fold_2
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: 1.1294
- Accuracy: 0.9517
- F1 Score: 0.9555
- Recall: 0.9524
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 |
|---|---|---|---|---|---|---|
| 2.9670 | 1.0 | 19 | 2.9365 | 0.2310 | 0.2142 | 0.2367 |
| 2.8214 | 2.0 | 38 | 2.7318 | 0.3138 | 0.2945 | 0.3061 |
| 2.5025 | 3.0 | 57 | 2.4670 | 0.4655 | 0.4676 | 0.4694 |
| 2.1809 | 4.0 | 76 | 2.1553 | 0.6448 | 0.6653 | 0.6767 |
| 1.7435 | 5.0 | 95 | 1.8033 | 0.8 | 0.8207 | 0.8246 |
| 1.4272 | 6.0 | 114 | 1.5291 | 0.8621 | 0.8763 | 0.8784 |
| 1.2204 | 7.0 | 133 | 1.3853 | 0.8897 | 0.8988 | 0.9015 |
| 1.1890 | 8.0 | 152 | 1.3082 | 0.9241 | 0.9281 | 0.9282 |
| 1.1161 | 9.0 | 171 | 1.2493 | 0.9276 | 0.9321 | 0.9331 |
| 1.0740 | 10.0 | 190 | 1.2355 | 0.9345 | 0.9384 | 0.9405 |
| 1.0635 | 11.0 | 209 | 1.2155 | 0.9379 | 0.9410 | 0.9417 |
| 1.0404 | 12.0 | 228 | 1.2010 | 0.9379 | 0.9418 | 0.9440 |
| 1.0244 | 13.0 | 247 | 1.2017 | 0.9448 | 0.9483 | 0.9526 |
| 0.9996 | 14.0 | 266 | 1.1814 | 0.9414 | 0.9443 | 0.9440 |
| 1.0016 | 15.0 | 285 | 1.1746 | 0.9310 | 0.9346 | 0.9330 |
| 1.0121 | 16.0 | 304 | 1.1656 | 0.9414 | 0.9452 | 0.9489 |
| 1.0118 | 17.0 | 323 | 1.1659 | 0.9379 | 0.9421 | 0.9453 |
| 0.9903 | 18.0 | 342 | 1.1524 | 0.9379 | 0.9417 | 0.9415 |
| 0.9853 | 19.0 | 361 | 1.1513 | 0.9414 | 0.9449 | 0.9452 |
| 0.9830 | 20.0 | 380 | 1.1551 | 0.9448 | 0.9484 | 0.9502 |
| 0.9801 | 21.0 | 399 | 1.1524 | 0.9414 | 0.9450 | 0.9453 |
| 0.9789 | 22.0 | 418 | 1.1515 | 0.9414 | 0.9452 | 0.9465 |
| 0.9704 | 23.0 | 437 | 1.1479 | 0.9448 | 0.9486 | 0.9488 |
| 0.9875 | 24.0 | 456 | 1.1507 | 0.9379 | 0.9416 | 0.9404 |
| 0.9705 | 25.0 | 475 | 1.1499 | 0.9379 | 0.9415 | 0.9442 |
| 0.9650 | 26.0 | 494 | 1.1488 | 0.9345 | 0.9386 | 0.9417 |
| 0.9642 | 27.0 | 513 | 1.1540 | 0.9345 | 0.9389 | 0.9429 |
| 0.9709 | 28.0 | 532 | 1.1403 | 0.9448 | 0.9485 | 0.9513 |
| 0.9636 | 29.0 | 551 | 1.1398 | 0.9379 | 0.9410 | 0.9417 |
| 0.9617 | 30.0 | 570 | 1.1364 | 0.9517 | 0.9549 | 0.9573 |
| 0.9724 | 31.0 | 589 | 1.1384 | 0.9483 | 0.9510 | 0.9539 |
| 0.9596 | 32.0 | 608 | 1.1324 | 0.9414 | 0.9449 | 0.9452 |
| 0.9802 | 33.0 | 627 | 1.1548 | 0.9379 | 0.9419 | 0.9466 |
| 0.9668 | 34.0 | 646 | 1.1408 | 0.9448 | 0.9498 | 0.9536 |
| 0.9645 | 35.0 | 665 | 1.1354 | 0.9483 | 0.9522 | 0.9524 |
| 0.9646 | 36.0 | 684 | 1.1294 | 0.9517 | 0.9555 | 0.9524 |
| 0.9598 | 37.0 | 703 | 1.1203 | 0.9517 | 0.9552 | 0.9574 |
| 0.9645 | 38.0 | 722 | 1.1170 | 0.9517 | 0.9550 | 0.9562 |
| 0.9643 | 39.0 | 741 | 1.1292 | 0.9448 | 0.9489 | 0.9501 |
| 0.9602 | 40.0 | 760 | 1.1349 | 0.9448 | 0.9482 | 0.9464 |
| 0.9616 | 41.0 | 779 | 1.1334 | 0.9483 | 0.9517 | 0.9539 |
| 0.9576 | 42.0 | 798 | 1.1323 | 0.9448 | 0.9491 | 0.9513 |
| 0.9651 | 43.0 | 817 | 1.1420 | 0.9414 | 0.9467 | 0.9500 |
| 0.9596 | 44.0 | 836 | 1.1366 | 0.9414 | 0.9458 | 0.9451 |
| 0.9606 | 45.0 | 855 | 1.1412 | 0.9414 | 0.9462 | 0.9488 |
| 0.9604 | 46.0 | 874 | 1.1442 | 0.9414 | 0.9458 | 0.9489 |
| 0.9628 | 47.0 | 893 | 1.1434 | 0.9379 | 0.9427 | 0.9453 |
| 0.9597 | 48.0 | 912 | 1.1399 | 0.9414 | 0.9460 | 0.9476 |
Framework versions
- Transformers 5.0.0
- Pytorch 2.10.0+cu128
- Datasets 5.0.0
- Tokenizers 0.22.2