hushem_40x_deit_small_rms_0001_fold3
This model is a fine-tuned version of facebook/deit-small-patch16-224 on the imagefolder dataset. It achieves the following results on the evaluation set:
- Loss: 2.3623
- Accuracy: 0.8605
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: 0.0001
- train_batch_size: 32
- eval_batch_size: 32
- seed: 42
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- lr_scheduler_warmup_ratio: 0.1
- num_epochs: 50
Training results
| Training Loss | Epoch | Step | Validation Loss | Accuracy |
|---|---|---|---|---|
| 0.162 | 1.0 | 217 | 0.7397 | 0.8140 |
| 0.0554 | 2.0 | 434 | 0.5902 | 0.8605 |
| 0.0178 | 3.0 | 651 | 1.1734 | 0.8605 |
| 0.0009 | 4.0 | 868 | 1.2319 | 0.8372 |
| 0.0013 | 5.0 | 1085 | 1.7982 | 0.7442 |
| 0.0274 | 6.0 | 1302 | 1.0518 | 0.8140 |
| 0.0022 | 7.0 | 1519 | 1.2789 | 0.7907 |
| 0.0002 | 8.0 | 1736 | 1.6091 | 0.7907 |
| 0.0002 | 9.0 | 1953 | 1.3608 | 0.7907 |
| 0.0001 | 10.0 | 2170 | 1.7662 | 0.7674 |
| 0.0001 | 11.0 | 2387 | 1.4719 | 0.8372 |
| 0.0001 | 12.0 | 2604 | 0.9802 | 0.8837 |
| 0.0537 | 13.0 | 2821 | 1.7727 | 0.8140 |
| 0.0 | 14.0 | 3038 | 1.4355 | 0.8372 |
| 0.0002 | 15.0 | 3255 | 1.2526 | 0.8140 |
| 0.0071 | 16.0 | 3472 | 1.9556 | 0.7674 |
| 0.0 | 17.0 | 3689 | 1.8517 | 0.7907 |
| 0.0016 | 18.0 | 3906 | 1.4335 | 0.8372 |
| 0.0124 | 19.0 | 4123 | 1.3513 | 0.7907 |
| 0.0235 | 20.0 | 4340 | 2.0239 | 0.7907 |
| 0.0 | 21.0 | 4557 | 1.2893 | 0.8605 |
| 0.0 | 22.0 | 4774 | 1.3114 | 0.8605 |
| 0.0 | 23.0 | 4991 | 1.3523 | 0.8605 |
| 0.0 | 24.0 | 5208 | 1.4204 | 0.8372 |
| 0.0 | 25.0 | 5425 | 1.5136 | 0.8372 |
| 0.0 | 26.0 | 5642 | 1.6287 | 0.8605 |
| 0.0 | 27.0 | 5859 | 1.7481 | 0.8605 |
| 0.0 | 28.0 | 6076 | 1.8569 | 0.8605 |
| 0.0 | 29.0 | 6293 | 1.9482 | 0.8605 |
| 0.0 | 30.0 | 6510 | 2.0219 | 0.8605 |
| 0.0 | 31.0 | 6727 | 2.0881 | 0.8605 |
| 0.0 | 32.0 | 6944 | 2.1406 | 0.8605 |
| 0.0 | 33.0 | 7161 | 2.1867 | 0.8605 |
| 0.0 | 34.0 | 7378 | 2.2231 | 0.8605 |
| 0.0 | 35.0 | 7595 | 2.2508 | 0.8605 |
| 0.0 | 36.0 | 7812 | 2.2725 | 0.8605 |
| 0.0 | 37.0 | 8029 | 2.2899 | 0.8605 |
| 0.0 | 38.0 | 8246 | 2.3039 | 0.8605 |
| 0.0 | 39.0 | 8463 | 2.3156 | 0.8605 |
| 0.0 | 40.0 | 8680 | 2.3253 | 0.8605 |
| 0.0 | 41.0 | 8897 | 2.3335 | 0.8605 |
| 0.0 | 42.0 | 9114 | 2.3403 | 0.8605 |
| 0.0 | 43.0 | 9331 | 2.3460 | 0.8605 |
| 0.0 | 44.0 | 9548 | 2.3507 | 0.8605 |
| 0.0 | 45.0 | 9765 | 2.3545 | 0.8605 |
| 0.0 | 46.0 | 9982 | 2.3575 | 0.8605 |
| 0.0 | 47.0 | 10199 | 2.3597 | 0.8605 |
| 0.0 | 48.0 | 10416 | 2.3612 | 0.8605 |
| 0.0 | 49.0 | 10633 | 2.3621 | 0.8605 |
| 0.0 | 50.0 | 10850 | 2.3623 | 0.8605 |
Framework versions
- Transformers 4.32.1
- Pytorch 2.1.0+cu121
- Datasets 2.12.0
- Tokenizers 0.13.2