[pymvpa] Pkg-ExpPsy-PyMVPA Digest, Vol 17, Issue 3

Matthew Cieslak mattcieslak at gmail.com
Thu Jul 9 16:21:25 UTC 2009

Hi Michael,

> >     smap = LinearNuSVMC().getSensitivityAnalyzer(transformer=N.abs)
> >     results = smap(data)
> This looks all fine to me and it should work -- does it?

Works great, a volume with an SVM weight in every brain voxel for every
included timepoint is the result of data.map2Nifti(results)

> However, in an earlier email you wrote:
> > > I just switched from using the transformWithBoxcar function in
> conjunction
> > > with fsl evfiles to make big files of concatenated samples with labels
> and
> > > chunks specified in an outside text file to using the ERNiftiDataset
> class
> > > (which is really nice!) with lists of Event objects and am having
> trouble
> > > running a searchlight on it.  I get the error
> > >
> > > ValueError: Searchlight only works with MappedDatasets that has metric
> > > assigned.
> > >
> > > Is it not possible to use a searchlight when samples have multiple time
> > > points? I had no trouble getting svm weights for the same
> ERNiftiDatasets,
> > > so I am a little confused
> With searchlights the situation is a little different. There are several
> possibilities to run a searchlight.
> 1. Pure-spatial searchlight that will run the analysis in a
>   sphere-shaped dataset spanning all volumes in time.
> 2. Spatio-temporal searchlight. In addition to 1. also limiting the
>   temporal window of the dataset.
> I am afraid that both scenarios are not well tested (probably nobody
> used that before), and I suspect that they also involve a bit of coding
> in PyMVPA itself. Especially 2. would make the runtime of a searchlight
> analysis even more ridiculus than it already is.

I was considering a searchlight where a sphere is centered in space at
(x,y,z) and an array of features is put together by taking fmri data in this
sphere at time t,t+1,..,t+N (with N small enough that it doesn't overlap
with another event) and pairing it with the label for the stimulus at time
t: Then running a cross-validation and mapping the transfer error back to
(x,y,z). This may not be the best way to go about including temporal
information in a searchlight, but I have a rapid event related paradigm and
I'm not sure what the best way to incorporate temporal information in the
feature set would be. I would really appreciate your thoughts on this..

Also, speaking of the long time it takes to run a searchlight, is there a
preferred method for doing analyses on the surface in PyMVPA?

Thanks again,
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.alioth.debian.org/pipermail/pkg-exppsy-pymvpa/attachments/20090709/9fcd1eb1/attachment.htm>

More information about the Pkg-ExpPsy-PyMVPA mailing list