InverseMatrixRichardson< VectorType > Class Template Reference

#include <deal.II/lac/matrix_lib.h>

Inheritance diagram for InverseMatrixRichardson< VectorType >:

Public Member Functions
	InverseMatrixRichardson (SolverControl &control, VectorMemory< VectorType > &mem)
	~InverseMatrixRichardson ()
template<typename MatrixType , typename PreconditionerType >
void	initialize (const MatrixType &, const PreconditionerType &)
SolverControl &	control () const
void	vmult (VectorType &, const VectorType &) const
void	vmult_add (VectorType &, const VectorType &) const
void	Tvmult (VectorType &, const VectorType &) const
void	Tvmult_add (VectorType &, const VectorType &) const
Public Member Functions inherited from Subscriptor
	Subscriptor ()
	Subscriptor (const Subscriptor &)
	Subscriptor (Subscriptor &&)
virtual	~Subscriptor ()
Subscriptor &	operator= (const Subscriptor &)
Subscriptor &	operator= (Subscriptor &&)
void	subscribe (const char *identifier=0) const
void	unsubscribe (const char *identifier=0) const
unsigned int	n_subscriptions () const
void	list_subscribers () const
template<class Archive >
void	serialize (Archive &ar, const unsigned int version)

Private Attributes
VectorMemory< VectorType > &	mem
SolverRichardson< VectorType >	solver
PointerMatrixBase< VectorType > *	matrix
PointerMatrixBase< VectorType > *	precondition

Additional Inherited Members
Static Public Member Functions inherited from Subscriptor
static ::ExceptionBase &	ExcInUse (int arg1, char *arg2, std::string &arg3)
static ::ExceptionBase &	ExcNoSubscriber (char *arg1, char *arg2)

Detailed Description

template<typename VectorType>
class InverseMatrixRichardson< VectorType >

Objects of this type represent the inverse of a matrix as computed approximately by using the SolverRichardson iterative solver. In other words, if you set up an object of the current type for a matrix \(A\), then calling the vmult() function with arguments \(v,w\) amounts to setting \(w=A^{-1}v\) by solving the linear system \(Aw=v\) using the Richardson solver with a preconditioner that can be chosen. Similarly, this class allows to also multiple with the transpose of the inverse (i.e., the inverse of the transpose) using the function SolverRichardson::Tsolve().

The functions vmult() and Tvmult() approximate the inverse iteratively starting with the vector dst. Functions vmult_add() and Tvmult_add() start the iteration with a zero vector. All of the matrix- vector multiplication functions expect that the Richardson solver with the given preconditioner actually converge. If the Richardson solver does not converge within the specified number of iterations, the exception that will result in the solver will simply be propagated to the caller of the member function of the current class.

Note

A more powerful version of this class is provided by the IterativeInverse class.

Instantiations for this template are provided for <float> and <double>; others can be generated in application programs (see the section on Template instantiations in the manual).

Deprecated:

If deal.II was configured with C++11 support, use the LinearOperator class instead, see the module on linear operators for further details.

Author

Guido Kanschat, 2005

Definition at line 440 of file matrix_lib.h.

Constructor & Destructor Documentation

InverseMatrixRichardson()

template<typename VectorType >

InverseMatrixRichardson< VectorType >::InverseMatrixRichardson	(	SolverControl &	control,
		VectorMemory< VectorType > &	mem
	)

Constructor, initializing the solver with a control and memory object. The inverted matrix and the preconditioner are added in initialize().

~InverseMatrixRichardson()

template<typename VectorType >

InverseMatrixRichardson< VectorType >::~InverseMatrixRichardson

(

)

Since we use two pointers, we must implement a destructor.

Member Function Documentation

initialize()

template<typename VectorType >

template<typename MatrixType , typename PreconditionerType >

void InverseMatrixRichardson< VectorType >::initialize ( const MatrixType &  m, const PreconditionerType &  p  )

inline

Initialization function. Provide a solver object, a matrix, and another preconditioner for this.

Definition at line 747 of file matrix_lib.h.

control()

template<typename VectorType >

SolverControl& InverseMatrixRichardson< VectorType >::control

(

)

const

Access to the SolverControl object used by the solver.

vmult()

template<typename VectorType >

void InverseMatrixRichardson< VectorType >::vmult	(	VectorType &	,
		const VectorType &
	)		const

Execute solver.

vmult_add()

template<typename VectorType >

void InverseMatrixRichardson< VectorType >::vmult_add	(	VectorType &	,
		const VectorType &
	)		const

Execute solver.

Tvmult()

template<typename VectorType >

void InverseMatrixRichardson< VectorType >::Tvmult	(	VectorType &	,
		const VectorType &
	)		const

Execute transpose solver.

Tvmult_add()

template<typename VectorType >

void InverseMatrixRichardson< VectorType >::Tvmult_add	(	VectorType &	,
		const VectorType &
	)		const

Execute transpose solver.

Member Data Documentation

mem

template<typename VectorType >

VectorMemory<VectorType>& InverseMatrixRichardson< VectorType >::mem

private

A reference to the provided VectorMemory object.

Definition at line 490 of file matrix_lib.h.

solver

template<typename VectorType >

SolverRichardson<VectorType> InverseMatrixRichardson< VectorType >::solver

mutableprivate

The solver object.

Definition at line 495 of file matrix_lib.h.

matrix

template<typename VectorType >

PointerMatrixBase<VectorType>* InverseMatrixRichardson< VectorType >::matrix

private

The matrix in use.

Definition at line 500 of file matrix_lib.h.

precondition

template<typename VectorType >

PointerMatrixBase<VectorType>* InverseMatrixRichardson< VectorType >::precondition

private

The preconditioner to use.

Definition at line 505 of file matrix_lib.h.

---

The documentation for this class was generated from the following file:

• deal.II/lac/matrix_lib.h