<div>Thank you guys,</div><div><br></div><div>Yes, I know that I can't predict using a portion of voxel. Let's say that I would like to train on full brain and test on a portion, putting out of ROI voxel intensity to zero. </div>
<div>I don't know if it makes sense conceptually because I would like to predict using a portion of features on a model built on multiple features. </div><div><br></div><div>Probably could be an sensitivity measure, e.g. building a classifier to predict from flag the country for example Liberarian flag and stars and stripes, If I use features from stripes part of the flag (common in both of the flags) the classifier isn't able to classify - well, using a feature selection probably those features were discarded - but using "stars" part as ROI the classifier identifies the flag, and so I will know where the classifier is more sensitive! (Hope I explained it clear).</div>
<div><br></div><div>I don't know if there is still a theoretical problem.</div><div><br></div><div>Thank you</div><div>R</div><div><br></div><div>PS: If I could help you to complete...</div><br><div class="gmail_quote">
On 31 January 2013 15:38, Yaroslav Halchenko <span dir="ltr"><<a href="mailto:debian@onerussian.com" target="_blank">debian@onerussian.com</a>></span> wrote:<br><blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex">
<div class="im"><br>
On Thu, 31 Jan 2013, Michael Hanke wrote:<br>
> As the subject "clearly" says, I would like to train a classifier using<br>
> classical methods for example using all voxels and then try to predict<br>
> using only a portion of ROI like a searchlight.<br>
> I've tried to do this using the classifier as data measure in<br>
> searchlight class, but obviously the features of the classifier are more<br>
> than those used in the searchlight.<br>
> How can I do?<br>
<br>
> This is less of a technical question, but more of a conceptual one. You<br>
> can't train an algorithm on one set of features and then run it on a<br>
> different one with a different number of features.<br>
> you need to have equally structured input in both training and testing<br>
> stage. This could be done (think e.g. PCA projection), but whether it<br>
> makes sense in you context is impossible to tell at this point.<br>
<br>
</div>indeed! But I guess it could be stretched to become a "technically<br>
legit" one in the case of kernel-based classifiers, where optimization<br>
and decision is done based on values within the kernel... theoretically<br>
it should be possible to get the solution for one kernel (estimated on<br>
full data) and then apply to another (estimated on subset of the<br>
features)... not sure how legit it would be, but at least possible<br>
technically -- I guess could become an improved "sensitivity"<br>
measure to complement existing ones ;-)<br>
<span class="HOEnZb"><font color="#888888"><br>
--<br>
Yaroslav O. Halchenko<br>
Postdoctoral Fellow, Department of Psychological and Brain Sciences<br>
Dartmouth College, 419 Moore Hall, Hinman Box 6207, Hanover, NH 03755<br>
Phone: <a href="tel:%2B1%20%28603%29%20646-9834" value="+16036469834">+1 (603) 646-9834</a> Fax: <a href="tel:%2B1%20%28603%29%20646-1419" value="+16036461419">+1 (603) 646-1419</a><br>
WWW: <a href="http://www.linkedin.com/in/yarik" target="_blank">http://www.linkedin.com/in/yarik</a><br>
<br>
_______________________________________________<br>
Pkg-ExpPsy-PyMVPA mailing list<br>
<a href="mailto:Pkg-ExpPsy-PyMVPA@lists.alioth.debian.org">Pkg-ExpPsy-PyMVPA@lists.alioth.debian.org</a><br>
<a href="http://lists.alioth.debian.org/cgi-bin/mailman/listinfo/pkg-exppsy-pymvpa" target="_blank">http://lists.alioth.debian.org/cgi-bin/mailman/listinfo/pkg-exppsy-pymvpa</a><br>
</font></span></blockquote></div><br>