https://github.com/swish78/diabeto-predictor

https://github.com/swish78/diabeto-predictor

Last synced: over 1 year ago
JSON representation

• Host: GitHub
• URL: https://github.com/swish78/diabeto-predictor
• Owner: Swish78
• License: mit
• Created: 2024-10-21T18:53:45.000Z (almost 2 years ago)
• Default Branch: main
• Last Pushed: 2024-10-21T19:15:23.000Z (almost 2 years ago)
• Last Synced: 2024-10-22T11:35:03.545Z (almost 2 years ago)
• Language: JavaScript
• Size: 253 KB
• Stars: 0
• Watchers: 1
• Forks: 0
• Open Issues: 0
• Metadata Files:
  • Readme: README.md
  • License: LICENSE

Awesome Lists containing this project

README

          
# Diabeto-Predictor

Diabeto-Predictor is a Flask-based API with a React frontend that predicts the likelihood of diabetes based on various input features. It provides data visualizations and predictions powered by a machine learning model. The backend is deployed on AWS EC2, with data stored on AWS RDS (MySQL) and static files hosted on AWS S3.

## Features

- **Diabetes Prediction API**: Get predictions on diabetes risk based on input features.

- **Data Visualization**: Explore visualizations like histograms, scatter plots, and box plots.

- **Machine Learning Model**: Powered by a RandomForest and GradientBoosting voting classifier.

- **AWS Integration**:

  - EC2 for backend deployment.

  - RDS (MySQL) for database management.

  - S3 for frontend hosting.

## Technology Stack

- **Backend**: Flask, Pandas, Joblib

- **Frontend**: React, Tailwind CSS

- **Machine Learning**: Scikit-learn, Imbalanced-learn (SMOTE), Voting Classifier (RandomForest, GradientBoosting)

- **Data Visualization**: Plotly, Seaborn, Matplotlib

- **AWS**:

  - EC2 for Flask server

  - RDS (MySQL) for database

  - S3 for static assets and frontend

## Prerequisites

- Python 3.x

- Node.js and npm

- AWS account with access to EC2, RDS, and S3 services

- MySQL installed locally or accessible via RDS

## Model Pipeline

The machine learning model is trained on the following features:

- **Numerical**: Age, BMI, FBS, HbA1c

- **Categorical**: Gender, Blood Pressure, Family History of Diabetes, Smoking, Diet, Exercise

The model uses a voting classifier composed of:

- RandomForestClassifier

- GradientBoostingClassifier

## License

This project is licensed under the MIT License.