hushem_1x_deit_base_rms_0001_fold3
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: 2.5728
- Accuracy: 0.5116
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 |
|---|---|---|---|---|
| No log | 1.0 | 6 | 1.5927 | 0.2558 |
| 1.8022 | 2.0 | 12 | 1.4288 | 0.2558 |
| 1.8022 | 3.0 | 18 | 1.5308 | 0.2558 |
| 1.5239 | 4.0 | 24 | 1.4740 | 0.2558 |
| 1.4801 | 5.0 | 30 | 2.2308 | 0.2558 |
| 1.4801 | 6.0 | 36 | 1.4153 | 0.2558 |
| 1.5953 | 7.0 | 42 | 1.4884 | 0.2326 |
| 1.5953 | 8.0 | 48 | 1.5775 | 0.2558 |
| 1.4262 | 9.0 | 54 | 1.4549 | 0.2558 |
| 1.4607 | 10.0 | 60 | 1.4401 | 0.2558 |
| 1.4607 | 11.0 | 66 | 1.3707 | 0.3256 |
| 1.3771 | 12.0 | 72 | 1.3285 | 0.3721 |
| 1.3771 | 13.0 | 78 | 1.2980 | 0.3721 |
| 1.2887 | 14.0 | 84 | 1.6464 | 0.2558 |
| 1.194 | 15.0 | 90 | 1.4572 | 0.2791 |
| 1.194 | 16.0 | 96 | 1.5171 | 0.3256 |
| 1.2311 | 17.0 | 102 | 1.4178 | 0.4186 |
| 1.2311 | 18.0 | 108 | 1.3427 | 0.3256 |
| 1.2234 | 19.0 | 114 | 1.3369 | 0.3721 |
| 1.1384 | 20.0 | 120 | 1.4187 | 0.3721 |
| 1.1384 | 21.0 | 126 | 1.5195 | 0.2558 |
| 1.0551 | 22.0 | 132 | 1.5010 | 0.3488 |
| 1.0551 | 23.0 | 138 | 1.4046 | 0.3721 |
| 0.958 | 24.0 | 144 | 1.2991 | 0.3721 |
| 0.9405 | 25.0 | 150 | 1.3687 | 0.4651 |
| 0.9405 | 26.0 | 156 | 1.8052 | 0.3256 |
| 0.7255 | 27.0 | 162 | 1.6391 | 0.4419 |
| 0.7255 | 28.0 | 168 | 1.3419 | 0.5116 |
| 0.6435 | 29.0 | 174 | 1.7678 | 0.4419 |
| 0.4939 | 30.0 | 180 | 2.2022 | 0.3256 |
| 0.4939 | 31.0 | 186 | 1.6603 | 0.5116 |
| 0.4434 | 32.0 | 192 | 1.5952 | 0.4884 |
| 0.4434 | 33.0 | 198 | 1.7642 | 0.5116 |
| 0.2864 | 34.0 | 204 | 2.0069 | 0.4651 |
| 0.1858 | 35.0 | 210 | 2.5370 | 0.4884 |
| 0.1858 | 36.0 | 216 | 2.2076 | 0.4651 |
| 0.1248 | 37.0 | 222 | 2.7233 | 0.5581 |
| 0.1248 | 38.0 | 228 | 2.6548 | 0.5349 |
| 0.0495 | 39.0 | 234 | 2.4864 | 0.5116 |
| 0.0526 | 40.0 | 240 | 2.5601 | 0.5116 |
| 0.0526 | 41.0 | 246 | 2.5731 | 0.5116 |
| 0.0219 | 42.0 | 252 | 2.5728 | 0.5116 |
| 0.0219 | 43.0 | 258 | 2.5728 | 0.5116 |
| 0.0327 | 44.0 | 264 | 2.5728 | 0.5116 |
| 0.0324 | 45.0 | 270 | 2.5728 | 0.5116 |
| 0.0324 | 46.0 | 276 | 2.5728 | 0.5116 |
| 0.0326 | 47.0 | 282 | 2.5728 | 0.5116 |
| 0.0326 | 48.0 | 288 | 2.5728 | 0.5116 |
| 0.0208 | 49.0 | 294 | 2.5728 | 0.5116 |
| 0.0503 | 50.0 | 300 | 2.5728 | 0.5116 |
Framework versions
- Transformers 4.35.2
- Pytorch 2.1.0+cu118
- Datasets 2.15.0
- Tokenizers 0.15.0