 7.2,
|
| 244 |
+
"learning_rate": 0.00027291666666666664,
|
| 245 |
+
"loss": 0.5918,
|
| 246 |
+
"step": 290
|
| 247 |
+
},
|
| 248 |
+
{
|
| 249 |
+
"epoch": 7.45,
|
| 250 |
+
"learning_rate": 0.0002708333333333333,
|
| 251 |
+
"loss": 0.5901,
|
| 252 |
+
"step": 300
|
| 253 |
+
},
|
| 254 |
+
{
|
| 255 |
+
"epoch": 7.7,
|
| 256 |
+
"learning_rate": 0.00026875,
|
| 257 |
+
"loss": 0.6415,
|
| 258 |
+
"step": 310
|
| 259 |
+
},
|
| 260 |
+
{
|
| 261 |
+
"epoch": 7.95,
|
| 262 |
+
"learning_rate": 0.0002666666666666666,
|
| 263 |
+
"loss": 0.5793,
|
| 264 |
+
"step": 320
|
| 265 |
+
},
|
| 266 |
+
{
|
| 267 |
+
"epoch": 8.0,
|
| 268 |
+
"eval_accuracy": 0.7963636363636364,
|
| 269 |
+
"eval_loss": 0.5241875648498535,
|
| 270 |
+
"eval_runtime": 1.3758,
|
| 271 |
+
"eval_samples_per_second": 399.758,
|
| 272 |
+
"eval_steps_per_second": 25.439,
|
| 273 |
+
"step": 322
|
| 274 |
+
},
|
| 275 |
+
{
|
| 276 |
+
"epoch": 8.2,
|
| 277 |
+
"learning_rate": 0.0002645833333333333,
|
| 278 |
+
"loss": 0.5718,
|
| 279 |
+
"step": 330
|
| 280 |
+
},
|
| 281 |
+
{
|
| 282 |
+
"epoch": 8.45,
|
| 283 |
+
"learning_rate": 0.0002625,
|
| 284 |
+
"loss": 0.5651,
|
| 285 |
+
"step": 340
|
| 286 |
+
},
|
| 287 |
+
{
|
| 288 |
+
"epoch": 8.7,
|
| 289 |
+
"learning_rate": 0.00026041666666666666,
|
| 290 |
+
"loss": 0.5828,
|
| 291 |
+
"step": 350
|
| 292 |
+
},
|
| 293 |
+
{
|
| 294 |
+
"epoch": 8.94,
|
| 295 |
+
"learning_rate": 0.00025833333333333334,
|
| 296 |
+
"loss": 0.5723,
|
| 297 |
+
"step": 360
|
| 298 |
+
},
|
| 299 |
+
{
|
| 300 |
+
"epoch": 8.99,
|
| 301 |
+
"eval_accuracy": 0.8309090909090909,
|
| 302 |
+
"eval_loss": 0.454416960477829,
|
| 303 |
+
"eval_runtime": 1.3498,
|
| 304 |
+
"eval_samples_per_second": 407.461,
|
| 305 |
+
"eval_steps_per_second": 25.929,
|
| 306 |
+
"step": 362
|
| 307 |
+
},
|
| 308 |
+
{
|
| 309 |
+
"epoch": 9.19,
|
| 310 |
+
"learning_rate": 0.00025624999999999997,
|
| 311 |
+
"loss": 0.5246,
|
| 312 |
+
"step": 370
|
| 313 |
+
},
|
| 314 |
+
{
|
| 315 |
+
"epoch": 9.44,
|
| 316 |
+
"learning_rate": 0.00025416666666666665,
|
| 317 |
+
"loss": 0.5618,
|
| 318 |
+
"step": 380
|
| 319 |
+
},
|
| 320 |
+
{
|
| 321 |
+
"epoch": 9.69,
|
| 322 |
+
"learning_rate": 0.0002520833333333333,
|
| 323 |
+
"loss": 0.5462,
|
| 324 |
+
"step": 390
|
| 325 |
+
},
|
| 326 |
+
{
|
| 327 |
+
"epoch": 9.94,
|
| 328 |
+
"learning_rate": 0.00025,
|
| 329 |
+
"loss": 0.5897,
|
| 330 |
+
"step": 400
|
| 331 |
+
},
|
| 332 |
+
{
|
| 333 |
+
"epoch": 9.99,
|
| 334 |
+
"eval_accuracy": 0.8272727272727273,
|
| 335 |
+
"eval_loss": 0.4509519636631012,
|
| 336 |
+
"eval_runtime": 1.3383,
|
| 337 |
+
"eval_samples_per_second": 410.963,
|
| 338 |
+
"eval_steps_per_second": 26.152,
|
| 339 |
+
"step": 402
|
| 340 |
+
},
|
| 341 |
+
{
|
| 342 |
+
"epoch": 10.19,
|
| 343 |
+
"learning_rate": 0.00024791666666666663,
|
| 344 |
+
"loss": 0.5194,
|
| 345 |
+
"step": 410
|
| 346 |
+
},
|
| 347 |
+
{
|
| 348 |
+
"epoch": 10.43,
|
| 349 |
+
"learning_rate": 0.0002458333333333333,
|
| 350 |
+
"loss": 0.4946,
|
| 351 |
+
"step": 420
|
| 352 |
+
},
|
| 353 |
+
{
|
| 354 |
+
"epoch": 10.68,
|
| 355 |
+
"learning_rate": 0.00024375,
|
| 356 |
+
"loss": 0.5858,
|
| 357 |
+
"step": 430
|
| 358 |
+
},
|
| 359 |
+
{
|
| 360 |
+
"epoch": 10.93,
|
| 361 |
+
"learning_rate": 0.00024166666666666664,
|
| 362 |
+
"loss": 0.5443,
|
| 363 |
+
"step": 440
|
| 364 |
+
},
|
| 365 |
+
{
|
| 366 |
+
"epoch": 10.98,
|
| 367 |
+
"eval_accuracy": 0.8054545454545454,
|
| 368 |
+
"eval_loss": 0.46854037046432495,
|
| 369 |
+
"eval_runtime": 1.3393,
|
| 370 |
+
"eval_samples_per_second": 410.65,
|
| 371 |
+
"eval_steps_per_second": 26.132,
|
| 372 |
+
"step": 442
|
| 373 |
+
},
|
| 374 |
+
{
|
| 375 |
+
"epoch": 11.18,
|
| 376 |
+
"learning_rate": 0.00023958333333333332,
|
| 377 |
+
"loss": 0.5482,
|
| 378 |
+
"step": 450
|
| 379 |
+
},
|
| 380 |
+
{
|
| 381 |
+
"epoch": 11.43,
|
| 382 |
+
"learning_rate": 0.00023749999999999997,
|
| 383 |
+
"loss": 0.4724,
|
| 384 |
+
"step": 460
|
| 385 |
+
},
