Metric Decision Tree

Navigate the landscape of evaluation metrics for classification, regression and ranking, including what to do when your classes are badly imbalanced.

It covers accuracy, precision, recall, F1, ROC-AUC, PR-AUC, MAE, RMSE, MAPE and the ranking metrics, with explicit handling of class imbalance. It is written for anyone about to report a number to a stakeholder who will act on it.

Every branch states the trade-off that decided it, so the recommendation you end on comes with the reasoning attached — something you can paste into a design note or defend in a review.