swin-base-patch4-window7-224-in22k-finetuned-brain-tumor-final_13
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: 0.0796
- Accuracy: 0.9784
- F1 Score: 0.9786
- Precision: 0.9787
- Sensitivity: 0.9790
- Specificity: 0.9946
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 |
|---|---|---|---|---|---|---|---|---|
| 1.2276 | 0.99 | 19 | 0.5721 | 0.7891 | 0.7955 | 0.8401 | 0.7933 | 0.9454 |
| 0.3873 | 1.97 | 38 | 0.2399 | 0.9195 | 0.9195 | 0.9207 | 0.9224 | 0.9796 |
| 0.1287 | 2.96 | 57 | 0.2204 | 0.9230 | 0.9237 | 0.9275 | 0.9261 | 0.9806 |
| 0.0882 | 4.0 | 77 | 0.1026 | 0.9647 | 0.9649 | 0.9647 | 0.9656 | 0.9911 |
| 0.0605 | 4.99 | 96 | 0.0898 | 0.9678 | 0.9683 | 0.9686 | 0.9685 | 0.9919 |
| 0.0439 | 5.97 | 115 | 0.0853 | 0.9741 | 0.9746 | 0.9748 | 0.9747 | 0.9935 |
| 0.0275 | 6.96 | 134 | 0.0941 | 0.9721 | 0.9724 | 0.9730 | 0.9730 | 0.9930 |
| 0.0186 | 8.0 | 154 | 0.0803 | 0.9764 | 0.9767 | 0.9770 | 0.9773 | 0.9941 |
| 0.0165 | 8.99 | 173 | 0.0740 | 0.9780 | 0.9782 | 0.9782 | 0.9786 | 0.9945 |
| 0.0106 | 9.87 | 190 | 0.0796 | 0.9784 | 0.9786 | 0.9787 | 0.9790 | 0.9946 |
Framework versions
- Transformers 4.29.2
- Pytorch 2.0.1+cu117
- Datasets 2.12.0
- Tokenizers 0.13.3