hushem_5x_deit_tiny_rms_001_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: 4.2103
- Accuracy: 0.4667
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 |
|---|---|---|---|---|
| 1.9281 | 1.0 | 27 | 1.5117 | 0.2444 |
| 1.5142 | 2.0 | 54 | 1.4322 | 0.2444 |
| 1.4521 | 3.0 | 81 | 1.4347 | 0.2667 |
| 1.4849 | 4.0 | 108 | 1.6669 | 0.2444 |
| 1.4295 | 5.0 | 135 | 1.4219 | 0.2444 |
| 1.5039 | 6.0 | 162 | 1.4488 | 0.2444 |
| 1.4071 | 7.0 | 189 | 1.5265 | 0.2667 |
| 1.3353 | 8.0 | 216 | 1.4657 | 0.2 |
| 1.2509 | 9.0 | 243 | 1.4974 | 0.2667 |
| 1.2656 | 10.0 | 270 | 1.3879 | 0.2667 |
| 1.1653 | 11.0 | 297 | 1.5762 | 0.3333 |
| 1.1362 | 12.0 | 324 | 1.6646 | 0.4 |
| 1.2223 | 13.0 | 351 | 1.6454 | 0.1556 |
| 1.027 | 14.0 | 378 | 1.7430 | 0.2222 |
| 1.0093 | 15.0 | 405 | 1.8846 | 0.3778 |
| 0.9584 | 16.0 | 432 | 2.4104 | 0.1333 |
| 0.9432 | 17.0 | 459 | 2.0298 | 0.2667 |
| 0.9525 | 18.0 | 486 | 2.0687 | 0.3111 |
| 0.8065 | 19.0 | 513 | 2.3917 | 0.1778 |
| 0.8858 | 20.0 | 540 | 1.9861 | 0.2222 |
| 0.761 | 21.0 | 567 | 2.1617 | 0.3111 |
| 0.7456 | 22.0 | 594 | 2.6510 | 0.2667 |
| 0.7215 | 23.0 | 621 | 2.5639 | 0.4 |
| 0.7957 | 24.0 | 648 | 2.1510 | 0.3111 |
| 0.6858 | 25.0 | 675 | 2.2884 | 0.5111 |
| 0.6662 | 26.0 | 702 | 2.5211 | 0.4667 |
| 0.6301 | 27.0 | 729 | 2.5983 | 0.4667 |
| 0.676 | 28.0 | 756 | 2.4047 | 0.3778 |
| 0.5592 | 29.0 | 783 | 2.9746 | 0.4444 |
| 0.5758 | 30.0 | 810 | 2.5122 | 0.3778 |
| 0.5241 | 31.0 | 837 | 2.9556 | 0.4222 |
| 0.5249 | 32.0 | 864 | 2.6136 | 0.4889 |
| 0.527 | 33.0 | 891 | 2.5736 | 0.4222 |
| 0.5079 | 34.0 | 918 | 2.8504 | 0.5333 |
| 0.4469 | 35.0 | 945 | 3.2380 | 0.4 |
| 0.4391 | 36.0 | 972 | 3.1267 | 0.4444 |
| 0.3925 | 37.0 | 999 | 3.3000 | 0.4667 |
| 0.3719 | 38.0 | 1026 | 3.1308 | 0.5111 |
| 0.3137 | 39.0 | 1053 | 3.2238 | 0.4889 |
| 0.2551 | 40.0 | 1080 | 3.3279 | 0.5333 |
| 0.2638 | 41.0 | 1107 | 3.4382 | 0.5111 |
| 0.2004 | 42.0 | 1134 | 3.7709 | 0.4667 |
| 0.1877 | 43.0 | 1161 | 3.7286 | 0.4222 |
| 0.1439 | 44.0 | 1188 | 3.9363 | 0.4889 |
| 0.1405 | 45.0 | 1215 | 4.0904 | 0.4667 |
| 0.1084 | 46.0 | 1242 | 4.2188 | 0.4889 |
| 0.0629 | 47.0 | 1269 | 4.1702 | 0.5111 |
| 0.0462 | 48.0 | 1296 | 4.2158 | 0.4667 |
| 0.0313 | 49.0 | 1323 | 4.2103 | 0.4667 |
| 0.0292 | 50.0 | 1350 | 4.2103 | 0.4667 |
Framework versions
- Transformers 4.35.2
- Pytorch 2.1.0+cu118
- Datasets 2.15.0
- Tokenizers 0.15.0