Re: [eigen] new API for Cholmod

[Christoph Hertzberg <chtz@xxxxxxxxxxxxxxxxxxxxxxxx>]

On 04.11.2010 19:23, Rodney Sparapani wrote:
> Definitely.  CHOLMOD is a brilliant piece of software.  But, one
> thing that has caught me with my pants down is that it will
> silently translate your problem into AA' if you forget to tell it that A
> is symmetric (and why is that anyways, then isn't the answer just A?).

Of course it's not just A, because it needs to be a triangular matrix. A
use case for that behavior is solving the normal equation:

  argmin_x |A'*x - b| = -(AA')^-1 * (A*b)

I'd agree that this should not happen implicitly (on the other hand
Cholesky should give a compile time error for non-symmetric matrices)

Christoph



-- 
----------------------------------------------
Dipl.-Inf. Christoph Hertzberg
Cartesium 0.051
Universität Bremen
Enrique-Schmidt-Straße 5
28359 Bremen

Tel: (+49) 421-218-64252
----------------------------------------------