Customer_Churn_rate

End-to-end customer churn prediction system using TensorFlow, Keras, and Streamlit. Features data preprocessing, ANN training with early stopping, and an interactive UI for real-time churn prediction and business insights.

🧠 Customer Churn Prediction System (ANN + Streamlit)

An end-to-end Deep Learning project built to predict telecom customer churn using TensorFlow, Keras, and Streamlit.
This project demonstrates the complete machine learning lifecycle — from data preprocessing and model training to deployment with an intuitive UI.

---

🚀 Features

• Built an Artificial Neural Network (ANN) from scratch using TensorFlow & Keras
• Achieved ~83% accuracy with dropout regularization and early stopping
• Data preprocessing pipeline: encoding, scaling, and missing value handling
• Streamlit web app for real-time churn prediction
• Model and scaler saved for easy deployment (.h5 and .pkl files)

---

🧩 Tech Stack

• Language: Python
• Libraries: TensorFlow, Keras, Pandas, NumPy, Scikit-learn, Matplotlib, Streamlit
• IDE/Tools: VS Code, Jupyter Notebook

---

📁 Project Structure