hushem_1x_deit_tiny_sgd_lr00001_fold2
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: 1.6351
- Accuracy: 0.1333
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: 1e-05
- 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 |
|---|---|---|---|---|
| No log | 1.0 | 6 | 1.6407 | 0.1333 |
| 1.6149 | 2.0 | 12 | 1.6404 | 0.1333 |
| 1.6149 | 3.0 | 18 | 1.6401 | 0.1333 |
| 1.588 | 4.0 | 24 | 1.6398 | 0.1333 |
| 1.6243 | 5.0 | 30 | 1.6396 | 0.1333 |
| 1.6243 | 6.0 | 36 | 1.6393 | 0.1333 |
| 1.6041 | 7.0 | 42 | 1.6390 | 0.1333 |
| 1.6041 | 8.0 | 48 | 1.6388 | 0.1333 |
| 1.5784 | 9.0 | 54 | 1.6386 | 0.1333 |
| 1.61 | 10.0 | 60 | 1.6383 | 0.1333 |
| 1.61 | 11.0 | 66 | 1.6381 | 0.1333 |
| 1.5857 | 12.0 | 72 | 1.6379 | 0.1333 |
| 1.5857 | 13.0 | 78 | 1.6377 | 0.1333 |
| 1.6282 | 14.0 | 84 | 1.6375 | 0.1333 |
| 1.5739 | 15.0 | 90 | 1.6373 | 0.1333 |
| 1.5739 | 16.0 | 96 | 1.6372 | 0.1333 |
| 1.5784 | 17.0 | 102 | 1.6370 | 0.1333 |
| 1.5784 | 18.0 | 108 | 1.6368 | 0.1333 |
| 1.6525 | 19.0 | 114 | 1.6367 | 0.1333 |
| 1.5978 | 20.0 | 120 | 1.6365 | 0.1333 |
| 1.5978 | 21.0 | 126 | 1.6364 | 0.1333 |
| 1.6239 | 22.0 | 132 | 1.6362 | 0.1333 |
| 1.6239 | 23.0 | 138 | 1.6361 | 0.1333 |
| 1.581 | 24.0 | 144 | 1.6360 | 0.1333 |
| 1.597 | 25.0 | 150 | 1.6359 | 0.1333 |
| 1.597 | 26.0 | 156 | 1.6358 | 0.1333 |
| 1.5864 | 27.0 | 162 | 1.6357 | 0.1333 |
| 1.5864 | 28.0 | 168 | 1.6356 | 0.1333 |
| 1.6236 | 29.0 | 174 | 1.6355 | 0.1333 |
| 1.6201 | 30.0 | 180 | 1.6354 | 0.1333 |
| 1.6201 | 31.0 | 186 | 1.6354 | 0.1333 |
| 1.6018 | 32.0 | 192 | 1.6353 | 0.1333 |
| 1.6018 | 33.0 | 198 | 1.6352 | 0.1333 |
| 1.5711 | 34.0 | 204 | 1.6352 | 0.1333 |
| 1.6003 | 35.0 | 210 | 1.6352 | 0.1333 |
| 1.6003 | 36.0 | 216 | 1.6351 | 0.1333 |
| 1.5762 | 37.0 | 222 | 1.6351 | 0.1333 |
| 1.5762 | 38.0 | 228 | 1.6351 | 0.1333 |
| 1.5979 | 39.0 | 234 | 1.6351 | 0.1333 |
| 1.6035 | 40.0 | 240 | 1.6351 | 0.1333 |
| 1.6035 | 41.0 | 246 | 1.6351 | 0.1333 |
| 1.5976 | 42.0 | 252 | 1.6351 | 0.1333 |
| 1.5976 | 43.0 | 258 | 1.6351 | 0.1333 |
| 1.5981 | 44.0 | 264 | 1.6351 | 0.1333 |
| 1.5912 | 45.0 | 270 | 1.6351 | 0.1333 |
| 1.5912 | 46.0 | 276 | 1.6351 | 0.1333 |
| 1.5981 | 47.0 | 282 | 1.6351 | 0.1333 |
| 1.5981 | 48.0 | 288 | 1.6351 | 0.1333 |
| 1.6158 | 49.0 | 294 | 1.6351 | 0.1333 |
| 1.593 | 50.0 | 300 | 1.6351 | 0.1333 |
Framework versions
- Transformers 4.35.0
- Pytorch 2.1.0+cu118
- Datasets 2.14.6
- Tokenizers 0.14.1