SerialMatrix< T > Class Template Reference

Class for managing a (serial) matrix stored in memory. More...

#include <SMatrix.h>

Public Member Functions
int	global2Local (const int &row, const int &col)
	SerialMatrix (const int &numRows, const int &numCols, const int &numBlasRows=0, const int &numBlasCols=0, const int &numBlocksRows=0, const int &numBlocksCols=0)
	Default SMatrix constructor. More...
	SerialMatrix ()
	Default constructor.
	~SerialMatrix ()
	Destructor, to delete raw buffer.
	SerialMatrix (const SerialMatrix< T > &that)
	Copy constructor.
SerialMatrix< T > &	operator= (const SerialMatrix< T > &that)
	Copy constructor.
std::vector< std::tuple< int, int > >	getAllLocalStates ()
	Find the global indices of the matrix elements that are stored locally by the current MPI process.
bool	indicesAreLocal (const int &row, const int &col)
	Returns true if the global indices (row,col) identify a matrix element stored by the MPI process.
int	rows () const
	Find global number of rows.
int	localRows () const
	Return local number of rows.
int	cols () const
	Find global number of columns.
int	localCols () const
	Return local number of rows.
size_t	size () const
	Find global number of matrix elements.
T &	operator() (const int &row, const int &col)
	Get and set operator.
const T &	operator() (const int &row, const int &col) const
	Const get and set operator.
void	enforcePositiveSemiDefinite ()
	A function to fix up a matrix with a bit of noise to be positive semi-definite, which currently does not have a SMatrix implementation, as it's only used by distributed scattering matrix calculations.
void	outputToHDF5 (const std::string &outFileName, const std::string &dataSetName)
	A dummy function not implemented, only for PMatrix currently. More...
SerialMatrix< T >	prod (const SerialMatrix< T > &that, const char &trans1=transN, const char &trans2=transN)
	Matrix-matrix multiplication. More...
SerialMatrix< T >	operator+= (const SerialMatrix< T > &m1)
	Matrix-matrix addition.
SerialMatrix< T >	operator-= (const SerialMatrix< T > &m1)
	Matrix-matrix subtraction.
SerialMatrix< T >	operator*= (const T &that)
	Matrix-scalar multiplication.
SerialMatrix< T >	operator/= (const T &that)
	Matrix-scalar division.
void	eye ()
	Sets this matrix as the identity. More...
std::tuple< std::vector< double >, SerialMatrix< T > >	diagonalize ()
	Diagonalize a complex-hermitian / real-symmetric matrix. More...
std::tuple< std::vector< double >, SerialMatrix< T > >	diagonalize (int numEigenvalues, bool checkNegativeEigenvalues=true)
	Diagonalize a complex-hermitian / real sym matrix for only some of its eigenvalues. More...
double	squaredNorm ()
	Computes the squared Frobenius norm of the matrix (or Euclidean norm, or L2 norm of the matrix)
double	norm ()
	Computes the Frobenius norm of the matrix (or Euclidean norm, or L2 norm of the matrix).
T	dot (const SerialMatrix< T > &that)
	Computes a "scalar product" between two matrices A and B, defined as \sum_ij A_ij * B_ij. More...
SerialMatrix< T >	operator- () const
	Unary negation.
void	symmetrize ()
	Symmetrize the matrix.
SerialMatrix< std::complex< double > >	prod (const SerialMatrix< std::complex< double >> &that, const char &trans1, const char &trans2)
std::tuple< std::vector< double >, SerialMatrix< std::complex< double > > >	diagonalize ()
double	norm ()
double	norm ()

Static Public Attributes
static const char	transN = 'N'
	Indicates that the matrix A is not modified: transN(A) = A.
static const char	transT = 'T'
	Indicates that the matrix A is taken as its transpose: transT(A) = A^T.
static const char	transC = 'C'
	Indicates that the matrix A is taken as its adjoint: transC(A) = A^+.

Detailed Description

template<typename T>
class SerialMatrix< T >

It mirrors the functionality of ParallelMatrix, only that in this case it doesn't use the scalapack library and instead relies on BLAS. Data is NOT distributed across different MPI processes. At the moment is used for making the code compile without MPI.

Templated matrix class with specialization for double and complex<double>.

Constructor & Destructor Documentation

SerialMatrix()

template<typename T >

SerialMatrix< T >::SerialMatrix	(	const int &	numRows,
		const int &	numCols,
		const int &	numBlasRows = 0,
		const int &	numBlasCols = 0,
		const int &	numBlocksRows = 0,
		const int &	numBlocksCols = 0
	)

SerialMatrix elements are set to zero upon initialization.

Parameters

numRows	number of rows of the matrix
numCols	number of columns of the matrix.
numBlocksRows,numBlocksCols	these parameters are ignored and are put here for mirroring the interface of ParallelMatrix.

Member Function Documentation

diagonalize() [1/2]

template<typename T >

std::tuple<std::vector<double>, SerialMatrix<T> > SerialMatrix< T >::diagonalize

(

)

Nota bene: we don't check if it's hermitian/symmetric.

diagonalize() [2/2]

template<typename T >

std::tuple<std::vector<double>, SerialMatrix<T> > SerialMatrix< T >::diagonalize	(	int	numEigenvalues,
		bool	checkNegativeEigenvalues = true
	)

Note: this isn't implemented for now and just calls the diagonalize() function

dot()

template<typename T >

T SerialMatrix< T >::dot

(

const SerialMatrix< T > &

that

)

For vectors, this reduces to the standard scalar product.

eye()

template<typename T >

void SerialMatrix< T >::eye

Deletes any previous content.

outputToHDF5()

template<typename T >

void SerialMatrix< T >::outputToHDF5 ( const std::string &  outFileName, const std::string &  dataSetName  )

inline

Parameters

outFileName	string naming the output file
dataSetName	keyname for data in the output file

prod()

template<typename T >

SerialMatrix<T> SerialMatrix< T >::prod	(	const SerialMatrix< T > &	that,
		const char &	trans1 = transN,
		const char &	trans2 = transN
	)

Computes result = trans1(*this) * trans2(that) where trans(1/2( can be "N" (matrix as is), "T" (transpose) or "C" adjoint (these flags are used so that the transposition/adjoint operation is never directly operated on the stored values)

Parameters

that	the matrix to multiply "this" with
trans2	controls transposition of "this" matrix
trans1	controls transposition of "that" matrix

Returns

result: a ParallelMatrix object.