|
From: | Yves Renard |
Subject: | Re: [Getfem-users] How to obtain determinant of a sparse matrix |
Date: | Mon, 19 Nov 2012 17:03:54 +0100 |
User-agent: | Mozilla/5.0 (X11; Linux x86_64; rv:16.0) Gecko/20121028 Thunderbird/16.0.2 |
Dear Tsai, The determinant of a large sparse matrix is useless 99.99% of the time because except if the eigenvalues are very close from 1, either the determinant is too small (represented by zero, even if all the eigenvalues are nonzero) or too large (> 1E308). So, for instance, if you want to test the invertibility of a system, the computation of the determinant is not appropriate. A better approach is for instance compute the smallest singluar value with a Lanczos or Von Mises algorithm. In fact, depending on the use of the determinant you need, you better have to try to find an indirect mean not to compute the determinant ... Yves. Le 19/11/2012 15:26, Finite Tsai a écrit :
-- Yves Renard (address@hidden) tel : (33) 04.72.43.87.08 Pole de Mathematiques, INSA-Lyon fax : (33) 04.72.43.85.29 20, rue Albert Einstein 69621 Villeurbanne Cedex, FRANCE http://math.univ-lyon1.fr/~renard --------- |
[Prev in Thread] | Current Thread | [Next in Thread] |