Hi,
is it possible that your system has some dodgy RAM? if run 6 2orks but
run 2 doesn't than this points to some memory related issue - but with
4GB all should be fine. Can you run some diagnostics?
If melodic crashes then yes, the report directory will be incomplete.
cheers
christian
On 25 Jul 2008, at 12:55, Sascha Purmann wrote:
> Hi Steve,
>
> thanks for your suggestions. This was my first idea as well, so I
> stopped
> using design matrix or contrast. With or without, same result.
>
> What I did so far for testing (all on the fsl course data;
> /fsl_course_data/melodic/tensor):
>
> 1. run tensor ICA with functional runs of all four subjects -> failed
> 2. run tensor ICA with functional runs of two subjects -> failed
> 3. split functional runs of subject 1 into two parts and run tensor
> ICA an
> these -> worked
> 4. split functional runs of subject 2 into two parts and run tensor
> ICA an
> these -> worked
>
> 5. run tensor ICA on part 1 of subject 1 and part 1 of subject 2 ->
> worked
> 6. run tensor ICA on all 4 parts
> (part1/subj1,part2/subj1,part1/subj2,part2/subj2) -> worked
>
> When tensor ICA failed, then with the error given in my previous
> message,
> which also means it does not give a full html-report (report/
> 00index.html
> does not show any components, TICA modes or PCA estimates).
>
> Especially 6 and 2 are interesting, because it's the same data but
> in 6
> split into pieces.
>
> I tried it on a PC running fedora and could successfully run tica on
> three
> subjects (again fsl course data). Including subject4 failed but
> without a
> "segmentation fault" error and reflect probably insufficient working
> memory.
> So it might be a debian related problem. Has anyone run a tensor ica
> on a
> debian system?
>
> Also, what alternatives do I have to bring single subject / single
> session
> icas together on the group level? I know Greicius et al. (2004). Does
> anyone know other papers?
>
> Thanks for your help,
>
> Cheers!
>
> Sascha
>
> fimlab.uni-hd.de
>
> On Tue, 22 Jul 2008 09:01:52 -0700, Steve Smith
> <[log in to unmask]> wrote:
>
>> Hi - I'm guessing that this means that the design matrix and/or
>> contrast that you're feeding into MELODIC (if you are doing that) are
>> not valid? If you could send those files we could have a look.
>>
>> Cheers.
>>
>> On 20 Jul 2008, at 07:26, Sascha Purmann wrote:
>>
>>> Dear FSL List:
>>>
>>> I am trying to run tica with melodic (Melodic Version 3.05). First
>>> try on my
>>> own data failed, so I tried it on the fsl course data. Again, no
>>> success. I
>>> tried it on different debian/ubuntu systems (e.g. ubuntu hardy,
>>> amd64bit,
>>> 4GB Ram) and get errors on all systems.
>>>
>>> I get a "segmentation fault" error messages in logs/gica:
>>>
>>> --snip--
>>>
>>> [...]
>>>
>>> Step no. 85 change : 0.000378202
>>> Rank-1 approximation of the time courses; change : 0.013905
>>> Convergence after 85 steps
>>>
>>> Sorting IC maps
>>>
>>> fdtr domain error
>>>
>>> [...]
>>>
>>> DOF cannot be zero or negative!
>>>
>>> stdtr domain error
>>>
>>> ndtri domain error
>>>
>>> [...]
>>>
>>> Segmentation fault (core dumped)
>>>
>>> --snap--
|