This project aims to develop a neural network model for classifying diabetic retinopathy using a dataset from the UCI Machine Learning Repository. The project addresses class imbalance and employs advanced techniques to enhance model performance, including hyperparameter tuning and early stopping.
- Data Preparation: Standardized and cleaned the dataset, handled class imbalance using SMOTE, and prepared data for training.
- Neural Network Development: Constructed and trained a neural network with early stopping to optimize performance.
- Performance Evaluation: Evaluated the model's accuracy and loss to ensure effectiveness.
The dataset is sourced from the UCI Machine Learning Repository and includes features related to diabetic retinopathy, with a target variable indicating the presence of the condition.
- TensorFlow: For building and training the neural network model.
- scikit-learn: For data preprocessing, splitting, and handling class imbalance.
- imblearn: For applying SMOTE to address class imbalance.
- Matplotlib & Seaborn: For data visualization.
To run this project, install the required Python libraries:
pip install numpy pandas matplotlib seaborn tensorflow scikit-learn imbalanced-learn ucimlrepo-
Clone the Repository:
git clone https://github.com/J3lly-Been/diabetic-retinopathy-classification.git cd diabetic-retinopathy-classification -
Prepare the Data and Train the Model:
Open the Jupyter notebook
diabetic-retinopathy-classification.ipynband follow the instructions within the notebook to load, clean, preprocess the dataset, build, train, and evaluate the neural network model.
- Test Accuracy: 78.9%
- Test Loss: 0.51
This project is licensed under the MIT License. See the LICENSE file for details.