Colin Raffel

This needs to be picked up again. I don't think anyone is planning on doing so.

I don't have a lot of bandwidth to discuss this now, but I can chime in with what I see as `mir_eval`'s philosophical approach. For the metrics we implement, we...

The key module was created to match the conventions used by MIREX. If there are other benchmarks that use a richer set of annotations, we welcome PRs to implement that...

Good idea!

Thanks for this discussion! > My student @ErikJohnsonCU and I have a working bss_eval_images pure python implementation which we could make ready for a PR into mir_eval. Great! A PR...

From a broader perspective, in `mir_eval` we try to include all metrics that are commonly used in the community,even if they are suboptimal. For example, the "goto" metric for beat...

> What I wanted to say was hence not really a matter of dropping mono evaluations, but rather of dropping this estimation of the best filter that is in bss_eval_sources....

I agree that it would make sense for Python bss-eval-ish metrics to all live in one place. I'm happy to have that be within `mir_eval`, and I don't see it...

I have similar thoughts as Brian on this.

Is this resolved?