hushem_1x_deit_small_sgd_0001_fold1
This model is a fine-tuned version of facebook/deit-small-patch16-224 on the imagefolder dataset. It achieves the following results on the evaluation set:
- Loss: 1.4520
- Accuracy: 0.2667
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 | 1.5076 | 0.2889 |
| 1.5379 | 2.0 | 12 | 1.5042 | 0.2889 |
| 1.5379 | 3.0 | 18 | 1.5010 | 0.2889 |
| 1.5099 | 4.0 | 24 | 1.4983 | 0.2889 |
| 1.521 | 5.0 | 30 | 1.4955 | 0.2889 |
| 1.521 | 6.0 | 36 | 1.4929 | 0.2889 |
| 1.4972 | 7.0 | 42 | 1.4902 | 0.2889 |
| 1.4972 | 8.0 | 48 | 1.4879 | 0.2889 |
| 1.5152 | 9.0 | 54 | 1.4855 | 0.2889 |
| 1.4839 | 10.0 | 60 | 1.4831 | 0.2889 |
| 1.4839 | 11.0 | 66 | 1.4810 | 0.2889 |
| 1.478 | 12.0 | 72 | 1.4788 | 0.2889 |
| 1.478 | 13.0 | 78 | 1.4768 | 0.2889 |
| 1.4972 | 14.0 | 84 | 1.4749 | 0.2889 |
| 1.4822 | 15.0 | 90 | 1.4732 | 0.2889 |
| 1.4822 | 16.0 | 96 | 1.4714 | 0.2889 |
| 1.4784 | 17.0 | 102 | 1.4698 | 0.2889 |
| 1.4784 | 18.0 | 108 | 1.4684 | 0.2889 |
| 1.4862 | 19.0 | 114 | 1.4671 | 0.2889 |
| 1.4536 | 20.0 | 120 | 1.4657 | 0.2889 |
| 1.4536 | 21.0 | 126 | 1.4645 | 0.2889 |
| 1.4751 | 22.0 | 132 | 1.4634 | 0.2889 |
| 1.4751 | 23.0 | 138 | 1.4621 | 0.2889 |
| 1.4645 | 24.0 | 144 | 1.4610 | 0.2889 |
| 1.4518 | 25.0 | 150 | 1.4602 | 0.2889 |
| 1.4518 | 26.0 | 156 | 1.4592 | 0.2889 |
| 1.4648 | 27.0 | 162 | 1.4583 | 0.2889 |
| 1.4648 | 28.0 | 168 | 1.4575 | 0.2667 |
| 1.47 | 29.0 | 174 | 1.4568 | 0.2667 |
| 1.4692 | 30.0 | 180 | 1.4560 | 0.2667 |
| 1.4692 | 31.0 | 186 | 1.4553 | 0.2667 |
| 1.4701 | 32.0 | 192 | 1.4547 | 0.2667 |
| 1.4701 | 33.0 | 198 | 1.4541 | 0.2667 |
| 1.4745 | 34.0 | 204 | 1.4536 | 0.2667 |
| 1.4582 | 35.0 | 210 | 1.4532 | 0.2667 |
| 1.4582 | 36.0 | 216 | 1.4528 | 0.2667 |
| 1.4443 | 37.0 | 222 | 1.4526 | 0.2667 |
| 1.4443 | 38.0 | 228 | 1.4523 | 0.2667 |
| 1.44 | 39.0 | 234 | 1.4522 | 0.2667 |
| 1.4727 | 40.0 | 240 | 1.4521 | 0.2667 |
| 1.4727 | 41.0 | 246 | 1.4520 | 0.2667 |
| 1.4651 | 42.0 | 252 | 1.4520 | 0.2667 |
| 1.4651 | 43.0 | 258 | 1.4520 | 0.2667 |
| 1.4764 | 44.0 | 264 | 1.4520 | 0.2667 |
| 1.4313 | 45.0 | 270 | 1.4520 | 0.2667 |
| 1.4313 | 46.0 | 276 | 1.4520 | 0.2667 |
| 1.4565 | 47.0 | 282 | 1.4520 | 0.2667 |
| 1.4565 | 48.0 | 288 | 1.4520 | 0.2667 |
| 1.4277 | 49.0 | 294 | 1.4520 | 0.2667 |
| 1.4569 | 50.0 | 300 | 1.4520 | 0.2667 |
Framework versions
- Transformers 4.35.0
- Pytorch 2.1.0+cu118
- Datasets 2.14.6
- Tokenizers 0.14.1