vit_fold_1_v3
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: 1.1599
- Accuracy: 0.9519
- F1 Score: 0.9528
- Recall: 0.9537
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 |
|---|---|---|---|---|---|---|
| 2.8856 | 1.0 | 20 | 2.9355 | 0.2308 | 0.2215 | 0.2173 |
| 2.6469 | 2.0 | 40 | 2.7349 | 0.3462 | 0.3300 | 0.3278 |
| 2.3026 | 3.0 | 60 | 2.4305 | 0.5385 | 0.5365 | 0.5293 |
| 1.9124 | 4.0 | 80 | 2.0391 | 0.7179 | 0.7186 | 0.7155 |
| 1.5476 | 5.0 | 100 | 1.6601 | 0.8109 | 0.8161 | 0.8111 |
| 1.3336 | 6.0 | 120 | 1.4040 | 0.8910 | 0.8952 | 0.8987 |
| 1.2139 | 7.0 | 140 | 1.2838 | 0.9199 | 0.9223 | 0.9207 |
| 1.1520 | 8.0 | 160 | 1.2277 | 0.9295 | 0.9316 | 0.9306 |
| 1.0852 | 9.0 | 180 | 1.2013 | 0.9455 | 0.9467 | 0.9476 |
| 1.0903 | 10.0 | 200 | 1.1819 | 0.9423 | 0.9436 | 0.9439 |
| 1.0677 | 11.0 | 220 | 1.1724 | 0.9487 | 0.9499 | 0.9513 |
| 1.0281 | 12.0 | 240 | 1.1847 | 0.9487 | 0.9499 | 0.9513 |
| 1.0140 | 13.0 | 260 | 1.1599 | 0.9519 | 0.9528 | 0.9537 |
| 1.0105 | 14.0 | 280 | 1.1549 | 0.9519 | 0.9528 | 0.9537 |
| 0.9910 | 15.0 | 300 | 1.1596 | 0.9519 | 0.9528 | 0.9537 |
| 0.9909 | 16.0 | 320 | 1.1607 | 0.9519 | 0.9525 | 0.9525 |
| 0.9890 | 17.0 | 340 | 1.1589 | 0.9487 | 0.9496 | 0.9501 |
| 0.9742 | 18.0 | 360 | 1.1694 | 0.9487 | 0.9496 | 0.9501 |
| 0.9855 | 19.0 | 380 | 1.1700 | 0.9455 | 0.9467 | 0.9476 |
| 0.9839 | 20.0 | 400 | 1.1701 | 0.9487 | 0.9496 | 0.9501 |
| 0.9736 | 21.0 | 420 | 1.1686 | 0.9487 | 0.9496 | 0.9501 |
| 0.9678 | 22.0 | 440 | 1.1670 | 0.9487 | 0.9496 | 0.9501 |
| 0.9718 | 23.0 | 460 | 1.1681 | 0.9519 | 0.9525 | 0.9525 |
| 0.9782 | 24.0 | 480 | 1.1579 | 0.9519 | 0.9525 | 0.9525 |
Framework versions
- Transformers 5.0.0
- Pytorch 2.10.0+cu128
- Datasets 5.0.0
- Tokenizers 0.22.2