Build the full pipeline (impute+indicator → scale → one-hot → logistic), fit it…

Machine Learning · from Pipelines and leakage

PipelineColumnTransformerleakagenamed_stepsset_paramsjoblib

Putting the kettle on…

Starting up…

Exercise 2

Build the full pipeline (impute+indicator → scale → one-hot → logistic), fit it, and return [list(named_steps), n_features_in_ of the classifier].

your answer
stuck? the lesson behind it
Pipelines and leakage
A 0.9989 AUC is a bug report, and a Pipeline is how you stop writing them