smids_1x_deit_tiny_rms_00001_fold3
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: 0.7866
- Accuracy: 0.8917
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 |
|---|---|---|---|---|
| 0.4569 | 1.0 | 75 | 0.3524 | 0.8733 |
| 0.257 | 2.0 | 150 | 0.3177 | 0.8783 |
| 0.228 | 3.0 | 225 | 0.2830 | 0.895 |
| 0.1874 | 4.0 | 300 | 0.2625 | 0.9133 |
| 0.0988 | 5.0 | 375 | 0.3112 | 0.8867 |
| 0.0547 | 6.0 | 450 | 0.3480 | 0.895 |
| 0.0671 | 7.0 | 525 | 0.4401 | 0.8783 |
| 0.0314 | 8.0 | 600 | 0.4835 | 0.8917 |
| 0.0373 | 9.0 | 675 | 0.4879 | 0.8983 |
| 0.007 | 10.0 | 750 | 0.5903 | 0.895 |
| 0.0283 | 11.0 | 825 | 0.5783 | 0.8867 |
| 0.0151 | 12.0 | 900 | 0.7372 | 0.8833 |
| 0.0012 | 13.0 | 975 | 0.6965 | 0.8783 |
| 0.0175 | 14.0 | 1050 | 0.6546 | 0.89 |
| 0.0013 | 15.0 | 1125 | 0.7058 | 0.8783 |
| 0.0001 | 16.0 | 1200 | 0.6811 | 0.8917 |
| 0.0007 | 17.0 | 1275 | 0.7469 | 0.8967 |
| 0.0153 | 18.0 | 1350 | 0.6408 | 0.8917 |
| 0.0082 | 19.0 | 1425 | 0.8396 | 0.8783 |
| 0.0026 | 20.0 | 1500 | 0.8283 | 0.8883 |
| 0.0001 | 21.0 | 1575 | 0.7596 | 0.89 |
| 0.0037 | 22.0 | 1650 | 0.8137 | 0.875 |
| 0.007 | 23.0 | 1725 | 0.7276 | 0.8833 |
| 0.0 | 24.0 | 1800 | 0.6779 | 0.9 |
| 0.0001 | 25.0 | 1875 | 0.7204 | 0.895 |
| 0.0081 | 26.0 | 1950 | 0.7595 | 0.8883 |
| 0.0 | 27.0 | 2025 | 0.7620 | 0.895 |
| 0.0 | 28.0 | 2100 | 0.7575 | 0.8867 |
| 0.0001 | 29.0 | 2175 | 0.7827 | 0.89 |
| 0.0 | 30.0 | 2250 | 0.7351 | 0.8917 |
| 0.0 | 31.0 | 2325 | 0.7715 | 0.89 |
| 0.0 | 32.0 | 2400 | 0.7652 | 0.8917 |
| 0.0 | 33.0 | 2475 | 0.7881 | 0.89 |
| 0.0066 | 34.0 | 2550 | 0.7810 | 0.89 |
| 0.0102 | 35.0 | 2625 | 0.8490 | 0.89 |
| 0.0026 | 36.0 | 2700 | 0.7973 | 0.885 |
| 0.0016 | 37.0 | 2775 | 0.7751 | 0.8983 |
| 0.0 | 38.0 | 2850 | 0.7861 | 0.8933 |
| 0.0 | 39.0 | 2925 | 0.7652 | 0.8917 |
| 0.0 | 40.0 | 3000 | 0.7874 | 0.8917 |
| 0.0 | 41.0 | 3075 | 0.7876 | 0.8883 |
| 0.0033 | 42.0 | 3150 | 0.7858 | 0.8917 |
| 0.0029 | 43.0 | 3225 | 0.7835 | 0.8917 |
| 0.0 | 44.0 | 3300 | 0.7876 | 0.8917 |
| 0.0023 | 45.0 | 3375 | 0.7887 | 0.8917 |
| 0.0023 | 46.0 | 3450 | 0.7887 | 0.8917 |
| 0.0053 | 47.0 | 3525 | 0.7882 | 0.8917 |
| 0.0 | 48.0 | 3600 | 0.7869 | 0.8917 |
| 0.0 | 49.0 | 3675 | 0.7873 | 0.8917 |
| 0.0047 | 50.0 | 3750 | 0.7866 | 0.8917 |
Framework versions
- Transformers 4.35.2
- Pytorch 2.1.0+cu118
- Datasets 2.15.0
- Tokenizers 0.15.0