<!-- This model card has been generated automatically according to the information the Trainer had access to. You should probably proofread and complete it, then remove this comment. -->
swin-base-patch4-window7-224-in22k-finetuned-brain-tumor-final_09
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: 0.2586
- Accuracy: 0.9077
- F1 Score: 0.9093
- Precision: 0.9184
- Sensitivity: 0.9071
- Specificity: 0.9766
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: 1e-05
- 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: 20
Training results
| Training Loss | Epoch | Step | Validation Loss | Accuracy | F1 Score | Precision | Sensitivity | Specificity |
|---|---|---|---|---|---|---|---|---|
| 1.4243 | 0.99 | 19 | 1.2818 | 0.4124 | 0.3570 | 0.4910 | 0.4019 | 0.8403 |
| 1.1046 | 1.97 | 38 | 0.8873 | 0.6658 | 0.6584 | 0.7235 | 0.6608 | 0.9117 |
| 0.5232 | 2.96 | 57 | 0.5753 | 0.7671 | 0.7654 | 0.8063 | 0.7631 | 0.9395 |
| 0.3235 | 4.0 | 77 | 0.4476 | 0.8256 | 0.8272 | 0.8496 | 0.8228 | 0.9549 |
| 0.2586 | 4.99 | 96 | 0.3886 | 0.8590 | 0.8608 | 0.8764 | 0.8567 | 0.9638 |
| 0.1986 | 5.97 | 115 | 0.3538 | 0.8641 | 0.8663 | 0.8816 | 0.8624 | 0.9652 |
| 0.166 | 6.96 | 134 | 0.3543 | 0.8649 | 0.8668 | 0.8849 | 0.8637 | 0.9655 |
| 0.1345 | 8.0 | 154 | 0.3729 | 0.8586 | 0.8610 | 0.8837 | 0.8571 | 0.9640 |
| 0.1197 | 8.99 | 173 | 0.2879 | 0.8975 | 0.8987 | 0.9098 | 0.8961 | 0.9740 |
| 0.1033 | 9.97 | 192 | 0.2810 | 0.8998 | 0.9013 | 0.9128 | 0.8983 | 0.9746 |
| 0.0957 | 10.96 | 211 | 0.3239 | 0.8802 | 0.8818 | 0.8988 | 0.8795 | 0.9696 |
| 0.085 | 12.0 | 231 | 0.2586 | 0.9077 | 0.9093 | 0.9184 | 0.9071 | 0.9766 |
| 0.0769 | 12.99 | 250 | 0.2662 | 0.9018 | 0.9036 | 0.9149 | 0.9011 | 0.9751 |
| 0.0758 | 13.97 | 269 | 0.2830 | 0.8951 | 0.8970 | 0.9102 | 0.8945 | 0.9734 |
| 0.068 | 14.96 | 288 | 0.2757 | 0.8967 | 0.8986 | 0.9113 | 0.8960 | 0.9738 |
| 0.0641 | 16.0 | 308 | 0.2743 | 0.8991 | 0.9008 | 0.9136 | 0.8984 | 0.9744 |
| 0.0623 | 16.99 | 327 | 0.2713 | 0.8987 | 0.9001 | 0.9127 | 0.8982 | 0.9743 |
| 0.0542 | 17.97 | 346 | 0.2650 | 0.8987 | 0.9005 | 0.9128 | 0.8980 | 0.9743 |
| 0.0573 | 18.96 | 365 | 0.2709 | 0.8963 | 0.8981 | 0.9112 | 0.8957 | 0.9737 |
| 0.058 | 19.74 | 380 | 0.2778 | 0.8947 | 0.8965 | 0.9101 | 0.8942 | 0.9733 |
Framework versions
- Transformers 4.29.2
- Pytorch 2.0.1+cu117
- Datasets 2.12.0
- Tokenizers 0.13.3
