Pick the Right Model the Right Way

Fix an evaluation setup that would ship the wrong classical model.

A beginner AI Engineering challenge worth 3 points. Solve it hands-on in a real AI Engineering environment in your browser - no local setup, no fake shells.

The Challenge

Scenario: A breast-cancer classifier is evaluated with a 2% non-stratified test split and models are ranked by raw accuracy alone - both are ways to fool yourself into shipping a bad model.

Fix the train/test split, compare a dummy baseline against a scaled logistic regression, report accuracy, precision, recall and F1, and select the model by recall (missing a real cancer case is far worse than a false alarm). Run check.