[Mne_analysis] Morph Deep Source reconstruction MEG

[Bear, Joshua]

        External Email - Use Caution        

The morph, BEM, etc. probably only need to be done once per subject (so 28 times) rather than separately for every trial (14,000 times). On brief glance of your script, it looks like you’re doing the latter?

Best,
Josh


Joshua Bear, MD, MA | Assistant Professor, Pediatric Neurology and Epilepsy
University of Colorado and Children’s Hospital Colorado
13123 East 16th Avenue, Box B-155 | Aurora, CO 80045 | Phone: 720-777-6138 | Fax: 720-777-7285

On Apr 14, 2020, at 07:52, Thomas THIERY <tthiery.univ at gmail.com<mailto:tthiery.univ at gmail.com>> wrote:


        External Email - Use Caution

Hello,

I'm trying to compute deep source reconstruction on 28 subjects (around 500 trials per subject), on 14 000 deep sources. I realize that the time needed to compute the morph is extremely long, and I was wondering whether there are some parameters I can change/modify to accelerate the morph without losing too much resolution ?

Here is my script, it works as is, but even when I use clusters (Compute Canada), it takes too long for now.

Best,

--
Thomas Thiery, Ph.D. Candidate in Cognitive Neuroscience @ University of Montreal, Canada
LinkedIn Profile: https://www.linkedin.com/in/thomas-thiery-92a7876a/

Twitter account: https://twitter.com/ThomayTHIERS/

Research Gate Profile:https://www.researchgate.net/profile/Thomas_Thiery

Google Scholar Profile: https://scholar.google.ca/citationsuser=ebPHcZ0AAAAJ&hl=en<https://scholar.google.ca/citations?user=ebPHcZ0AAAAJ&hl=en>
<04_compute_sources.py>_______________________________________________
Mne_analysis mailing list
Mne_analysis at nmr.mgh.harvard.edu<mailto: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/20200414/35d82c6e/attachment.html