hushem_5x_deit_small_sgd_001_fold4
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: 1.0586
- Accuracy: 0.5
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 |
|---|---|---|---|---|
| 1.487 | 1.0 | 28 | 1.4172 | 0.3095 |
| 1.4655 | 2.0 | 56 | 1.3829 | 0.3333 |
| 1.3519 | 3.0 | 84 | 1.3645 | 0.2619 |
| 1.3235 | 4.0 | 112 | 1.3524 | 0.2619 |
| 1.3019 | 5.0 | 140 | 1.3421 | 0.2857 |
| 1.2683 | 6.0 | 168 | 1.3287 | 0.2857 |
| 1.2448 | 7.0 | 196 | 1.3147 | 0.2857 |
| 1.2154 | 8.0 | 224 | 1.3011 | 0.2619 |
| 1.1886 | 9.0 | 252 | 1.2876 | 0.3571 |
| 1.1547 | 10.0 | 280 | 1.2739 | 0.3810 |
| 1.1374 | 11.0 | 308 | 1.2618 | 0.3810 |
| 1.1111 | 12.0 | 336 | 1.2488 | 0.3810 |
| 1.1298 | 13.0 | 364 | 1.2398 | 0.4048 |
| 1.0797 | 14.0 | 392 | 1.2302 | 0.4048 |
| 1.0414 | 15.0 | 420 | 1.2217 | 0.4286 |
| 1.061 | 16.0 | 448 | 1.2120 | 0.4286 |
| 1.0634 | 17.0 | 476 | 1.2016 | 0.4524 |
| 1.0054 | 18.0 | 504 | 1.1928 | 0.4524 |
| 0.9762 | 19.0 | 532 | 1.1844 | 0.4524 |
| 1.0106 | 20.0 | 560 | 1.1764 | 0.4524 |
| 0.9235 | 21.0 | 588 | 1.1685 | 0.4524 |
| 0.9458 | 22.0 | 616 | 1.1599 | 0.4762 |
| 0.9326 | 23.0 | 644 | 1.1543 | 0.5 |
| 0.9222 | 24.0 | 672 | 1.1465 | 0.5 |
| 0.8846 | 25.0 | 700 | 1.1391 | 0.5 |
| 0.8795 | 26.0 | 728 | 1.1307 | 0.4762 |
| 0.8711 | 27.0 | 756 | 1.1242 | 0.5238 |
| 0.8921 | 28.0 | 784 | 1.1184 | 0.5238 |
| 0.8796 | 29.0 | 812 | 1.1133 | 0.5238 |
| 0.8567 | 30.0 | 840 | 1.1054 | 0.5238 |
| 0.8632 | 31.0 | 868 | 1.1011 | 0.5238 |
| 0.8179 | 32.0 | 896 | 1.0965 | 0.5238 |
| 0.8418 | 33.0 | 924 | 1.0917 | 0.5238 |
| 0.8097 | 34.0 | 952 | 1.0866 | 0.5238 |
| 0.8474 | 35.0 | 980 | 1.0818 | 0.5238 |
| 0.7989 | 36.0 | 1008 | 1.0776 | 0.5238 |
| 0.7935 | 37.0 | 1036 | 1.0750 | 0.5238 |
| 0.8104 | 38.0 | 1064 | 1.0725 | 0.5238 |
| 0.8018 | 39.0 | 1092 | 1.0698 | 0.5238 |
| 0.797 | 40.0 | 1120 | 1.0673 | 0.5238 |
| 0.8004 | 41.0 | 1148 | 1.0654 | 0.5238 |
| 0.775 | 42.0 | 1176 | 1.0641 | 0.5 |
| 0.7606 | 43.0 | 1204 | 1.0623 | 0.5 |
| 0.7649 | 44.0 | 1232 | 1.0613 | 0.5 |
| 0.7627 | 45.0 | 1260 | 1.0601 | 0.5 |
| 0.7807 | 46.0 | 1288 | 1.0595 | 0.5 |
| 0.7697 | 47.0 | 1316 | 1.0588 | 0.5 |
| 0.7683 | 48.0 | 1344 | 1.0586 | 0.5 |
| 0.783 | 49.0 | 1372 | 1.0586 | 0.5 |
| 0.7862 | 50.0 | 1400 | 1.0586 | 0.5 |
Framework versions
- Transformers 4.35.2
- Pytorch 2.1.0+cu118
- Datasets 2.15.0
- Tokenizers 0.15.0