vit-superhero-villain
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: 1.2902
- Accuracy: 0.7363
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: 5e-05
- train_batch_size: 64
- eval_batch_size: 8
- seed: 42
- optimizer: Use OptimizerNames.ADAMW_TORCH with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
- lr_scheduler_type: linear
- num_epochs: 20
- mixed_precision_training: Native AMP
Training results
| Training Loss | Epoch | Step | Validation Loss | Accuracy |
|---|---|---|---|---|
| No log | 1.0 | 26 | 1.4140 | 0.735 |
| 1.2713 | 2.0 | 52 | 1.3908 | 0.735 |
| 1.2713 | 3.0 | 78 | 1.3709 | 0.735 |
| 1.2028 | 4.0 | 104 | 1.3544 | 0.74 |
| 1.2028 | 5.0 | 130 | 1.3359 | 0.74 |
| 1.1776 | 6.0 | 156 | 1.3219 | 0.74 |
| 1.1776 | 7.0 | 182 | 1.3078 | 0.74 |
| 1.1515 | 8.0 | 208 | 1.2952 | 0.74 |
| 1.1515 | 9.0 | 234 | 1.2841 | 0.74 |
| 1.1519 | 10.0 | 260 | 1.2733 | 0.745 |
| 1.1519 | 11.0 | 286 | 1.2637 | 0.745 |
| 1.107 | 12.0 | 312 | 1.2557 | 0.745 |
| 1.107 | 13.0 | 338 | 1.2495 | 0.745 |
| 1.0611 | 14.0 | 364 | 1.2441 | 0.745 |
| 1.0611 | 15.0 | 390 | 1.2388 | 0.745 |
| 1.0748 | 16.0 | 416 | 1.2347 | 0.745 |
| 1.0748 | 17.0 | 442 | 1.2317 | 0.745 |
| 1.0563 | 18.0 | 468 | 1.2294 | 0.745 |
| 1.0563 | 19.0 | 494 | 1.2280 | 0.745 |
| 1.062 | 20.0 | 520 | 1.2277 | 0.745 |
Framework versions
- Transformers 4.50.3
- Pytorch 2.5.1+cu121
- Datasets 3.5.0
- Tokenizers 0.21.1