Training workspace
A dedicated notebook separates model development from the prediction interface.
House Price Prediction
Making housing data useful.
Explore the repository ↗Why this project?
Housing prices depend on multiple property characteristics. This project uses Ames Housing data to explore regression and presents a prediction interface for entering property features.
From input to output.
The project combines a training notebook, dataset files, a serialized model, and a Streamlit application. The repository describes Random Forest and XGBoost experimentation, feature engineering, cross-validation, and hyperparameter tuning.
01Property features
02Feature preparation
03Trained XGBoost model
04Price estimate
A simplified overview of the repository’s application flow.
Six input features
The application uses total square footage, lot area, year built, bedrooms above ground, total bathrooms, and garage capacity.
Interactive prediction
A Streamlit interface connects property inputs to a saved model, with deployment instructions in the repository.