About

This canvas builds an end-to-end customer churn prediction pipeline for a telecom company, starting with data exploration and preprocessing of the Telco dataset, training a logistic regression model, persisting it for reuse, and finally demonstrating inference capabilities with both sample predictions and a deployment-ready API function. The workflow chains five Python blocks that load data, clean and encode features, train and evaluate the model, save it for production, and test both inference methods to validate the churn prediction system is ready for deployment.