UMFPACK Mean solver class. More...

#include <umfpack.hpp>

Inheritance diagram for FEAT::Solver::UmfpackMean:

Public Types
typedef SolverBase< VectorType >	BaseClass
	our base class More...

typedef LAFEM::SparseMatrixCSR< double, Index >	MatrixType
	compatible matrix type More...

typedef LAFEM::DenseVector< double, Index >	VectorType
	compatible vector type More...

Public Member Functions
	UmfpackMean (const MatrixType &system_matrix, const LAFEM::MeanFilter< double, Index > &mean_filter)
	Constructor. More...

	UmfpackMean (const MatrixType &system_matrix, const VectorType &weight_vector)
	Constructor. More...

virtual Status	apply (LAFEM::DenseVector< double, Index > &vec_cor, const LAFEM::DenseVector< double, Index > &vec_def)=0
	Solver application method. More...

virtual Status	apply (VectorType &vec_sol, const VectorType &vec_rhs) override
	Solves a linear system with the factorized system matrix. More...

virtual void	done ()
	Finalization method. More...

virtual void	done_numeric () override
	Numeric finalization method. More...

virtual void	done_symbolic () override
	Symbolic finalization method. More...

virtual void	init ()
	Initialization method. More...

virtual void	init_numeric () override
	Numeric initialization method. More...

virtual void	init_symbolic () override
	Symbolic initialization method. More...

virtual String	name () const override
	Returns the name of the solver. More...

Private Attributes
MatrixType	_solver_matrix
	our extended system matrix More...

const MatrixType &	_system_matrix
	system matrix More...

Umfpack	_umfpack
	our internal Umfpack solver object More...

VectorType	_vec_b

VectorType	_vec_x
	two temporary extended vectors More...

const VectorType &	_weight_vector
	weight vector More...

Detailed Description

UMFPACK Mean solver class.

This class implements a variant of the Umfpack solver, which is capable of solving linear systems including the integral mean constraint or any other constraint that can be expressed as a scalar Lagrange multiplier to the original system matrix.

As Umfpack is a direct solver, it cannot directly utilize the MeanFilter class, which is used by iterative solvers to enforce the integral mean constraint. Therefore, this class implements an algorithm, which extends the original system matrix by a Lagrange multiplier vector and applied the Umfpack solver onto this extended linear system.

This class offers two constructors:

A CTOR which takes a SparseMatrixCSR and a MeanFilter as input
A CTOR which takes a SparseMatrixCSR and a DenseVector representing the Lagrange multiplier as input.

Author: Peter Zajac

Definition at line 140 of file umfpack.hpp.

Member Typedef Documentation

◆ BaseClass

typedef SolverBase<VectorType> FEAT::Solver::UmfpackMean::BaseClass

our base class

Definition at line 150 of file umfpack.hpp.

◆ MatrixType

typedef LAFEM::SparseMatrixCSR<double, Index> FEAT::Solver::UmfpackMean::MatrixType

compatible matrix type

Definition at line 145 of file umfpack.hpp.

◆ VectorType

typedef LAFEM::DenseVector<double, Index> FEAT::Solver::UmfpackMean::VectorType

compatible vector type

Definition at line 147 of file umfpack.hpp.

Constructor & Destructor Documentation

◆ UmfpackMean() [1/2]

FEAT::Solver::UmfpackMean::UmfpackMean	(	const MatrixType &	system_matrix,
		const VectorType &	weight_vector
	)

explicit

Constructor.

Parameters

[in]	system_matrix	A reference to the system matrix to be factorized.
[in]	weight_vector	The weight vector to be used as a Lagrange multiplier.

◆ UmfpackMean() [2/2]

FEAT::Solver::UmfpackMean::UmfpackMean	(	const MatrixType &	system_matrix,
		const LAFEM::MeanFilter< double, Index > &	mean_filter
	)

inlineexplicit

Constructor.

Parameters

[in]	system_matrix	A reference to the system matrix to be factorized.
[in]	mean_filter	A reference to the mean filter containing the weight vector.

Definition at line 185 of file umfpack.hpp.

Member Function Documentation

◆ apply() [1/2]

virtual Status FEAT::Solver::SolverBase< LAFEM::DenseVector< double, Index > >::apply	(	LAFEM::DenseVector< double, Index > &	vec_cor,
		const LAFEM::DenseVector< double, Index > &	vec_def
	)

pure virtualinherited

Solver application method.

This method applies the solver represented by this object onto a given defect vector and returns the corresponding correction vector.

Note: Solvers which derive from the IterativeSolver base class also provide a correct() method which corrects an initial solution instead of starting with the null vector.

