skin-cancer-model
This model is a fine-tuned version of microsoft/swin-tiny-patch4-window7-224 on the imagefolder dataset. It achieves the following results on the evaluation set:
- Loss: 0.8460
- Accuracy: 0.7455
- F1 Macro: 0.5959
- F1 Weighted: 0.7645
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: 0.0002
- train_batch_size: 16
- eval_batch_size: 8
- seed: 42
- optimizer: Use OptimizerNames.ADAMW_TORCH_FUSED with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
- lr_scheduler_type: linear
- num_epochs: 15
- mixed_precision_training: Native AMP
Training results
| Training Loss | Epoch | Step | Validation Loss | Accuracy | F1 Macro | F1 Weighted |
|---|---|---|---|---|---|---|
| 1.8576 | 1.0 | 564 | 1.7057 | 0.2844 | 0.1599 | 0.3736 |
| 1.4609 | 2.0 | 1128 | 1.5769 | 0.4501 | 0.2754 | 0.5106 |
| 1.502 | 3.0 | 1692 | 1.2726 | 0.6637 | 0.4165 | 0.6799 |
| 1.7705 | 4.0 | 2256 | 1.2379 | 0.6208 | 0.4232 | 0.6596 |
| 1.5335 | 5.0 | 2820 | 1.3475 | 0.5788 | 0.4288 | 0.6291 |
| 1.1427 | 6.0 | 3384 | 1.3055 | 0.5878 | 0.4028 | 0.6421 |
| 1.1915 | 7.0 | 3948 | 1.1663 | 0.6537 | 0.4802 | 0.6864 |
| 0.9543 | 8.0 | 4512 | 1.0841 | 0.6836 | 0.4763 | 0.7054 |
| 1.2565 | 9.0 | 5076 | 0.9467 | 0.6078 | 0.5395 | 0.6484 |
| 1.1679 | 10.0 | 5640 | 0.9393 | 0.6776 | 0.5634 | 0.7090 |
| 1.19 | 11.0 | 6204 | 0.8992 | 0.6677 | 0.5324 | 0.7011 |
| 0.9055 | 12.0 | 6768 | 0.8849 | 0.6876 | 0.5730 | 0.7156 |
| 0.6657 | 13.0 | 7332 | 0.8501 | 0.7146 | 0.5919 | 0.7420 |
| 0.7657 | 14.0 | 7896 | 0.8460 | 0.7455 | 0.5959 | 0.7645 |
| 0.5182 | 15.0 | 8460 | 0.8811 | 0.7305 | 0.5851 | 0.7538 |
Framework versions
- Transformers 4.56.1
- Pytorch 2.8.0+cu126
- Datasets 4.1.1
- Tokenizers 0.22.0