Matrix is singluar to working precision - Freesurfer

9 Oct 2009


      Hi Freesurfers,
I'm trying to selxavg3 my data.
My design includes 6 conditions including a 0 rest condition. 3 contrasts
have been defined using mkcontrast. The analysis too using mkanalysis.
I use this command line :
[/homes/11/chaumon/data] (nmr-std-env) selxavg3-sess -analysis
"Analyse_Simple_Confidence" -s S003 -overwrite
Which returns what's in the log file attached.
and ends with
Creating Design Matrix
Warning: Matrix is singular to working precision.
...
In fast_selxavg3 at 212
Warning: Matrix is singular to working precision.
...
In fast_selxavg3 at 212
Warning: Matrix is singular to working precision.
...
In fast_selxavg3 at 212
Warning: Matrix is singular to working precision.
...
In fast_selxavg3 at 212
Warning: Matrix is singular to working precision.
...
In fast_selxavg3 at 212
Warning: Matrix is singular to working precision.
...
In fast_selxavg3 at 212
Warning: Matrix is singular to working precision.
...
In fast_selxavg3 at 212
Warning: Matrix is singular to working precision.
...
In fast_selxavg3 at 212
Warning: Matrix is singular to working precision.
...
In fast_selxavg3 at 212
ntptot = 693, nX = 15, DOF = 678
Saving X matrix to
/autofs/space/bomba_016/users/chaumon/S003/bold/Analyse_Simple_Confidence/Xtmp.mat
??? Error using ==> svd
Input to SVD must not contain NaN or Inf.
Error in ==> cond at 40
   s = svd(A);
Error in ==> fast_selxavg3 at 248
  XCond = cond(XtX);
...
...
09-Oct-2009 18:03:13
Salut !
------------------------------------------
ERROR: fast_selxavg3() failed\n
I'm puzzled. I've been checking and rechecking my design. The problem does
not come from the contrasts because I get the same error if I remove them...
Any help will be greatly appreciated.
Thanks,
Max