smids_5x_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: 0.6720
- Accuracy: 0.7563
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 |
|---|---|---|---|---|
| 0.9519 | 1.0 | 376 | 0.9699 | 0.4808 |
| 0.8617 | 2.0 | 752 | 0.8618 | 0.5392 |
| 0.8149 | 3.0 | 1128 | 0.8048 | 0.5893 |
| 0.8075 | 4.0 | 1504 | 0.7999 | 0.5609 |
| 0.9135 | 5.0 | 1880 | 0.7865 | 0.6160 |
| 0.783 | 6.0 | 2256 | 0.8586 | 0.5893 |
| 0.725 | 7.0 | 2632 | 0.8054 | 0.6227 |
| 0.6972 | 8.0 | 3008 | 0.7248 | 0.6444 |
| 0.72 | 9.0 | 3384 | 0.7167 | 0.6661 |
| 0.7292 | 10.0 | 3760 | 0.7657 | 0.6795 |
| 0.645 | 11.0 | 4136 | 0.6894 | 0.6861 |
| 0.7059 | 12.0 | 4512 | 0.7066 | 0.6928 |
| 0.7086 | 13.0 | 4888 | 0.7125 | 0.6995 |
| 0.6705 | 14.0 | 5264 | 0.6700 | 0.7078 |
| 0.6566 | 15.0 | 5640 | 0.6881 | 0.6861 |
| 0.5734 | 16.0 | 6016 | 0.7052 | 0.6694 |
| 0.5199 | 17.0 | 6392 | 0.7378 | 0.6628 |
| 0.659 | 18.0 | 6768 | 0.6486 | 0.7112 |
| 0.6288 | 19.0 | 7144 | 0.7161 | 0.6528 |
| 0.566 | 20.0 | 7520 | 0.6171 | 0.7212 |
| 0.6474 | 21.0 | 7896 | 0.6184 | 0.7262 |
| 0.5542 | 22.0 | 8272 | 0.6826 | 0.6861 |
| 0.5759 | 23.0 | 8648 | 0.6131 | 0.7229 |
| 0.6266 | 24.0 | 9024 | 0.6647 | 0.7112 |
| 0.6436 | 25.0 | 9400 | 0.6298 | 0.7078 |
| 0.5378 | 26.0 | 9776 | 0.6147 | 0.7229 |
| 0.534 | 27.0 | 10152 | 0.6258 | 0.7179 |
| 0.4794 | 28.0 | 10528 | 0.6515 | 0.7095 |
| 0.5282 | 29.0 | 10904 | 0.6735 | 0.6912 |
| 0.4828 | 30.0 | 11280 | 0.6279 | 0.7179 |
| 0.5597 | 31.0 | 11656 | 0.6003 | 0.7295 |
| 0.5931 | 32.0 | 12032 | 0.6323 | 0.7362 |
| 0.4604 | 33.0 | 12408 | 0.6185 | 0.7446 |
| 0.473 | 34.0 | 12784 | 0.6171 | 0.7396 |
| 0.5357 | 35.0 | 13160 | 0.6139 | 0.7279 |
| 0.5273 | 36.0 | 13536 | 0.6022 | 0.7379 |
| 0.446 | 37.0 | 13912 | 0.6164 | 0.7362 |
| 0.5051 | 38.0 | 14288 | 0.6160 | 0.7329 |
| 0.5127 | 39.0 | 14664 | 0.6147 | 0.7629 |
| 0.5424 | 40.0 | 15040 | 0.5988 | 0.7579 |
| 0.4672 | 41.0 | 15416 | 0.6152 | 0.7613 |
| 0.4259 | 42.0 | 15792 | 0.6298 | 0.7429 |
| 0.4313 | 43.0 | 16168 | 0.6086 | 0.7462 |
| 0.4716 | 44.0 | 16544 | 0.6307 | 0.7496 |
| 0.4303 | 45.0 | 16920 | 0.6176 | 0.7513 |
| 0.3889 | 46.0 | 17296 | 0.6198 | 0.7479 |
| 0.4191 | 47.0 | 17672 | 0.6340 | 0.7563 |
| 0.3752 | 48.0 | 18048 | 0.6420 | 0.7596 |
| 0.3744 | 49.0 | 18424 | 0.6614 | 0.7529 |
| 0.3137 | 50.0 | 18800 | 0.6720 | 0.7563 |
Framework versions
- Transformers 4.32.1
- Pytorch 2.1.1+cu121
- Datasets 2.12.0
- Tokenizers 0.13.2