vit-base-patch16-224_rice-leaf-disease-augmented-v4_v5_pft

This model is a fine-tuned version of google/vit-base-patch16-224 on an unknown dataset. It achieves the following results on the evaluation set:

• Loss: 0.4332
• Accuracy: 0.8456

Model description

More information needed

Intended uses & limitations

More information needed

Training and evaluation data

More information needed

Training procedure

Training hyperparameters

The following hyperparameters were used during training:

• learning_rate: 0.0003
• train_batch_size: 64
• eval_batch_size: 64
• seed: 42
• optimizer: Use OptimizerNames.ADAMW_TORCH with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
• lr_scheduler_type: cosine_with_restarts
• lr_scheduler_warmup_steps: 256
• num_epochs: 30
• mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy
2.1701	0.5	64	1.9984	0.2349
1.878	1.0	128	1.6113	0.4698
1.4921	1.5	192	1.2635	0.6174
1.2126	2.0	256	1.0178	0.6812
0.9922	2.5	320	0.8928	0.7148
0.8858	3.0	384	0.7883	0.7483
0.7966	3.5	448	0.7408	0.7517
0.7427	4.0	512	0.6912	0.7651
0.7077	4.5	576	0.6795	0.7718
0.6796	5.0	640	0.6647	0.7785
0.6597	5.5	704	0.6684	0.7752
0.6652	6.0	768	0.6535	0.7752
0.6762	6.5	832	0.6533	0.7752
0.6277	7.0	896	0.6356	0.7886
0.6264	7.5	960	0.6008	0.7987
0.5906	8.0	1024	0.5791	0.8154
0.5596	8.5	1088	0.5789	0.8054
0.5619	9.0	1152	0.5785	0.7987
0.5381	9.5	1216	0.5532	0.8121
0.5275	10.0	1280	0.5591	0.8087
0.5183	10.5	1344	0.5555	0.8054
0.5236	11.0	1408	0.5551	0.8087
0.5198	11.5	1472	0.5649	0.8020
0.5128	12.0	1536	0.5312	0.8356
0.4932	12.5	1600	0.5238	0.8054
0.4854	13.0	1664	0.5234	0.8121
0.4674	13.5	1728	0.5142	0.8221
0.4614	14.0	1792	0.5109	0.8154
0.4558	14.5	1856	0.5095	0.8289
0.4419	15.0	1920	0.5043	0.8188
0.4362	15.5	1984	0.5034	0.8221
0.4496	16.0	2048	0.5032	0.8221
0.4484	16.5	2112	0.5017	0.8221
0.4325	17.0	2176	0.5015	0.8289
0.428	17.5	2240	0.4967	0.8221
0.4091	18.0	2304	0.4704	0.8356
0.405	18.5	2368	0.4792	0.8289
0.4012	19.0	2432	0.4750	0.8322
0.3887	19.5	2496	0.4750	0.8289
0.3986	20.0	2560	0.4711	0.8255
0.3983	20.5	2624	0.4713	0.8255
0.3857	21.0	2688	0.4750	0.8289
0.3925	21.5	2752	0.4506	0.8456
0.3787	22.0	2816	0.4622	0.8255
0.368	22.5	2880	0.4583	0.8389
0.3702	23.0	2944	0.4479	0.8423
0.3591	23.5	3008	0.4485	0.8389
0.3588	24.0	3072	0.4534	0.8356
0.3517	24.5	3136	0.4496	0.8356
0.3546	25.0	3200	0.4482	0.8389
0.3636	25.5	3264	0.4518	0.8356
0.3435	26.0	3328	0.4495	0.8322
0.3423	26.5	3392	0.4427	0.8322
0.3477	27.0	3456	0.4365	0.8423
0.3405	27.5	3520	0.4380	0.8389
0.3254	28.0	3584	0.4366	0.8389
0.3245	28.5	3648	0.4316	0.8423
0.3265	29.0	3712	0.4305	0.8423
0.3193	29.5	3776	0.4339	0.8456
0.3244	30.0	3840	0.4332	0.8456

Framework versions

• Transformers 4.48.3
• Pytorch 2.5.1+cu124
• Datasets 3.3.2
• Tokenizers 0.21.1