Machine learning resources

Knowledge

• Course: scikit-learn MOOC by Inria End-to-end predictive modeling with tabular data, preprocessing, validation, and model interpretation. Use for: primary course reading and executable examples.
• Guide: scikit-learn, Getting Started Official estimator, preprocessing, pipeline, evaluation, and parameter-search workflow. Use for: API patterns used in lessons.
• Guide: scikit-learn, Cross-validation Official guidance on held-out evaluation, cross-validation strategies, groups, and time-aware splits. Use for: choosing trustworthy validation.
• Guide: scikit-learn, Model evaluation Official definitions and APIs for classification and regression metrics. Use for: selecting metrics from decision costs.
• Guide: scikit-learn, Pipelines and composite estimators Official patterns for joining preprocessing and models while avoiding leakage. Use for: reproducible mixed-column workflows.
• Course: Google Machine Learning Crash Course Practical explanations and exercises covering models, numerical and categorical data, and generalization. Use for: concept refreshers.
• Module: Google, Datasets, generalization, and overfitting Data quality, splitting, leakage risks, and generalization diagnostics. Use for: data and evaluation judgment.
• Dataset: UCI Bank Marketing Public tabular classification data about direct-marketing outcomes. Use for: extending exercises into a realistic project.
• Checklist: Google, Rules of Machine Learning Production-minded rules emphasizing metrics, simple first models, data pipelines, and monitoring. Use for: model handoff decisions.

Wisdom (Communities)

• Cross Validated Moderated statistics and machine-learning Q&A. Use for: validation design, metric choice, and interpretation questions with reproducible detail.
• scikit-learn Discussions Project community for usage patterns and library-specific questions. Use for: checking pipeline and estimator choices after consulting documentation.