// the find
amueller/scipy-2017-sklearn
Scipy 2017 scikit-learn tutorial by Alex Gramfort and Andreas Mueller
A two-part scikit-learn tutorial built for SciPy 2017, walking through supervised and unsupervised learning via numbered Jupyter notebooks. Aimed at people new to scikit-learn who want a structured, hands-on intro rather than scattered blog posts.
The notebook sequence is genuinely well-ordered, going from basic classification/regression through pipelines, grid search, and into less commonly taught topics like out-of-core text classification and anomaly detection in one coherent arc. Exercise solutions are included separately so you can attempt problems before peeking. The check_env.ipynb and fetch_data.py scripts are a nice touch for conference use, catching broken environments and missing datasets before people get stuck mid-tutorial.
Dead since 2017 — no commits in 9 years, and scikit-learn's API has moved substantially since then (cross_validation was folded into model_selection right around this release, among other breaking changes), so a chunk of these notebooks will throw import errors or deprecation failures on any modern sklearn install. Pinned to Python 2.7/3.4-3.6, both long past end-of-life. There's no content outside the notebooks themselves — no written explanations, just slides-as-code, so it only works well if you're following along live or willing to reconstruct context from code comments.