A small example. I took also the source (I hope the latest one) and a commented the balancing out which gives then correct results.
Best Regards -----Original Message----- From: Brian Gough [mailto:[email protected]] Sent: Sonntag, 8. November 2009 21:32 To: Mezger Anton-Chr. Cc: [email protected] Subject: Re: [Bug-gsl] bug report gsl_multifit_wlinear_svd At Fri, 6 Nov 2009 09:47:51 +0100, Mezger Anton-Chr. wrote: > The introduction of balancing the matrix does not seem to account for a > rank reduction. The routine gsl_multifit_wlinear_svd performs well when > no rank reduction is taken place. When the rank gets reduced however the > results are wrong and when the balancing code is taken out the results > are correct. Do I see this correctly and when yes is it possible to > correct it ? Thanks for the bug report. Could you send a small example program which demonstrates the problem -- thank you. -- Brian Gough GNU Scientific Library - http://www.gnu.org/software/gsl/
gsl_wlinear_svd.c
Description: gsl_wlinear_svd.c
_______________________________________________ Bug-gsl mailing list [email protected] http://lists.gnu.org/mailman/listinfo/bug-gsl
