smids_10x_deit_tiny_adamax_0001_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: 0.7859
- Accuracy: 0.9182
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 |
|---|---|---|---|---|
| 0.2608 | 1.0 | 751 | 0.2844 | 0.8898 |
| 0.1304 | 2.0 | 1502 | 0.3294 | 0.8765 |
| 0.111 | 3.0 | 2253 | 0.3516 | 0.9015 |
| 0.1003 | 4.0 | 3004 | 0.4446 | 0.8932 |
| 0.034 | 5.0 | 3755 | 0.5205 | 0.8982 |
| 0.0067 | 6.0 | 4506 | 0.6326 | 0.9015 |
| 0.0227 | 7.0 | 5257 | 0.8411 | 0.8815 |
| 0.0262 | 8.0 | 6008 | 0.8754 | 0.8865 |
| 0.0488 | 9.0 | 6759 | 0.7139 | 0.9098 |
| 0.0361 | 10.0 | 7510 | 0.7866 | 0.8948 |
| 0.0107 | 11.0 | 8261 | 0.8081 | 0.9048 |
| 0.0056 | 12.0 | 9012 | 0.7555 | 0.8998 |
| 0.0 | 13.0 | 9763 | 0.8196 | 0.9015 |
| 0.0016 | 14.0 | 10514 | 0.8589 | 0.9032 |
| 0.0 | 15.0 | 11265 | 0.8346 | 0.9098 |
| 0.0 | 16.0 | 12016 | 0.7703 | 0.9115 |
| 0.0 | 17.0 | 12767 | 0.8587 | 0.9032 |
| 0.0 | 18.0 | 13518 | 0.8122 | 0.9115 |
| 0.0 | 19.0 | 14269 | 0.8002 | 0.9048 |
| 0.0 | 20.0 | 15020 | 0.8446 | 0.9115 |
| 0.0 | 21.0 | 15771 | 0.8926 | 0.9048 |
| 0.0 | 22.0 | 16522 | 0.8190 | 0.9065 |
| 0.0 | 23.0 | 17273 | 0.7943 | 0.9098 |
| 0.0 | 24.0 | 18024 | 0.7616 | 0.9098 |
| 0.0 | 25.0 | 18775 | 0.7566 | 0.9149 |
| 0.0 | 26.0 | 19526 | 0.7309 | 0.9149 |
| 0.0 | 27.0 | 20277 | 0.7760 | 0.9032 |
| 0.0 | 28.0 | 21028 | 0.7849 | 0.9132 |
| 0.008 | 29.0 | 21779 | 0.7826 | 0.9149 |
| 0.0 | 30.0 | 22530 | 0.7666 | 0.9199 |
| 0.0 | 31.0 | 23281 | 0.7402 | 0.9199 |
| 0.0 | 32.0 | 24032 | 0.7484 | 0.9199 |
| 0.0 | 33.0 | 24783 | 0.7616 | 0.9165 |
| 0.0 | 34.0 | 25534 | 0.7803 | 0.9149 |
| 0.0 | 35.0 | 26285 | 0.7685 | 0.9199 |
| 0.0 | 36.0 | 27036 | 0.7685 | 0.9165 |
| 0.0 | 37.0 | 27787 | 0.7687 | 0.9199 |
| 0.0 | 38.0 | 28538 | 0.7876 | 0.9199 |
| 0.0 | 39.0 | 29289 | 0.7749 | 0.9215 |
| 0.0 | 40.0 | 30040 | 0.7734 | 0.9165 |
| 0.0 | 41.0 | 30791 | 0.7803 | 0.9199 |
| 0.0 | 42.0 | 31542 | 0.7799 | 0.9182 |
| 0.0 | 43.0 | 32293 | 0.7798 | 0.9182 |
| 0.0 | 44.0 | 33044 | 0.7789 | 0.9182 |
| 0.0 | 45.0 | 33795 | 0.7827 | 0.9199 |
| 0.0 | 46.0 | 34546 | 0.7810 | 0.9182 |
| 0.0 | 47.0 | 35297 | 0.7840 | 0.9182 |
| 0.0 | 48.0 | 36048 | 0.7837 | 0.9199 |
| 0.0 | 49.0 | 36799 | 0.7839 | 0.9199 |
| 0.0 | 50.0 | 37550 | 0.7859 | 0.9182 |
Framework versions
- Transformers 4.32.1
- Pytorch 2.1.1+cu121
- Datasets 2.12.0
- Tokenizers 0.13.2