Micro Free Course — 100% Free Learning
All lessons in this module are free to learn. Sign in with Google to save your progress.
M2 — Boosting + Calibrated Evaluation
RF vs XGB/LGBM case, PR-AUC/Brier/threshold article, boosting quiz, campaign classifier + error-slice submissions.
Module Content
RF vs XGB/LGBM Case Walkthrough
Outcome: compare RF vs boosting on campaign data. Curated video (6:04, verified 2026-09-24 via yt-dlp): Random Forest vs XGBoost: Which One Should You Actually Use? (IvyProSchool) Watch: https://www.youtube.com/watch?v=HKtHezjnncQ Notes: Bagging vs boosting framing; RF vs XGB use cases, performance + data-handling comparison. Apply to campaign-response data: compare, then calibrate before deciding. Key moments: 0:00 Intro; 0:33 Bagging vs Boosting; 1:16 RF use case; 1:49 XGB use case; 2:26 Performance; 3:02 Data handling; 4:15 Final comparison; 5:31 Best approach.
Calibration: PR-AUC/Brier + Threshold-to-Decision
Cost-matrix thresholds; PR-AUC + Brier. Cheatsheet playbook; WE W3-W4.
Campaign Classifier w/ Calibrated Thresholds
Calibrated classifier + threshold memo. Temporal trap; 100-pt rubric inside.
Quiz: Boosting + Eval
10 scenario MCQs (PR-AUC-vs-ROC, Brier, calibration). Pass >= 80.
Error-Analysis + Slice Writeup
3 error slices + fix proposals. Slice template inside.