vit-base-rice-disease-results
This model is a fine-tuned version of google/vit-base-patch16-224 on the rice-diseases-image-dataset dataset. It achieves the following results on the evaluation set:
- Loss: 0.8920
- Accuracy: 0.9655
- F1 Score: 0.9688
- Recall: 0.9710
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: 1e-05
- train_batch_size: 64
- eval_batch_size: 64
- seed: 42
- optimizer: Use OptimizerNames.ADAMW_TORCH_FUSED with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
- lr_scheduler_type: cosine
- lr_scheduler_warmup_steps: 0.2
- num_epochs: 80
- label_smoothing_factor: 0.1
Training results
| Training Loss | Epoch | Step | Validation Loss | Accuracy | F1 Score | Recall |
|---|---|---|---|---|---|---|
| 2.9327 | 1.0 | 19 | 2.9028 | 0.3241 | 0.2608 | 0.2838 |
| 2.8530 | 2.0 | 38 | 2.7941 | 0.3448 | 0.2745 | 0.2963 |
| 2.6606 | 3.0 | 57 | 2.6310 | 0.3931 | 0.3394 | 0.3463 |
| 2.4447 | 4.0 | 76 | 2.4248 | 0.4966 | 0.4610 | 0.4586 |
| 2.1496 | 5.0 | 95 | 2.1769 | 0.6138 | 0.5982 | 0.5839 |
| 1.8553 | 6.0 | 114 | 1.8718 | 0.8 | 0.8113 | 0.8024 |
| 1.4903 | 7.0 | 133 | 1.5533 | 0.8759 | 0.8884 | 0.8879 |
| 1.2108 | 8.0 | 152 | 1.2939 | 0.8966 | 0.9076 | 0.9097 |
| 1.0539 | 9.0 | 171 | 1.1431 | 0.9310 | 0.9397 | 0.9465 |
| 0.9732 | 10.0 | 190 | 1.0671 | 0.9310 | 0.9397 | 0.9465 |
| 0.9214 | 11.0 | 209 | 1.0210 | 0.9310 | 0.9397 | 0.9465 |
| 0.8485 | 12.0 | 228 | 1.0281 | 0.9310 | 0.9397 | 0.9465 |
| 0.8726 | 13.0 | 247 | 0.9690 | 0.9517 | 0.9571 | 0.9612 |
| 0.8211 | 14.0 | 266 | 0.9582 | 0.9517 | 0.9571 | 0.9612 |
| 0.8426 | 15.0 | 285 | 0.9242 | 0.9517 | 0.9571 | 0.9612 |
| 0.7897 | 16.0 | 304 | 0.9435 | 0.9517 | 0.9571 | 0.9612 |
| 0.7950 | 17.0 | 323 | 0.9363 | 0.9586 | 0.9629 | 0.9661 |
| 0.7542 | 18.0 | 342 | 0.9241 | 0.9586 | 0.9629 | 0.9661 |
| 0.7658 | 19.0 | 361 | 0.9148 | 0.9586 | 0.9629 | 0.9661 |
| 0.7524 | 20.0 | 380 | 0.9399 | 0.9517 | 0.9571 | 0.9612 |
| 0.7312 | 21.0 | 399 | 0.9408 | 0.9517 | 0.9571 | 0.9612 |
| 0.7440 | 22.0 | 418 | 0.8920 | 0.9655 | 0.9688 | 0.9710 |
| 0.7229 | 23.0 | 437 | 0.9685 | 0.9379 | 0.9455 | 0.9514 |
| 0.7308 | 24.0 | 456 | 0.8833 | 0.9655 | 0.9688 | 0.9710 |
| 0.7232 | 25.0 | 475 | 0.9071 | 0.9655 | 0.9688 | 0.9710 |
| 0.7350 | 26.0 | 494 | 0.9186 | 0.9655 | 0.9688 | 0.9710 |
| 0.7203 | 27.0 | 513 | 0.8773 | 0.9655 | 0.9688 | 0.9710 |
| 0.7180 | 28.0 | 532 | 0.8870 | 0.9655 | 0.9688 | 0.9710 |
| 0.7211 | 29.0 | 551 | 0.9010 | 0.9655 | 0.9688 | 0.9710 |
| 0.7583 | 30.0 | 570 | 0.8894 | 0.9655 | 0.9688 | 0.9710 |
| 0.7169 | 31.0 | 589 | 0.8782 | 0.9655 | 0.9688 | 0.9710 |
| 0.7125 | 32.0 | 608 | 0.8701 | 0.9655 | 0.9688 | 0.9710 |
Framework versions
- Transformers 5.0.0
- Pytorch 2.10.0+cu128
- Datasets 5.0.0
- Tokenizers 0.22.2