metadata
library_name: transformers
license: apache-2.0
base_model: WinKawaks/vit-tiny-patch16-224
tags:
- generated_from_trainer
metrics:
- accuracy
model-index:
- name: vit-tiny-patch16-224_rice-leaf-disease-augmented-v4_fft
results: []
vit-tiny-patch16-224_rice-leaf-disease-augmented-v4_fft
This model is a fine-tuned version of WinKawaks/vit-tiny-patch16-224 on an unknown dataset. It achieves the following results on the evaluation set:
- Loss: 0.3674
- Accuracy: 0.9262
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: 5e-05
- 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 | Accuracy | Validation Loss |
|---|---|---|---|---|
| 2.0564 | 0.5 | 64 | 0.4899 | 1.4541 |
| 1.0767 | 1.0 | 128 | 0.7651 | 0.6909 |
| 0.4917 | 1.5 | 192 | 0.8322 | 0.4307 |
| 0.285 | 2.0 | 256 | 0.9027 | 0.2932 |
| 0.0902 | 2.5 | 320 | 0.8993 | 0.3134 |
| 0.0588 | 3.0 | 384 | 0.9161 | 0.3076 |
| 0.0155 | 3.5 | 448 | 0.9396 | 0.2627 |
| 0.0066 | 4.0 | 512 | 0.9295 | 0.2992 |
| 0.0017 | 4.5 | 576 | 0.9228 | 0.2936 |
| 0.0009 | 5.0 | 640 | 0.9228 | 0.2961 |
| 0.0006 | 5.5 | 704 | 0.9228 | 0.3005 |
| 0.0005 | 6.0 | 768 | 0.9228 | 0.3004 |
| 0.0005 | 6.5 | 832 | 0.9262 | 0.2867 |
| 0.0004 | 7.0 | 896 | 0.9295 | 0.2977 |
| 0.0003 | 7.5 | 960 | 0.9295 | 0.2944 |
| 0.0002 | 8.0 | 1024 | 0.9295 | 0.3074 |
| 0.0002 | 8.5 | 1088 | 0.9329 | 0.3053 |
| 0.0002 | 9.0 | 1152 | 0.9295 | 0.3098 |
| 0.0001 | 9.5 | 1216 | 0.9295 | 0.3102 |
| 0.0001 | 10.0 | 1280 | 0.9262 | 0.3105 |
| 0.0001 | 10.5 | 1344 | 0.9262 | 0.3105 |
| 0.0001 | 11.0 | 1408 | 0.9262 | 0.3202 |
| 0.0001 | 11.5 | 1472 | 0.9295 | 0.3183 |
| 0.0001 | 12.0 | 1536 | 0.9329 | 0.3131 |
| 0.0001 | 12.5 | 1600 | 0.9295 | 0.3157 |
| 0.0001 | 13.0 | 1664 | 0.9228 | 0.3238 |
| 0.0001 | 13.5 | 1728 | 0.9228 | 0.3220 |
| 0.0001 | 14.0 | 1792 | 0.9228 | 0.3266 |
| 0.0001 | 14.5 | 1856 | 0.9228 | 0.3274 |
| 0.0001 | 15.0 | 1920 | 0.9228 | 0.3269 |
| 0.0001 | 15.5 | 1984 | 0.3267 | 0.9262 |
| 0.0001 | 16.0 | 2048 | 0.3298 | 0.9228 |
| 0.0001 | 16.5 | 2112 | 0.3330 | 0.9228 |
| 0.0001 | 17.0 | 2176 | 0.3337 | 0.9228 |
| 0.0001 | 17.5 | 2240 | 0.3337 | 0.9228 |
| 0.0001 | 18.0 | 2304 | 0.3355 | 0.9228 |
| 0.0 | 18.5 | 2368 | 0.3346 | 0.9228 |
| 0.0 | 19.0 | 2432 | 0.3360 | 0.9228 |
| 0.0 | 19.5 | 2496 | 0.3368 | 0.9228 |
| 0.0 | 20.0 | 2560 | 0.3365 | 0.9228 |
| 0.0 | 20.5 | 2624 | 0.3364 | 0.9228 |
| 0.0 | 21.0 | 2688 | 0.3412 | 0.9228 |
| 0.0 | 21.5 | 2752 | 0.3414 | 0.9228 |
| 0.0 | 22.0 | 2816 | 0.3435 | 0.9262 |
| 0.0 | 22.5 | 2880 | 0.3557 | 0.9228 |
| 0.0 | 23.0 | 2944 | 0.3490 | 0.9295 |
| 0.0 | 23.5 | 3008 | 0.3564 | 0.9262 |
| 0.0 | 24.0 | 3072 | 0.3545 | 0.9295 |
| 0.0 | 24.5 | 3136 | 0.3577 | 0.9262 |
| 0.0 | 25.0 | 3200 | 0.3597 | 0.9262 |
| 0.0 | 25.5 | 3264 | 0.3632 | 0.9262 |
| 0.0 | 26.0 | 3328 | 0.3627 | 0.9262 |
| 0.0 | 26.5 | 3392 | 0.3650 | 0.9262 |
| 0.0 | 27.0 | 3456 | 0.3664 | 0.9262 |
| 0.0 | 27.5 | 3520 | 0.3664 | 0.9262 |
| 0.0 | 28.0 | 3584 | 0.3666 | 0.9262 |
| 0.0 | 28.5 | 3648 | 0.3666 | 0.9262 |
| 0.0 | 29.0 | 3712 | 0.3670 | 0.9262 |
| 0.0 | 29.5 | 3776 | 0.3673 | 0.9262 |
| 0.0 | 30.0 | 3840 | 0.3674 | 0.9262 |
Framework versions
- Transformers 4.48.3
- Pytorch 2.5.1+cu124
- Datasets 3.3.2
- Tokenizers 0.21.1