swin-base-patch4-window7-224-in22k-finetuned-brain-tumor-final_11
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: 0.0783
- Accuracy: 0.9800
- F1 Score: 0.9796
- Precision: 0.9797
- Sensitivity: 0.9796
- Specificity: 0.9950
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: 100
- eval_batch_size: 100
- seed: 42
- gradient_accumulation_steps: 4
- total_train_batch_size: 400
- 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 |
|---|---|---|---|---|---|---|---|---|
| 1.2367 | 0.99 | 19 | 0.3560 | 0.8649 | 0.8629 | 0.8649 | 0.8629 | 0.9648 |
| 0.2911 | 1.97 | 38 | 0.2087 | 0.9297 | 0.9290 | 0.9335 | 0.9277 | 0.9822 |
| 0.1681 | 2.96 | 57 | 0.1393 | 0.9564 | 0.9558 | 0.9558 | 0.9562 | 0.9890 |
| 0.0923 | 4.0 | 77 | 0.1106 | 0.9643 | 0.9639 | 0.9637 | 0.9647 | 0.9910 |
| 0.0602 | 4.99 | 96 | 0.1510 | 0.9505 | 0.9494 | 0.9512 | 0.9504 | 0.9875 |
| 0.0388 | 5.97 | 115 | 0.1145 | 0.9666 | 0.9667 | 0.9670 | 0.9672 | 0.9916 |
| 0.0197 | 6.96 | 134 | 0.0783 | 0.9800 | 0.9796 | 0.9797 | 0.9796 | 0.9950 |
| 0.0172 | 8.0 | 154 | 0.1032 | 0.9713 | 0.9713 | 0.9715 | 0.9718 | 0.9928 |
| 0.0169 | 8.99 | 173 | 0.0854 | 0.9776 | 0.9772 | 0.9771 | 0.9774 | 0.9944 |
| 0.0104 | 9.87 | 190 | 0.0948 | 0.9753 | 0.9750 | 0.9749 | 0.9753 | 0.9938 |
Framework versions
- Transformers 4.29.2
- Pytorch 2.0.1+cu117
- Datasets 2.12.0
- Tokenizers 0.13.3