smids_10x_deit_tiny_sgd_00001_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: 0.8572
- Accuracy: 0.6210
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 |
|---|---|---|---|---|
| 1.1945 | 1.0 | 751 | 1.0811 | 0.4240 |
| 1.055 | 2.0 | 1502 | 1.0655 | 0.4407 |
| 1.0679 | 3.0 | 2253 | 1.0517 | 0.4608 |
| 1.0096 | 4.0 | 3004 | 1.0390 | 0.4808 |
| 1.0116 | 5.0 | 3755 | 1.0270 | 0.4908 |
| 1.0321 | 6.0 | 4506 | 1.0157 | 0.4975 |
| 1.0032 | 7.0 | 5257 | 1.0051 | 0.5175 |
| 0.974 | 8.0 | 6008 | 0.9949 | 0.5225 |
| 1.0026 | 9.0 | 6759 | 0.9855 | 0.5275 |
| 0.9625 | 10.0 | 7510 | 0.9766 | 0.5309 |
| 1.0189 | 11.0 | 8261 | 0.9684 | 0.5359 |
| 0.9634 | 12.0 | 9012 | 0.9606 | 0.5459 |
| 0.9198 | 13.0 | 9763 | 0.9533 | 0.5492 |
| 0.8927 | 14.0 | 10514 | 0.9463 | 0.5543 |
| 0.913 | 15.0 | 11265 | 0.9398 | 0.5626 |
| 0.8696 | 16.0 | 12016 | 0.9337 | 0.5659 |
| 0.9094 | 17.0 | 12767 | 0.9279 | 0.5710 |
| 0.8979 | 18.0 | 13518 | 0.9225 | 0.5826 |
| 0.9196 | 19.0 | 14269 | 0.9174 | 0.5860 |
| 0.8982 | 20.0 | 15020 | 0.9126 | 0.5876 |
| 0.8408 | 21.0 | 15771 | 0.9081 | 0.5893 |
| 0.8801 | 22.0 | 16522 | 0.9038 | 0.5927 |
| 0.8134 | 23.0 | 17273 | 0.8998 | 0.5927 |
| 0.8902 | 24.0 | 18024 | 0.8960 | 0.5943 |
| 0.7916 | 25.0 | 18775 | 0.8925 | 0.5977 |
| 0.9125 | 26.0 | 19526 | 0.8892 | 0.5977 |
| 0.8433 | 27.0 | 20277 | 0.8861 | 0.5977 |
| 0.8267 | 28.0 | 21028 | 0.8831 | 0.6010 |
| 0.8301 | 29.0 | 21779 | 0.8804 | 0.6027 |
| 0.8483 | 30.0 | 22530 | 0.8778 | 0.6043 |
| 0.8001 | 31.0 | 23281 | 0.8755 | 0.6043 |
| 0.8847 | 32.0 | 24032 | 0.8733 | 0.6043 |
| 0.8518 | 33.0 | 24783 | 0.8712 | 0.6043 |
| 0.8101 | 34.0 | 25534 | 0.8693 | 0.6060 |
| 0.8163 | 35.0 | 26285 | 0.8676 | 0.6060 |
| 0.793 | 36.0 | 27036 | 0.8660 | 0.6110 |
| 0.8627 | 37.0 | 27787 | 0.8646 | 0.6127 |
| 0.845 | 38.0 | 28538 | 0.8633 | 0.6160 |
| 0.8148 | 39.0 | 29289 | 0.8622 | 0.6177 |
| 0.8547 | 40.0 | 30040 | 0.8611 | 0.6177 |
| 0.8088 | 41.0 | 30791 | 0.8603 | 0.6177 |
| 0.7818 | 42.0 | 31542 | 0.8595 | 0.6194 |
| 0.8609 | 43.0 | 32293 | 0.8589 | 0.6210 |
| 0.7903 | 44.0 | 33044 | 0.8583 | 0.6210 |
| 0.8473 | 45.0 | 33795 | 0.8579 | 0.6210 |
| 0.8128 | 46.0 | 34546 | 0.8576 | 0.6210 |
| 0.7919 | 47.0 | 35297 | 0.8574 | 0.6210 |
| 0.8252 | 48.0 | 36048 | 0.8573 | 0.6210 |
| 0.8055 | 49.0 | 36799 | 0.8572 | 0.6210 |
| 0.8385 | 50.0 | 37550 | 0.8572 | 0.6210 |
Framework versions
- Transformers 4.32.1
- Pytorch 2.1.0+cu121
- Datasets 2.12.0
- Tokenizers 0.13.2