Hello all Totally new beginner here and just trying to read nifti files.
Don't know if the issue has been raised before (sorry if it's been), but I have a series of 3d nifti files (>1000) to be read into the dataset and I ran into three problems: (1) 'Too many open files' error that arises from nibabel.load function (cf http://mail.scipy.org/pipermail/nipy-devel/2011-February/005428.html) (2) Wrong voxel dimensions ('voxel_eldim') (3) Wrong time coordinates ('time_coords') I checked the source code and found out the fmri_dataset relies on get_zooms(), which returns an 3-element array for 3D files and 4-element array for 4D files. It took me a bit of time to find out the problem. To save more troubles by other beginners like me, I suggest that fmri_dataset (a) forbid loading series of 3D files (which can be converted to 4D easily in SPM anyway), and (b) allow the optional parameter of specifying TR length. Thanks! Hang-yee
_______________________________________________ Pkg-ExpPsy-PyMVPA mailing list [email protected] http://lists.alioth.debian.org/cgi-bin/mailman/listinfo/pkg-exppsy-pymvpa