|
| 386 |
+
{
|
| 387 |
+
"epoch": 11.68,
|
| 388 |
+
"learning_rate": 0.00023541666666666665,
|
| 389 |
+
"loss": 0.5285,
|
| 390 |
+
"step": 470
|
| 391 |
+
},
|
| 392 |
+
{
|
| 393 |
+
"epoch": 11.93,
|
| 394 |
+
"learning_rate": 0.0002333333333333333,
|
| 395 |
+
"loss": 0.4959,
|
| 396 |
+
"step": 480
|
| 397 |
+
},
|
| 398 |
+
{
|
| 399 |
+
"epoch": 12.0,
|
| 400 |
+
"eval_accuracy": 0.8254545454545454,
|
| 401 |
+
"eval_loss": 0.42480796575546265,
|
| 402 |
+
"eval_runtime": 1.4113,
|
| 403 |
+
"eval_samples_per_second": 389.699,
|
| 404 |
+
"eval_steps_per_second": 24.799,
|
| 405 |
+
"step": 483
|
| 406 |
+
},
|
| 407 |
+
{
|
| 408 |
+
"epoch": 12.17,
|
| 409 |
+
"learning_rate": 0.00023124999999999998,
|
| 410 |
+
"loss": 0.6238,
|
| 411 |
+
"step": 490
|
| 412 |
+
},
|
| 413 |
+
{
|
| 414 |
+
"epoch": 12.42,
|
| 415 |
+
"learning_rate": 0.00022916666666666664,
|
| 416 |
+
"loss": 0.5296,
|
| 417 |
+
"step": 500
|
| 418 |
+
},
|
| 419 |
+
{
|
| 420 |
+
"epoch": 12.67,
|
| 421 |
+
"learning_rate": 0.00022708333333333331,
|
| 422 |
+
"loss": 0.4759,
|
| 423 |
+
"step": 510
|
| 424 |
+
},
|
| 425 |
+
{
|
| 426 |
+
"epoch": 12.92,
|
| 427 |
+
"learning_rate": 0.000225,
|
| 428 |
+
"loss": 0.4743,
|
| 429 |
+
"step": 520
|
| 430 |
+
},
|
| 431 |
+
{
|
| 432 |
+
"epoch": 12.99,
|
| 433 |
+
"eval_accuracy": 0.8309090909090909,
|
| 434 |
+
"eval_loss": 0.4307559132575989,
|
| 435 |
+
"eval_runtime": 1.3628,
|
| 436 |
+
"eval_samples_per_second": 403.57,
|
| 437 |
+
"eval_steps_per_second": 25.682,
|
| 438 |
+
"step": 523
|
| 439 |
+
},
|
| 440 |
+
{
|
| 441 |
+
"epoch": 13.17,
|
| 442 |
+
"learning_rate": 0.00022291666666666665,
|
| 443 |
+
"loss": 0.6277,
|
| 444 |
+
"step": 530
|
| 445 |
+
},
|
| 446 |
+
{
|
| 447 |
+
"epoch": 13.42,
|
| 448 |
+
"learning_rate": 0.00022083333333333333,
|
| 449 |
+
"loss": 0.5356,
|
| 450 |
+
"step": 540
|
| 451 |
+
},
|
| 452 |
+
{
|
| 453 |
+
"epoch": 13.66,
|
| 454 |
+
"learning_rate": 0.00021874999999999998,
|
| 455 |
+
"loss": 0.5112,
|
| 456 |
+
"step": 550
|
| 457 |
+
},
|
| 458 |
+
{
|
| 459 |
+
"epoch": 13.91,
|
| 460 |
+
"learning_rate": 0.00021666666666666666,
|
| 461 |
+
"loss": 0.4679,
|
| 462 |
+
"step": 560
|
| 463 |
+
},
|
| 464 |
+
{
|
| 465 |
+
"epoch": 13.99,
|
| 466 |
+
"eval_accuracy": 0.84,
|
| 467 |
+
"eval_loss": 0.39991825819015503,
|
| 468 |
+
"eval_runtime": 1.3308,
|
| 469 |
+
"eval_samples_per_second": 413.283,
|
| 470 |
+
"eval_steps_per_second": 26.3,
|
| 471 |
+
"step": 563
|
| 472 |
+
},
|
| 473 |
+
{
|
| 474 |
+
"epoch": 14.16,
|
| 475 |
+
"learning_rate": 0.0002145833333333333,
|
| 476 |
+
"loss": 0.4224,
|
| 477 |
+
"step": 570
|
| 478 |
+
},
|
| 479 |
+
{
|
| 480 |
+
"epoch": 14.41,
|
| 481 |
+
"learning_rate": 0.0002125,
|
| 482 |
+
"loss": 0.4699,
|
| 483 |
+
"step": 580
|
| 484 |
+
},
|
| 485 |
+
{
|
| 486 |
+
"epoch": 14.66,
|
| 487 |
+
"learning_rate": 0.00021041666666666664,
|
| 488 |
+
"loss": 0.4727,
|
| 489 |
+
"step": 590
|
| 490 |
+
},
|
| 491 |
+
{
|
| 492 |
+
"epoch": 14.91,
|
| 493 |
+
"learning_rate": 0.00020833333333333332,
|
| 494 |
+
"loss": 0.4997,
|
| 495 |
+
"step": 600
|
| 496 |
+
},
|
| 497 |
+
{
|
| 498 |
+
"epoch": 14.98,
|
| 499 |
+
"eval_accuracy": 0.8381818181818181,
|
| 500 |
+
"eval_loss": 0.41513749957084656,
|
| 501 |
+
"eval_runtime": 1.3563,
|
| 502 |
+
"eval_samples_per_second": 405.507,
|
| 503 |
+
"eval_steps_per_second": 25.805,
|
| 504 |
+
"step": 603
|
| 505 |
+
},
|
| 506 |
+
{
|
| 507 |
+
"epoch": 15.16,
|
| 508 |
+
"learning_rate": 0.00020624999999999997,
|
| 509 |
+
"loss": 0.4647,
|
| 510 |
+
"step": 610
|
| 511 |
+
},
|
| 512 |
+
{
|
| 513 |
+
"epoch": 15.4,
|
| 514 |
+
"learning_rate": 0.00020416666666666665,
|
| 515 |
+
"loss": 0.507,
|
| 516 |
+
"step": 620
|
| 517 |
+
},
|
| 518 |
+
{
|
| 519 |
+
"epoch": 15.65,
|
| 520 |
+
"learning_rate": 0.00020208333333333333,
|
| 521 |
+
"loss": 0.4575,
|
| 522 |
+
"step": 630
|
| 523 |
+
},
|
| 524 |
+
{
|
| 525 |
+
"epoch": 15.9,
|
| 526 |
+
"learning_rate": 0.00019999999999999998,
