fujelhrx/california-house-model ? reverse-engineered prompt

Build me a California house price prediction project that can take housing data and estimate the price of a home from things like location, median income, rooms, and other basic features.

I want it to include a notebook that walks through the whole process, from cleaning the data and handling missing values to exploring the dataset, training a few regression models, and comparing how well they do. Please also add a simple Python script version so I can run the training outside the notebook.

If it makes sense, include a reusable preprocessing pipeline and a couple of custom transformers for things like scaling and feature engineering. Keep the code easy to follow, with clear comments and a straightforward way to run everything from start to finish. Use the California housing dataset, and if you need to check current scikit learn docs online for best practices, go ahead.