efficientnet-b2_rice-leaf-disease-augmented-v4_v5_fft
This model is a fine-tuned version of google/efficientnet-b2 on an unknown dataset. It achieves the following results on the evaluation set:
- Loss: 0.2236
- Accuracy: 0.9362
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 |
|---|---|---|---|---|
| 2.0775 | 0.5 | 64 | 2.0200 | 0.2282 |
| 1.9215 | 1.0 | 128 | 1.7530 | 0.5336 |
| 1.5941 | 1.5 | 192 | 1.3576 | 0.6309 |
| 1.1312 | 2.0 | 256 | 0.8490 | 0.7517 |
| 0.6799 | 2.5 | 320 | 0.5743 | 0.8221 |
| 0.4743 | 3.0 | 384 | 0.4281 | 0.8624 |
| 0.2937 | 3.5 | 448 | 0.3946 | 0.8893 |
| 0.2342 | 4.0 | 512 | 0.3713 | 0.8758 |
| 0.1563 | 4.5 | 576 | 0.3339 | 0.8893 |
| 0.1296 | 5.0 | 640 | 0.2886 | 0.9128 |
| 0.1026 | 5.5 | 704 | 0.3032 | 0.8926 |
| 0.1009 | 6.0 | 768 | 0.2951 | 0.8893 |
| 0.0956 | 6.5 | 832 | 0.2795 | 0.9128 |
| 0.0817 | 7.0 | 896 | 0.3031 | 0.9094 |
| 0.0591 | 7.5 | 960 | 0.2778 | 0.9195 |
| 0.0444 | 8.0 | 1024 | 0.2435 | 0.9060 |
| 0.0268 | 8.5 | 1088 | 0.2506 | 0.9228 |
| 0.0198 | 9.0 | 1152 | 0.2692 | 0.8993 |
| 0.0139 | 9.5 | 1216 | 0.2384 | 0.9195 |
| 0.0164 | 10.0 | 1280 | 0.2712 | 0.9195 |
| 0.0118 | 10.5 | 1344 | 0.2868 | 0.9128 |
| 0.0139 | 11.0 | 1408 | 0.2262 | 0.9295 |
| 0.0122 | 11.5 | 1472 | 0.2492 | 0.9128 |
| 0.0132 | 12.0 | 1536 | 0.2751 | 0.9128 |
| 0.0084 | 12.5 | 1600 | 0.3184 | 0.8993 |
| 0.0082 | 13.0 | 1664 | 0.2596 | 0.9228 |
| 0.0093 | 13.5 | 1728 | 0.2636 | 0.9228 |
| 0.0059 | 14.0 | 1792 | 0.2501 | 0.9262 |
| 0.0060 | 14.5 | 1856 | 0.3249 | 0.9027 |
| 0.0036 | 15.0 | 1920 | 0.2584 | 0.9228 |
| 0.0051 | 15.5 | 1984 | 0.2501 | 0.9161 |
| 0.0049 | 16.0 | 2048 | 0.2698 | 0.9228 |
| 0.0042 | 16.5 | 2112 | 0.2403 | 0.9262 |
| 0.0054 | 17.0 | 2176 | 0.2536 | 0.9262 |
| 0.0056 | 17.5 | 2240 | 0.2506 | 0.9228 |
| 0.0031 | 18.0 | 2304 | 0.3199 | 0.9027 |
| 0.0038 | 18.5 | 2368 | 0.3303 | 0.9228 |
| 0.0029 | 19.0 | 2432 | 0.2250 | 0.9295 |
| 0.0024 | 19.5 | 2496 | 0.2577 | 0.9161 |
| 0.0023 | 20.0 | 2560 | 0.2365 | 0.9396 |
| 0.0026 | 20.5 | 2624 | 0.2501 | 0.9295 |
| 0.0034 | 21.0 | 2688 | 0.2283 | 0.9262 |
| 0.0034 | 21.5 | 2752 | 0.2608 | 0.9195 |
| 0.0045 | 22.0 | 2816 | 0.3040 | 0.9094 |
| 0.0023 | 22.5 | 2880 | 0.2782 | 0.9228 |
| 0.0026 | 23.0 | 2944 | 0.2520 | 0.9161 |
| 0.0015 | 23.5 | 3008 | 0.2440 | 0.9228 |
| 0.0015 | 24.0 | 3072 | 0.2341 | 0.9362 |
| 0.0019 | 24.5 | 3136 | 0.2779 | 0.9161 |
| 0.0018 | 25.0 | 3200 | 0.2662 | 0.9362 |
| 0.0016 | 25.5 | 3264 | 0.2244 | 0.9362 |
| 0.0014 | 26.0 | 3328 | 0.3105 | 0.9161 |
| 0.0016 | 26.5 | 3392 | 0.2696 | 0.9195 |
| 0.0022 | 27.0 | 3456 | 0.2696 | 0.9262 |
| 0.0012 | 27.5 | 3520 | 0.2382 | 0.9362 |
| 0.0019 | 28.0 | 3584 | 0.2430 | 0.9329 |
| 0.0010 | 28.5 | 3648 | 0.2321 | 0.9396 |
| 0.0010 | 29.0 | 3712 | 0.2566 | 0.9161 |
| 0.0017 | 29.5 | 3776 | 0.2965 | 0.9295 |
| 0.0013 | 30.0 | 3840 | 0.2236 | 0.9362 |
Framework versions
- Transformers 4.48.3
- Pytorch 2.5.1+cu124
- Datasets 3.3.2
- Tokenizers 0.21.1