Rent Price Prediction (End-to-End ML + MLOps)

An end-to-end machine learning system for predicting apartment rent prices using structured housing data. The project demonstrates production-grade ML engineering, including data preprocessing, model training, hyperparameter tuning, experiment tracking with MLflow, and reusable inference pipelines.

Problem Statement

Accurately predicting rental prices is challenging due to: Mixed numerical & categorical features Location-driven price variation High variance across property types

This project addresses these challenges using a robust ML pipeline with cross-validated model selection and experiment tracking.

Key Features

End-to-end ML pipeline (not a notebook) Automated model comparison with cross-validation Hyperparameter tuning (GridSearchCV) Experiment tracking with MLflow Model versioning & reproducibility Saved preprocessing + model pipeline Modular, production-ready codebase