AI & ML / README.md

Model evaluation

Updated 1 min read index source

Model evaluation

Metrics are where modelling meets the business. Most questions here are really “do you understand what your number means to someone who isn’t an ML engineer”.

# File The question it answers
01 Confusion matrix and accuracy the four cells everything derives from; why accuracy misleads
02 Precision, recall and F1 the trade-off, and that it’s a threshold choice not a model property
03 ROC-AUC vs PR-AUC the discriminator question — when ROC-AUC flatters a bad model
04 Regression metrics MAE vs RMSE, why MAPE is a trap, asymmetric costs
05 Calibration when the probability has to be a real probability
06 Choosing the metric working backwards from the decision to the metric

The two ideas that carry most of this

The threshold is not part of the model. Train once, choose the operating point separately from a business constraint — review capacity, a precision floor, or a cost ratio. Leaving it at 0.5 is a decision by default.

ROC-AUC hides false positives under class imbalance. FPR divides by the total actual negatives, so a huge easy-negative pool absorbs enormous numbers of false alarms with barely any movement. PR-AUC has no true-negative term and reflects what the person reviewing the alerts experiences.

Contents 6