hushem_5x_deit_tiny_sgd_00001_fold4
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: 1.6392
- Accuracy: 0.2857
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 |
|---|---|---|---|---|
| 1.5126 | 1.0 | 28 | 1.6960 | 0.2857 |
| 1.4875 | 2.0 | 56 | 1.6934 | 0.2857 |
| 1.484 | 3.0 | 84 | 1.6908 | 0.2857 |
| 1.5224 | 4.0 | 112 | 1.6883 | 0.2857 |
| 1.464 | 5.0 | 140 | 1.6861 | 0.2857 |
| 1.514 | 6.0 | 168 | 1.6836 | 0.2857 |
| 1.4795 | 7.0 | 196 | 1.6813 | 0.2857 |
| 1.4849 | 8.0 | 224 | 1.6790 | 0.2857 |
| 1.4832 | 9.0 | 252 | 1.6769 | 0.2857 |
| 1.5231 | 10.0 | 280 | 1.6747 | 0.2857 |
| 1.5146 | 11.0 | 308 | 1.6728 | 0.2857 |
| 1.4709 | 12.0 | 336 | 1.6708 | 0.2857 |
| 1.5002 | 13.0 | 364 | 1.6689 | 0.2857 |
| 1.4731 | 14.0 | 392 | 1.6671 | 0.2857 |
| 1.4733 | 15.0 | 420 | 1.6654 | 0.2857 |
| 1.4938 | 16.0 | 448 | 1.6637 | 0.2857 |
| 1.517 | 17.0 | 476 | 1.6621 | 0.2857 |
| 1.4904 | 18.0 | 504 | 1.6604 | 0.2857 |
| 1.4813 | 19.0 | 532 | 1.6589 | 0.2857 |
| 1.4788 | 20.0 | 560 | 1.6576 | 0.2857 |
| 1.476 | 21.0 | 588 | 1.6562 | 0.2857 |
| 1.5095 | 22.0 | 616 | 1.6550 | 0.2857 |
| 1.4801 | 23.0 | 644 | 1.6537 | 0.2857 |
| 1.4778 | 24.0 | 672 | 1.6525 | 0.2857 |
| 1.4526 | 25.0 | 700 | 1.6513 | 0.2857 |
| 1.4781 | 26.0 | 728 | 1.6502 | 0.2857 |
| 1.4923 | 27.0 | 756 | 1.6492 | 0.2857 |
| 1.4977 | 28.0 | 784 | 1.6482 | 0.2857 |
| 1.4558 | 29.0 | 812 | 1.6472 | 0.2857 |
| 1.4785 | 30.0 | 840 | 1.6464 | 0.2857 |
| 1.4962 | 31.0 | 868 | 1.6455 | 0.2857 |
| 1.4638 | 32.0 | 896 | 1.6447 | 0.2857 |
| 1.5095 | 33.0 | 924 | 1.6440 | 0.2857 |
| 1.4775 | 34.0 | 952 | 1.6433 | 0.2857 |
| 1.4595 | 35.0 | 980 | 1.6427 | 0.2857 |
| 1.4656 | 36.0 | 1008 | 1.6421 | 0.2857 |
| 1.4403 | 37.0 | 1036 | 1.6416 | 0.2857 |
| 1.4824 | 38.0 | 1064 | 1.6412 | 0.2857 |
| 1.5225 | 39.0 | 1092 | 1.6408 | 0.2857 |
| 1.4471 | 40.0 | 1120 | 1.6404 | 0.2857 |
| 1.4797 | 41.0 | 1148 | 1.6401 | 0.2857 |
| 1.4455 | 42.0 | 1176 | 1.6399 | 0.2857 |
| 1.4823 | 43.0 | 1204 | 1.6397 | 0.2857 |
| 1.4547 | 44.0 | 1232 | 1.6395 | 0.2857 |
| 1.4676 | 45.0 | 1260 | 1.6394 | 0.2857 |
| 1.4731 | 46.0 | 1288 | 1.6393 | 0.2857 |
| 1.4786 | 47.0 | 1316 | 1.6393 | 0.2857 |
| 1.4805 | 48.0 | 1344 | 1.6392 | 0.2857 |
| 1.5172 | 49.0 | 1372 | 1.6392 | 0.2857 |
| 1.4525 | 50.0 | 1400 | 1.6392 | 0.2857 |
Framework versions
- Transformers 4.35.2
- Pytorch 2.1.0+cu118
- Datasets 2.15.0
- Tokenizers 0.15.0