Skip to content

Latest commit

 

History

History

churn_model

README.md

📞 Customer Churn Prediction Model

🎯 Overview

This project demonstrates a Customer Churn Prediction Model using XGBoost and the JFrogML platform. It showcases multiple deployment strategies and training approaches for production-ready churn prediction systems.

📋 Prerequisites

Before starting, ensure you have:


🚀 Quick Start

☁️ ML App Code → Build → Deploy

┌─────────────────┐    ┌─────────────────┐    ┌─────────────────┐
│   💻 Local ML   │ -> │   🏗️ Build      │ -> │   🚀 Deploy     │
│   App Code      │    │    Process      │    │   ML Serving    │
│   (or GitHub)   │    │(w/ Training Job)│    │   API Endpoint  │
└─────────────────┘    └─────────────────┘    └─────────────────┘
     Local IDE              FrogML CLI           FrogML CLI

Complete workflow: 🚀 Remote Training & Deployment Guide

Best for: Standardized, replicable, production-ready workflows with integrated training and serving


📁 Project Structure

churn_model_new/
├── main/                       # Main directory containing core code
│   ├── __init__.py             # Python package initialization
│   ├── model.py                # FrogMLModel with churn prediction logic
│   ├── data.csv                # Training dataset
│   └── conda.yml               # Environment dependencies
├── tests/                      # Integration tests
│   └── it/
│       └── test_churn.py       # Integration test for churn model
├── test_model_code_locally.py  # Script to test the model locally
├── test_live_endpoint.py       # Script to test live deployment endpoint
├── test_batch_endpoint.py      # Script to test batch inference endpoint
├── README.md                   # This documentation
└── remote-training-and-deployment.md  # Deployment guide

🔗 Related Resources

📚 Next Steps

  1. Choose your deployment path from the guides above
  2. Follow the step-by-step instructions in your chosen guide
  3. Customize the model for your specific churn prediction needs
  4. Scale up with larger datasets and more complex feature engineering

Ready to get started? Pick a guide above and begin your churn prediction journey! 🚀