smids_10x_deit_tiny_sgd_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: 0.3411
- Accuracy: 0.8715
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 |
|---|---|---|---|---|
| 0.5089 | 1.0 | 751 | 0.6112 | 0.7429 |
| 0.3967 | 2.0 | 1502 | 0.4763 | 0.7930 |
| 0.2926 | 3.0 | 2253 | 0.3983 | 0.8314 |
| 0.2707 | 4.0 | 3004 | 0.3655 | 0.8431 |
| 0.3017 | 5.0 | 3755 | 0.3388 | 0.8598 |
| 0.2883 | 6.0 | 4506 | 0.3239 | 0.8581 |
| 0.2607 | 7.0 | 5257 | 0.3147 | 0.8681 |
| 0.2345 | 8.0 | 6008 | 0.3166 | 0.8564 |
| 0.2342 | 9.0 | 6759 | 0.3080 | 0.8765 |
| 0.2448 | 10.0 | 7510 | 0.3026 | 0.8715 |
| 0.2489 | 11.0 | 8261 | 0.3046 | 0.8664 |
| 0.2253 | 12.0 | 9012 | 0.3048 | 0.8731 |
| 0.1846 | 13.0 | 9763 | 0.3033 | 0.8681 |
| 0.1475 | 14.0 | 10514 | 0.3005 | 0.8698 |
| 0.1911 | 15.0 | 11265 | 0.2987 | 0.8731 |
| 0.1765 | 16.0 | 12016 | 0.3037 | 0.8664 |
| 0.2371 | 17.0 | 12767 | 0.3023 | 0.8698 |
| 0.1758 | 18.0 | 13518 | 0.3012 | 0.8731 |
| 0.2258 | 19.0 | 14269 | 0.3050 | 0.8698 |
| 0.1902 | 20.0 | 15020 | 0.3050 | 0.8681 |
| 0.1077 | 21.0 | 15771 | 0.3133 | 0.8698 |
| 0.1647 | 22.0 | 16522 | 0.3182 | 0.8715 |
| 0.1389 | 23.0 | 17273 | 0.3149 | 0.8731 |
| 0.1858 | 24.0 | 18024 | 0.3136 | 0.8664 |
| 0.1711 | 25.0 | 18775 | 0.3139 | 0.8631 |
| 0.191 | 26.0 | 19526 | 0.3138 | 0.8648 |
| 0.181 | 27.0 | 20277 | 0.3209 | 0.8715 |
| 0.2185 | 28.0 | 21028 | 0.3215 | 0.8748 |
| 0.1776 | 29.0 | 21779 | 0.3185 | 0.8715 |
| 0.135 | 30.0 | 22530 | 0.3255 | 0.8681 |
| 0.1282 | 31.0 | 23281 | 0.3235 | 0.8715 |
| 0.1784 | 32.0 | 24032 | 0.3244 | 0.8698 |
| 0.2006 | 33.0 | 24783 | 0.3311 | 0.8581 |
| 0.106 | 34.0 | 25534 | 0.3280 | 0.8715 |
| 0.1426 | 35.0 | 26285 | 0.3297 | 0.8681 |
| 0.1309 | 36.0 | 27036 | 0.3304 | 0.8614 |
| 0.2379 | 37.0 | 27787 | 0.3292 | 0.8748 |
| 0.1297 | 38.0 | 28538 | 0.3352 | 0.8698 |
| 0.0635 | 39.0 | 29289 | 0.3365 | 0.8664 |
| 0.1936 | 40.0 | 30040 | 0.3376 | 0.8698 |
| 0.1143 | 41.0 | 30791 | 0.3368 | 0.8731 |
| 0.1251 | 42.0 | 31542 | 0.3385 | 0.8664 |
| 0.1338 | 43.0 | 32293 | 0.3411 | 0.8681 |
| 0.1093 | 44.0 | 33044 | 0.3395 | 0.8681 |
| 0.1208 | 45.0 | 33795 | 0.3395 | 0.8715 |
| 0.1467 | 46.0 | 34546 | 0.3416 | 0.8698 |
| 0.0894 | 47.0 | 35297 | 0.3413 | 0.8698 |
| 0.1318 | 48.0 | 36048 | 0.3412 | 0.8681 |
| 0.0983 | 49.0 | 36799 | 0.3410 | 0.8715 |
| 0.1226 | 50.0 | 37550 | 0.3411 | 0.8715 |
Framework versions
- Transformers 4.32.1
- Pytorch 2.1.1+cu121
- Datasets 2.12.0
- Tokenizers 0.13.2