hushem_5x_deit_small_adamax_0001_fold5
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: 0.5287
- Accuracy: 0.9024
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 |
|---|---|---|---|---|
| 0.662 | 1.0 | 28 | 0.4044 | 0.8537 |
| 0.1186 | 2.0 | 56 | 0.4284 | 0.8049 |
| 0.014 | 3.0 | 84 | 0.4888 | 0.9024 |
| 0.0339 | 4.0 | 112 | 0.4623 | 0.8780 |
| 0.0015 | 5.0 | 140 | 0.4160 | 0.8537 |
| 0.0005 | 6.0 | 168 | 0.4980 | 0.8537 |
| 0.0004 | 7.0 | 196 | 0.4944 | 0.9024 |
| 0.0002 | 8.0 | 224 | 0.4584 | 0.9024 |
| 0.0002 | 9.0 | 252 | 0.4587 | 0.9024 |
| 0.0002 | 10.0 | 280 | 0.4614 | 0.9024 |
| 0.0002 | 11.0 | 308 | 0.4658 | 0.9024 |
| 0.0001 | 12.0 | 336 | 0.4673 | 0.9024 |
| 0.0001 | 13.0 | 364 | 0.4741 | 0.9024 |
| 0.0001 | 14.0 | 392 | 0.4749 | 0.9024 |
| 0.0001 | 15.0 | 420 | 0.4791 | 0.9024 |
| 0.0001 | 16.0 | 448 | 0.4817 | 0.9024 |
| 0.0001 | 17.0 | 476 | 0.4846 | 0.9024 |
| 0.0001 | 18.0 | 504 | 0.4881 | 0.9024 |
| 0.0001 | 19.0 | 532 | 0.4907 | 0.9024 |
| 0.0001 | 20.0 | 560 | 0.4932 | 0.9024 |
| 0.0001 | 21.0 | 588 | 0.4952 | 0.9024 |
| 0.0001 | 22.0 | 616 | 0.4973 | 0.9024 |
| 0.0001 | 23.0 | 644 | 0.4995 | 0.9024 |
| 0.0001 | 24.0 | 672 | 0.5025 | 0.9024 |
| 0.0001 | 25.0 | 700 | 0.5047 | 0.9024 |
| 0.0001 | 26.0 | 728 | 0.5054 | 0.9024 |
| 0.0001 | 27.0 | 756 | 0.5078 | 0.9024 |
| 0.0001 | 28.0 | 784 | 0.5090 | 0.9024 |
| 0.0001 | 29.0 | 812 | 0.5119 | 0.9024 |
| 0.0001 | 30.0 | 840 | 0.5133 | 0.9024 |
| 0.0001 | 31.0 | 868 | 0.5148 | 0.9024 |
| 0.0001 | 32.0 | 896 | 0.5157 | 0.9024 |
| 0.0001 | 33.0 | 924 | 0.5187 | 0.9024 |
| 0.0001 | 34.0 | 952 | 0.5193 | 0.9024 |
| 0.0001 | 35.0 | 980 | 0.5205 | 0.9024 |
| 0.0001 | 36.0 | 1008 | 0.5218 | 0.9024 |
| 0.0 | 37.0 | 1036 | 0.5225 | 0.9024 |
| 0.0 | 38.0 | 1064 | 0.5237 | 0.9024 |
| 0.0 | 39.0 | 1092 | 0.5248 | 0.9024 |
| 0.0 | 40.0 | 1120 | 0.5253 | 0.9024 |
| 0.0 | 41.0 | 1148 | 0.5262 | 0.9024 |
| 0.0 | 42.0 | 1176 | 0.5266 | 0.9024 |
| 0.0 | 43.0 | 1204 | 0.5275 | 0.9024 |
| 0.0 | 44.0 | 1232 | 0.5280 | 0.9024 |
| 0.0 | 45.0 | 1260 | 0.5281 | 0.9024 |
| 0.0 | 46.0 | 1288 | 0.5285 | 0.9024 |
| 0.0 | 47.0 | 1316 | 0.5286 | 0.9024 |
| 0.0 | 48.0 | 1344 | 0.5287 | 0.9024 |
| 0.0 | 49.0 | 1372 | 0.5287 | 0.9024 |
| 0.0 | 50.0 | 1400 | 0.5287 | 0.9024 |
Framework versions
- Transformers 4.35.2
- Pytorch 2.1.0+cu118
- Datasets 2.15.0
- Tokenizers 0.15.0