finalProject
This model is a fine-tuned version of microsoft/swin-base-patch4-window7-224-in22k on the imagefolder dataset. It achieves the following results on the evaluation set:
- Loss: 0.0411
- Accuracy: 0.9890
- F1 Score: 0.9892
- Precision: 0.9894
- Sensitivity: 0.9891
- Specificity: 0.9972
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: 0.0001
- train_batch_size: 64
- eval_batch_size: 64
- seed: 42
- gradient_accumulation_steps: 4
- total_train_batch_size: 256
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- lr_scheduler_warmup_ratio: 0.1
- num_epochs: 10
Training results
| Training Loss | Epoch | Step | Validation Loss | Accuracy | F1 Score | Precision | Sensitivity | Specificity |
|---|---|---|---|---|---|---|---|---|
| 0.3384 | 1.0 | 30 | 0.2387 | 0.9144 | 0.9163 | 0.9197 | 0.9146 | 0.9781 |
| 0.1608 | 2.0 | 60 | 0.1635 | 0.9466 | 0.9476 | 0.9485 | 0.9474 | 0.9865 |
| 0.0953 | 3.0 | 90 | 0.0915 | 0.9698 | 0.9703 | 0.9706 | 0.9706 | 0.9924 |
| 0.0573 | 4.0 | 120 | 0.1125 | 0.9607 | 0.9617 | 0.9634 | 0.9621 | 0.9901 |
| 0.0335 | 5.0 | 150 | 0.0536 | 0.9827 | 0.9831 | 0.9837 | 0.9826 | 0.9957 |
| 0.0185 | 6.0 | 180 | 0.0543 | 0.9827 | 0.9830 | 0.9837 | 0.9825 | 0.9957 |
| 0.0226 | 7.0 | 210 | 0.0478 | 0.9859 | 0.9861 | 0.9866 | 0.9856 | 0.9965 |
| 0.0131 | 8.0 | 240 | 0.0468 | 0.9843 | 0.9846 | 0.9847 | 0.9846 | 0.9961 |
| 0.0087 | 9.0 | 270 | 0.0411 | 0.9890 | 0.9892 | 0.9894 | 0.9891 | 0.9972 |
| 0.0043 | 10.0 | 300 | 0.0376 | 0.9886 | 0.9888 | 0.9890 | 0.9887 | 0.9971 |
Framework versions
- Transformers 4.30.2
- Pytorch 2.0.1+cu117
- Datasets 2.13.0
- Tokenizers 0.13.3