smids_1x_beit_base_adamax_001_fold4
This model is a fine-tuned version of microsoft/beit-base-patch16-224 on the imagefolder dataset. It achieves the following results on the evaluation set:
- Loss: 1.7646
- Accuracy: 0.775
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 |
|---|---|---|---|---|
| 0.9222 | 1.0 | 75 | 0.8216 | 0.5567 |
| 0.8457 | 2.0 | 150 | 0.8398 | 0.57 |
| 0.8147 | 3.0 | 225 | 0.7493 | 0.6333 |
| 0.7701 | 4.0 | 300 | 0.7606 | 0.6117 |
| 0.8026 | 5.0 | 375 | 0.8189 | 0.565 |
| 0.6963 | 6.0 | 450 | 0.6808 | 0.665 |
| 0.7638 | 7.0 | 525 | 0.6641 | 0.7017 |
| 0.6601 | 8.0 | 600 | 0.6495 | 0.6833 |
| 0.6719 | 9.0 | 675 | 0.7134 | 0.66 |
| 0.5461 | 10.0 | 750 | 0.5791 | 0.7483 |
| 0.547 | 11.0 | 825 | 0.5859 | 0.7633 |
| 0.4912 | 12.0 | 900 | 0.5937 | 0.735 |
| 0.5352 | 13.0 | 975 | 0.5233 | 0.7667 |
| 0.4434 | 14.0 | 1050 | 0.5543 | 0.7617 |
| 0.4927 | 15.0 | 1125 | 0.7581 | 0.6767 |
| 0.4312 | 16.0 | 1200 | 0.5587 | 0.7667 |
| 0.3899 | 17.0 | 1275 | 0.6422 | 0.7633 |
| 0.3786 | 18.0 | 1350 | 0.6068 | 0.7783 |
| 0.4006 | 19.0 | 1425 | 0.6778 | 0.7617 |
| 0.3094 | 20.0 | 1500 | 0.6494 | 0.775 |
| 0.3319 | 21.0 | 1575 | 0.6363 | 0.765 |
| 0.2928 | 22.0 | 1650 | 0.7276 | 0.7817 |
| 0.2846 | 23.0 | 1725 | 0.8156 | 0.7733 |
| 0.1736 | 24.0 | 1800 | 0.7838 | 0.785 |
| 0.2416 | 25.0 | 1875 | 0.8283 | 0.775 |
| 0.1805 | 26.0 | 1950 | 0.8042 | 0.7867 |
| 0.1895 | 27.0 | 2025 | 1.0411 | 0.7933 |
| 0.0832 | 28.0 | 2100 | 1.0766 | 0.7983 |
| 0.099 | 29.0 | 2175 | 1.1178 | 0.7683 |
| 0.0916 | 30.0 | 2250 | 1.3040 | 0.775 |
| 0.128 | 31.0 | 2325 | 1.2237 | 0.7983 |
| 0.0775 | 32.0 | 2400 | 1.1999 | 0.79 |
| 0.0706 | 33.0 | 2475 | 1.4034 | 0.78 |
| 0.0546 | 34.0 | 2550 | 1.4009 | 0.785 |
| 0.0453 | 35.0 | 2625 | 1.2357 | 0.7917 |
| 0.0136 | 36.0 | 2700 | 1.4685 | 0.79 |
| 0.0534 | 37.0 | 2775 | 1.8215 | 0.7717 |
| 0.0751 | 38.0 | 2850 | 1.6150 | 0.7833 |
| 0.0013 | 39.0 | 2925 | 1.7207 | 0.7917 |
| 0.0466 | 40.0 | 3000 | 1.4737 | 0.785 |
| 0.0122 | 41.0 | 3075 | 1.5635 | 0.7783 |
| 0.0071 | 42.0 | 3150 | 1.6935 | 0.7783 |
| 0.0119 | 43.0 | 3225 | 1.6935 | 0.7833 |
| 0.0065 | 44.0 | 3300 | 1.7015 | 0.7883 |
| 0.0254 | 45.0 | 3375 | 1.7329 | 0.7867 |
| 0.0205 | 46.0 | 3450 | 1.6886 | 0.785 |
| 0.0082 | 47.0 | 3525 | 1.7094 | 0.7833 |
| 0.0134 | 48.0 | 3600 | 1.7793 | 0.78 |
| 0.005 | 49.0 | 3675 | 1.7866 | 0.7767 |
| 0.0132 | 50.0 | 3750 | 1.7646 | 0.775 |
Framework versions
- Transformers 4.35.2
- Pytorch 2.1.0+cu118
- Datasets 2.15.0
- Tokenizers 0.15.0