these messages can be ignored. they originate in the bowels of some fortran code of a minimization routine in a 3rd party library we use.
n.
On Fri, 2010-09-10 at 17:10 -0400, Zoltick, Brad (NIH/NIMH) [E] wrote:
Hi, I have been running subjects in Version 5, and notice there are consistently weird error messages in the skull strip section of recon-all.log. These look like:
IFLAG= -1 LINE SEARCH FAILED. SEE DOCUMENTATION OF ROUTINE MCSRCH ERROR RETURN OF LINE SEARCH: INFO= 3 POSSIBLE CAUSES: FUNCTION OR GRADIENT ARE INCORRECT OR INCORRECT TOLERANCESoutof QuasiNewtonEMA: 006: -log(p) = 10319.2 tol 0.000010
Despite the log containing a few of these warnings, recon-all itself almost always completes “without errors, ” and the subjects appear to be fully processed. However, I have noticed a greater prevalence of bad skull strips compared to version 4.5.
Does anyone know what this error means?
Thanks,
Brad Zoltick and Aaron Goldman
Freesurfer mailing list Freesurfer@nmr.mgh.harvard.edu https://mail.nmr.mgh.harvard.edu/mailman/listinfo/freesurfer