Oriol Nieto

Thanks! Yeah, C-NMF can be demanding. Probably the Checkerboard kernel (`foote`) is the fastest at finding boundaries. And 2D-FMC (`fmc2d`) the fastest at labeling the segments.

This module provides several functions to plot the results of segmentation algorithms (see the [demo notebook](https://github.com/urinieto/msaf/blob/master/examples/Run%20MSAF.ipynb)). I agree that you shouldn't need to have matplotlib to run this as long...

Thanks for reporting this. I started playing around with this, and seems like there might be some small differences in the PCP feature computation maybe due to `numpy`/`scipy` that may...

That's what I thought too, but `tag_open` still requires time and duration with the current API (or perhaps I'm doing something wrong?). Ideally, I'd like to do something like this:...

It still feels a bit cumbersome to _having_ to add these values at the observation level (and have them repeated along the whole raw JSON annotation), when you just want...

Sounds good. Thanks for your help!

Nice, I like this idea. ISMIR always sparking great conversations!

Could we simply add a new identifier field in the annotator dictionary that is basically a unique hash produced by the annotator name, email, affiliation, etc? On Thu, Aug 18,...

My (embarrassingly belated) two cents: I think having the identifiers as a sandbox in the JAMS schema already allows for this kind of `URI`s insertion. We could add some sort...

I will be able to help, but after I write my msaf LBD paper :/ On Mon, Sep 21, 2015 at 11:56 AM, Justin Salamon [email protected] wrote: > I probably...