template<typename _MatrixType, typename _Preconditioner> class Eigen::DGMRES< _MatrixType, _Preconditioner >
A Restarted GMRES with deflation. This class implements a modification of the GMRES solver for sparse linear systems. The basis is built with modified Gram-Schmidt. At each restart, a few approximated eigenvectors corresponding to the smallest eigenvalues are used to build a preconditioner for the next cycle. This preconditioner for deflation can be combined with any other preconditioner, the IncompleteLUT for instance. The preconditioner is applied at right of the matrix and the combination is multiplicative.
Template Parameters
_MatrixType
the type of the sparse matrix A, can be a dense or a sparse matrix.
SparseMatrix<double> A;
VectorXd x, b;
//Fill A and b ...
DGMRES<SparseMatrix<double> > solver;
solver.set_restart(30); // Set restarting value
solver.setEigenv(1); // Set the number of eigenvalues to deflate
solver.compute(A);
x = solver.solve(b);
DGMRES can also be used in a matrix-free context, see the following example .
References : [1] D. NUENTSA WAKAM and F. PACULL, Memory Efficient Hybrid Algebraic Solvers for Linear Systems Arising from Compressible Flows, Computers and Fluids, In Press, https://doi.org/10.1016/j.compfluid.2012.03.023 [2] K. Burrage and J. Erhel, On the performance of various adaptive preconditioned GMRES strategies, 5(1998), 101-121. [3] J. Erhel, K. Burrage and B. Pohl, Restarted GMRES preconditioned by deflation,J. Computational and Applied Mathematics, 69(1996), 303-318.
Initialize the solver with matrix A for further Ax=b solving.
This constructor is a shortcut for the default constructor followed by a call to compute().
Warning
this class stores a reference to the matrix A as well as some precomputed values that depend on it. Therefore, if A is changed this class becomes invalid. Call compute() to update it with the new matrix A, or modify a copy of A.
