Scilab Function
Last update : 7/2/2005
pcg - precondioned conjugate gradient
Calling Sequence
- [x, flag, err, iter, res] = pcg(A,
b, tol, maxit, M, x)
Parameters
-
A: symmetric positive definite matrix or function
returning A*x
-
b: right hand side vector (size: n)
-
tol: error tolerance (default: 1000*%eps)
-
maxi: maximum number of iterations (default: n)
-
M: preconditioner: matrix or function returning
M*x (default: none)
-
x0: initial guess vector (default: zeros(n,1))
-
x: solution vector
-
flag: 0 if pcg converged to the desired
tolerance within maxi iterations, 1 else
-
err: final relative norm of the residual
-
iter: number of iterations performed
-
res: vector of the residual norms
Description
Solves the linear system Ax=b using the conjugate
gradient method with preconditioning.
The A matrix must be a symmetric positive definite
matrix.
Examples
See Also
pcg,
Author
Sage Group (IRISA, 2004)