Classes | |
struct | AdditionalData |
Public Member Functions | |
SolverFGMRES (SolverControl &cn, VectorMemory< VECTOR > &mem, const AdditionalData &data=AdditionalData()) | |
SolverFGMRES (SolverControl &cn, const AdditionalData &data=AdditionalData()) | |
template<class MATRIX , class PRECONDITIONER > | |
void | solve (const MATRIX &A, VECTOR &x, const VECTOR &b, const PRECONDITIONER &precondition) |
Private Attributes | |
AdditionalData | additional_data |
FullMatrix< double > | H |
FullMatrix< double > | H1 |
Implementation of the Generalized minimal residual method with flexible preconditioning method.
This version of the GMRES method allows for the use of a different preconditioner in each iteration step. Therefore, it is also more robust with respect to inaccurate evaluation of the preconditioner. An important application is also the use of a Krylov space method inside the preconditioner.
FGMRES needs two vectors in each iteration steps yielding a total of 2*AdditionalDatamax_basis_size+1
auxiliary vectors.
Caveat: documentation of this class is not up to date. There are also a few parameters of GMRES we would like to introduce here.
SolverFGMRES< VECTOR >::SolverFGMRES | ( | SolverControl & | cn, | |
VectorMemory< VECTOR > & | mem, | |||
const AdditionalData & | data = AdditionalData() | |||
) |
Constructor.
SolverFGMRES< VECTOR >::SolverFGMRES | ( | SolverControl & | cn, | |
const AdditionalData & | data = AdditionalData() | |||
) |
Constructor. Use an object of type GrowingVectorMemory as a default to allocate memory.
void SolverFGMRES< VECTOR >::solve | ( | const MATRIX & | A, | |
VECTOR & | x, | |||
const VECTOR & | b, | |||
const PRECONDITIONER & | precondition | |||
) | [inline] |
Solve the linear system for x.
Referenced by SolverSelector< VECTOR >::solve().
AdditionalData SolverFGMRES< VECTOR >::additional_data [private] |
Additional flags.
FullMatrix<double> SolverFGMRES< VECTOR >::H [private] |
Projected system matrix
FullMatrix<double> SolverFGMRES< VECTOR >::H1 [private] |
Auxiliary matrix for inverting H