Segmentation Measures missing for Chord Recognition

[fdlm]

The segmentation measures (directed hamming distance) as defined by Harte 2010 (Sec. 8.3) and used in MIREX (MeanSeg, UnderSeg, OverSeg) are missing. I have Python code lying around for these, and am happy to contribute it. Before I start, however, what are the thoughts on this from the maintainers?

[craffel]

@ejhumphrey @mattmcvicar @bmcfee What do you think?

[mattmcvicar]

I think additional metrics can only be a good thing 👍

…

On Fri, Sep 29, 2017 at 4:35 PM, Colin Raffel ***@***.***> wrote: @ejhumphrey <https://github.com/ejhumphrey> @mattmcvicar <https://github.com/mattmcvicar> @bmcfee <https://github.com/bmcfee> What do you think? — You are receiving this because you were mentioned. Reply to this email directly, view it on GitHub <#262 (comment)>, or mute the thread <https://github.com/notifications/unsubscribe-auth/ACfGsmoxW--qUUkhGWY3II494b7xq91rks5snQ5IgaJpZM4Poalc> .

-- Matt McVicar Machine Learning Researcher Jukedeck contact: mattjamesmcvicar@gmail.com blog/website: www.mattmcvicar.com

[bmcfee]

SGTM. 👍