To follow up on this question - It was a pre-preocessing glitch (wrong
number of files used to create 4D) and inspection of the log files
revealed lots of "wrong sized matrix" error messages.
Oops!
Christian Lambert wrote:
> Hi all,
>
> I've been running a bedpostx batch over christmas, and am now finding
> one subject slightly problematic. Essentially BedpostX has run
> producing data_slice_00->79 and nodif_mask_slice 00->79 in the host
> directory, and in the created bedpostx directory I have a additional
> directory "diff_slices" containing dataslice 00->79. The problem comes
> in that there are no merged files. I have attempted to re-run it (as
> all the slices are essentially pre-processed) but with no joy. The
> error message that appears in the post-processing stage is:
>
> Error: /home/fsl/fsl/bin/bedpostx Line 220 14432 Terminated
> ${subdir}.bedpostx/monitor
>
> To clarify I'm using FSL_vm5_64 with 2 CPUs dedicated and ~10GB RAM.
> The subjects prior and after have processed without a problem.
>
> Any suggestions much appreciated. Is there a command line to try this
> manually?
>
> Thanks,
>
> - Chris
>
--
Dr Christian Lambert,
Wellcome Trust Clinical Research Fellow,
Wellcome Department of Imaging Neuroscience,
Institute of Neurology,
12 Queen Square,
London, WC1N 3BG.
|