Q: problems with cross-validating model

15 views
Skip to first unread message

Yezhe

unread,
Mar 24, 2021, 5:51:00 AM3/24/21
to kend...@post.harvard.edu, glmde...@googlegroups.com

Hi Kendrick,

Thank you for the nice package! I encountered the following error message when running the script:

The default parameters are used in the GLMdenoisedata function. The input BOLD data (frame=375, tr=0.8s) is processed with motion correction and smoothing and registered to the standard template. Apertures data (200x200x375) is processed to be consistent with BOLD data. What is the possible reason for this error? Can you give some suggestions?



Kendrick Kay

unread,
Mar 24, 2021, 9:11:25 AM3/24/21
to Yezhe, GLMdenoise
Hi Yezhe, 

The "GLMdenoise" feature of analyzePRF is a bit experimental (and likely delivers minimal benefits for block/continuous-type experimental designs)...  so I think the easiest solution is to just turn it off, e.g. 'wantglmdenoise',0 (or just allow the default to take hold (the default is 0)).

The longer answer: In order to diagnose, I would need to make some inspection of your apertures (for your four runs).  If you want to send me that, I could take a look.

Hope this helps,
Kendrick

--
Kendrick Kay (k...@umn.edu)
Assistant Professor
Center for Magnetic Resonance Research
University of Minnesota

On Mar 24, 2021, at 4:50 AM, Yezhe <yezh...@163.com> wrote:

Hi Kendrick,

Thank you for the nice package! I encountered the following error message when running the script:

<kampndcdcnhgnmbp.png>

The default parameters are used in the GLMdenoisedata function. The input BOLD data (frame=375, tr=0.8s) is processed with motion correction and smoothing and registered to the standard template. Apertures data (200x200x375) is processed to be consistent with BOLD data. What is the possible reason for this error? Can you give some suggestions?




--
You received this message because you are subscribed to the Google Groups "GLMdenoise" group.
To unsubscribe from this group and stop receiving emails from it, send an email to glmdenoise+...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/glmdenoise/edf586e1-cb37-b5fc-8be1-23ab4d3533aa%40163.com.

Reply all
Reply to author
Forward
0 new messages