Inpsect and try to interpret your scikit-learn machine-learning models
summary
Hands‑on tutorial that walks through notebooks and code showing how to scikit-learn models are inspected not summarised: model‑specific vs model‑agnostic methods, partial dependence and permutation importance, plus SHAP caveats. Concrete demos use pipelines, ColumnTransformer, OneHotEncoder, logistic/ridge/lasso, five‑fold CV and report a mean absolute error ≈2.14; shows lasso instability with correlated age/experience, explains scaling and data‑leakage traps, and notes practical binder issues (numpy 1.22 vs 1.21).
AI-written summary · about
chapters (11)
- 0:00Start
- 13:06Evaluate Your Model
- 14:48Causal Graph
- 20:44Feature Importances
- 24:50Cross-Validate
- 32:51Data Leakage
- 39:03Marginal Representation
- 52:35The Mean Absolute Error
- 1:05:12Cross Validation
- 1:12:59How Do You Mitigate the Frequency for Categories When a Category Is Uncommon
- 1:19:50Lasso Cv
related talks
01h 33mUnderstanding and diagnosing your machine-learning models
15mInterpretable Machine Learning with R
33mUsing Scikit-Learn's interface for turning Spaghetti Data Science into Maintainable Software
27mPyTesting the Limits of Machine Learning
57mHow To Evaluate and Improve Machine Learning Models
34mInterpreting Machine Learning Models: Why and How!
01h 27mLearning scikit learn - An Introduction to Machine Learning in Python
43mVisualising Machine Learning Models