swin-tiny-rice-disease-results
This model is a fine-tuned version of microsoft/swin-tiny-patch4-window7-224 on the rice-diseases-image-dataset dataset. It achieves the following results on the evaluation set:
- Loss: 0.5802
- Accuracy: 0.8898
- F1 Score: 0.8879
- Recall: 0.8885
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: 0.00015
- 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: linear
- lr_scheduler_warmup_steps: 100
- num_epochs: 25
Training results
| Training Loss | Epoch | Step | Validation Loss | Accuracy | F1 Score | Recall |
|---|---|---|---|---|---|---|
| 2.2794 | 1.0 | 31 | 1.9852 | 0.5633 | 0.5269 | 0.5496 |
| 1.6545 | 2.0 | 62 | 1.3268 | 0.7388 | 0.7268 | 0.7358 |
| 1.2607 | 3.0 | 93 | 1.3093 | 0.7429 | 0.7350 | 0.7437 |
| 1.2308 | 4.0 | 124 | 0.9915 | 0.7796 | 0.7731 | 0.7713 |
| 1.0737 | 5.0 | 155 | 0.9625 | 0.7918 | 0.7874 | 0.7869 |
| 0.9774 | 6.0 | 186 | 0.8665 | 0.8286 | 0.8236 | 0.8251 |
| 0.9999 | 7.0 | 217 | 0.8700 | 0.8327 | 0.8298 | 0.8274 |
| 0.8674 | 8.0 | 248 | 0.8487 | 0.8245 | 0.8169 | 0.8196 |
| 0.8330 | 9.0 | 279 | 1.3820 | 0.7633 | 0.7485 | 0.7582 |
| 0.8289 | 10.0 | 310 | 0.9742 | 0.8490 | 0.8425 | 0.8438 |
| 0.7983 | 11.0 | 341 | 0.9251 | 0.8408 | 0.8397 | 0.8428 |
| 0.7231 | 12.0 | 372 | 0.8026 | 0.8408 | 0.8397 | 0.8458 |
| 0.6888 | 13.0 | 403 | 0.8638 | 0.8490 | 0.8466 | 0.8520 |
| 0.6415 | 14.0 | 434 | 0.7139 | 0.8612 | 0.8592 | 0.8624 |
| 0.6002 | 15.0 | 465 | 0.6743 | 0.8939 | 0.8946 | 0.8984 |
| 0.5523 | 16.0 | 496 | 0.6339 | 0.8857 | 0.8835 | 0.8837 |
| 0.5171 | 17.0 | 527 | 0.6958 | 0.8776 | 0.8757 | 0.8759 |
| 0.3976 | 18.0 | 558 | 0.6692 | 0.8735 | 0.8714 | 0.8725 |
| 0.4916 | 19.0 | 589 | 0.6365 | 0.8816 | 0.8793 | 0.8841 |
| 0.3095 | 20.0 | 620 | 0.5939 | 0.9143 | 0.9129 | 0.9165 |
| 0.3086 | 21.0 | 651 | 0.5382 | 0.9102 | 0.9087 | 0.9127 |
| 0.2511 | 22.0 | 682 | 0.5993 | 0.8980 | 0.8963 | 0.8974 |
| 0.2288 | 23.0 | 713 | 0.5829 | 0.8816 | 0.8797 | 0.8792 |
| 0.2095 | 24.0 | 744 | 0.5896 | 0.8816 | 0.8798 | 0.8783 |
| 0.1754 | 25.0 | 775 | 0.5802 | 0.8898 | 0.8879 | 0.8885 |
Framework versions
- Transformers 5.0.0
- Pytorch 2.10.0+cu128
- Datasets 5.0.0
- Tokenizers 0.22.2