hushem_1x_deit_tiny_sgd_lr0001_fold1
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: 1.5010
- Accuracy: 0.2444
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 | 1.5317 | 0.2667 |
| 1.5768 | 2.0 | 12 | 1.5295 | 0.2444 |
| 1.5768 | 3.0 | 18 | 1.5277 | 0.2444 |
| 1.5217 | 4.0 | 24 | 1.5259 | 0.2444 |
| 1.5806 | 5.0 | 30 | 1.5243 | 0.2444 |
| 1.5806 | 6.0 | 36 | 1.5226 | 0.2444 |
| 1.5608 | 7.0 | 42 | 1.5211 | 0.2444 |
| 1.5608 | 8.0 | 48 | 1.5198 | 0.2444 |
| 1.5538 | 9.0 | 54 | 1.5184 | 0.2444 |
| 1.5354 | 10.0 | 60 | 1.5172 | 0.2444 |
| 1.5354 | 11.0 | 66 | 1.5159 | 0.2444 |
| 1.5529 | 12.0 | 72 | 1.5148 | 0.2444 |
| 1.5529 | 13.0 | 78 | 1.5137 | 0.2444 |
| 1.5094 | 14.0 | 84 | 1.5127 | 0.2444 |
| 1.5228 | 15.0 | 90 | 1.5118 | 0.2444 |
| 1.5228 | 16.0 | 96 | 1.5108 | 0.2444 |
| 1.5295 | 17.0 | 102 | 1.5100 | 0.2444 |
| 1.5295 | 18.0 | 108 | 1.5092 | 0.2444 |
| 1.5298 | 19.0 | 114 | 1.5084 | 0.2444 |
| 1.5372 | 20.0 | 120 | 1.5077 | 0.2444 |
| 1.5372 | 21.0 | 126 | 1.5071 | 0.2444 |
| 1.5336 | 22.0 | 132 | 1.5065 | 0.2444 |
| 1.5336 | 23.0 | 138 | 1.5059 | 0.2444 |
| 1.5077 | 24.0 | 144 | 1.5053 | 0.2444 |
| 1.5022 | 25.0 | 150 | 1.5049 | 0.2444 |
| 1.5022 | 26.0 | 156 | 1.5044 | 0.2444 |
| 1.5158 | 27.0 | 162 | 1.5040 | 0.2444 |
| 1.5158 | 28.0 | 168 | 1.5036 | 0.2444 |
| 1.4961 | 29.0 | 174 | 1.5032 | 0.2444 |
| 1.5155 | 30.0 | 180 | 1.5029 | 0.2444 |
| 1.5155 | 31.0 | 186 | 1.5025 | 0.2444 |
| 1.5093 | 32.0 | 192 | 1.5022 | 0.2444 |
| 1.5093 | 33.0 | 198 | 1.5020 | 0.2444 |
| 1.4596 | 34.0 | 204 | 1.5017 | 0.2444 |
| 1.4894 | 35.0 | 210 | 1.5015 | 0.2444 |
| 1.4894 | 36.0 | 216 | 1.5014 | 0.2444 |
| 1.5058 | 37.0 | 222 | 1.5012 | 0.2444 |
| 1.5058 | 38.0 | 228 | 1.5011 | 0.2444 |
| 1.4675 | 39.0 | 234 | 1.5010 | 0.2444 |
| 1.4822 | 40.0 | 240 | 1.5010 | 0.2444 |
| 1.4822 | 41.0 | 246 | 1.5010 | 0.2444 |
| 1.5008 | 42.0 | 252 | 1.5010 | 0.2444 |
| 1.5008 | 43.0 | 258 | 1.5010 | 0.2444 |
| 1.5075 | 44.0 | 264 | 1.5010 | 0.2444 |
| 1.5338 | 45.0 | 270 | 1.5010 | 0.2444 |
| 1.5338 | 46.0 | 276 | 1.5010 | 0.2444 |
| 1.5016 | 47.0 | 282 | 1.5010 | 0.2444 |
| 1.5016 | 48.0 | 288 | 1.5010 | 0.2444 |
| 1.4777 | 49.0 | 294 | 1.5010 | 0.2444 |
| 1.4813 | 50.0 | 300 | 1.5010 | 0.2444 |
Framework versions
- Transformers 4.35.0
- Pytorch 2.1.0+cu118
- Datasets 2.14.6
- Tokenizers 0.14.1