hushem_40x_deit_small_adamax_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: 0.7579
- Accuracy: 0.9302
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 |
|---|---|---|---|---|
| 0.0315 | 1.0 | 217 | 0.2572 | 0.9070 |
| 0.0049 | 2.0 | 434 | 0.4551 | 0.8837 |
| 0.0004 | 3.0 | 651 | 0.3965 | 0.8837 |
| 0.0001 | 4.0 | 868 | 0.4995 | 0.9070 |
| 0.0 | 5.0 | 1085 | 0.3370 | 0.9535 |
| 0.0 | 6.0 | 1302 | 0.4294 | 0.9302 |
| 0.0 | 7.0 | 1519 | 0.4525 | 0.9302 |
| 0.0 | 8.0 | 1736 | 0.4672 | 0.9302 |
| 0.0 | 9.0 | 1953 | 0.4797 | 0.9302 |
| 0.0 | 10.0 | 2170 | 0.4904 | 0.9302 |
| 0.0 | 11.0 | 2387 | 0.4947 | 0.9302 |
| 0.0 | 12.0 | 2604 | 0.5020 | 0.9302 |
| 0.0 | 13.0 | 2821 | 0.5084 | 0.9302 |
| 0.0 | 14.0 | 3038 | 0.5153 | 0.9302 |
| 0.0 | 15.0 | 3255 | 0.5246 | 0.9302 |
| 0.0 | 16.0 | 3472 | 0.5296 | 0.9302 |
| 0.0 | 17.0 | 3689 | 0.5346 | 0.9302 |
| 0.0 | 18.0 | 3906 | 0.5408 | 0.9302 |
| 0.0 | 19.0 | 4123 | 0.5469 | 0.9302 |
| 0.0 | 20.0 | 4340 | 0.5538 | 0.9302 |
| 0.0 | 21.0 | 4557 | 0.5570 | 0.9302 |
| 0.0 | 22.0 | 4774 | 0.5610 | 0.9302 |
| 0.0 | 23.0 | 4991 | 0.5712 | 0.9302 |
| 0.0 | 24.0 | 5208 | 0.5753 | 0.9302 |
| 0.0 | 25.0 | 5425 | 0.5846 | 0.9302 |
| 0.0 | 26.0 | 5642 | 0.5887 | 0.9302 |
| 0.0 | 27.0 | 5859 | 0.5949 | 0.9302 |
| 0.0 | 28.0 | 6076 | 0.6007 | 0.9302 |
| 0.0 | 29.0 | 6293 | 0.6068 | 0.9302 |
| 0.0 | 30.0 | 6510 | 0.6184 | 0.9302 |
| 0.0 | 31.0 | 6727 | 0.6280 | 0.9302 |
| 0.0 | 32.0 | 6944 | 0.6394 | 0.9302 |
| 0.0 | 33.0 | 7161 | 0.6407 | 0.9302 |
| 0.0 | 34.0 | 7378 | 0.6480 | 0.9302 |
| 0.0 | 35.0 | 7595 | 0.6588 | 0.9302 |
| 0.0 | 36.0 | 7812 | 0.6700 | 0.9302 |
| 0.0 | 37.0 | 8029 | 0.6709 | 0.9302 |
| 0.0 | 38.0 | 8246 | 0.6850 | 0.9302 |
| 0.0 | 39.0 | 8463 | 0.6933 | 0.9302 |
| 0.0 | 40.0 | 8680 | 0.7079 | 0.9302 |
| 0.0 | 41.0 | 8897 | 0.7123 | 0.9302 |
| 0.0 | 42.0 | 9114 | 0.7231 | 0.9302 |
| 0.0 | 43.0 | 9331 | 0.7313 | 0.9302 |
| 0.0 | 44.0 | 9548 | 0.7417 | 0.9302 |
| 0.0 | 45.0 | 9765 | 0.7473 | 0.9302 |
| 0.0 | 46.0 | 9982 | 0.7513 | 0.9302 |
| 0.0 | 47.0 | 10199 | 0.7551 | 0.9302 |
| 0.0 | 48.0 | 10416 | 0.7564 | 0.9302 |
| 0.0 | 49.0 | 10633 | 0.7578 | 0.9302 |
| 0.0 | 50.0 | 10850 | 0.7579 | 0.9302 |
Framework versions
- Transformers 4.32.1
- Pytorch 2.1.0+cu121
- Datasets 2.12.0
- Tokenizers 0.13.2