EfficientNet Skin Cancer Classification Model

Overview

This project uses Keras to implement an EfficientNet-B0 model for classifying skin cancer images as either benign or malignant.

Key Features

• Utilizes EfficientNet-B0 as the base model
• Fine-tunes the model on a skin cancer dataset
• Achieves a test accuracy of 85.02%

Usage

1. Download the skin cancer dataset from Kaggle.
2. Run the provided Python script to train and fine-tune the EfficientNet model.
3. Evaluate the model on the test dataset.

Results

• Test Loss: 34.56%
• Test Accuracy: 85.02269%

Model Details

• Base Model: EfficientNet-B0
• Fine-tuned Layers: 20
• Epochs: 30
• Batch Size: 1024

Code

The code is written in Python and utilizes the Keras API. It includes data loading, model building, fine-tuning, and evaluation.