At Wed, 3 Sep 2008 15:51:12 +0200,
Geelen Bert wrote:
> Shouldn't the dimensions of U and S instead be as follows:
> "A general rectangular M-by-N matrix A has a singular value
> decomposition (svd) into the product of an M-by-M orthogonal matrix U,
> an M-by-N diagonal matrix of singular values S and the transpose of an
> N-by-N orthogonal square matrix V"
Thanks for your email. The routines compute the "thin" version of the
SVD as opposed to the "full" SVD. I will add a note to the manual to
make that clearer.
--
Brian Gough
GNU Scientific Library -
http://www.gnu.org/software/gsl/