|
| 527 |
+
"loss": 0.4602,
|
| 528 |
+
"step": 640
|
| 529 |
+
},
|
| 530 |
+
{
|
| 531 |
+
"epoch": 16.0,
|
| 532 |
+
"eval_accuracy": 0.8218181818181818,
|
| 533 |
+
"eval_loss": 0.41115185618400574,
|
| 534 |
+
"eval_runtime": 1.3393,
|
| 535 |
+
"eval_samples_per_second": 410.657,
|
| 536 |
+
"eval_steps_per_second": 26.133,
|
| 537 |
+
"step": 644
|
| 538 |
+
},
|
| 539 |
+
{
|
| 540 |
+
"epoch": 16.15,
|
| 541 |
+
"learning_rate": 0.00019791666666666663,
|
| 542 |
+
"loss": 0.4685,
|
| 543 |
+
"step": 650
|
| 544 |
+
},
|
| 545 |
+
{
|
| 546 |
+
"epoch": 16.4,
|
| 547 |
+
"learning_rate": 0.00019583333333333331,
|
| 548 |
+
"loss": 0.4071,
|
| 549 |
+
"step": 660
|
| 550 |
+
},
|
| 551 |
+
{
|
| 552 |
+
"epoch": 16.65,
|
| 553 |
+
"learning_rate": 0.00019375,
|
| 554 |
+
"loss": 0.4358,
|
| 555 |
+
"step": 670
|
| 556 |
+
},
|
| 557 |
+
{
|
| 558 |
+
"epoch": 16.89,
|
| 559 |
+
"learning_rate": 0.00019166666666666665,
|
| 560 |
+
"loss": 0.4459,
|
| 561 |
+
"step": 680
|
| 562 |
+
},
|
| 563 |
+
{
|
| 564 |
+
"epoch": 16.99,
|
| 565 |
+
"eval_accuracy": 0.8345454545454546,
|
| 566 |
+
"eval_loss": 0.4195989668369293,
|
| 567 |
+
"eval_runtime": 1.4108,
|
| 568 |
+
"eval_samples_per_second": 389.839,
|
| 569 |
+
"eval_steps_per_second": 24.808,
|
| 570 |
+
"step": 684
|
| 571 |
+
},
|
| 572 |
+
{
|
| 573 |
+
"epoch": 17.14,
|
| 574 |
+
"learning_rate": 0.0001895833333333333,
|
| 575 |
+
"loss": 0.4683,
|
| 576 |
+
"step": 690
|
| 577 |
+
},
|
| 578 |
+
{
|
| 579 |
+
"epoch": 17.39,
|
| 580 |
+
"learning_rate": 0.00018749999999999998,
|
| 581 |
+
"loss": 0.4322,
|
| 582 |
+
"step": 700
|
| 583 |
+
},
|
| 584 |
+
{
|
| 585 |
+
"epoch": 17.64,
|
| 586 |
+
"learning_rate": 0.00018541666666666666,
|
| 587 |
+
"loss": 0.4099,
|
| 588 |
+
"step": 710
|
| 589 |
+
},
|
| 590 |
+
{
|
| 591 |
+
"epoch": 17.89,
|
| 592 |
+
"learning_rate": 0.00018333333333333334,
|
| 593 |
+
"loss": 0.4668,
|
| 594 |
+
"step": 720
|
| 595 |
+
},
|
| 596 |
+
{
|
| 597 |
+
"epoch": 17.99,
|
| 598 |
+
"eval_accuracy": 0.8290909090909091,
|
| 599 |
+
"eval_loss": 0.4041709899902344,
|
| 600 |
+
"eval_runtime": 1.5199,
|
| 601 |
+
"eval_samples_per_second": 361.876,
|
| 602 |
+
"eval_steps_per_second": 23.028,
|
| 603 |
+
"step": 724
|
| 604 |
+
},
|
| 605 |
+
{
|
| 606 |
+
"epoch": 18.14,
|
| 607 |
+
"learning_rate": 0.00018124999999999996,
|
| 608 |
+
"loss": 0.5344,
|
| 609 |
+
"step": 730
|
| 610 |
+
},
|
| 611 |
+
{
|
| 612 |
+
"epoch": 18.39,
|
| 613 |
+
"learning_rate": 0.00017916666666666664,
|
| 614 |
+
"loss": 0.4369,
|
| 615 |
+
"step": 740
|
| 616 |
+
},
|
| 617 |
+
{
|
| 618 |
+
"epoch": 18.63,
|
| 619 |
+
"learning_rate": 0.00017708333333333332,
|
| 620 |
+
"loss": 0.4014,
|
| 621 |
+
"step": 750
|
| 622 |
+
},
|
| 623 |
+
{
|
| 624 |
+
"epoch": 18.88,
|
| 625 |
+
"learning_rate": 0.000175,
|
| 626 |
+
"loss": 0.41,
|
| 627 |
+
"step": 760
|
| 628 |
+
},
|
| 629 |
+
{
|
| 630 |
+
"epoch": 18.98,
|
| 631 |
+
"eval_accuracy": 0.8436363636363636,
|
| 632 |
+
"eval_loss": 0.4112071990966797,
|
| 633 |
+
"eval_runtime": 1.3473,
|
| 634 |
+
"eval_samples_per_second": 408.216,
|
| 635 |
+
"eval_steps_per_second": 25.977,
|
| 636 |
+
"step": 764
|
| 637 |
+
},
|
| 638 |
+
{
|
| 639 |
+
"epoch": 19.13,
|
| 640 |
+
"learning_rate": 0.00017291666666666662,
|
| 641 |
+
"loss": 0.3628,
|
| 642 |
+
"step": 770
|
| 643 |
+
},
|
| 644 |
+
{
|
| 645 |
+
"epoch": 19.38,
|
| 646 |
+
"learning_rate": 0.0001708333333333333,
|
| 647 |
+
"loss": 0.3968,
|
| 648 |
+
"step": 780
|
| 649 |
+
},
|
| 650 |
+
{
|
| 651 |
+
"epoch": 19.63,
|
| 652 |
+
"learning_rate": 0.00016874999999999998,
|
| 653 |
+
"loss": 0.4301,
|
| 654 |
+
"step": 790
|
| 655 |
+
},
|
| 656 |
+
{
|
| 657 |
+
"epoch": 19.88,
|
| 658 |
+
"learning_rate": 0.00016666666666666666,
|
| 659 |
+
"loss": 0.4349,
|
| 660 |
+
"step": 800
|
| 661 |
+
},
|
| 662 |
+
{
|
| 663 |
+
"epoch": 20.0,
|
| 664 |
+
"eval_accuracy": 0.8418181818181818,
|
| 665 |
+
"eval_loss": 0.4119725525379181,
|
| 666 |
+
"eval_runtime": 1.3453,
|
