Docs: multipitch.evaluation macro scores

[Laubeee]

I'm confused by the docs: https://github.com/craffel/mir_eval/blob/4d52264c90ec6f69d5dcfa013eba98d1479ae7af/mir_eval/multipitch.py#L348

in what sense is this the "macro" level? I thought this is what is called the 'micro-average', i.e. average across every single instance, whereas "macro" means per-class scores that are then averaged. Are the docs wrong here?

[craffel]

@rabitt