Hi,
I’ve been able to generate results from your data on a test machine here - can you let me know what OS and FSL version you are using as well as how much RAM your machine has?

Kind Regards
Matthew
--------------------------------
Dr Matthew Webster
FMRIB Centre 
John Radcliffe Hospital
University of Oxford

On 18 Jun 2019, at 18:39, Moataz Assem <[log in to unmask]> wrote:

Hi,

 

I get the following fsl error related to the beta estimations (while running film_gls):

 

Prewhitening and Computing PEs...

Percentage done:

1,2,3,4,5,6,7,8,9,10,11,12,13,14,15,16,17,18,19,20,21,22,23,24,25,26,27,28,29,30,31,32,33,34,35,36,37,38,39,

error: pinv(): svd failed

 

pinv(): svd failed

 

In my case, film_gls is being run as part of the HCP task analysis pipelines which involves splitting a CIFTI file and running the estimation for the subcortical volumetric data and each surface separately. The codes are fine as I only get this error

for one run (in one subject) and while estimating the left surface (subcortical estimation ran fine). The design matrix doesn’t contain any NaNs or empty EVs neither does the timeseries for that hemisphere (also std(timeseries) did not give any zeros).

 

The way film_gls is called is as follows (filenames shortened for simplicity):

 

film_gls

--rn=../dir.feat/L_SurfaceStats

--sa --ms=15 --epith=5

--in2=../dir/S23.L.midthickness.32k_fs_LR.surf.gii

--in=../dir.feat/func_file.func.gii

--pd=../dir.feat/design.mat

--con=../dir.feat/design.con

--mode=surface

 

The HCP list recommended to remove the ‘--sa --ms=15 --epith=5’ to probe the error further and this did actually solve the problem. Any ideas what would this indicate?

 

Thanks

 

Moataz

 



To unsubscribe from the FSL list, click the following link:
https://www.jiscmail.ac.uk/cgi-bin/webadmin?SUBED1=FSL&A=1




To unsubscribe from the FSL list, click the following link:
https://www.jiscmail.ac.uk/cgi-bin/webadmin?SUBED1=FSL&A=1