Yep - melodic is definitely segmentation faulting - I guess we'll have to
work out what's wrong with the new compiler once we start compiling with
that for the new release - probably just some compatibility issue with an
updated c++ definition or something - thanks for pointing it out!
Cheers.
On Wed, 7 Jul 2004, Zrinka Bilusic wrote:
> Hello,
> "melodic --help" did display the help info on melodic - so it did run.
> However, I got "segmentation fault" error while running the feeds data
> using melodic. The only thing I can find in the output folder in the
> mask image, and that one seems fine (compared to the mask image in the
> corresponding ica folder in results folder). So I am guessing that
> something happened during compilation? (there were no errors reported).
> any ideas? Thanks!!
>
> here is the output from running melodic:
>
> melodic -i feeds/data/fmri -v --report
>
> Melodic Version 2.0
>
> Melodic results will be in feeds/data/fmri.ica+
>
> Reading data file feeds/data/fmri ... done
> Create mask ... Calling BET: /Users/nina/fsl_gcc_G5opt/bin/bet
> /var/tmp/tmp.0.xCk5ot /var/tmp/tmp.0.xCk5ot_brain -m -f 0.4 > /dev/null
> done
> Excluding voxels with constant value
> Data size : 180 x 16911
> Removing mean image ... done
> Estimating data covariance ... done
> Perform variance-normalisation ... done
> Removing mean time course ... done
> Segmentation fault
>
> On Jul 7, 2004, at 3:44 AM, Christian F. Beckmann wrote:
>
> > Hi Zrinka,
> >
> > feeds is a very simple script which runs different parts of FSL. If
> > melodic does not complete feeds this is quite likely to be due to
> > compilation issues. First, you should try and see if the different
> > binaries run at all (just type melodic --help).
> > If the binaries are OK, run melodic on the feeds example fmri data set
> >
> > melodic -i /PATH_TO_FEEDS/data/fmri -v --report
> >
> > and see if that tells you more about what is going on.
> > cheers
> > christian
> >
> >
> > On 30 Jun 2004, at 18:48, Zrinka Bilusic wrote:
> >
> >> Hello...
> >>
> >> I have just recompiled FSL again with the latest version of gcc (3.3)
> >> using also the G5 optimization flags (suggested by apple and
> >> applicable
> >> only for gcc 3.3 I believe) and I ran feeds (I wanted to compare the
> >> performance of the G5 optimized gcc vs IBM xlc). However, it failed
> >> while testing Melodic (see below for the output). I checked the
> >> compilation and it completed without any errors and almost no
> >> warnings.
> >> What does it exactly mean when feeds test fails? What should I be
> >> looking for to fix this?
> >>
> >> Thanks!
> >> zrinka
> >>
> >> FSL Evaluation and Example Data Suite v3.1
> >>
> >> start time = Tue Jun 29 15:09:14 PDT 2004
> >> hostname = nina.local
> >> os = Darwin nina.local 7.4.0 Darwin Kernel Version 7.4.0: Wed May 12
> >> 16:58:24 PDT 2004; root:xnu/xnu-517.7.7.obj~7/RELEASE_PPC Power
> >> Macintosh powerpc
> >>
> >>
> >> /bin/rm -rf /Users/nina/feeds/results /Users/nina/feeds/data/fmri+*
> >>
> >>
> >> mkdir /Users/nina/feeds/results
> >>
> >>
> >> Starting SUSAN at Tue Jun 29 15:09:14 PDT 2004
> >> % error = 0.0
> >>
> >> Starting PRELUDE & FUGUE at Tue Jun 29 15:12:53 PDT 2004
> >> % error = 0.0
> >> % error = 0.0
> >>
> >> Starting SIENAX (including testing BET and FLIRT and FAST) at Tue Jun
> >> 29 15:12:56 PDT 2004
> >> checking error on BET:
> >> % error = 0.0
> >> checking error on FLIRT:
> >> % error = 0.322037594668
> >> checking error on FAST:
> >> checking error on single-image binary segmentation:
> >> % error = 0.345114163645
> >> checking error on partial volume images:
> >> % error = 0.393971552914
> >> % error = 0.281726013818
> >> % error = 0.264880778174
> >> checking error on SIENAX volume outputs:
> >> % error = 0.0351231718211
> >> % error = 0.00218948572482
> >> % error = 0.00546535143417
> >> % error = 0.0313234567567
> >> % error = 0.0367888081908
> >>
> >> Starting FEAT at Tue Jun 29 15:24:52 PDT 2004
> >> checking error on filtered functional data:
> >> % error = 0.179583640438
> >> checking error on raw Z stat images:
> >> % error = 0.311126636832
> >> % error = 0.202559213588
> >> % error = 0.303332786639
> >> checking error on thresholded Z stat images:
> >> % error = 0.388013690501
> >> % error = 0.196759687433
> >> % error = 0.335210397996
> >> checking error on position of largest cluster of Talairached zfstat1:
> >> % error = 0.04
> >> % error = 0.28
> >> % error = 0.184
> >> % error = 0.032
> >> % error = 0.14
> >> % error = 0.028
> >>
> >> Starting MELODIC at Tue Jun 29 15:39:48 PDT 2004
> >> % error = 6.66666666667
> >> Warning - test failed!
> >>
> >> Warning - not all tests passed
> >>
> >> end time = Tue Jun 29 15:40:24 PDT 2004
> >>
> >>
> > --
> > Christian F. Beckmann
> > Oxford University Centre for Functional
> > Magnetic Resonance Imaging of the Brain,
> > John Radcliffe Hospital, Headington, Oxford OX3 9DU, UK
> > Email: [log in to unmask] - http://www.fmrib.ox.ac.uk/~beckmann/
> > Phone: +44(0)1865 222782 Fax: +44(0)1865 222717
> >
>
Stephen M. Smith DPhil
Associate Director, FMRIB and Analysis Research Coordinator
Oxford University Centre for Functional MRI of the Brain
John Radcliffe Hospital, Headington, Oxford OX3 9DU, UK
+44 (0) 1865 222726 (fax 222717)
[log in to unmask] http://www.fmrib.ox.ac.uk/~steve
|