Housing Price Prediction System
An end-to-end machine learning pipeline for predicting housing prices, built with ZenML for orchestration and MLflow for experiment tracking and model management.
Problem Statement
Predicting housing prices accurately requires handling complex feature interactions, missing data, and regional variations. Many ML projects lack reproducibility and proper experiment tracking, making it difficult to iterate on models and deploy reliably.
Methodology
Designed an end-to-end ML pipeline using ZenML for orchestration and MLflow for experiment tracking. Implemented data ingestion, cleaning, feature engineering, model training, and evaluation as modular pipeline steps. Tested multiple regression algorithms and used pytest for pipeline validation.
Results
Achieved a fully reproducible pipeline with automated experiment tracking. The modular architecture allows easy swapping of data sources and model algorithms. MLflow integration provides clear model versioning and comparison across experiments.