Machine Learning in Python with scikit-learn

Practical machine learning with scikit-learn — building predictive pipelines, cross-validation, model selection, and hyperparameter tuning across linear and tree-based models.

Download certificate (PDF) Verify

Skills covered

“Machine Learning in Python with scikit-learn” — a hands-on MOOC developed by the Inria Learning Lab (with scikit-learn @ La Fondation Inria and probabl) and delivered on FUN-MOOC. It works through the full predictive-modelling workflow in scikit-learn:

• Machine learning concepts — the predictive-modelling framing, train/test methodology, and what a model actually learns
• The predictive modelling pipeline — data exploration, numerical and categorical preprocessing, and chaining steps with Pipeline and ColumnTransformer to avoid data leakage
• Selecting the best model — overfitting vs underfitting, validation and learning curves, and the bias–variance trade-off
• Hyperparameter tuning — manual tuning plus automated grid-search and randomized-search
• Linear models — regression and classification with feature engineering and regularisation
• Decision trees — classification, regression, and tree hyperparameters
• Ensembles — bagging, random forests, AdaBoost, and gradient boosting
• Evaluating model performance — baselines, cross-validation strategies, and classification/regression metrics

Issued

INRIA (Inria Learning Lab) — 2023-01-19