fold_1
This model is a fine-tuned version of google/vit-base-patch16-224 on the imagefolder dataset. It achieves the following results on the evaluation set:
- Loss: 1.0725
- Accuracy: 0.9793
- F1 Score: 0.9805
- Recall: 0.9804
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: 150
- num_epochs: 100
- label_smoothing_factor: 0.15
Training results
| Training Loss | Epoch | Step | Validation Loss | Accuracy | F1 Score | Recall |
|---|---|---|---|---|---|---|
| 2.9237 | 1.0 | 19 | 2.7962 | 0.3138 | 0.2811 | 0.3322 |
| 2.7282 | 2.0 | 38 | 2.5914 | 0.4172 | 0.3767 | 0.4048 |
| 2.4043 | 3.0 | 57 | 2.3162 | 0.6034 | 0.5791 | 0.5798 |
| 2.0554 | 4.0 | 76 | 1.9844 | 0.7552 | 0.7492 | 0.7470 |
| 1.6234 | 5.0 | 95 | 1.6522 | 0.8276 | 0.8322 | 0.8284 |
| 1.3940 | 6.0 | 114 | 1.4163 | 0.9069 | 0.9144 | 0.9180 |
| 1.2166 | 7.0 | 133 | 1.2960 | 0.9345 | 0.9396 | 0.9426 |
| 1.1550 | 8.0 | 152 | 1.2153 | 0.9586 | 0.9616 | 0.9608 |
| 1.1312 | 9.0 | 171 | 1.1790 | 0.9621 | 0.9641 | 0.9632 |
| 1.0838 | 10.0 | 190 | 1.1564 | 0.9690 | 0.9706 | 0.9706 |
| 1.0738 | 11.0 | 209 | 1.1289 | 0.9655 | 0.9673 | 0.9669 |
| 1.0487 | 12.0 | 228 | 1.1151 | 0.9690 | 0.9711 | 0.9706 |
| 1.0299 | 13.0 | 247 | 1.1217 | 0.9724 | 0.9745 | 0.9767 |
| 1.0164 | 14.0 | 266 | 1.1016 | 0.9759 | 0.9774 | 0.9779 |
| 0.9999 | 15.0 | 285 | 1.1021 | 0.9621 | 0.9647 | 0.9632 |
| 0.9976 | 16.0 | 304 | 1.0900 | 0.9724 | 0.9744 | 0.9755 |
| 0.9823 | 17.0 | 323 | 1.0973 | 0.9586 | 0.9614 | 0.9596 |
| 0.9854 | 18.0 | 342 | 1.0897 | 0.9621 | 0.9644 | 0.9620 |
| 0.9863 | 19.0 | 361 | 1.0825 | 0.9759 | 0.9774 | 0.9779 |
| 0.9700 | 20.0 | 380 | 1.0744 | 0.9759 | 0.9774 | 0.9779 |
| 0.9759 | 21.0 | 399 | 1.0824 | 0.9655 | 0.9677 | 0.9657 |
| 0.9796 | 22.0 | 418 | 1.0725 | 0.9793 | 0.9805 | 0.9804 |
| 0.9706 | 23.0 | 437 | 1.0655 | 0.9724 | 0.9731 | 0.9729 |
| 0.9635 | 24.0 | 456 | 1.0727 | 0.9759 | 0.9759 | 0.9742 |
| 0.9672 | 25.0 | 475 | 1.0723 | 0.9724 | 0.9731 | 0.9729 |
| 0.9662 | 26.0 | 494 | 1.0667 | 0.9724 | 0.9731 | 0.9729 |
| 0.9655 | 27.0 | 513 | 1.0770 | 0.9724 | 0.9730 | 0.9717 |
| 0.9641 | 28.0 | 532 | 1.0689 | 0.9759 | 0.9762 | 0.9766 |
| 0.9675 | 29.0 | 551 | 1.0696 | 0.9690 | 0.9698 | 0.9680 |
| 0.9609 | 30.0 | 570 | 1.0618 | 0.9690 | 0.9700 | 0.9693 |
| 0.9661 | 31.0 | 589 | 1.0660 | 0.9724 | 0.9731 | 0.9729 |
| 0.9630 | 32.0 | 608 | 1.0684 | 0.9724 | 0.9731 | 0.9729 |
| 0.9706 | 33.0 | 627 | 1.0765 | 0.9724 | 0.9730 | 0.9717 |
| 0.9575 | 34.0 | 646 | 1.0753 | 0.9655 | 0.9669 | 0.9668 |
| 0.9607 | 35.0 | 665 | 1.0766 | 0.9690 | 0.9700 | 0.9693 |
| 0.9589 | 36.0 | 684 | 1.0685 | 0.9759 | 0.9761 | 0.9754 |
| 0.9569 | 37.0 | 703 | 1.0672 | 0.9759 | 0.9761 | 0.9754 |
| 0.9572 | 38.0 | 722 | 1.0670 | 0.9724 | 0.9732 | 0.9742 |
| 0.9606 | 39.0 | 741 | 1.0680 | 0.9690 | 0.9700 | 0.9693 |
| 0.9552 | 40.0 | 760 | 1.0706 | 0.9724 | 0.9732 | 0.9729 |
Framework versions
- Transformers 5.0.0
- Pytorch 2.10.0+cu128
- Datasets 5.0.0
- Tokenizers 0.22.2