Hi All,
I'm here at PyCon listening to Olivier Grisel's excellent Scikit-Learn
tutorial. I'm reminded that PLSRegression does not have a VIP (i.e.,
feature importance) attribute.
Is there a particular reason? It is fairly straight forward to calculate
and it would be very handy.
I'm not a programmer, but could put together a draft, if someone is willing
to work with me to actually create a decent pull request.
For those not familiar with PLS VIP, Section 3.1.3 of this article shows
the math:
http://www.researchgate.net/profile/Lars_Snipen/publication/233748490_A_review_of_variable_selection_methods_in_Partial_Least_Squares_Regression/links/0912f50bca8a9cc896000000.pdf
Here's an example implementation in matlab:
http://www.mathworks.com/matlabcentral/fileexchange/47767-libpls-1-95-zip/content/libPLS_1.95/vip.m
Thoughts?
Thanks,
Walter
------------------------------------------------------------------------------
BPM Camp - Free Virtual Workshop May 6th at 10am PDT/1PM EDT
Develop your own process in accordance with the BPMN 2 standard
Learn Process modeling best practices with Bonita BPM through live exercises
http://www.bonitasoft.com/be-part-of-it/events/bpm-camp-virtual- event?utm_
source=Sourceforge_BPM_Camp_5_6_15&utm_medium=email&utm_campaign=VA_SF
_______________________________________________
Scikit-learn-general mailing list
Scikit-learn-general@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/scikit-learn-general
