File: cs_chol.m

package info (click to toggle)
suitesparse 1%3A5.12.0%2Bdfsg-2
  • links: PTS, VCS
  • area: main
  • in suites: bookworm
  • size: 176,720 kB
  • sloc: ansic: 1,193,914; cpp: 31,704; makefile: 6,638; fortran: 1,927; java: 1,826; csh: 765; ruby: 725; sh: 529; python: 333; perl: 225; sed: 164; awk: 35
file content (18 lines) | stat: -rw-r--r-- 791 bytes parent folder | download | duplicates (6)
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
function [L,p] = cs_chol (A,drop)                                           %#ok
%CS_CHOL sparse Cholesky factorization.
%   L = cs_chol(A) is the same as L = chol(A)', using triu(A).
%   [L,p] = cs_chol(A) first orders A with p=cs_amd(A), so that L*L' = A(p,p).
%   A second optional input argument controls whether or not numerically zero
%   entries are removed from L.  cs_chol(A) and cs_chol(A,1) drop them;
%   cs_chol(A,0) keeps them.  They must be kept for cs_updown to work properly.
%
%   Example:
%       Prob = ssget ('HB/bcsstk01') ; A = Prob.A ; [L,p] = cs_chol (A) ;
%       cspy (A (p,p)) ;
%       cspy (L) ;
%
%   See also CS_AMD, CS_UPDOWN, CHOL, AMD, SYMAMD.

% Copyright 2006-2012, Timothy A. Davis, http://www.suitesparse.com

error ('cs_chol mexFunction not found') ;