[pymvpa] Train and test on different classes from a dataset

Yaroslav Halchenko debian at onerussian.com
Tue Feb 5 13:52:32 UTC 2013



On Tue, 05 Feb 2013, MS Al-Rawi wrote:

> Interesting discussion, 

> I have similar expectations to those of Michael, and I find this 4% hard to justify. Could this be due to the randomization algorithm used to permute the training set? Non-perfect randomization could be one reason for this difference. 

since we had only a limited set of permutations/samples -- there is an
error of estimation as well.  But my concern was primarily triggered
by widening of chance distribution with growing SNR.

-- 
Yaroslav O. Halchenko
Postdoctoral Fellow,   Department of Psychological and Brain Sciences
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        



More information about the Pkg-ExpPsy-PyMVPA mailing list