<div class="gmail_quote"><div>Hi Michael,<br> <br></div><blockquote class="gmail_quote" style="border-left: 1px solid rgb(204, 204, 204); margin: 0pt 0pt 0pt 0.8ex; padding-left: 1ex;"><br>
> smap = LinearNuSVMC().getSensitivityAnalyzer(transformer=N.abs)<br>
> results = smap(data)<br>
<br>
This looks all fine to me and it should work -- does it?</blockquote><div><br>Works great, a volume with an SVM weight in every brain voxel for every included timepoint is the result of data.map2Nifti(results)<br></div><blockquote class="gmail_quote" style="border-left: 1px solid rgb(204, 204, 204); margin: 0pt 0pt 0pt 0.8ex; padding-left: 1ex;">
<br>
However, in an earlier email you wrote:<br>
> > I just switched from using the transformWithBoxcar function in conjunction<br>
> > with fsl evfiles to make big files of concatenated samples with labels and<br>
> > chunks specified in an outside text file to using the ERNiftiDataset class<br>
> > (which is really nice!) with lists of Event objects and am having trouble<br>
> > running a searchlight on it. I get the error<br>
> ><br>
> > ValueError: Searchlight only works with MappedDatasets that has metric<br>
> > assigned.<br>
> ><br>
> > Is it not possible to use a searchlight when samples have multiple time<br>
> > points? I had no trouble getting svm weights for the same ERNiftiDatasets,<br>
> > so I am a little confused<br>
<br>
With searchlights the situation is a little different. There are several<br>
possibilities to run a searchlight.<br>
<br>
1. Pure-spatial searchlight that will run the analysis in a<br>
sphere-shaped dataset spanning all volumes in time.<br>
<br>
2. Spatio-temporal searchlight. In addition to 1. also limiting the<br>
temporal window of the dataset.<br>
<br>
I am afraid that both scenarios are not well tested (probably nobody<br>
used that before), and I suspect that they also involve a bit of coding<br>
in PyMVPA itself. Especially 2. would make the runtime of a searchlight<br>
analysis even more ridiculus than it already is.</blockquote><div><br>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..<br>
<br>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?<br></div><br>Thanks again,<br>Matt<br></div><br>