[pymvpa] Sensitivity analysis with GNB

[Yaroslav Halchenko]

On Fri, 12 Dec 2014, Thomas Nickson wrote:

>    On Fri, Dec 12, 2014 at 1:42 PM, Yaroslav Halchenko
>    <debian at onerussian.com> wrote:

>      On Fri, 12 Dec 2014, Thomas Nickson wrote:

>      >A  A  I've trained a GNB classifier on a dataset with moderate success
>      and would
>      >A  A  like to look at the regions that the classifier finds most
>      interesting.

>      just use sphere_gnbsearchlightA  ? ;)

>    I wanted to do whole brain but I'm doing both anyway.

GNB is naive for a reason -- each voxel estimate is independent from any
other.  So even if you do full brain GNB and some kind of its
sensitivity -- those would be highly "univariate".  searchlight would
give you full brain coverage and "concert" effect of multiple voxels
within easch searchlight.

-- 
Yaroslav O. Halchenko, Ph.D.
http://neuro.debian.net http://www.pymvpa.org http://www.fail2ban.org
Research Scientist,            Psychological and Brain Sciences Dept.
Dartmouth College, 419 Moore Hall, Hinman Box 6207, Hanover, NH 03755
Phone: +1 (603) 646-9834                       Fax: +1 (603) 646-1419
WWW:   http://www.linkedin.com/in/yarik