Fullstack Customer Churn Prediction System

A complete end-to-end Machine Learning project that covers data preprocessing, model training, API development, and frontend integration. The system uses a Random Forest Classifier to predict customer churn based on tenure and billing features. The trained model is exposed via a FastAPI backend, containerized using Docker, and deployed on Render. A responsive React frontend allows users to input customer data and receive real-time predictions with probability visualization. The project demonstrates practical MLOps skills including model serving, API integration, and fullstack deployment.

Key Features

• Machine Learning Model: Built and trained a Random Forest Classifier for churn prediction.
• Data Preprocessing Pipeline: Implemented feature engineering and handling of missing values using Pandas.
• REST API Development: Designed a FastAPI backend to serve real-time predictions.
• Frontend Integration: Developed an interactive React UI for user input and visualization of churn probability.
• Containerization: Dockerized the backend to ensure consistent deployment environments.
• Cloud Deployment: Deployed backend on Render and connected it with the frontend for public access.
• Real-time Predictions: Users can input data and instantly receive churn probability and classification.
• API Communication: Integrated frontend and backend via HTTP requests with proper error handling.

Tech Stack

Screenshots