Hello:
I got this error message when I first upgraded to FSL4.0, and after
several experiments, I found out it is because I attempted to run the
tbss_2_reg command simultaneously on several terminals.
FSL4.0 requires SGE to run parallel. As for the older version, it is ok
to run the registration at several terminals at once by simply typing the
same command. Not ok with FSL4.0.
See if it applies to your case.
Yi-Shin
> Dear FSL users,
>
> While running the command tbss_2_reg -T, I received the following error
> after a few days of
> processing:
>
> sed: can't read .commands: No such file or directory
>
> The program had already created *FA_FAi_to_target_affine.dof,
> *FA_FAi_to_target_nonlinear.dof
> *FA_FAi_to_target_nonlinear.msf, *_FA_FAi_to_target_nonlinear.nii.gz on 21
> subjects when it
> stopped. Interestingly, when I re-ran it, it stopped at a different
> location (i.e., after the 23rd
> subject). Also of note is when I attempted to run this command on
> additional computers after
> starting one successfully, I receive the following error immediately and I
> am unable to run the
> program:
>
> wc: .commands: No such file or directory
>
> Incidentally, these subjects have been registered before. I am only
> re-registering them because I
> am running a different analysis where only a sub-group of the original set
> is included. Is there a
> better approach for doing re-analyses like these? For instance, can I
> simply take subjects out of
> analyses by designating them with zeroes in the contrast matrix?
>
> Thanks in advance for your suggestions,
> Francesca
>
Sheu, Yi-Shin
Research Assistant
McLean Hospital
Developmental Biopsychiatry Research Program
Tel: 617-855-2942
Fax: 617-855-3712
email: [log in to unmask]
|