smids_3x_deit_tiny_rms_001_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: 2.0122
- Accuracy: 0.8030
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: 0.001
- 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 |
|---|---|---|---|---|
| 0.8369 | 1.0 | 226 | 0.8757 | 0.5359 |
| 0.8499 | 2.0 | 452 | 0.8448 | 0.5776 |
| 0.7754 | 3.0 | 678 | 0.9313 | 0.5142 |
| 0.8771 | 4.0 | 904 | 0.7652 | 0.6194 |
| 0.7781 | 5.0 | 1130 | 0.7375 | 0.6711 |
| 0.7866 | 6.0 | 1356 | 0.7599 | 0.6394 |
| 0.664 | 7.0 | 1582 | 0.7260 | 0.6711 |
| 0.6855 | 8.0 | 1808 | 0.8844 | 0.5376 |
| 0.6098 | 9.0 | 2034 | 0.7118 | 0.6778 |
| 0.6338 | 10.0 | 2260 | 0.6856 | 0.6962 |
| 0.5812 | 11.0 | 2486 | 0.6665 | 0.6962 |
| 0.5909 | 12.0 | 2712 | 0.8126 | 0.6260 |
| 0.5272 | 13.0 | 2938 | 0.6279 | 0.7329 |
| 0.5688 | 14.0 | 3164 | 0.7483 | 0.6494 |
| 0.5214 | 15.0 | 3390 | 0.6410 | 0.7312 |
| 0.5581 | 16.0 | 3616 | 0.6042 | 0.7412 |
| 0.4723 | 17.0 | 3842 | 0.6758 | 0.7145 |
| 0.5595 | 18.0 | 4068 | 0.6233 | 0.7412 |
| 0.5549 | 19.0 | 4294 | 0.6152 | 0.7329 |
| 0.5078 | 20.0 | 4520 | 0.6278 | 0.7195 |
| 0.5707 | 21.0 | 4746 | 0.5335 | 0.7780 |
| 0.4944 | 22.0 | 4972 | 0.6366 | 0.7396 |
| 0.5416 | 23.0 | 5198 | 0.5752 | 0.7663 |
| 0.5022 | 24.0 | 5424 | 0.5999 | 0.7479 |
| 0.5615 | 25.0 | 5650 | 0.5710 | 0.7596 |
| 0.5132 | 26.0 | 5876 | 0.5875 | 0.7730 |
| 0.3982 | 27.0 | 6102 | 0.5830 | 0.7763 |
| 0.4012 | 28.0 | 6328 | 0.7036 | 0.7563 |
| 0.37 | 29.0 | 6554 | 0.6641 | 0.7429 |
| 0.4588 | 30.0 | 6780 | 0.6124 | 0.7613 |
| 0.3873 | 31.0 | 7006 | 0.6238 | 0.7646 |
| 0.3153 | 32.0 | 7232 | 0.6857 | 0.7613 |
| 0.3038 | 33.0 | 7458 | 0.7385 | 0.7730 |
| 0.2793 | 34.0 | 7684 | 0.6805 | 0.7846 |
| 0.2405 | 35.0 | 7910 | 0.7592 | 0.7846 |
| 0.2843 | 36.0 | 8136 | 0.8044 | 0.7746 |
| 0.2771 | 37.0 | 8362 | 0.7613 | 0.7813 |
| 0.2263 | 38.0 | 8588 | 0.8328 | 0.7679 |
| 0.1499 | 39.0 | 8814 | 0.9707 | 0.7696 |
| 0.1482 | 40.0 | 9040 | 1.0206 | 0.7896 |
| 0.1303 | 41.0 | 9266 | 1.1237 | 0.7947 |
| 0.0595 | 42.0 | 9492 | 1.3060 | 0.7763 |
| 0.0163 | 43.0 | 9718 | 1.4374 | 0.7830 |
| 0.0383 | 44.0 | 9944 | 1.5230 | 0.7863 |
| 0.0303 | 45.0 | 10170 | 1.5896 | 0.7947 |
| 0.0051 | 46.0 | 10396 | 1.8469 | 0.7896 |
| 0.0006 | 47.0 | 10622 | 1.9434 | 0.7880 |
| 0.0004 | 48.0 | 10848 | 2.0244 | 0.7947 |
| 0.0004 | 49.0 | 11074 | 1.9864 | 0.7997 |
| 0.0002 | 50.0 | 11300 | 2.0122 | 0.8030 |
Framework versions
- Transformers 4.32.1
- Pytorch 2.1.1+cu121
- Datasets 2.12.0
- Tokenizers 0.13.2