Hello Rainer,
Can you confirm if you are using fsl 5.0.8 or 5..0.9 ( the latest version ) as this bug should have been fixed in FSL 5.0.9.
Kind Regards
Matthew
> Dear FSL Experts,
>
> I am having some problems with fsl_glm when using Dual Regression.
> The output of Dual Regression is really odd in stage 1 (HEX numbers I think???) and the NIFTI files of stage 2 contain only NaNs.
>
> The text files with the time courses from stage 1 of dual regression seem to contain only HEX numbers.
>
> It seems that this has started to happen after installing the new FSL version 5.08.
>
> I compared with a previous version:
> Here are the first few entries of a dual regression stage 1 txt-file with the previous version
>
> 475.529 536.112 -8.61474 437.769 382.946 100.037 407.096 194.886 443.785 30.356 415.762 177.114 328.565 164.314 319.926 344.078 -81.3183 342.733 318.474 425.6 195.076
> and this is the new one
>
> 0x1.db876ecdfaf19p+8 0x1.0c0e5ddeafd51p+9 -0x1.13abfc766b54ep+3 0x1.b5c4cccf9770ap+8 0x1.7ef22db78922fp+8 0x1.90262387f9da5p+6 0x1.97188308c1598p+8 0x1.85c5a8b4f2af5p+7 0x1.bbc8ed6d46dcap+8 0x1.e5b235503eb37p+4 0x1.9fc3288e846d1p+8 0x1.623a395b2bc35p+7 0x1.4890c037b0bd2p+8 0x1.48a0b4fda8f21p+7 0x1.3fed18ed16494p+8 0x1.5813fa222d039p+8 -0x1.4545faaf00c4cp+6
>
> (Same data in the same place on the hard drive only changed the command.)
>
> Any ideas what might have gone wrong?
> Is my version of fsl_glm somehow corrupt since the new installation?
> The response on the command line seems fine.
>
> How could I find out?
>
> Thank you in advance!
> Rainer
>
>
|