smids_10x_deit_small_sgd_00001_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: 0.9431
- Accuracy: 0.5543
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 |
|---|---|---|---|---|
| 1.0703 | 1.0 | 751 | 1.0730 | 0.4341 |
| 1.028 | 2.0 | 1502 | 1.0687 | 0.4341 |
| 1.0258 | 3.0 | 2253 | 1.0640 | 0.4374 |
| 1.0492 | 4.0 | 3004 | 1.0592 | 0.4391 |
| 1.036 | 5.0 | 3755 | 1.0544 | 0.4407 |
| 1.0185 | 6.0 | 4506 | 1.0494 | 0.4457 |
| 1.0044 | 7.0 | 5257 | 1.0445 | 0.4524 |
| 1.0106 | 8.0 | 6008 | 1.0397 | 0.4524 |
| 1.0008 | 9.0 | 6759 | 1.0351 | 0.4558 |
| 0.983 | 10.0 | 7510 | 1.0305 | 0.4624 |
| 0.9888 | 11.0 | 8261 | 1.0261 | 0.4691 |
| 0.98 | 12.0 | 9012 | 1.0217 | 0.4758 |
| 0.9777 | 13.0 | 9763 | 1.0175 | 0.4775 |
| 0.9805 | 14.0 | 10514 | 1.0134 | 0.4825 |
| 0.9554 | 15.0 | 11265 | 1.0095 | 0.4875 |
| 0.9727 | 16.0 | 12016 | 1.0055 | 0.4942 |
| 0.9405 | 17.0 | 12767 | 1.0016 | 0.4992 |
| 0.9669 | 18.0 | 13518 | 0.9980 | 0.5042 |
| 0.9407 | 19.0 | 14269 | 0.9944 | 0.5042 |
| 0.9487 | 20.0 | 15020 | 0.9909 | 0.5075 |
| 0.9336 | 21.0 | 15771 | 0.9876 | 0.5092 |
| 0.9505 | 22.0 | 16522 | 0.9843 | 0.5109 |
| 0.9425 | 23.0 | 17273 | 0.9812 | 0.5125 |
| 0.9422 | 24.0 | 18024 | 0.9782 | 0.5175 |
| 0.9397 | 25.0 | 18775 | 0.9753 | 0.5209 |
| 0.9277 | 26.0 | 19526 | 0.9725 | 0.5225 |
| 0.9248 | 27.0 | 20277 | 0.9699 | 0.5326 |
| 0.915 | 28.0 | 21028 | 0.9674 | 0.5342 |
| 0.9341 | 29.0 | 21779 | 0.9650 | 0.5376 |
| 0.9201 | 30.0 | 22530 | 0.9628 | 0.5392 |
| 0.8994 | 31.0 | 23281 | 0.9606 | 0.5376 |
| 0.9167 | 32.0 | 24032 | 0.9586 | 0.5392 |
| 0.8872 | 33.0 | 24783 | 0.9568 | 0.5426 |
| 0.8983 | 34.0 | 25534 | 0.9550 | 0.5426 |
| 0.8839 | 35.0 | 26285 | 0.9534 | 0.5442 |
| 0.9018 | 36.0 | 27036 | 0.9519 | 0.5476 |
| 0.8955 | 37.0 | 27787 | 0.9506 | 0.5492 |
| 0.8964 | 38.0 | 28538 | 0.9493 | 0.5492 |
| 0.9005 | 39.0 | 29289 | 0.9482 | 0.5492 |
| 0.8988 | 40.0 | 30040 | 0.9472 | 0.5526 |
| 0.8967 | 41.0 | 30791 | 0.9463 | 0.5543 |
| 0.8873 | 42.0 | 31542 | 0.9455 | 0.5543 |
| 0.9048 | 43.0 | 32293 | 0.9449 | 0.5543 |
| 0.8665 | 44.0 | 33044 | 0.9443 | 0.5543 |
| 0.8925 | 45.0 | 33795 | 0.9439 | 0.5543 |
| 0.8934 | 46.0 | 34546 | 0.9435 | 0.5543 |
| 0.8656 | 47.0 | 35297 | 0.9433 | 0.5543 |
| 0.9144 | 48.0 | 36048 | 0.9431 | 0.5543 |
| 0.9081 | 49.0 | 36799 | 0.9431 | 0.5543 |
| 0.8986 | 50.0 | 37550 | 0.9431 | 0.5543 |
Framework versions
- Transformers 4.32.1
- Pytorch 2.1.0+cu121
- Datasets 2.12.0
- Tokenizers 0.13.2