| 667 |
+
"eval_samples_per_second": 408.821,
|
| 668 |
+
"eval_steps_per_second": 26.016,
|
| 669 |
+
"step": 805
|
| 670 |
+
},
|
| 671 |
+
{
|
| 672 |
+
"epoch": 20.12,
|
| 673 |
+
"learning_rate": 0.00016458333333333334,
|
| 674 |
+
"loss": 0.3921,
|
| 675 |
+
"step": 810
|
| 676 |
+
},
|
| 677 |
+
{
|
| 678 |
+
"epoch": 20.37,
|
| 679 |
+
"learning_rate": 0.00016249999999999997,
|
| 680 |
+
"loss": 0.4351,
|
| 681 |
+
"step": 820
|
| 682 |
+
},
|
| 683 |
+
{
|
| 684 |
+
"epoch": 20.62,
|
| 685 |
+
"learning_rate": 0.00016041666666666664,
|
| 686 |
+
"loss": 0.4144,
|
| 687 |
+
"step": 830
|
| 688 |
+
},
|
| 689 |
+
{
|
| 690 |
+
"epoch": 20.87,
|
| 691 |
+
"learning_rate": 0.00015833333333333332,
|
| 692 |
+
"loss": 0.4558,
|
| 693 |
+
"step": 840
|
| 694 |
+
},
|
| 695 |
+
{
|
| 696 |
+
"epoch": 20.99,
|
| 697 |
+
"eval_accuracy": 0.8218181818181818,
|
| 698 |
+
"eval_loss": 0.4554423391819,
|
| 699 |
+
"eval_runtime": 1.3368,
|
| 700 |
+
"eval_samples_per_second": 411.426,
|
| 701 |
+
"eval_steps_per_second": 26.182,
|
| 702 |
+
"step": 845
|
| 703 |
+
},
|
| 704 |
+
{
|
| 705 |
+
"epoch": 21.12,
|
| 706 |
+
"learning_rate": 0.00015625,
|
| 707 |
+
"loss": 0.4053,
|
| 708 |
+
"step": 850
|
| 709 |
+
},
|
| 710 |
+
{
|
| 711 |
+
"epoch": 21.37,
|
| 712 |
+
"learning_rate": 0.00015416666666666663,
|
| 713 |
+
"loss": 0.3483,
|
| 714 |
+
"step": 860
|
| 715 |
+
},
|
| 716 |
+
{
|
| 717 |
+
"epoch": 21.61,
|
| 718 |
+
"learning_rate": 0.0001520833333333333,
|
| 719 |
+
"loss": 0.4139,
|
| 720 |
+
"step": 870
|
| 721 |
+
},
|
| 722 |
+
{
|
| 723 |
+
"epoch": 21.86,
|
| 724 |
+
"learning_rate": 0.00015,
|
| 725 |
+
"loss": 0.3952,
|
| 726 |
+
"step": 880
|
| 727 |
+
},
|
| 728 |
+
{
|
| 729 |
+
"epoch": 21.99,
|
| 730 |
+
"eval_accuracy": 0.850909090909091,
|
| 731 |
+
"eval_loss": 0.3774945139884949,
|
| 732 |
+
"eval_runtime": 1.3708,
|
| 733 |
+
"eval_samples_per_second": 401.212,
|
| 734 |
+
"eval_steps_per_second": 25.532,
|
| 735 |
+
"step": 885
|
| 736 |
+
},
|
| 737 |
+
{
|
| 738 |
+
"epoch": 22.11,
|
| 739 |
+
"learning_rate": 0.00014791666666666667,
|
| 740 |
+
"loss": 0.3661,
|
| 741 |
+
"step": 890
|
| 742 |
+
},
|
| 743 |
+
{
|
| 744 |
+
"epoch": 22.36,
|
| 745 |
+
"learning_rate": 0.00014583333333333332,
|
| 746 |
+
"loss": 0.3706,
|
| 747 |
+
"step": 900
|
| 748 |
+
},
|
| 749 |
+
{
|
| 750 |
+
"epoch": 22.61,
|
| 751 |
+
"learning_rate": 0.00014375,
|
| 752 |
+
"loss": 0.4127,
|
| 753 |
+
"step": 910
|
| 754 |
+
},
|
| 755 |
+
{
|
| 756 |
+
"epoch": 22.86,
|
| 757 |
+
"learning_rate": 0.00014166666666666665,
|
| 758 |
+
"loss": 0.3395,
|
| 759 |
+
"step": 920
|
| 760 |
+
},
|
| 761 |
+
{
|
| 762 |
+
"epoch": 22.98,
|
| 763 |
+
"eval_accuracy": 0.84,
|
| 764 |
+
"eval_loss": 0.4027109742164612,
|
| 765 |
+
"eval_runtime": 1.4218,
|
| 766 |
+
"eval_samples_per_second": 386.825,
|
| 767 |
+
"eval_steps_per_second": 24.616,
|
| 768 |
+
"step": 925
|
| 769 |
+
},
|
| 770 |
+
{
|
| 771 |
+
"epoch": 23.11,
|
| 772 |
+
"learning_rate": 0.00013958333333333333,
|
| 773 |
+
"loss": 0.3821,
|
| 774 |
+
"step": 930
|
| 775 |
+
},
|
| 776 |
+
{
|
| 777 |
+
"epoch": 23.35,
|
| 778 |
+
"learning_rate": 0.00013749999999999998,
|
| 779 |
+
"loss": 0.3408,
|
| 780 |
+
"step": 940
|
| 781 |
+
},
|
| 782 |
+
{
|
| 783 |
+
"epoch": 23.6,
|
| 784 |
+
"learning_rate": 0.00013541666666666666,
|
| 785 |
+
"loss": 0.3344,
|
| 786 |
+
"step": 950
|
| 787 |
+
},
|
| 788 |
+
{
|
| 789 |
+
"epoch": 23.85,
|
| 790 |
+
"learning_rate": 0.0001333333333333333,
|
| 791 |
+
"loss": 0.3525,
|
| 792 |
+
"step": 960
|
| 793 |
+
},
|
| 794 |
+
{
|
| 795 |
+
"epoch": 24.0,
|
| 796 |
+
"eval_accuracy": 0.850909090909091,
|
| 797 |
+
"eval_loss": 0.3956778645515442,
|
| 798 |
+
"eval_runtime": 1.3433,
|
| 799 |
+
"eval_samples_per_second": 409.433,
|
| 800 |
+
"eval_steps_per_second": 26.055,
|
| 801 |
+
"step": 966
|
| 802 |
+
},
|
| 803 |
+
{
|
| 804 |
+
"epoch": 24.1,
|
| 805 |
+
"learning_rate": 0.00013125,
|
| 806 |
+
"loss": 0.3803,
|
| 807 |
+
"step": 970
|
| 808 |
+
