This is the mail archive of the
gsl-discuss@sources.redhat.com
mailing list for the GSL project.
Re: gsl_linalg_complex_householder_transform
- From: Brian Gough <bjg at network-theory dot co dot uk>
- To: Mario Pernici <Mario dot Pernici at mi dot infn dot it>
- Cc: gsl-discuss at sources dot redhat dot com
- Date: Wed, 28 May 2003 15:56:57 +0100
- Subject: Re: gsl_linalg_complex_householder_transform
- References: <Pine.LNX.4.44.0305261456150.9827-101000@ptt.mi.infn.it>
Mario Pernici writes:
>
> Dear Sirs,
> it seems that gsl_linalg_complex_householder_transform
> does not follow Golub and Van Loan; is it done
> on purpose or there is a bug?
Hi,
I think I used the LAPACK method described in Lapack Working Note 72
"The Computation of Elementary Matrices", as it is supposed to have
better properties than the one in Golub and Van Loan.
I don't remember the details now -- I should have put some comments in
the code.
I will take another look at it to make sure it is ok. Thanks for the
bug reports, keep them coming.
--
Brian