How tahanadeem112299/customer_churn_model works
This repository is a small telecom churn prediction application with a separate model-training script and a Streamlit inference interface. Training reads the CSV, fits a preprocessing-and-classification pipeline, and saves a model artifact; the interface collects customer attributes, loads that artifact, and displays a churn probability. The excerpts show these file-level relationships directly. The ZIP archive and helper script are not part of the demonstrated runtime workflow.
Main entry points: testing.py Prediction interface · model_training.py Model training · telecom_model.pkl Trained model · training_data.csv Training data
Components 6
Model training
- Training data
training_data.csv - Model training
model_training.py - Trained model
telecom_model.pkl
Churn inference
- Prediction interface
testing.py
Other components
- Operator
- Customer user
Connections 5
- Model training → Training data: reads
model_training.py - Model training → Trained model: saves
model_training.py - Customer user → Prediction interface: enters attributes
testing.py - Prediction interface → Trained model: loads and predicts
testing.py - Prediction interface → Customer user: displays result
testing.py