hushem_1x_deit_tiny_sgd_lr0001_fold2
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: 1.5853
- Accuracy: 0.1333
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 | 1.6380 | 0.1333 |
| 1.612 | 2.0 | 12 | 1.6350 | 0.1333 |
| 1.612 | 3.0 | 18 | 1.6320 | 0.1333 |
| 1.5793 | 4.0 | 24 | 1.6293 | 0.1333 |
| 1.6085 | 5.0 | 30 | 1.6268 | 0.1333 |
| 1.6085 | 6.0 | 36 | 1.6242 | 0.1333 |
| 1.5833 | 7.0 | 42 | 1.6217 | 0.1333 |
| 1.5833 | 8.0 | 48 | 1.6197 | 0.1333 |
| 1.5532 | 9.0 | 54 | 1.6175 | 0.1333 |
| 1.5785 | 10.0 | 60 | 1.6153 | 0.1333 |
| 1.5785 | 11.0 | 66 | 1.6132 | 0.1333 |
| 1.5506 | 12.0 | 72 | 1.6113 | 0.1333 |
| 1.5506 | 13.0 | 78 | 1.6095 | 0.1333 |
| 1.5868 | 14.0 | 84 | 1.6078 | 0.1333 |
| 1.532 | 15.0 | 90 | 1.6062 | 0.1333 |
| 1.532 | 16.0 | 96 | 1.6045 | 0.1333 |
| 1.5321 | 17.0 | 102 | 1.6029 | 0.1333 |
| 1.5321 | 18.0 | 108 | 1.6015 | 0.1333 |
| 1.5965 | 19.0 | 114 | 1.6001 | 0.1333 |
| 1.5428 | 20.0 | 120 | 1.5987 | 0.1333 |
| 1.5428 | 21.0 | 126 | 1.5975 | 0.1333 |
| 1.5622 | 22.0 | 132 | 1.5964 | 0.1333 |
| 1.5622 | 23.0 | 138 | 1.5951 | 0.1333 |
| 1.5259 | 24.0 | 144 | 1.5941 | 0.1333 |
| 1.5339 | 25.0 | 150 | 1.5932 | 0.1333 |
| 1.5339 | 26.0 | 156 | 1.5923 | 0.1333 |
| 1.5237 | 27.0 | 162 | 1.5914 | 0.1333 |
| 1.5237 | 28.0 | 168 | 1.5907 | 0.1333 |
| 1.5539 | 29.0 | 174 | 1.5899 | 0.1333 |
| 1.5487 | 30.0 | 180 | 1.5891 | 0.1333 |
| 1.5487 | 31.0 | 186 | 1.5885 | 0.1333 |
| 1.5317 | 32.0 | 192 | 1.5879 | 0.1333 |
| 1.5317 | 33.0 | 198 | 1.5874 | 0.1333 |
| 1.4989 | 34.0 | 204 | 1.5869 | 0.1333 |
| 1.5301 | 35.0 | 210 | 1.5865 | 0.1333 |
| 1.5301 | 36.0 | 216 | 1.5862 | 0.1333 |
| 1.5061 | 37.0 | 222 | 1.5859 | 0.1333 |
| 1.5061 | 38.0 | 228 | 1.5857 | 0.1333 |
| 1.5205 | 39.0 | 234 | 1.5855 | 0.1333 |
| 1.5267 | 40.0 | 240 | 1.5854 | 0.1333 |
| 1.5267 | 41.0 | 246 | 1.5854 | 0.1333 |
| 1.5211 | 42.0 | 252 | 1.5853 | 0.1333 |
| 1.5211 | 43.0 | 258 | 1.5853 | 0.1333 |
| 1.524 | 44.0 | 264 | 1.5853 | 0.1333 |
| 1.5163 | 45.0 | 270 | 1.5853 | 0.1333 |
| 1.5163 | 46.0 | 276 | 1.5853 | 0.1333 |
| 1.5253 | 47.0 | 282 | 1.5853 | 0.1333 |
| 1.5253 | 48.0 | 288 | 1.5853 | 0.1333 |
| 1.5384 | 49.0 | 294 | 1.5853 | 0.1333 |
| 1.5175 | 50.0 | 300 | 1.5853 | 0.1333 |
Framework versions
- Transformers 4.35.0
- Pytorch 2.1.0+cu118
- Datasets 2.14.6
- Tokenizers 0.14.1