hushem_5x_deit_tiny_adamax_0001_fold1
This model is a fine-tuned version of facebook/deit-tiny-patch16-224 on the imagefolder dataset. It achieves the following results on the evaluation set:
- Loss: 1.7993
- Accuracy: 0.6667
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 |
|---|---|---|---|---|
| 1.0199 | 1.0 | 27 | 1.3645 | 0.4 |
| 0.2821 | 2.0 | 54 | 1.2055 | 0.5778 |
| 0.0646 | 3.0 | 81 | 1.3812 | 0.6 |
| 0.0073 | 4.0 | 108 | 1.4600 | 0.6889 |
| 0.0019 | 5.0 | 135 | 1.3896 | 0.6444 |
| 0.0006 | 6.0 | 162 | 1.4491 | 0.6667 |
| 0.0004 | 7.0 | 189 | 1.4950 | 0.6444 |
| 0.0002 | 8.0 | 216 | 1.5645 | 0.6444 |
| 0.0002 | 9.0 | 243 | 1.5792 | 0.6667 |
| 0.0001 | 10.0 | 270 | 1.5954 | 0.6667 |
| 0.0001 | 11.0 | 297 | 1.6100 | 0.6667 |
| 0.0001 | 12.0 | 324 | 1.6341 | 0.6667 |
| 0.0001 | 13.0 | 351 | 1.6463 | 0.6667 |
| 0.0001 | 14.0 | 378 | 1.6524 | 0.6667 |
| 0.0001 | 15.0 | 405 | 1.6584 | 0.6667 |
| 0.0001 | 16.0 | 432 | 1.6769 | 0.6667 |
| 0.0001 | 17.0 | 459 | 1.6829 | 0.6667 |
| 0.0001 | 18.0 | 486 | 1.6898 | 0.6667 |
| 0.0001 | 19.0 | 513 | 1.7029 | 0.6667 |
| 0.0001 | 20.0 | 540 | 1.7109 | 0.6667 |
| 0.0001 | 21.0 | 567 | 1.7134 | 0.6667 |
| 0.0001 | 22.0 | 594 | 1.7169 | 0.6667 |
| 0.0001 | 23.0 | 621 | 1.7239 | 0.6667 |
| 0.0001 | 24.0 | 648 | 1.7396 | 0.6667 |
| 0.0001 | 25.0 | 675 | 1.7377 | 0.6667 |
| 0.0001 | 26.0 | 702 | 1.7493 | 0.6667 |
| 0.0001 | 27.0 | 729 | 1.7500 | 0.6667 |
| 0.0001 | 28.0 | 756 | 1.7483 | 0.6667 |
| 0.0001 | 29.0 | 783 | 1.7547 | 0.6667 |
| 0.0001 | 30.0 | 810 | 1.7688 | 0.6667 |
| 0.0 | 31.0 | 837 | 1.7675 | 0.6667 |
| 0.0001 | 32.0 | 864 | 1.7692 | 0.6667 |
| 0.0 | 33.0 | 891 | 1.7739 | 0.6667 |
| 0.0 | 34.0 | 918 | 1.7755 | 0.6667 |
| 0.0 | 35.0 | 945 | 1.7806 | 0.6667 |
| 0.0 | 36.0 | 972 | 1.7812 | 0.6667 |
| 0.0 | 37.0 | 999 | 1.7859 | 0.6667 |
| 0.0 | 38.0 | 1026 | 1.7853 | 0.6667 |
| 0.0 | 39.0 | 1053 | 1.7884 | 0.6667 |
| 0.0 | 40.0 | 1080 | 1.7925 | 0.6667 |
| 0.0 | 41.0 | 1107 | 1.7936 | 0.6667 |
| 0.0 | 42.0 | 1134 | 1.7942 | 0.6667 |
| 0.0 | 43.0 | 1161 | 1.7955 | 0.6667 |
| 0.0 | 44.0 | 1188 | 1.7973 | 0.6667 |
| 0.0 | 45.0 | 1215 | 1.7990 | 0.6667 |
| 0.0 | 46.0 | 1242 | 1.7986 | 0.6667 |
| 0.0 | 47.0 | 1269 | 1.7991 | 0.6667 |
| 0.0 | 48.0 | 1296 | 1.7993 | 0.6667 |
| 0.0 | 49.0 | 1323 | 1.7993 | 0.6667 |
| 0.0 | 50.0 | 1350 | 1.7993 | 0.6667 |
Framework versions
- Transformers 4.35.2
- Pytorch 2.1.0+cu118
- Datasets 2.15.0
- Tokenizers 0.15.0