Build a leakage-free scikit-learn pipeline with mixed feature types
Interview insights are lockedSee how often this is asked and which companies asked it with Buncho Pro.A dataset has numeric columns age, income (with missing values) and categorical columns city, plan (with unseen categories at prediction time).
Build a single estimator that imputes, scales, one-hot encodes and fits a logistic regression, and evaluate it with 5-fold stratified cross-validation on an imbalanced target. Explain why a Pipeline matters.