smids_5x_deit_small_rms_001_fold1
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: 0.5438
- Accuracy: 0.7663
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 |
|---|---|---|---|---|
| 0.9114 | 1.0 | 376 | 0.8706 | 0.5559 |
| 0.8509 | 2.0 | 752 | 1.2414 | 0.3456 |
| 0.8099 | 3.0 | 1128 | 1.0576 | 0.4007 |
| 0.8085 | 4.0 | 1504 | 0.8246 | 0.5442 |
| 0.8886 | 5.0 | 1880 | 0.8245 | 0.5376 |
| 0.7819 | 6.0 | 2256 | 0.7875 | 0.5977 |
| 0.7498 | 7.0 | 2632 | 0.8002 | 0.6344 |
| 0.7083 | 8.0 | 3008 | 0.8113 | 0.6027 |
| 0.7609 | 9.0 | 3384 | 0.7440 | 0.6594 |
| 0.7953 | 10.0 | 3760 | 0.7639 | 0.5993 |
| 0.694 | 11.0 | 4136 | 0.7065 | 0.6594 |
| 0.7315 | 12.0 | 4512 | 0.7188 | 0.6277 |
| 0.7192 | 13.0 | 4888 | 0.6863 | 0.7229 |
| 0.6504 | 14.0 | 5264 | 0.6661 | 0.6828 |
| 0.6524 | 15.0 | 5640 | 0.6777 | 0.6661 |
| 0.5701 | 16.0 | 6016 | 0.7272 | 0.6561 |
| 0.5543 | 17.0 | 6392 | 0.7125 | 0.6878 |
| 0.6439 | 18.0 | 6768 | 0.6430 | 0.7028 |
| 0.648 | 19.0 | 7144 | 0.6863 | 0.6928 |
| 0.5899 | 20.0 | 7520 | 0.6226 | 0.7162 |
| 0.6393 | 21.0 | 7896 | 0.6018 | 0.7312 |
| 0.5884 | 22.0 | 8272 | 0.5610 | 0.7412 |
| 0.5288 | 23.0 | 8648 | 0.5975 | 0.7379 |
| 0.5965 | 24.0 | 9024 | 0.6473 | 0.7028 |
| 0.58 | 25.0 | 9400 | 0.5765 | 0.7396 |
| 0.5899 | 26.0 | 9776 | 0.6331 | 0.7245 |
| 0.5507 | 27.0 | 10152 | 0.5858 | 0.7396 |
| 0.5002 | 28.0 | 10528 | 0.5674 | 0.7396 |
| 0.5229 | 29.0 | 10904 | 0.5711 | 0.7629 |
| 0.5096 | 30.0 | 11280 | 0.5570 | 0.7312 |
| 0.5311 | 31.0 | 11656 | 0.5601 | 0.7396 |
| 0.5742 | 32.0 | 12032 | 0.6065 | 0.7346 |
| 0.4585 | 33.0 | 12408 | 0.5565 | 0.7462 |
| 0.5294 | 34.0 | 12784 | 0.5555 | 0.7446 |
| 0.5171 | 35.0 | 13160 | 0.5723 | 0.7462 |
| 0.4899 | 36.0 | 13536 | 0.5748 | 0.7279 |
| 0.4582 | 37.0 | 13912 | 0.5789 | 0.7396 |
| 0.5149 | 38.0 | 14288 | 0.5146 | 0.7679 |
| 0.4968 | 39.0 | 14664 | 0.6020 | 0.7613 |
| 0.5645 | 40.0 | 15040 | 0.5459 | 0.7546 |
| 0.4741 | 41.0 | 15416 | 0.5562 | 0.7479 |
| 0.4423 | 42.0 | 15792 | 0.5487 | 0.7412 |
| 0.4186 | 43.0 | 16168 | 0.5329 | 0.7479 |
| 0.4763 | 44.0 | 16544 | 0.5469 | 0.7462 |
| 0.4775 | 45.0 | 16920 | 0.5538 | 0.7496 |
| 0.4053 | 46.0 | 17296 | 0.5298 | 0.7613 |
| 0.429 | 47.0 | 17672 | 0.5338 | 0.7663 |
| 0.4194 | 48.0 | 18048 | 0.5631 | 0.7496 |
| 0.3965 | 49.0 | 18424 | 0.5407 | 0.7629 |
| 0.356 | 50.0 | 18800 | 0.5438 | 0.7663 |
Framework versions
- Transformers 4.32.1
- Pytorch 2.1.0+cu121
- Datasets 2.12.0
- Tokenizers 0.13.2