Dear all, I've another question about spatiotemporal analysis. In my dataset I have 6 runs per task with approximately a hundred volumes each task each run and running a classification on only 6 examples is very overfitting-oriented, how can I do to reduce the overfitting? I've tried with a leave 3 run out crossvalidation in order to have more testing runs and a very general model fitted only on 3 runs (I use SVM)? Are there some other strategies?
Thank you, Roberto On 19 July 2012 13:01, Roberto Guidotti <[email protected]> wrote: > Dear all, > > I'm working on spatiotemporal analysis of fMRI data, in this case the > number of features increases drammatically and I want to perform a Feature > Selection. > In the toolbox the Feature Selection of event related datasets is done > watching the spatiotemporal features variation across condition or checking > the single feature. I think it could be useful to select voxels that > temporally varies across experiment. > > It is possible to perform a sort of spatiotemporal feature selection, if > not yet implemented? Or I'm asking a question theoretically wrong? > > Thank you > Roberto. >
_______________________________________________ Pkg-ExpPsy-PyMVPA mailing list [email protected] http://lists.alioth.debian.org/cgi-bin/mailman/listinfo/pkg-exppsy-pymvpa
