hushem_5x_deit_small_adamax_001_fold2
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: 5.1428
- Accuracy: 0.5556
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: 0.001
- 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 |
|---|---|---|---|---|
| 1.3869 | 1.0 | 27 | 1.3961 | 0.2444 |
| 1.1374 | 2.0 | 54 | 1.4468 | 0.4222 |
| 0.928 | 3.0 | 81 | 1.5182 | 0.5111 |
| 1.0235 | 4.0 | 108 | 1.5182 | 0.5333 |
| 0.797 | 5.0 | 135 | 1.2483 | 0.3778 |
| 0.8238 | 6.0 | 162 | 1.2902 | 0.5333 |
| 0.7166 | 7.0 | 189 | 1.5715 | 0.4889 |
| 0.6354 | 8.0 | 216 | 1.5461 | 0.4 |
| 0.5304 | 9.0 | 243 | 2.1851 | 0.5111 |
| 0.5677 | 10.0 | 270 | 2.3132 | 0.4222 |
| 0.4857 | 11.0 | 297 | 2.4291 | 0.5556 |
| 0.3582 | 12.0 | 324 | 2.5091 | 0.4444 |
| 0.3326 | 13.0 | 351 | 2.6635 | 0.5556 |
| 0.3365 | 14.0 | 378 | 2.7106 | 0.6 |
| 0.3145 | 15.0 | 405 | 3.0588 | 0.5556 |
| 0.1837 | 16.0 | 432 | 3.1156 | 0.5778 |
| 0.277 | 17.0 | 459 | 2.8497 | 0.5778 |
| 0.1371 | 18.0 | 486 | 3.1885 | 0.6222 |
| 0.2815 | 19.0 | 513 | 3.0392 | 0.5778 |
| 0.1153 | 20.0 | 540 | 3.8749 | 0.5778 |
| 0.2394 | 21.0 | 567 | 2.9618 | 0.5111 |
| 0.2767 | 22.0 | 594 | 2.7586 | 0.5333 |
| 0.0906 | 23.0 | 621 | 4.0821 | 0.5333 |
| 0.1246 | 24.0 | 648 | 3.7041 | 0.4889 |
| 0.0582 | 25.0 | 675 | 4.8478 | 0.5778 |
| 0.0262 | 26.0 | 702 | 3.9021 | 0.5556 |
| 0.0436 | 27.0 | 729 | 4.3474 | 0.5778 |
| 0.0484 | 28.0 | 756 | 4.4662 | 0.5778 |
| 0.0031 | 29.0 | 783 | 4.8306 | 0.5778 |
| 0.0039 | 30.0 | 810 | 4.6722 | 0.5778 |
| 0.023 | 31.0 | 837 | 4.3620 | 0.5556 |
| 0.0011 | 32.0 | 864 | 4.9095 | 0.5556 |
| 0.0004 | 33.0 | 891 | 4.8409 | 0.5333 |
| 0.0003 | 34.0 | 918 | 5.0267 | 0.5556 |
| 0.0001 | 35.0 | 945 | 5.0410 | 0.5556 |
| 0.0001 | 36.0 | 972 | 5.0651 | 0.5556 |
| 0.0001 | 37.0 | 999 | 5.0839 | 0.5556 |
| 0.0001 | 38.0 | 1026 | 5.0966 | 0.5556 |
| 0.0001 | 39.0 | 1053 | 5.1068 | 0.5556 |
| 0.0 | 40.0 | 1080 | 5.1156 | 0.5556 |
| 0.0001 | 41.0 | 1107 | 5.1226 | 0.5556 |
| 0.0001 | 42.0 | 1134 | 5.1286 | 0.5556 |
| 0.0001 | 43.0 | 1161 | 5.1335 | 0.5556 |
| 0.0 | 44.0 | 1188 | 5.1362 | 0.5556 |
| 0.0 | 45.0 | 1215 | 5.1386 | 0.5556 |
| 0.0001 | 46.0 | 1242 | 5.1410 | 0.5556 |
| 0.0 | 47.0 | 1269 | 5.1424 | 0.5556 |
| 0.0001 | 48.0 | 1296 | 5.1428 | 0.5556 |
| 0.0 | 49.0 | 1323 | 5.1428 | 0.5556 |
| 0.0001 | 50.0 | 1350 | 5.1428 | 0.5556 |
Framework versions
- Transformers 4.35.2
- Pytorch 2.1.0+cu118
- Datasets 2.15.0
- Tokenizers 0.15.0