hushem_5x_deit_small_adamax_001_fold3
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: 2.7808
- Accuracy: 0.6744
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 |
|---|---|---|---|---|
| 1.4231 | 1.0 | 28 | 1.5192 | 0.2558 |
| 1.3519 | 2.0 | 56 | 1.1068 | 0.5581 |
| 1.0181 | 3.0 | 84 | 1.0826 | 0.5581 |
| 0.9214 | 4.0 | 112 | 0.9767 | 0.5814 |
| 0.6936 | 5.0 | 140 | 1.4687 | 0.3488 |
| 0.9125 | 6.0 | 168 | 1.1689 | 0.5116 |
| 0.8268 | 7.0 | 196 | 1.1523 | 0.5581 |
| 0.7335 | 8.0 | 224 | 0.7001 | 0.6744 |
| 0.527 | 9.0 | 252 | 0.6807 | 0.6977 |
| 0.5031 | 10.0 | 280 | 1.2372 | 0.5814 |
| 0.3869 | 11.0 | 308 | 1.0194 | 0.6744 |
| 0.4401 | 12.0 | 336 | 1.2579 | 0.6279 |
| 0.3701 | 13.0 | 364 | 0.9873 | 0.7209 |
| 0.2336 | 14.0 | 392 | 1.2562 | 0.5581 |
| 0.1629 | 15.0 | 420 | 1.2297 | 0.6744 |
| 0.1635 | 16.0 | 448 | 1.3203 | 0.7674 |
| 0.1352 | 17.0 | 476 | 1.7615 | 0.6977 |
| 0.1062 | 18.0 | 504 | 1.7045 | 0.6977 |
| 0.0801 | 19.0 | 532 | 2.1745 | 0.6047 |
| 0.0683 | 20.0 | 560 | 1.8826 | 0.6977 |
| 0.0931 | 21.0 | 588 | 2.7349 | 0.6047 |
| 0.0235 | 22.0 | 616 | 2.1002 | 0.6744 |
| 0.05 | 23.0 | 644 | 1.7777 | 0.7442 |
| 0.0038 | 24.0 | 672 | 1.7249 | 0.6977 |
| 0.0183 | 25.0 | 700 | 2.2930 | 0.6977 |
| 0.023 | 26.0 | 728 | 1.7975 | 0.7209 |
| 0.0109 | 27.0 | 756 | 2.0767 | 0.7442 |
| 0.0003 | 28.0 | 784 | 2.6313 | 0.6977 |
| 0.0001 | 29.0 | 812 | 2.7110 | 0.6512 |
| 0.0 | 30.0 | 840 | 2.7098 | 0.6744 |
| 0.0 | 31.0 | 868 | 2.7164 | 0.6744 |
| 0.0 | 32.0 | 896 | 2.7474 | 0.6744 |
| 0.0 | 33.0 | 924 | 2.7506 | 0.6744 |
| 0.0 | 34.0 | 952 | 2.7511 | 0.6744 |
| 0.0 | 35.0 | 980 | 2.7547 | 0.6744 |
| 0.0 | 36.0 | 1008 | 2.7577 | 0.6744 |
| 0.0 | 37.0 | 1036 | 2.7609 | 0.6744 |
| 0.0 | 38.0 | 1064 | 2.7631 | 0.6744 |
| 0.0 | 39.0 | 1092 | 2.7673 | 0.6744 |
| 0.0 | 40.0 | 1120 | 2.7700 | 0.6744 |
| 0.0 | 41.0 | 1148 | 2.7723 | 0.6744 |
| 0.0 | 42.0 | 1176 | 2.7743 | 0.6744 |
| 0.0 | 43.0 | 1204 | 2.7761 | 0.6744 |
| 0.0 | 44.0 | 1232 | 2.7776 | 0.6744 |
| 0.0 | 45.0 | 1260 | 2.7788 | 0.6744 |
| 0.0 | 46.0 | 1288 | 2.7800 | 0.6744 |
| 0.0 | 47.0 | 1316 | 2.7805 | 0.6744 |
| 0.0 | 48.0 | 1344 | 2.7808 | 0.6744 |
| 0.0 | 49.0 | 1372 | 2.7808 | 0.6744 |
| 0.0 | 50.0 | 1400 | 2.7808 | 0.6744 |
Framework versions
- Transformers 4.35.2
- Pytorch 2.1.0+cu118
- Datasets 2.15.0
- Tokenizers 0.15.0