https://github.com/nivindulakshitha/skin-cancer-type

Skin Cancer Classification model with Convolutional Neural Network (CNN)
https://github.com/nivindulakshitha/skin-cancer-type

machine-learning python

Last synced: 4 months ago
JSON representation

Skin Cancer Classification model with Convolutional Neural Network (CNN)

• Host: GitHub
• URL: https://github.com/nivindulakshitha/skin-cancer-type
• Owner: nivindulakshitha
• License: cc0-1.0
• Created: 2024-03-06T06:12:05.000Z (over 1 year ago)
• Default Branch: Master
• Last Pushed: 2024-03-06T06:20:38.000Z (over 1 year ago)
• Last Synced: 2025-01-01T18:26:27.359Z (6 months ago)
• Topics: machine-learning, python
• Language: Jupyter Notebook
• Homepage:
• Size: 11.7 KB
• Stars: 2
• Watchers: 1
• Forks: 0
• Open Issues: 0
• Metadata Files:
  • Readme: README.md
  • License: LICENSE

Awesome Lists containing this project

README

        
## Skin Cancer Classification - Convolutional Neural Network (CNN)

This project implements a Convolutional Neural Network (CNN) to classify skin cancer images as malignant or benign. 

### Project Overview

* **Goal:** Develop a CNN model to accurately classify skin cancer images.

* **Data:** Used a dataset of skin cancer images from [Kaggle](/kaggle/input/skin-cancer-malignant-vs-benign). 

* **Model:** Implemented a CNN architecture with convolutional and dense layers.

* **Evaluation:** Trained the model on a portion of the data and evaluated its performance on unseen images.

### Technical Details

* **Libraries:** TensorFlow, Keras, NumPy, Matplotlib

* **Preprocessing:** Rescaled images, applied random shearing/zooming/flipping (training only).

* **Model Architecture:**

    * Sequential CNN with convolutional and pooling layers for feature extraction.

    * Dense layers for classification.

* **Training:** Used Adam optimizer, binary cross-entropy loss, and accuracy metric.

### Results

* Trained the model for X epochs. (Replace with actual number)

* Achieved an accuracy of Y% on the test set. (Replace with actual accuracy)

### Future Improvements

* Explore more advanced CNN architectures (e.g., deeper networks, transfer learning).

* Experiment with different hyperparameters (learning rate, batch size).

* Include data augmentation techniques for a more robust model.