CGAL::Eigen_sparse_matrix< T > Struct Template Reference

#include <CGAL/Eigen_sparse_matrix.h>

Inherited by CGAL::Eigen_sparse_symmetric_matrix< T >.

Definition

template<class T>
struct CGAL::Eigen_sparse_matrix< T >

The class Eigen_sparse_matrix is a wrapper around Eigen matrix type Eigen::SparseMatrix that represents general matrices, be they symmetric or not.

Is model of

SparseLinearAlgebraTraits_d::Matrix

Template Parameters

T	Number type.

See also

CGAL::Eigen_vector<T>

CGAL::Eigen_matrix<T>

CGAL::Eigen_sparse_symmetric_matrix<T>

Public Member Functions
	Eigen_sparse_matrix (const EigenType &et)
	Eigen_sparse_matrix ()
	Eigen_sparse_matrix (std::size_t dim, bool is_symmetric=false)
	Create a square matrix initialized with zeros.
	Eigen_sparse_matrix (int dim, bool is_symmetric=false)
	Create a square matrix initialized with zeros.
	Eigen_sparse_matrix (std::size_t rows, std::size_t columns, bool is_symmetric=false)
	Create a rectangular matrix initialized with zeros.
void	swap (Eigen_sparse_matrix &other)
	~Eigen_sparse_matrix ()
	Delete this object and the wrapped matrix.
	Eigen_sparse_matrix (int rows, int columns, bool is_symmetric=false)
	Create a rectangular matrix initialized with zeros.
int	row_dimension () const
	Return the matrix number of rows.
int	column_dimension () const
	Return the matrix number of columns.
void	set_coef (std::size_t i_, std::size_t j_, T val, bool new_coef=false)
	Write access to a matrix coefficient: a_ij <- val.
void	add_coef (std::size_t i_, std::size_t j_, T val)
	Write access to a matrix coefficient: a_ij <- a_ij + val.
NT	get_coef (std::size_t i_, std::size_t j_) const
	Read access to a matrix coefficient.
const EigenType &	eigen_object () const
	Return the internal matrix, with type EigenType.
EigenType &	eigen_object ()
	Return the internal matrix, with type EigenType.

Types
typedef Eigen::SparseMatrix< T >	EigenType
	The internal matrix type from Eigen.
typedef T	NT

Constructor & Destructor Documentation

Eigen_sparse_matrix() [1/4]

template<class T>

CGAL::Eigen_sparse_matrix< T >::Eigen_sparse_matrix	(	std::size_t	dim,
		bool	is_symmetric = false )

Create a square matrix initialized with zeros.

Parameters

dim	Matrix dimension.
is_symmetric	Symmetric/hermitian?

Eigen_sparse_matrix() [2/4]

template<class T>

CGAL::Eigen_sparse_matrix< T >::Eigen_sparse_matrix	(	int	dim,
		bool	is_symmetric = false )

Create a square matrix initialized with zeros.

Parameters

dim	Matrix dimension.
is_symmetric	Symmetric/hermitian?

Eigen_sparse_matrix() [3/4]

template<class T>

CGAL::Eigen_sparse_matrix< T >::Eigen_sparse_matrix	(	std::size_t	rows,
		std::size_t	columns,
		bool	is_symmetric = false )

Create a rectangular matrix initialized with zeros.

Precondition

rows == columns if is_symmetric is true.

Parameters

rows	Number of rows.
columns	Number of columns.
is_symmetric	Symmetric/hermitian?

Eigen_sparse_matrix() [4/4]

template<class T>

CGAL::Eigen_sparse_matrix< T >::Eigen_sparse_matrix	(	int	rows,
		int	columns,
		bool	is_symmetric = false )

Create a rectangular matrix initialized with zeros.

Precondition

rows == columns if is_symmetric is true.

Parameters

rows	Number of rows.
columns	Number of columns.
is_symmetric	Symmetric/hermitian?

Member Function Documentation

add_coef()

template<class T>

void CGAL::Eigen_sparse_matrix< T >::add_coef	(	std::size_t	i_,
		std::size_t	j_,
		T	val )

Write access to a matrix coefficient: a_ij <- a_ij + val.

Warning

For symmetric matrices, Eigen_sparse_matrix only stores the lower triangle add_coef() does nothing if (i, j) belongs to the upper triangle.

Precondition

0 <= i < row_dimension().

0 <= j < column_dimension().

get_coef()

template<class T>

NT CGAL::Eigen_sparse_matrix< T >::get_coef	(	std::size_t	i_,
		std::size_t	j_ ) const

Read access to a matrix coefficient.

Warning

Complexity:

• \(O(log(n))\) if the matrix is already built.
• \(O(n)\) if the matrix is not built. n being the number of entries in the matrix.

Precondition

0 <= i < row_dimension().

0 <= j < column_dimension().

set_coef()

template<class T>

void CGAL::Eigen_sparse_matrix< T >::set_coef	(	std::size_t	i_,
		std::size_t	j_,
		T	val,
		bool	new_coef = false )

Write access to a matrix coefficient: a_ij <- val.

Users can optimize calls to this function by setting 'new_coef' to true if the coefficient does not already exist in the matrix.

Warning

For symmetric matrices, Eigen_sparse_matrix only stores the lower triangle and set_coef() does nothing if (i, j) belongs to the upper triangle.

Precondition

0 <= i < row_dimension().

0 <= j < column_dimension().