hushem_5x_deit_base_adamax_001_fold1
This model is a fine-tuned version of facebook/deit-base-patch16-224 on the imagefolder dataset. It achieves the following results on the evaluation set:
- Loss: 5.3716
- Accuracy: 0.4444
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 |
|---|---|---|---|---|
| 1.4051 | 1.0 | 27 | 1.4119 | 0.2667 |
| 1.4631 | 2.0 | 54 | 1.4906 | 0.2444 |
| 1.1049 | 3.0 | 81 | 1.5988 | 0.2889 |
| 1.1893 | 4.0 | 108 | 1.5499 | 0.3778 |
| 0.8095 | 5.0 | 135 | 1.4187 | 0.3778 |
| 0.7867 | 6.0 | 162 | 1.5747 | 0.4444 |
| 0.8062 | 7.0 | 189 | 1.7621 | 0.3778 |
| 0.6185 | 8.0 | 216 | 1.6443 | 0.3778 |
| 0.6622 | 9.0 | 243 | 2.0501 | 0.4 |
| 0.7761 | 10.0 | 270 | 1.5776 | 0.3778 |
| 0.7478 | 11.0 | 297 | 1.4437 | 0.3556 |
| 0.6339 | 12.0 | 324 | 1.4821 | 0.4 |
| 0.6965 | 13.0 | 351 | 1.6662 | 0.3556 |
| 0.5829 | 14.0 | 378 | 1.5844 | 0.3778 |
| 0.4908 | 15.0 | 405 | 2.2476 | 0.3778 |
| 0.5234 | 16.0 | 432 | 1.3445 | 0.4222 |
| 0.4872 | 17.0 | 459 | 2.4054 | 0.3556 |
| 0.466 | 18.0 | 486 | 1.6395 | 0.4 |
| 0.4427 | 19.0 | 513 | 1.8006 | 0.4 |
| 0.3436 | 20.0 | 540 | 1.6496 | 0.4 |
| 0.3213 | 21.0 | 567 | 2.4490 | 0.3778 |
| 0.3024 | 22.0 | 594 | 1.9317 | 0.4222 |
| 0.3831 | 23.0 | 621 | 2.1520 | 0.3556 |
| 0.2575 | 24.0 | 648 | 2.1832 | 0.4 |
| 0.3022 | 25.0 | 675 | 2.7719 | 0.4222 |
| 0.2889 | 26.0 | 702 | 2.2904 | 0.3778 |
| 0.1794 | 27.0 | 729 | 2.3541 | 0.3778 |
| 0.1256 | 28.0 | 756 | 3.3430 | 0.4444 |
| 0.1586 | 29.0 | 783 | 2.5549 | 0.4 |
| 0.1387 | 30.0 | 810 | 2.5659 | 0.3778 |
| 0.0828 | 31.0 | 837 | 3.1241 | 0.4 |
| 0.0717 | 32.0 | 864 | 3.9424 | 0.4222 |
| 0.0424 | 33.0 | 891 | 4.2735 | 0.4444 |
| 0.0451 | 34.0 | 918 | 4.8532 | 0.4222 |
| 0.0483 | 35.0 | 945 | 5.1314 | 0.4444 |
| 0.0395 | 36.0 | 972 | 4.8009 | 0.4 |
| 0.0349 | 37.0 | 999 | 4.7979 | 0.4222 |
| 0.003 | 38.0 | 1026 | 5.0139 | 0.4222 |
| 0.0007 | 39.0 | 1053 | 5.1670 | 0.4222 |
| 0.0004 | 40.0 | 1080 | 5.3634 | 0.4444 |
| 0.0028 | 41.0 | 1107 | 5.3467 | 0.4667 |
| 0.0002 | 42.0 | 1134 | 5.3534 | 0.4667 |
| 0.0001 | 43.0 | 1161 | 5.3434 | 0.4667 |
| 0.0001 | 44.0 | 1188 | 5.3497 | 0.4444 |
| 0.0001 | 45.0 | 1215 | 5.3599 | 0.4444 |
| 0.0001 | 46.0 | 1242 | 5.3641 | 0.4444 |
| 0.0001 | 47.0 | 1269 | 5.3694 | 0.4444 |
| 0.0001 | 48.0 | 1296 | 5.3716 | 0.4444 |
| 0.0001 | 49.0 | 1323 | 5.3716 | 0.4444 |
| 0.0001 | 50.0 | 1350 | 5.3716 | 0.4444 |
Framework versions
- Transformers 4.35.2
- Pytorch 2.1.0+cu118
- Datasets 2.15.0
- Tokenizers 0.15.0