smids_10x_deit_small_rms_0001_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: 1.4626
- Accuracy: 0.885
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 |
|---|---|---|---|---|
| 0.2601 | 1.0 | 750 | 0.3716 | 0.8583 |
| 0.1216 | 2.0 | 1500 | 0.6123 | 0.855 |
| 0.1115 | 3.0 | 2250 | 0.5264 | 0.8667 |
| 0.066 | 4.0 | 3000 | 0.5528 | 0.8733 |
| 0.0447 | 5.0 | 3750 | 0.7645 | 0.8633 |
| 0.0487 | 6.0 | 4500 | 0.8232 | 0.8767 |
| 0.0554 | 7.0 | 5250 | 0.7277 | 0.865 |
| 0.0304 | 8.0 | 6000 | 0.8549 | 0.8667 |
| 0.0413 | 9.0 | 6750 | 0.8526 | 0.865 |
| 0.0142 | 10.0 | 7500 | 1.0441 | 0.8567 |
| 0.0561 | 11.0 | 8250 | 1.0164 | 0.8633 |
| 0.0352 | 12.0 | 9000 | 0.8537 | 0.875 |
| 0.0141 | 13.0 | 9750 | 0.9173 | 0.8617 |
| 0.0343 | 14.0 | 10500 | 1.0250 | 0.86 |
| 0.0048 | 15.0 | 11250 | 0.9231 | 0.8617 |
| 0.0038 | 16.0 | 12000 | 1.1476 | 0.8617 |
| 0.0308 | 17.0 | 12750 | 0.9914 | 0.885 |
| 0.0168 | 18.0 | 13500 | 1.0050 | 0.8783 |
| 0.0001 | 19.0 | 14250 | 1.0610 | 0.8667 |
| 0.0 | 20.0 | 15000 | 1.0251 | 0.86 |
| 0.0232 | 21.0 | 15750 | 1.1692 | 0.855 |
| 0.0026 | 22.0 | 16500 | 0.9562 | 0.8833 |
| 0.0001 | 23.0 | 17250 | 1.0914 | 0.8733 |
| 0.0042 | 24.0 | 18000 | 1.0684 | 0.8767 |
| 0.0 | 25.0 | 18750 | 0.9724 | 0.8833 |
| 0.0001 | 26.0 | 19500 | 1.0636 | 0.86 |
| 0.0001 | 27.0 | 20250 | 1.1239 | 0.86 |
| 0.0015 | 28.0 | 21000 | 1.1692 | 0.8683 |
| 0.0308 | 29.0 | 21750 | 1.1241 | 0.875 |
| 0.0263 | 30.0 | 22500 | 1.0816 | 0.8867 |
| 0.0 | 31.0 | 23250 | 0.9644 | 0.8867 |
| 0.0 | 32.0 | 24000 | 1.1653 | 0.875 |
| 0.0 | 33.0 | 24750 | 1.2370 | 0.8833 |
| 0.0117 | 34.0 | 25500 | 1.3585 | 0.8767 |
| 0.0014 | 35.0 | 26250 | 1.1826 | 0.885 |
| 0.0 | 36.0 | 27000 | 1.2030 | 0.8867 |
| 0.0 | 37.0 | 27750 | 1.4012 | 0.8717 |
| 0.0 | 38.0 | 28500 | 1.3242 | 0.8717 |
| 0.0 | 39.0 | 29250 | 1.2640 | 0.8833 |
| 0.0 | 40.0 | 30000 | 1.3613 | 0.8817 |
| 0.0 | 41.0 | 30750 | 1.3467 | 0.8883 |
| 0.0 | 42.0 | 31500 | 1.3505 | 0.8917 |
| 0.0 | 43.0 | 32250 | 1.3990 | 0.8867 |
| 0.0 | 44.0 | 33000 | 1.4081 | 0.8917 |
| 0.0 | 45.0 | 33750 | 1.4275 | 0.8883 |
| 0.0 | 46.0 | 34500 | 1.4373 | 0.8867 |
| 0.0 | 47.0 | 35250 | 1.4471 | 0.8867 |
| 0.0 | 48.0 | 36000 | 1.4562 | 0.885 |
| 0.0 | 49.0 | 36750 | 1.4611 | 0.885 |
| 0.0 | 50.0 | 37500 | 1.4626 | 0.885 |
Framework versions
- Transformers 4.32.1
- Pytorch 2.1.0+cu121
- Datasets 2.12.0
- Tokenizers 0.13.2