metadata
tags:
- model_hub_mixin
- pytorch_model_hub_mixin
license: mit
base_model:
- microsoft/cvt-13
CvTSkinCancer for Skin Cancer Classification
Model Details
- Model Architecture: CvT (Convolutional Vision Transformer)
- Framework: PyTorch
- Input Shape: 224x224 RGB images
- Number of Parameters: (Based on CvT-13 Model)
- Output: Multi-class classification (9 classes)
Model Description
This model leverages the CvT (Convolutional Vision Transformer) from the microsoft/cvt-13 pretrained model for skin cancer classification. The model is fine-tuned by adding a custom fully connected layer that includes SiLU activations, Dropout, and Batch Normalization to improve generalization and reduce overfitting. Only the classifier layer is trainable, while the convolutional layers of the CvT model are frozen to retain pretrained features.
The final model performs multi-class classification with 9 output classes corresponding to different skin cancer types.
Training Details
- Optimizer: Adam
- Batch Size: 64
- Loss Function: Cross-Entropy Loss
- Number of Epochs: 10
- Dataset: Skin Cancer 9-Class Dataset
Metrics (Validation Set)
| Class | Precision | Recall | F1-Score |
|---|---|---|---|
| 0 | 0.78 | 0.44 | 0.56 |
| 1 | 0.50 | 0.56 | 0.53 |
| 2 | 0.67 | 0.38 | 0.48 |
| 3 | 0.50 | 0.25 | 0.33 |
| 4 | 0.44 | 0.88 | 0.58 |
| 5 | 0.81 | 0.81 | 0.81 |
| 6 | 0.43 | 1.00 | 0.60 |
| 7 | 0.40 | 0.38 | 0.39 |
| 8 | 0.50 | 0.67 | 0.57 |
- Overall Accuracy: 0.54
- Macro Average Precision: 0.56
- Macro Average Recall: 0.59
- Macro Average F1-Score: 0.54
- Weighted Average Precision: 0.58
- Weighted Average Recall: 0.54
- Weighted Average F1-Score: 0.53
License
This model is released under the MIT License.
This model has been pushed to the Hub using the PytorchModelHubMixin integration:
- Library: [More Information Needed]
- Docs: [More Information Needed]