What does Hamming loss measure that exact-match multilabel accuracy does not?
Classic MLLevel: IntermediateLabel-level error
What does Hamming loss measure that exact-match multilabel accuracy does not?
Hints
- Count label decisions across rows and columns.
- Compare one wrong label with all labels wrong.
Answer
It measures the fraction of individual label decisions that are wrong. It gives partial credit when some labels of a sample are correct, unlike all-or-nothing subset accuracy.
Source and licence: scikit-learn 1.7.2 — model evaluation · BSD-3-Clause · answer: Quantoros editorial