Brian McFee

This is actually a more subtle point than it may seem. However, I'll preface the following by agreeing that the documentation examples should be audited and corrected to include sampling...

Adding a tag to 0.10.1 for the docstring example audit. For the bigger picture, I think it deserves some more time and thought. However, I do lean toward the assumption...

> #1573 And please reconsider this too! What's to reconsider? I have not seen much (anything?) in the way of counter-arguments to [this post](https://github.com/librosa/librosa/issues/1573#issuecomment-1248107136). As I said before, if people...

> It would just be nice to be able to use librosa in an "explicit" mode as well that has no fallback and throws an exception when forgetting to pass-in...

Ok, after a bunch of back and forth with ye olde chatbot, I think I have a docstring scraper that more or less does this job of identifying docstring examples...

This is related to #1364 I think. Maybe @lostanlen has thoughts? Meanwhile, I think you could probably pull this off by passing in explicit `x_coords=` to specshow, though you might...

Closing this out as a duplicate of #1364, but noting here also that this is blocked upstream by matplotlib and not currently feasible.

Taking a first whack at this, failures are plentiful: ``` librosa/display.py:152: error: Item "None" of "Axis | _DummyAxis | _AxisWrapper | None" has no attribute "get_data_interval" [union-attr] _, dmax =...

Sure, seems like a worthwhile endeavor. Thanks! We may need to be a bit strategic about prioritizing which functions to test though. The package is entirely Python (no c/cython extensions),...

I believe this is only an issue for tempogram display, not the feature calculation. But yeah, we should expand the api to expose the min and max tempo parameters.