Car Price Prediction with Machine Learning

An end-to-end machine learning regression pipeline that predicts car prices from vehicle specifications, structured to answer one question: does a manual ML workflow outperform an automated one? The Kaggle CarPrice dataset is processed through three phases: manual feature engineering with hand-picked models (Linear Regression, Random Forest, Gradient Boosting), automated model selection via PyCaret, and interpretation through both white-box methods (Linear Regression coefficients, Decision Tree splits) and black-box methods (SHAP, LIME).

Built locally in VS Code as part of Masterschool’s AI Enhanced Productivity project series, which explores how automation tools shape analytical productivity across different ML problem types.