Different lengths of runs

[Paul Glad Mihai]

Hi Kendrick,

I have one subject where the first run is slightly shorter (a couple of volumes) than the remaining four. The algorithm ran into trouble with this configuration, I suppose at the cross validation step due to different matrix sizes. One way to avoid this trouble is to just use the four equal length runs instead of five. Is there another way?

Regards,
Glad

[Kendrick Kay]

The code should be able to handle runs with different lengths. It might be that your design matrices and your run lengths are not commensurate, so you should check that.

Seeing the error message might be informative. Also, checking the size and dimensionality of all your inputs to function would be good.

--
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 post to this group, send email to glmde...@googlegroups.com.
Visit this group at https://groups.google.com/group/glmdenoise.
For more options, visit https://groups.google.com/d/optout.

[Paul Glad Mihai]

It works, indeed! I made a scripting error.