vit_small_patch16_224.augreg_in21k_rice-leaf-disease-augmented-v4_v5_fft
This model is a fine-tuned version of timm/vit_small_patch16_224.augreg_in21k on an unknown dataset. It achieves the following results on the evaluation set:
- Loss: 0.2267
- Accuracy: 0.9463
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 | Validation Loss | Accuracy |
|---|---|---|---|---|
| 2.1289 | 0.5 | 64 | 1.5575 | 0.4799 |
| 1.0829 | 1.0 | 128 | 0.6383 | 0.8020 |
| 0.4038 | 1.5 | 192 | 0.3332 | 0.8792 |
| 0.1843 | 2.0 | 256 | 0.2121 | 0.9362 |
| 0.0424 | 2.5 | 320 | 0.2280 | 0.9295 |
| 0.0314 | 3.0 | 384 | 0.2349 | 0.9262 |
| 0.0064 | 3.5 | 448 | 0.1955 | 0.9295 |
| 0.0028 | 4.0 | 512 | 0.2161 | 0.9430 |
| 0.0011 | 4.5 | 576 | 0.2301 | 0.9396 |
| 0.0007 | 5.0 | 640 | 0.2116 | 0.9362 |
| 0.0004 | 5.5 | 704 | 0.2168 | 0.9396 |
| 0.0004 | 6.0 | 768 | 0.2157 | 0.9396 |
| 0.0004 | 6.5 | 832 | 0.2157 | 0.9396 |
| 0.0004 | 7.0 | 896 | 0.2119 | 0.9396 |
| 0.0003 | 7.5 | 960 | 0.2073 | 0.9396 |
| 0.0002 | 8.0 | 1024 | 0.2137 | 0.9430 |
| 0.0002 | 8.5 | 1088 | 0.2117 | 0.9430 |
| 0.0002 | 9.0 | 1152 | 0.2186 | 0.9430 |
| 0.0001 | 9.5 | 1216 | 0.2160 | 0.9430 |
| 0.0001 | 10.0 | 1280 | 0.2149 | 0.9430 |
| 0.0001 | 10.5 | 1344 | 0.2146 | 0.9430 |
| 0.0001 | 11.0 | 1408 | 0.2149 | 0.9430 |
| 0.0001 | 11.5 | 1472 | 0.2171 | 0.9430 |
| 0.0001 | 12.0 | 1536 | 0.2166 | 0.9463 |
| 0.0001 | 12.5 | 1600 | 0.2188 | 0.9430 |
| 0.0001 | 13.0 | 1664 | 0.2153 | 0.9463 |
| 0.0001 | 13.5 | 1728 | 0.2152 | 0.9463 |
| 0.0001 | 14.0 | 1792 | 0.2166 | 0.9463 |
| 0.0001 | 14.5 | 1856 | 0.2157 | 0.9463 |
| 0.0001 | 15.0 | 1920 | 0.2170 | 0.9463 |
| 0.0001 | 15.5 | 1984 | 0.2172 | 0.9463 |
| 0.0001 | 16.0 | 2048 | 0.2171 | 0.9463 |
| 0.0001 | 16.5 | 2112 | 0.2163 | 0.9463 |
| 0.0001 | 17.0 | 2176 | 0.2212 | 0.9463 |
| 0.0001 | 17.5 | 2240 | 0.2194 | 0.9463 |
| 0.0001 | 18.0 | 2304 | 0.2188 | 0.9463 |
| 0.0 | 18.5 | 2368 | 0.2191 | 0.9463 |
| 0.0 | 19.0 | 2432 | 0.2203 | 0.9463 |
| 0.0 | 19.5 | 2496 | 0.2209 | 0.9463 |
| 0.0 | 20.0 | 2560 | 0.2206 | 0.9463 |
| 0.0 | 20.5 | 2624 | 0.2204 | 0.9463 |
| 0.0 | 21.0 | 2688 | 0.2202 | 0.9463 |
| 0.0 | 21.5 | 2752 | 0.2178 | 0.9463 |
| 0.0 | 22.0 | 2816 | 0.2239 | 0.9463 |
| 0.0 | 22.5 | 2880 | 0.2207 | 0.9430 |
| 0.0 | 23.0 | 2944 | 0.2199 | 0.9463 |
| 0.0 | 23.5 | 3008 | 0.2238 | 0.9430 |
| 0.0 | 24.0 | 3072 | 0.2219 | 0.9463 |
| 0.0 | 24.5 | 3136 | 0.2222 | 0.9430 |
| 0.0 | 25.0 | 3200 | 0.2222 | 0.9430 |
| 0.0 | 25.5 | 3264 | 0.2245 | 0.9463 |
| 0.0 | 26.0 | 3328 | 0.2237 | 0.9430 |
| 0.0 | 26.5 | 3392 | 0.2235 | 0.9463 |
| 0.0 | 27.0 | 3456 | 0.2248 | 0.9430 |
| 0.0 | 27.5 | 3520 | 0.2267 | 0.9463 |
| 0.0 | 28.0 | 3584 | 0.2241 | 0.9463 |
| 0.0 | 28.5 | 3648 | 0.2247 | 0.9463 |
| 0.0 | 29.0 | 3712 | 0.2262 | 0.9463 |
| 0.0 | 29.5 | 3776 | 0.2268 | 0.9463 |
| 0.0 | 30.0 | 3840 | 0.2267 | 0.9463 |
Framework versions
- Transformers 4.48.3
- Pytorch 2.5.1+cu124
- Datasets 3.3.2
- Tokenizers 0.21.1