},
|
| 809 |
+
{
|
| 810 |
+
"epoch": 24.35,
|
| 811 |
+
"learning_rate": 0.00012916666666666667,
|
| 812 |
+
"loss": 0.3213,
|
| 813 |
+
"step": 980
|
| 814 |
+
},
|
| 815 |
+
{
|
| 816 |
+
"epoch": 24.6,
|
| 817 |
+
"learning_rate": 0.00012708333333333332,
|
| 818 |
+
"loss": 0.3845,
|
| 819 |
+
"step": 990
|
| 820 |
+
},
|
| 821 |
+
{
|
| 822 |
+
"epoch": 24.84,
|
| 823 |
+
"learning_rate": 0.000125,
|
| 824 |
+
"loss": 0.3497,
|
| 825 |
+
"step": 1000
|
| 826 |
+
},
|
| 827 |
+
{
|
| 828 |
+
"epoch": 24.99,
|
| 829 |
+
"eval_accuracy": 0.8527272727272728,
|
| 830 |
+
"eval_loss": 0.4020462930202484,
|
| 831 |
+
"eval_runtime": 1.3428,
|
| 832 |
+
"eval_samples_per_second": 409.587,
|
| 833 |
+
"eval_steps_per_second": 26.065,
|
| 834 |
+
"step": 1006
|
| 835 |
+
},
|
| 836 |
+
{
|
| 837 |
+
"epoch": 25.09,
|
| 838 |
+
"learning_rate": 0.00012291666666666665,
|
| 839 |
+
"loss": 0.375,
|
| 840 |
+
"step": 1010
|
| 841 |
+
},
|
| 842 |
+
{
|
| 843 |
+
"epoch": 25.34,
|
| 844 |
+
"learning_rate": 0.00012083333333333332,
|
| 845 |
+
"loss": 0.3141,
|
| 846 |
+
"step": 1020
|
| 847 |
+
},
|
| 848 |
+
{
|
| 849 |
+
"epoch": 25.59,
|
| 850 |
+
"learning_rate": 0.00011874999999999999,
|
| 851 |
+
"loss": 0.3899,
|
| 852 |
+
"step": 1030
|
| 853 |
+
},
|
| 854 |
+
{
|
| 855 |
+
"epoch": 25.84,
|
| 856 |
+
"learning_rate": 0.00011666666666666665,
|
| 857 |
+
"loss": 0.3625,
|
| 858 |
+
"step": 1040
|
| 859 |
+
},
|
| 860 |
+
{
|
| 861 |
+
"epoch": 25.99,
|
| 862 |
+
"eval_accuracy": 0.8563636363636363,
|
| 863 |
+
"eval_loss": 0.39101672172546387,
|
| 864 |
+
"eval_runtime": 1.3918,
|
| 865 |
+
"eval_samples_per_second": 395.16,
|
| 866 |
+
"eval_steps_per_second": 25.147,
|
| 867 |
+
"step": 1046
|
| 868 |
+
},
|
| 869 |
+
{
|
| 870 |
+
"epoch": 26.09,
|
| 871 |
+
"learning_rate": 0.00011458333333333332,
|
| 872 |
+
"loss": 0.3777,
|
| 873 |
+
"step": 1050
|
| 874 |
+
},
|
| 875 |
+
{
|
| 876 |
+
"epoch": 26.34,
|
| 877 |
+
"learning_rate": 0.0001125,
|
| 878 |
+
"loss": 0.3572,
|
| 879 |
+
"step": 1060
|
| 880 |
+
},
|
| 881 |
+
{
|
| 882 |
+
"epoch": 26.58,
|
| 883 |
+
"learning_rate": 0.00011041666666666666,
|
| 884 |
+
"loss": 0.3486,
|
| 885 |
+
"step": 1070
|
| 886 |
+
},
|
| 887 |
+
{
|
| 888 |
+
"epoch": 26.83,
|
| 889 |
+
"learning_rate": 0.00010833333333333333,
|
| 890 |
+
"loss": 0.3008,
|
| 891 |
+
"step": 1080
|
| 892 |
+
},
|
| 893 |
+
{
|
| 894 |
+
"epoch": 26.98,
|
| 895 |
+
"eval_accuracy": 0.8581818181818182,
|
| 896 |
+
"eval_loss": 0.3789851665496826,
|
| 897 |
+
"eval_runtime": 1.4113,
|
| 898 |
+
"eval_samples_per_second": 389.701,
|
| 899 |
+
"eval_steps_per_second": 24.799,
|
| 900 |
+
"step": 1086
|
| 901 |
+
},
|
| 902 |
+
{
|
| 903 |
+
"epoch": 27.08,
|
| 904 |
+
"learning_rate": 0.00010625,
|
| 905 |
+
"loss": 0.32,
|
| 906 |
+
"step": 1090
|
| 907 |
+
},
|
| 908 |
+
{
|
| 909 |
+
"epoch": 27.33,
|
| 910 |
+
"learning_rate": 0.00010416666666666666,
|
| 911 |
+
"loss": 0.356,
|
| 912 |
+
"step": 1100
|
| 913 |
+
},
|
| 914 |
+
{
|
| 915 |
+
"epoch": 27.58,
|
| 916 |
+
"learning_rate": 0.00010208333333333333,
|
| 917 |
+
"loss": 0.3053,
|
| 918 |
+
"step": 1110
|
| 919 |
+
},
|
| 920 |
+
{
|
| 921 |
+
"epoch": 27.83,
|
| 922 |
+
"learning_rate": 9.999999999999999e-05,
|
| 923 |
+
"loss": 0.2907,
|
| 924 |
+
"step": 1120
|
| 925 |
+
},
|
| 926 |
+
{
|
| 927 |
+
"epoch": 28.0,
|
| 928 |
+
"eval_accuracy": 0.8563636363636363,
|
| 929 |
+
"eval_loss": 0.38305726647377014,
|
| 930 |
+
"eval_runtime": 1.3603,
|
| 931 |
+
"eval_samples_per_second": 404.315,
|
| 932 |
+
"eval_steps_per_second": 25.729,
|
| 933 |
+
"step": 1127
|
| 934 |
+
},
|
| 935 |
+
{
|
| 936 |
+
"epoch": 28.07,
|
| 937 |
+
"learning_rate": 9.791666666666666e-05,
|
| 938 |
+
"loss": 0.3384,
|
| 939 |
+
"step": 1130
|
| 940 |
+
},
|
| 941 |
+
{
|
| 942 |
+
"epoch": 28.32,
|
| 943 |
+
"learning_rate": 9.583333333333332e-05,
|
| 944 |
+
"loss": 0.2956,
|
| 945 |
+
"step": 1140
|
| 946 |
+
},
|
| 947 |
+
{
|
| 948 |
+
"epoch": 28.57,
|
| 949 |
+
