Activity: Create the algorithm portfolio and selection memo

Estimated active time: 45-75 minutes

Scenario

Use the synthetic learner-support dataset. The fictional platform wants to decide which model families deserve a place in an algorithm portfolio.

Task

Assemble the portfolio: bring every module's result into one table and write the selection memo that follows from it.

Required sections

1. Problem, decision, and the criteria set in Module 1.
2. One table with every algorithm, its validation score, its cost, and its interpretability note.
3. Both baselines, in the same table.
4. Your selection, with the trade-off you accepted stated explicitly.
5. What you would need to see before deploying it.
6. Limitation note covering data, evaluation, and stability.

Rules

• Every row must come from the same split. Re-run anything that does not.
• If the winner beats logistic regression by less than your Module 1 margin, recommend logistic regression and say why.
• The memo is the deliverable. A table with no recommendation is not a decision.

Output

Use the relevant learner-resource template from learner-resources/ or write a one-page memo with the same sections.