problem with gave and gcov functions in mne_process_raw

Hi,

I'm having a problem getting the --gave and --gcov functions to work
as part of mne_process_raw. The routine generates -ave and -cov files
for each run without a problem, but then it terminates without
generating the grandave and grand-cov files.

The problem is specific to 3 subjects. For the others, everything
worked fine. I'm not sure where to start looking to identify the
problem. I'm attaching a -cov.log file in case that helps.

Thanks for any help you can provide,

Linnaea
-------------- next part --------------
A non-text attachment was scrubbed...
Name: CA10_02-cov.log
Type: application/octet-stream
Size: 9891 bytes
Desc: not available
Url : http://mail.nmr.mgh.harvard.edu/pipermail/mne_analysis/attachments/20081202/5e390e33/attachment.obj

Hi Linnaea,

I wasn't able to tell anything from this (maybe, someone else can),
but what would be very helpful is to re-run the whole thing and look
at the output to the terminal window. There should be an error
message describing what happened.

Dan

?

Hi,

I'm having a problem getting the --gave and --gcov functions to work
as part of mne_process_raw. The routine generates -ave and -cov files
for each run without a problem, but then it terminates without
generating the grandave and grand-cov files.

The problem is specific to 3 subjects. For the others, everything
worked fine. I'm not sure where to start looking to identify the
problem. I'm attaching a -cov.log file in case that helps.

Thanks for any help you can provide,

Linnaea<CA10_02-
cov.log>_______________________________________________
Mne_analysis mailing list
Mne_analysis at nmr.mgh.harvard.edu
https://mail.nmr.mgh.harvard.edu/mailman/listinfo/mne_analysis

-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mail.nmr.mgh.harvard.edu/pipermail/mne_analysis/attachments/20081202/7f21bc1f/attachment.html
-------------- next part --------------
A non-text attachment was scrubbed...
Name: DGWakeman.vcf
Type: text/directory
Size: 792 bytes
Desc: not available
Url : http://mail.nmr.mgh.harvard.edu/pipermail/mne_analysis/attachments/20081202/7f21bc1f/attachment.bin
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mail.nmr.mgh.harvard.edu/pipermail/mne_analysis/attachments/20081202/7f21bc1f/attachment-0001.html

Hi Linnaea,

I am guessing the computation of the covariance matrix fails because you
have too few samples in some of your runs.

Can you try generating only the gave files (without computing the gcov)
and see if it works? For the gcov you could exclude the shorter runs and
see if it works. Again..I am just guessing, so not sure if this is your
solution.

~N&ita