smids_3x_deit_small_rms_00001_fold4
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: 1.2799
- Accuracy: 0.8783
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 |
|---|---|---|---|---|
| 0.2819 | 1.0 | 225 | 0.3374 | 0.875 |
| 0.0966 | 2.0 | 450 | 0.3633 | 0.8733 |
| 0.0536 | 3.0 | 675 | 0.4166 | 0.8633 |
| 0.0743 | 4.0 | 900 | 0.4615 | 0.88 |
| 0.0231 | 5.0 | 1125 | 0.6512 | 0.865 |
| 0.0353 | 6.0 | 1350 | 0.6888 | 0.8783 |
| 0.0422 | 7.0 | 1575 | 0.8522 | 0.8717 |
| 0.0395 | 8.0 | 1800 | 0.9258 | 0.865 |
| 0.0246 | 9.0 | 2025 | 0.8665 | 0.8717 |
| 0.0065 | 10.0 | 2250 | 0.9925 | 0.8683 |
| 0.0002 | 11.0 | 2475 | 0.9127 | 0.875 |
| 0.0053 | 12.0 | 2700 | 0.9170 | 0.88 |
| 0.0078 | 13.0 | 2925 | 1.0774 | 0.8683 |
| 0.0094 | 14.0 | 3150 | 1.0384 | 0.8717 |
| 0.0815 | 15.0 | 3375 | 0.9472 | 0.8767 |
| 0.0 | 16.0 | 3600 | 1.0185 | 0.87 |
| 0.0 | 17.0 | 3825 | 1.1297 | 0.8733 |
| 0.0 | 18.0 | 4050 | 1.0917 | 0.865 |
| 0.0 | 19.0 | 4275 | 1.2078 | 0.865 |
| 0.0 | 20.0 | 4500 | 1.1817 | 0.8717 |
| 0.0328 | 21.0 | 4725 | 1.2494 | 0.8667 |
| 0.0221 | 22.0 | 4950 | 1.1544 | 0.87 |
| 0.0 | 23.0 | 5175 | 1.1616 | 0.87 |
| 0.0 | 24.0 | 5400 | 1.0994 | 0.875 |
| 0.0001 | 25.0 | 5625 | 1.1448 | 0.8767 |
| 0.0 | 26.0 | 5850 | 1.1095 | 0.88 |
| 0.0 | 27.0 | 6075 | 1.1096 | 0.88 |
| 0.0 | 28.0 | 6300 | 1.1094 | 0.8783 |
| 0.0 | 29.0 | 6525 | 1.1571 | 0.875 |
| 0.0 | 30.0 | 6750 | 1.1384 | 0.8833 |
| 0.0 | 31.0 | 6975 | 1.1678 | 0.875 |
| 0.0 | 32.0 | 7200 | 1.1864 | 0.8733 |
| 0.0006 | 33.0 | 7425 | 1.1549 | 0.8817 |
| 0.0 | 34.0 | 7650 | 1.1715 | 0.8833 |
| 0.0 | 35.0 | 7875 | 1.2200 | 0.8817 |
| 0.0 | 36.0 | 8100 | 1.1957 | 0.8817 |
| 0.0 | 37.0 | 8325 | 1.1943 | 0.8833 |
| 0.0 | 38.0 | 8550 | 1.1984 | 0.88 |
| 0.0 | 39.0 | 8775 | 1.2416 | 0.8817 |
| 0.0 | 40.0 | 9000 | 1.2319 | 0.8783 |
| 0.0 | 41.0 | 9225 | 1.2497 | 0.8783 |
| 0.0028 | 42.0 | 9450 | 1.2526 | 0.8783 |
| 0.0 | 43.0 | 9675 | 1.2621 | 0.8767 |
| 0.0026 | 44.0 | 9900 | 1.2618 | 0.875 |
| 0.0 | 45.0 | 10125 | 1.2702 | 0.8767 |
| 0.0 | 46.0 | 10350 | 1.2698 | 0.88 |
| 0.0 | 47.0 | 10575 | 1.2756 | 0.88 |
| 0.0 | 48.0 | 10800 | 1.2780 | 0.8783 |
| 0.0 | 49.0 | 11025 | 1.2789 | 0.8783 |
| 0.0 | 50.0 | 11250 | 1.2799 | 0.8783 |
Framework versions
- Transformers 4.32.1
- Pytorch 2.1.0+cu121
- Datasets 2.12.0
- Tokenizers 0.13.2