"learning_rate": 9.374999999999999e-05,
|
| 950 |
+
"loss": 0.2923,
|
| 951 |
+
"step": 1150
|
| 952 |
+
},
|
| 953 |
+
{
|
| 954 |
+
"epoch": 28.82,
|
| 955 |
+
"learning_rate": 9.166666666666667e-05,
|
| 956 |
+
"loss": 0.3165,
|
| 957 |
+
"step": 1160
|
| 958 |
+
},
|
| 959 |
+
{
|
| 960 |
+
"epoch": 28.99,
|
| 961 |
+
"eval_accuracy": 0.8618181818181818,
|
| 962 |
+
"eval_loss": 0.4005330204963684,
|
| 963 |
+
"eval_runtime": 1.3778,
|
| 964 |
+
"eval_samples_per_second": 399.18,
|
| 965 |
+
"eval_steps_per_second": 25.402,
|
| 966 |
+
"step": 1167
|
| 967 |
+
},
|
| 968 |
+
{
|
| 969 |
+
"epoch": 29.07,
|
| 970 |
+
"learning_rate": 8.958333333333332e-05,
|
| 971 |
+
"loss": 0.302,
|
| 972 |
+
"step": 1170
|
| 973 |
+
},
|
| 974 |
+
{
|
| 975 |
+
"epoch": 29.32,
|
| 976 |
+
"learning_rate": 8.75e-05,
|
| 977 |
+
"loss": 0.2723,
|
| 978 |
+
"step": 1180
|
| 979 |
+
},
|
| 980 |
+
{
|
| 981 |
+
"epoch": 29.57,
|
| 982 |
+
"learning_rate": 8.541666666666665e-05,
|
| 983 |
+
"loss": 0.2953,
|
| 984 |
+
"step": 1190
|
| 985 |
+
},
|
| 986 |
+
{
|
| 987 |
+
"epoch": 29.81,
|
| 988 |
+
"learning_rate": 8.333333333333333e-05,
|
| 989 |
+
"loss": 0.3231,
|
| 990 |
+
"step": 1200
|
| 991 |
+
},
|
| 992 |
+
{
|
| 993 |
+
"epoch": 29.99,
|
| 994 |
+
"eval_accuracy": 0.8327272727272728,
|
| 995 |
+
"eval_loss": 0.4512370228767395,
|
| 996 |
+
"eval_runtime": 1.3443,
|
| 997 |
+
"eval_samples_per_second": 409.127,
|
| 998 |
+
"eval_steps_per_second": 26.035,
|
| 999 |
+
"step": 1207
|
| 1000 |
+
},
|
| 1001 |
+
{
|
| 1002 |
+
"epoch": 30.06,
|
| 1003 |
+
"learning_rate": 8.124999999999998e-05,
|
| 1004 |
+
"loss": 0.3583,
|
| 1005 |
+
"step": 1210
|
| 1006 |
+
},
|
| 1007 |
+
{
|
| 1008 |
+
"epoch": 30.31,
|
| 1009 |
+
"learning_rate": 7.916666666666666e-05,
|
| 1010 |
+
"loss": 0.3356,
|
| 1011 |
+
"step": 1220
|
| 1012 |
+
},
|
| 1013 |
+
{
|
| 1014 |
+
"epoch": 30.56,
|
| 1015 |
+
"learning_rate": 7.708333333333331e-05,
|
| 1016 |
+
"loss": 0.2687,
|
| 1017 |
+
"step": 1230
|
| 1018 |
+
},
|
| 1019 |
+
{
|
| 1020 |
+
"epoch": 30.81,
|
| 1021 |
+
"learning_rate": 7.5e-05,
|
| 1022 |
+
"loss": 0.2819,
|
| 1023 |
+
"step": 1240
|
| 1024 |
+
},
|
| 1025 |
+
{
|
| 1026 |
+
"epoch": 30.98,
|
| 1027 |
+
"eval_accuracy": 0.8454545454545455,
|
| 1028 |
+
"eval_loss": 0.4227891266345978,
|
| 1029 |
+
"eval_runtime": 1.4114,
|
| 1030 |
+
"eval_samples_per_second": 389.695,
|
| 1031 |
+
"eval_steps_per_second": 24.799,
|
| 1032 |
+
"step": 1247
|
| 1033 |
+
},
|
| 1034 |
+
{
|
| 1035 |
+
"epoch": 31.06,
|
| 1036 |
+
"learning_rate": 7.291666666666666e-05,
|
| 1037 |
+
"loss": 0.2949,
|
| 1038 |
+
"step": 1250
|
| 1039 |
+
},
|
| 1040 |
+
{
|
| 1041 |
+
"epoch": 31.3,
|
| 1042 |
+
"learning_rate": 7.083333333333332e-05,
|
| 1043 |
+
"loss": 0.2654,
|
| 1044 |
+
"step": 1260
|
| 1045 |
+
},
|
| 1046 |
+
{
|
| 1047 |
+
"epoch": 31.55,
|
| 1048 |
+
"learning_rate": 6.874999999999999e-05,
|
| 1049 |
+
"loss": 0.2745,
|
| 1050 |
+
"step": 1270
|
| 1051 |
+
},
|
| 1052 |
+
{
|
| 1053 |
+
"epoch": 31.8,
|
| 1054 |
+
"learning_rate": 6.666666666666666e-05,
|
| 1055 |
+
"loss": 0.2704,
|
| 1056 |
+
"step": 1280
|
| 1057 |
+
},
|
| 1058 |
+
{
|
| 1059 |
+
"epoch": 32.0,
|
| 1060 |
+
"eval_accuracy": 0.86,
|
| 1061 |
+
"eval_loss": 0.4074237644672394,
|
| 1062 |
+
"eval_runtime": 1.3463,
|
| 1063 |
+
"eval_samples_per_second": 408.52,
|
| 1064 |
+
"eval_steps_per_second": 25.997,
|
| 1065 |
+
"step": 1288
|
| 1066 |
+
},
|
| 1067 |
+
{
|
| 1068 |
+
"epoch": 32.05,
|
| 1069 |
+
"learning_rate": 6.458333333333334e-05,
|
| 1070 |
+
"loss": 0.3105,
|
| 1071 |
+
"step": 1290
|
| 1072 |
+
},
|
| 1073 |
+
{
|
| 1074 |
+
"epoch": 32.3,
|
| 1075 |
+
"learning_rate": 6.25e-05,
|
| 1076 |
+
"loss": 0.2614,
|
| 1077 |
+
"step": 1300
|
| 1078 |
+
},
|
| 1079 |
+
{
|
| 1080 |
+
"epoch": 32.55,
|
| 1081 |
+
"learning_rate": 6.041666666666666e-05,
|
| 1082 |
+
"loss": 0.2842,
|
| 1083 |
+
"step": 1310
|
| 1084 |
+
},
|
| 1085 |
+
{
|
| 1086 |
+
"epoch": 32.8,
|
| 1087 |
+
"learning_rate": 5.8333333333333326e-05,
