hushem_5x_deit_small_sgd_00001_fold3
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: 1.5763
- Accuracy: 0.2093
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 |
|---|---|---|---|---|
| 1.5114 | 1.0 | 28 | 1.6002 | 0.1860 |
| 1.516 | 2.0 | 56 | 1.5991 | 0.2093 |
| 1.448 | 3.0 | 84 | 1.5981 | 0.2093 |
| 1.5009 | 4.0 | 112 | 1.5971 | 0.2093 |
| 1.646 | 5.0 | 140 | 1.5960 | 0.2093 |
| 1.55 | 6.0 | 168 | 1.5950 | 0.2093 |
| 1.5383 | 7.0 | 196 | 1.5941 | 0.2093 |
| 1.4892 | 8.0 | 224 | 1.5932 | 0.2093 |
| 1.5021 | 9.0 | 252 | 1.5922 | 0.2093 |
| 1.5057 | 10.0 | 280 | 1.5913 | 0.2093 |
| 1.5326 | 11.0 | 308 | 1.5905 | 0.2093 |
| 1.5462 | 12.0 | 336 | 1.5897 | 0.2093 |
| 1.5032 | 13.0 | 364 | 1.5889 | 0.2093 |
| 1.4978 | 14.0 | 392 | 1.5881 | 0.2093 |
| 1.4757 | 15.0 | 420 | 1.5874 | 0.2093 |
| 1.4813 | 16.0 | 448 | 1.5867 | 0.2093 |
| 1.4635 | 17.0 | 476 | 1.5860 | 0.2093 |
| 1.511 | 18.0 | 504 | 1.5853 | 0.2093 |
| 1.4905 | 19.0 | 532 | 1.5847 | 0.2093 |
| 1.4302 | 20.0 | 560 | 1.5841 | 0.2093 |
| 1.5108 | 21.0 | 588 | 1.5835 | 0.2093 |
| 1.504 | 22.0 | 616 | 1.5829 | 0.2093 |
| 1.4924 | 23.0 | 644 | 1.5824 | 0.2093 |
| 1.5112 | 24.0 | 672 | 1.5819 | 0.2093 |
| 1.5256 | 25.0 | 700 | 1.5814 | 0.2093 |
| 1.4493 | 26.0 | 728 | 1.5809 | 0.2093 |
| 1.4989 | 27.0 | 756 | 1.5805 | 0.2093 |
| 1.5477 | 28.0 | 784 | 1.5801 | 0.2093 |
| 1.5116 | 29.0 | 812 | 1.5797 | 0.2093 |
| 1.4275 | 30.0 | 840 | 1.5793 | 0.2093 |
| 1.453 | 31.0 | 868 | 1.5790 | 0.2093 |
| 1.524 | 32.0 | 896 | 1.5786 | 0.2093 |
| 1.5026 | 33.0 | 924 | 1.5783 | 0.2093 |
| 1.4458 | 34.0 | 952 | 1.5780 | 0.2093 |
| 1.4708 | 35.0 | 980 | 1.5778 | 0.2093 |
| 1.5117 | 36.0 | 1008 | 1.5776 | 0.2093 |
| 1.5594 | 37.0 | 1036 | 1.5773 | 0.2093 |
| 1.5028 | 38.0 | 1064 | 1.5771 | 0.2093 |
| 1.4691 | 39.0 | 1092 | 1.5770 | 0.2093 |
| 1.5214 | 40.0 | 1120 | 1.5768 | 0.2093 |
| 1.5285 | 41.0 | 1148 | 1.5767 | 0.2093 |
| 1.4667 | 42.0 | 1176 | 1.5766 | 0.2093 |
| 1.4652 | 43.0 | 1204 | 1.5765 | 0.2093 |
| 1.4952 | 44.0 | 1232 | 1.5765 | 0.2093 |
| 1.4825 | 45.0 | 1260 | 1.5764 | 0.2093 |
| 1.4816 | 46.0 | 1288 | 1.5764 | 0.2093 |
| 1.4911 | 47.0 | 1316 | 1.5764 | 0.2093 |
| 1.5027 | 48.0 | 1344 | 1.5763 | 0.2093 |
| 1.4424 | 49.0 | 1372 | 1.5763 | 0.2093 |
| 1.4881 | 50.0 | 1400 | 1.5763 | 0.2093 |
Framework versions
- Transformers 4.35.2
- Pytorch 2.1.0+cu118
- Datasets 2.15.0
- Tokenizers 0.15.0