Hi Everyone,<br><br>I searched for this in the list archives but didn't see anything regarding this so here goes...<br><br>Has there been any success using PyMVPA with fast event-related data? <br><br>Looking at the sensitivity map in AFNI (mapped back into original space with map2nifti), the high sensitivity areas look completely random. They seem to mainly be on the edges of grey matter/in spaces. I wonder if this is because, with fast event-related data, there is just too little information for the classifier, or is there something wrong with what I have been doing?<br>
<br>We <font class="PhorumMessage" color="#000000">used 3dDeconvolve's -stim_times_IM
option to estimate the amplitude of the response to each individual stimulus, plus this in PyMVPA:<br><br>from mvpa.suite import *<br><br>datasetfile = 'allstim.nii.gz'<br>attrfile = 'attr.txt'<br>attr = SampleAttributes(attrfile)<br>
ds = fmri_dataset(datasetfile, targets = attr.targets, chunks = attr.chunks, mask ='resampledmask.nii.gz')<br><br>zscore(ds, chunks_attr='chunks')<br>#no detrending...done in afni<br><br>clf = LinearCSVMC()<br>
fsel = SensitivityBasedFeatureSelection(OneWayAnova(), FractionTailSelector(0.08, mode='select', tail='upper'))<br>fclf = FeatureSelectionClassifier(clf, fsel)<br><br>sclf = SplitClassifier(fclf, enable_ca=['stats'])<br>
cv_sensana = sclf.get_sensitivity_analyzer()<br>sens = cv_sensana(ds)<br><br>sens_comb = sens.get_mapped(maxofabs_sample())<br>map2nifti(avgds, sens_comb).to_filename('test.nii.gz')<br><br>Anybody have any ideas or info...?<br>
Thanks,<br>Kim<br></font>