|
| 1088 |
+
"loss": 0.2429,
|
| 1089 |
+
"step": 1320
|
| 1090 |
+
},
|
| 1091 |
+
{
|
| 1092 |
+
"epoch": 32.99,
|
| 1093 |
+
"eval_accuracy": 0.8545454545454545,
|
| 1094 |
+
"eval_loss": 0.4405260682106018,
|
| 1095 |
+
"eval_runtime": 1.6104,
|
| 1096 |
+
"eval_samples_per_second": 341.534,
|
| 1097 |
+
"eval_steps_per_second": 21.734,
|
| 1098 |
+
"step": 1328
|
| 1099 |
+
},
|
| 1100 |
+
{
|
| 1101 |
+
"epoch": 33.04,
|
| 1102 |
+
"learning_rate": 5.625e-05,
|
| 1103 |
+
"loss": 0.2956,
|
| 1104 |
+
"step": 1330
|
| 1105 |
+
},
|
| 1106 |
+
{
|
| 1107 |
+
"epoch": 33.29,
|
| 1108 |
+
"learning_rate": 5.4166666666666664e-05,
|
| 1109 |
+
"loss": 0.2767,
|
| 1110 |
+
"step": 1340
|
| 1111 |
+
},
|
| 1112 |
+
{
|
| 1113 |
+
"epoch": 33.54,
|
| 1114 |
+
"learning_rate": 5.208333333333333e-05,
|
| 1115 |
+
"loss": 0.2676,
|
| 1116 |
+
"step": 1350
|
| 1117 |
+
},
|
| 1118 |
+
{
|
| 1119 |
+
"epoch": 33.79,
|
| 1120 |
+
"learning_rate": 4.9999999999999996e-05,
|
| 1121 |
+
"loss": 0.2421,
|
| 1122 |
+
"step": 1360
|
| 1123 |
+
},
|
| 1124 |
+
{
|
| 1125 |
+
"epoch": 33.99,
|
| 1126 |
+
"eval_accuracy": 0.8527272727272728,
|
| 1127 |
+
"eval_loss": 0.43370744585990906,
|
| 1128 |
+
"eval_runtime": 1.4724,
|
| 1129 |
+
"eval_samples_per_second": 373.551,
|
| 1130 |
+
"eval_steps_per_second": 23.771,
|
| 1131 |
+
"step": 1368
|
| 1132 |
+
},
|
| 1133 |
+
{
|
| 1134 |
+
"epoch": 34.04,
|
| 1135 |
+
"learning_rate": 4.791666666666666e-05,
|
| 1136 |
+
"loss": 0.3275,
|
| 1137 |
+
"step": 1370
|
| 1138 |
+
},
|
| 1139 |
+
{
|
| 1140 |
+
"epoch": 34.29,
|
| 1141 |
+
"learning_rate": 4.5833333333333334e-05,
|
| 1142 |
+
"loss": 0.2475,
|
| 1143 |
+
"step": 1380
|
| 1144 |
+
},
|
| 1145 |
+
{
|
| 1146 |
+
"epoch": 34.53,
|
| 1147 |
+
"learning_rate": 4.375e-05,
|
| 1148 |
+
"loss": 0.288,
|
| 1149 |
+
"step": 1390
|
| 1150 |
+
},
|
| 1151 |
+
{
|
| 1152 |
+
"epoch": 34.78,
|
| 1153 |
+
"learning_rate": 4.1666666666666665e-05,
|
| 1154 |
+
"loss": 0.3039,
|
| 1155 |
+
"step": 1400
|
| 1156 |
+
},
|
| 1157 |
+
{
|
| 1158 |
+
"epoch": 34.98,
|
| 1159 |
+
"eval_accuracy": 0.8472727272727273,
|
| 1160 |
+
"eval_loss": 0.46279987692832947,
|
| 1161 |
+
"eval_runtime": 1.3718,
|
| 1162 |
+
"eval_samples_per_second": 400.919,
|
| 1163 |
+
"eval_steps_per_second": 25.513,
|
| 1164 |
+
"step": 1408
|
| 1165 |
+
},
|
| 1166 |
+
{
|
| 1167 |
+
"epoch": 35.03,
|
| 1168 |
+
"learning_rate": 3.958333333333333e-05,
|
| 1169 |
+
"loss": 0.2737,
|
| 1170 |
+
"step": 1410
|
| 1171 |
+
},
|
| 1172 |
+
{
|
| 1173 |
+
"epoch": 35.28,
|
| 1174 |
+
"learning_rate": 3.75e-05,
|
| 1175 |
+
"loss": 0.2746,
|
| 1176 |
+
"step": 1420
|
| 1177 |
+
},
|
| 1178 |
+
{
|
| 1179 |
+
"epoch": 35.53,
|
| 1180 |
+
"learning_rate": 3.541666666666666e-05,
|
| 1181 |
+
"loss": 0.2906,
|
| 1182 |
+
"step": 1430
|
| 1183 |
+
},
|
| 1184 |
+
{
|
| 1185 |
+
"epoch": 35.78,
|
| 1186 |
+
"learning_rate": 3.333333333333333e-05,
|
| 1187 |
+
"loss": 0.2677,
|
| 1188 |
+
"step": 1440
|
| 1189 |
+
},
|
| 1190 |
+
{
|
| 1191 |
+
"epoch": 36.0,
|
| 1192 |
+
"eval_accuracy": 0.8545454545454545,
|
| 1193 |
+
"eval_loss": 0.4411250948905945,
|
| 1194 |
+
"eval_runtime": 1.3953,
|
| 1195 |
+
"eval_samples_per_second": 394.172,
|
| 1196 |
+
"eval_steps_per_second": 25.084,
|
| 1197 |
+
"step": 1449
|
| 1198 |
+
},
|
| 1199 |
+
{
|
| 1200 |
+
"epoch": 36.02,
|
| 1201 |
+
"learning_rate": 3.125e-05,
|
| 1202 |
+
"loss": 0.2467,
|
| 1203 |
+
"step": 1450
|
| 1204 |
+
},
|
| 1205 |
+
{
|
| 1206 |
+
"epoch": 36.27,
|
| 1207 |
+
"learning_rate": 2.9166666666666663e-05,
|
| 1208 |
+
"loss": 0.2513,
|
| 1209 |
+
"step": 1460
|
| 1210 |
+
},
|
| 1211 |
+
{
|
| 1212 |
+
"epoch": 36.52,
|
| 1213 |
+
"learning_rate": 2.7083333333333332e-05,
|
| 1214 |
+
"loss": 0.271,
|
| 1215 |
+
"step": 1470
|
| 1216 |
+
},
|
| 1217 |
+
{
|
| 1218 |
+
"epoch": 36.77,
|
| 1219 |
+
"learning_rate": 2.4999999999999998e-05,
|
| 1220 |
+
"loss": 0.2171,
|
| 1221 |
+
"step": 1480
|
| 1222 |
+
},
|
| 1223 |
+
{
|
| 1224 |
+
"epoch": 36.99,