Parameters

[out]	vec_cor	The vector that shall receive the solution of the linear system. It is assumed to be allocated, but its numerical contents may be undefined upon calling this method.
[in]	vec_def	The vector that represents the right-hand-side of the linear system to be solved.

Attention: vec_cor and vec_def must not refer to the same vector object!

Returns: A Status code that represents the status of the solution step.

◆ apply() [2/2]

virtual Status FEAT::Solver::UmfpackMean::apply	(	VectorType &	vec_sol,
		const VectorType &	vec_rhs
	)

overridevirtual

Solves a linear system with the factorized system matrix.

Parameters

[in,out]	vec_sol	A reference to the solution vector. The vector must be allocated to the correct length, but its initial contents are ignored.
[in]	vec_rhs	A reference to the right-hand-side of the linear system.

◆ done()

virtual void FEAT::Solver::SolverBase< LAFEM::DenseVector< double, Index > >::done ( )

inlinevirtualinherited

Finalization method.

This function performs both the symbolic and numeric finalization, i.e. it simply performs

this->done_numeric();
this->done_symbolic();

Definition at line 283 of file base.hpp.

◆ done_numeric()

virtual void FEAT::Solver::UmfpackMean::done_numeric ( )

overridevirtual

Numeric finalization method.

This method is called to release any data allocated in the numeric initialization step.

Reimplemented from FEAT::Solver::SolverBase< LAFEM::DenseVector< double, Index > >.

◆ done_symbolic()

virtual void FEAT::Solver::UmfpackMean::done_symbolic ( )

overridevirtual

Symbolic finalization method.

This method is called to release any data allocated in the symbolic initialization step.

Reimplemented from FEAT::Solver::SolverBase< LAFEM::DenseVector< double, Index > >.

◆ init()

virtual void FEAT::Solver::SolverBase< LAFEM::DenseVector< double, Index > >::init ( )

inlinevirtualinherited

Initialization method.

This function performs both the symbolic and numeric initialization, i.e. it simply performs

this->init_symbolic();
this->init_numeric();

Definition at line 268 of file base.hpp.

◆ init_numeric()

virtual void FEAT::Solver::UmfpackMean::init_numeric ( )

overridevirtual

Numeric initialization method.

This method is called to perform numeric initialization of the solver.
Before this function can be called, the symbolic initialization must be performed.

Reimplemented from FEAT::Solver::SolverBase< LAFEM::DenseVector< double, Index > >.

◆ init_symbolic()

virtual void FEAT::Solver::UmfpackMean::init_symbolic ( )

overridevirtual

Symbolic initialization method.

This method is called to perform symbolic initialization of the solver.

Reimplemented from FEAT::Solver::SolverBase< LAFEM::DenseVector< double, Index > >.

◆ name()

virtual String FEAT::Solver::UmfpackMean::name ( ) const

inlineoverridevirtual

Returns the name of the solver.

Implements FEAT::Solver::SolverBase< LAFEM::DenseVector< double, Index > >.

Definition at line 193 of file umfpack.hpp.

Member Data Documentation

◆ _solver_matrix

MatrixType FEAT::Solver::UmfpackMean::_solver_matrix

private

our extended system matrix

Definition at line 158 of file umfpack.hpp.

◆ _system_matrix

const MatrixType& FEAT::Solver::UmfpackMean::_system_matrix

private

system matrix

Definition at line 154 of file umfpack.hpp.

◆ _umfpack

Umfpack FEAT::Solver::UmfpackMean::_umfpack

private

our internal Umfpack solver object

Definition at line 162 of file umfpack.hpp.

◆ _vec_b

VectorType FEAT::Solver::UmfpackMean::_vec_b

private

Definition at line 160 of file umfpack.hpp.

◆ _vec_x

VectorType FEAT::Solver::UmfpackMean::_vec_x

private

two temporary extended vectors

Definition at line 160 of file umfpack.hpp.

◆ _weight_vector

const VectorType& FEAT::Solver::UmfpackMean::_weight_vector

private

weight vector

Definition at line 156 of file umfpack.hpp.

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

kernel/solver/umfpack.hpp

Public Types

Public Member Functions

Private Attributes

Detailed Description

Member Typedef Documentation

◆ BaseClass

◆ MatrixType

◆ VectorType

Constructor & Destructor Documentation

◆ UmfpackMean() [1/2]

◆ UmfpackMean() [2/2]

Member Function Documentation

◆ apply() [1/2]

◆ apply() [2/2]

◆ done()

◆ done_numeric()

◆ done_symbolic()

◆ init()

◆ init_numeric()

◆ init_symbolic()

◆ name()

Member Data Documentation

◆ _solver_matrix

◆ _system_matrix

◆ _umfpack

◆ _vec_b

◆ _vec_x

◆ _weight_vector