Count what changes

Machine Learning · from Class imbalance

class_weightthreshold tuningSMOTEresamplingprecision/recall trade

Putting the kettle on…

Starting up…

Exercise 2

Count what changes. Return [missed_plain, missed_balanced] — the false negatives with and without class_weight="balanced", as ints.

your answer
stuck? the lesson behind it
Class imbalance
One keyword takes recall from 0.44 to 0.78 — and accuracy below the dummy