// the find
wesm/pydata-book
Materials and IPython notebooks for "Python for Data Analysis" by Wes McKinney, published by O'Reilly Media
The companion notebooks for Wes McKinney's 'Python for Data Analysis', the book that essentially taught a generation of developers how to use pandas — which McKinney created. It covers NumPy, pandas, matplotlib, and data wrangling end to end. Aimed squarely at people new to data work in Python, not at practitioners who are already comfortable with the stack.
The datasets are bundled in the repo, so the notebooks actually run without hunting down files. The 3rd edition was updated for pandas 2.0, which is a real commitment given how much the API changed. The uv setup option is a nice touch — one command and you're in Jupyter with all deps pinned. Chapter coverage is logical: NumPy fundamentals before pandas, time series before modeling, which reflects how the knowledge actually builds.
This is a book companion, not a library — there's nothing to install or extend, so its value decays as the book itself ages. The notebooks have no tests, so cells go stale silently between pandas releases; running them top-to-bottom on a newer pandas will likely produce deprecation warnings or outright failures in spots. The modeling chapter (ch12) remains shallow by design — statsmodels and scikit-learn get a surface tour that leaves you needing other resources immediately. The repo hasn't been touched since late 2025, and pandas 3.x is already shipping breaking changes the notebooks don't account for.