SVD whit GESVD

anacondgame — Thu, 03 Jan 2008 15:35:32 GMT

I calculate the SVD of a real matrix A. I used the LAPACK ?gesvd, but with this function I have get the following decomposition:

A = T*S*V^H

But I want to get the decomposition:

A = D*S*V^T

How can I do? What function should I use?

Re: SVD whit GESVD

Michael_C_Intel4 — Wed, 18 Jun 2008 15:21:02 GMT

SVD is usually written as

A = U*S*V(H)

where U and V are orthogonal (unitary) matrices, S is diagonal with the singular values on the diagonal.

It's the onlyform of SVD. What do you mean by matrices T and D?

BTW, transposed and hermitian transposed matrix V is the same in case of real matrix A.

Michael.