hushem_1x_deit_small_adamax_0001_fold1
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: 1.3048
- Accuracy: 0.5778
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 | 1.3083 | 0.3111 |
| 1.2395 | 2.0 | 12 | 1.1116 | 0.5556 |
| 1.2395 | 3.0 | 18 | 0.9370 | 0.6444 |
| 0.5222 | 4.0 | 24 | 0.8939 | 0.6222 |
| 0.1257 | 5.0 | 30 | 1.0613 | 0.6222 |
| 0.1257 | 6.0 | 36 | 1.1544 | 0.6667 |
| 0.0192 | 7.0 | 42 | 1.0970 | 0.6222 |
| 0.0192 | 8.0 | 48 | 1.3834 | 0.5778 |
| 0.0034 | 9.0 | 54 | 1.4273 | 0.6222 |
| 0.0011 | 10.0 | 60 | 1.2955 | 0.6222 |
| 0.0011 | 11.0 | 66 | 1.1578 | 0.6222 |
| 0.0006 | 12.0 | 72 | 1.1209 | 0.6 |
| 0.0006 | 13.0 | 78 | 1.1439 | 0.6 |
| 0.0005 | 14.0 | 84 | 1.1840 | 0.6 |
| 0.0004 | 15.0 | 90 | 1.2222 | 0.5778 |
| 0.0004 | 16.0 | 96 | 1.2485 | 0.5778 |
| 0.0003 | 17.0 | 102 | 1.2638 | 0.5778 |
| 0.0003 | 18.0 | 108 | 1.2689 | 0.5778 |
| 0.0003 | 19.0 | 114 | 1.2732 | 0.5778 |
| 0.0003 | 20.0 | 120 | 1.2771 | 0.5778 |
| 0.0003 | 21.0 | 126 | 1.2803 | 0.5778 |
| 0.0003 | 22.0 | 132 | 1.2805 | 0.5778 |
| 0.0003 | 23.0 | 138 | 1.2805 | 0.5778 |
| 0.0002 | 24.0 | 144 | 1.2807 | 0.5778 |
| 0.0002 | 25.0 | 150 | 1.2825 | 0.5778 |
| 0.0002 | 26.0 | 156 | 1.2850 | 0.5778 |
| 0.0002 | 27.0 | 162 | 1.2856 | 0.5778 |
| 0.0002 | 28.0 | 168 | 1.2878 | 0.5778 |
| 0.0002 | 29.0 | 174 | 1.2904 | 0.5778 |
| 0.0002 | 30.0 | 180 | 1.2922 | 0.5778 |
| 0.0002 | 31.0 | 186 | 1.2931 | 0.5778 |
| 0.0002 | 32.0 | 192 | 1.2945 | 0.5778 |
| 0.0002 | 33.0 | 198 | 1.2963 | 0.5778 |
| 0.0002 | 34.0 | 204 | 1.2983 | 0.5778 |
| 0.0002 | 35.0 | 210 | 1.2995 | 0.5778 |
| 0.0002 | 36.0 | 216 | 1.3007 | 0.5778 |
| 0.0002 | 37.0 | 222 | 1.3018 | 0.5778 |
| 0.0002 | 38.0 | 228 | 1.3034 | 0.5778 |
| 0.0002 | 39.0 | 234 | 1.3042 | 0.5778 |
| 0.0002 | 40.0 | 240 | 1.3046 | 0.5778 |
| 0.0002 | 41.0 | 246 | 1.3047 | 0.5778 |
| 0.0002 | 42.0 | 252 | 1.3048 | 0.5778 |
| 0.0002 | 43.0 | 258 | 1.3048 | 0.5778 |
| 0.0002 | 44.0 | 264 | 1.3048 | 0.5778 |
| 0.0002 | 45.0 | 270 | 1.3048 | 0.5778 |
| 0.0002 | 46.0 | 276 | 1.3048 | 0.5778 |
| 0.0002 | 47.0 | 282 | 1.3048 | 0.5778 |
| 0.0002 | 48.0 | 288 | 1.3048 | 0.5778 |
| 0.0002 | 49.0 | 294 | 1.3048 | 0.5778 |
| 0.0002 | 50.0 | 300 | 1.3048 | 0.5778 |
Framework versions
- Transformers 4.35.0
- Pytorch 2.1.0+cu118
- Datasets 2.14.6
- Tokenizers 0.14.1