|
| 1225 |
+
"eval_accuracy": 0.8618181818181818,
|
| 1226 |
+
"eval_loss": 0.4755066931247711,
|
| 1227 |
+
"eval_runtime": 1.4153,
|
| 1228 |
+
"eval_samples_per_second": 388.599,
|
| 1229 |
+
"eval_steps_per_second": 24.729,
|
| 1230 |
+
"step": 1489
|
| 1231 |
+
},
|
| 1232 |
+
{
|
| 1233 |
+
"epoch": 37.02,
|
| 1234 |
+
"learning_rate": 2.2916666666666667e-05,
|
| 1235 |
+
"loss": 0.2207,
|
| 1236 |
+
"step": 1490
|
| 1237 |
+
},
|
| 1238 |
+
{
|
| 1239 |
+
"epoch": 37.27,
|
| 1240 |
+
"learning_rate": 2.0833333333333333e-05,
|
| 1241 |
+
"loss": 0.2346,
|
| 1242 |
+
"step": 1500
|
| 1243 |
+
},
|
| 1244 |
+
{
|
| 1245 |
+
"epoch": 37.52,
|
| 1246 |
+
"learning_rate": 1.875e-05,
|
| 1247 |
+
"loss": 0.2055,
|
| 1248 |
+
"step": 1510
|
| 1249 |
+
},
|
| 1250 |
+
{
|
| 1251 |
+
"epoch": 37.76,
|
| 1252 |
+
"learning_rate": 1.6666666666666664e-05,
|
| 1253 |
+
"loss": 0.2268,
|
| 1254 |
+
"step": 1520
|
| 1255 |
+
},
|
| 1256 |
+
{
|
| 1257 |
+
"epoch": 37.99,
|
| 1258 |
+
"eval_accuracy": 0.86,
|
| 1259 |
+
"eval_loss": 0.45060980319976807,
|
| 1260 |
+
"eval_runtime": 1.4394,
|
| 1261 |
+
"eval_samples_per_second": 382.115,
|
| 1262 |
+
"eval_steps_per_second": 24.316,
|
| 1263 |
+
"step": 1529
|
| 1264 |
+
},
|
| 1265 |
+
{
|
| 1266 |
+
"epoch": 38.01,
|
| 1267 |
+
"learning_rate": 1.4583333333333331e-05,
|
| 1268 |
+
"loss": 0.2844,
|
| 1269 |
+
"step": 1530
|
| 1270 |
+
},
|
| 1271 |
+
{
|
| 1272 |
+
"epoch": 38.26,
|
| 1273 |
+
"learning_rate": 1.2499999999999999e-05,
|
| 1274 |
+
"loss": 0.2419,
|
| 1275 |
+
"step": 1540
|
| 1276 |
+
},
|
| 1277 |
+
{
|
| 1278 |
+
"epoch": 38.51,
|
| 1279 |
+
"learning_rate": 1.0416666666666666e-05,
|
| 1280 |
+
"loss": 0.256,
|
| 1281 |
+
"step": 1550
|
| 1282 |
+
},
|
| 1283 |
+
{
|
| 1284 |
+
"epoch": 38.76,
|
| 1285 |
+
"learning_rate": 8.333333333333332e-06,
|
| 1286 |
+
"loss": 0.2378,
|
| 1287 |
+
"step": 1560
|
| 1288 |
+
},
|
| 1289 |
+
{
|
| 1290 |
+
"epoch": 38.98,
|
| 1291 |
+
"eval_accuracy": 0.8527272727272728,
|
| 1292 |
+
"eval_loss": 0.4633024334907532,
|
| 1293 |
+
"eval_runtime": 1.3843,
|
| 1294 |
+
"eval_samples_per_second": 397.304,
|
| 1295 |
+
"eval_steps_per_second": 25.283,
|
| 1296 |
+
"step": 1569
|
| 1297 |
+
},
|
| 1298 |
+
{
|
| 1299 |
+
"epoch": 39.01,
|
| 1300 |
+
"learning_rate": 6.2499999999999995e-06,
|
| 1301 |
+
"loss": 0.3098,
|
| 1302 |
+
"step": 1570
|
| 1303 |
+
},
|
| 1304 |
+
{
|
| 1305 |
+
"epoch": 39.25,
|
| 1306 |
+
"learning_rate": 4.166666666666666e-06,
|
| 1307 |
+
"loss": 0.2416,
|
| 1308 |
+
"step": 1580
|
| 1309 |
+
},
|
| 1310 |
+
{
|
| 1311 |
+
"epoch": 39.5,
|
| 1312 |
+
"learning_rate": 2.083333333333333e-06,
|
| 1313 |
+
"loss": 0.2014,
|
| 1314 |
+
"step": 1590
|
| 1315 |
+
},
|
| 1316 |
+
{
|
| 1317 |
+
"epoch": 39.75,
|
| 1318 |
+
"learning_rate": 0.0,
|
| 1319 |
+
"loss": 0.2021,
|
| 1320 |
+
"step": 1600
|
| 1321 |
+
},
|
| 1322 |
+
{
|
| 1323 |
+
"epoch": 39.75,
|
| 1324 |
+
"eval_accuracy": 0.8563636363636363,
|
| 1325 |
+
"eval_loss": 0.4492277503013611,
|
| 1326 |
+
"eval_runtime": 1.3993,
|
| 1327 |
+
"eval_samples_per_second": 393.042,
|
| 1328 |
+
"eval_steps_per_second": 25.012,
|
| 1329 |
+
"step": 1600
|
| 1330 |
+
},
|
| 1331 |
+
{
|
| 1332 |
+
"epoch": 39.75,
|
| 1333 |
+
"step": 1600,
|
| 1334 |
+
"total_flos": 2.7935287357114368e+17,
|
| 1335 |
+
"train_loss": 0.4774747739732266,
|
| 1336 |
+
"train_runtime": 572.0649,
|
| 1337 |
+
"train_samples_per_second": 179.21,
|
| 1338 |
+
"train_steps_per_second": 2.797
|
| 1339 |
+
}
|
| 1340 |
+
],
|
| 1341 |
+
"logging_steps": 10,
|
| 1342 |
+
"max_steps": 1600,
|
| 1343 |
+
"num_input_tokens_seen": 0,
|
| 1344 |
+
"num_train_epochs": 40,
|
| 1345 |
+
"save_steps": 500,
|
| 1346 |
+
"total_flos": 2.7935287357114368e+17,
|
| 1347 |
+
"train_batch_size": 16,
|
| 1348 |
+
"trial_name": null,
|
| 1349 |
+
"trial_params": null
|
| 1350 |
+
}
|
training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:230b9d0a95216c9edd1498f6766a875856cb31f6733b459f5774e3ecb24cd957
|
| 3 |
+
size 4728
|