California Housing Price Prediction

A Jupyter Notebook that trains an XGBoost regression model to estimate median house values from the California Housing dataset.

Workflow

• Load the dataset from scikit-learn
• Explore feature distributions and correlations
• Split the data into training and test sets
• Train an XGBRegressor
• Evaluate predictions with R-squared and mean absolute error
• Compare predicted and actual values visually

Dataset

The California Housing dataset contains eight numerical features:

MedInc HouseAge AveRooms AveBedrms Population AveOccup Latitude Longitude

The target is the median house value for each California district.

Run

pip install pandas matplotlib seaborn scikit-learn xgboost jupyter
jupyter notebook PROJECT3_HOUSEPRIZE.ipynb

Run the notebook cells in order. The dataset is downloaded through scikit-learn.

Built with

Python pandas scikit-learn XGBoost Matplotlib Seaborn