Lapack++: LaGenMatDouble Class Reference

typedef double LaGenMatDouble::value_type

The type of the value elements.

typedef LaGenMatDouble LaGenMatDouble::matrix_type

Convenience typedef of this class to itself to make common function definitions easier. (New in lapackpp-2.4.5)

typedef VectorDouble LaGenMatDouble::vec_type

Internal wrapper type; don't use that in an application.

LaGenMatDouble::LaGenMatDouble ( )

Constructs a null 0x0 matrix.

LaGenMatDouble::LaGenMatDouble	(	int	m,
		int	n
	)

Constructs a column-major matrix of size $m\times n$ . Matrix elements are NOT initialized!

LaGenMatDouble::LaGenMatDouble	(	double *	v,
		int	m,
		int	n,
		bool	row_ordering = `false`
	)

Constructs an $m\times n$ matrix by using the values from the one-dimensional C array v of length m*n.

Note:: If row_ordering is false, then the data will not be copied but instead the C array will be shared (shallow copy). In that case, you must not delete the C array as long as you use this newly created matrix. Also, if you need a copy (deep copy), construct one matrix A by this constructor, and then copy this content into a second matrix by B.copy(A). On the other hand, if row_ordering is true, then the data will be copied immediately (deep copy).

Parameters:

	v	The one-dimensional C array of size `mn` whose data should be used. If `row_ordering` is `false`, then the data will not* be copied but shared (shallow copy). If `row_ordering` is `true`, then the data will be copied (deep copy).
	m	The number of rows in the new matrix.
	n	The number of columns in the new matrix.
	row_ordering	If `false`, then the C array is used in column-order, i.e. the first `m` elements of `v` are used as the first column of the matrix, the next `m` elements are the second column and so on. (This is the default and this is also the internal storage format in order to be compatible with the underlying Fortran subroutines.) If this is `true`, then the C array is used in row-order, i.e. the first `n` elements of `v` are used as the first row of the matrix, the next `n` elements are the second row and so on. (Internally, this is achieved by allocating a new copy of the array and copying the array into the internal ordering.)

LaGenMatDouble::LaGenMatDouble ( const LaGenMatDouble & )

Create a new matrix from an existing one by copying.

Watch out! Due to the C++ "named return value optimization" you cannot use this as an alias for copy() when declaring a variable if the right-side is a return value of operator(). More precisely, you cannot write the following:

       LaGenMatDouble x( y(LaIndex(),LaIndex()) ); // erroneous reference copy!

Instead, if the initialization should create a new copy of the right-side matrix, you have to write it this way:

       LaGenMatDouble x( y(LaIndex(),LaIndex()).copy() ); // correct deep-copy

Or this way:

       LaGenMatDouble x;
       x = y(LaIndex(),LaIndex()); // correct deep-copy

LaGenMatDouble::LaGenMatDouble ( const LaGenMatFloat & ) [explicit]

Create a new matrix from an existing one by copying and converting each element from a float to a double.

virtual LaGenMatDouble::~LaGenMatDouble ( ) [virtual]

Destroy matrix and reclaim vector memory space if this is the only structure using it.

LaGenMatDouble& LaGenMatDouble::resize	(	int	m,
		int	n
	)

Resize to a new matrix of size m x n. The element values of the new matrix are uninitialized, even if resizing to a smaller matrix.

Reimplemented in LaVectorDouble.

LaGenMatDouble& LaGenMatDouble::resize ( const LaGenMatDouble & s )

Resize to a new matrix of the same size as the given matrix s. The element values of the new matrix are uninitialized, even if resizing to a smaller matrix.

bool LaGenMatDouble::is_zero ( ) const

Returns true if this is an all-zero matrix. (New in lapackpp-2.4.5)

bool LaGenMatDouble::is_submatrixview ( ) const [inline]

Returns true if this matrix is only a submatrix view of another (larger) matrix. (New in lapackpp-2.4.4)

bool LaGenMatDouble::has_unitstride ( ) const [inline]

Returns true if this matrix has unit stride.

This is a necessary condition for not being a submatrix view, but it's not sufficient. (New in lapackpp-2.4.4)

bool LaGenMatDouble::equal_to ( const LaGenMatDouble & mat ) const

Returns true if the given matrix mat is exactly equal to this object. (New in lapackpp-2.4.5)

int LaGenMatDouble::size ( int d ) const [inline]

Returns the length n of the dth dimension, i.e. for a M x N matrix, size(0) returns M and size(1) returns N.

int LaGenMatDouble::cols ( ) const [inline]

Returns the number of columns, i.e. for a M x N matrix this returns N. New in lapackpp-2.4.4.

int LaGenMatDouble::rows ( ) const [inline]

Returns the number of rows, i.e. for a M x N matrix this returns M. New in lapackpp-2.4.4.

int LaGenMatDouble::inc ( int d ) const [inline]

Returns the distance between memory locations (in terms of number of elements) between consecutive elements along dimension d. For example, if inc(d) returns 1, then elements along the dth dimension are contiguous in memory.

int LaGenMatDouble::gdim ( int d ) const [inline]

Returns the global dimensions of the (possibly larger) matrix owning this space. This will only differ from size(d) if the current matrix is actually a submatrix view of some larger matrix.

int LaGenMatDouble::start ( int d ) const [inline]

If the memory space used by this matrix is viewed as a linear array, start(d) returns the starting offset of the first element in dimension d. (See LaIndex class.)

int LaGenMatDouble::end ( int d ) const [inline]

If the memory space used by this matrix is viewed as a linear array, end(d) returns the starting offset of the last element in dimension d. (See LaIndex class.)

LaIndex LaGenMatDouble::index ( int d ) const [inline]

Returns the index specifying this submatrix view in dimension d. (See LaIndex class.) This will only differ from a unit-stride index if the current matrix is actually a submatrix view of some larger matrix.

int LaGenMatDouble::ref_count ( ) const [inline]

Returns the number of data objects which utilize the same (or portions of the same) memory space used by this matrix.

double * LaGenMatDouble::addr ( ) const [inline]

Returns the memory address of the first element of the matrix. G.addr() is equivalent to &G(0,0) .

double & LaGenMatDouble::operator()	(	int	i,
		int	j
	)			`[inline]`

Returns the $(i,j)$ th element of this matrix, with the indices i and j starting at zero (zero-based offset). This means you have

$A_{n\times m} = \left(\begin{array}{ccc} a_{11} & & a_{1m} \\ & \ddots & \\ a_{n1} & & a_{nm} \end{array}\right)$

but for accessing the element $a_{11}$ you have to write A(0,0).

Optional runtime bounds checking (0<=i<m, 0<=j<n) is set by the compile time macro LA_BOUNDS_CHECK.

double & LaGenMatDouble::operator()	(	int	i,
		int	j
	)			const `[inline]`

Returns the $(i,j)$ th element of this matrix, with the indices i and j starting at zero (zero-based offset). This means you have

$A_{n\times m} = \left(\begin{array}{ccc} a_{11} & & a_{1m} \\ & \ddots & \\ a_{n1} & & a_{nm} \end{array}\right)$

but for accessing the element $a_{11}$ you have to write A(0,0).

Optional runtime bounds checking (0<=i<m, 0<=j<n) is set by the compile time macro LA_BOUNDS_CHECK.

LaGenMatDouble LaGenMatDouble::operator()	(	const LaIndex &	I,
		const LaIndex &	J
	)

Return a submatrix view specified by the indices I and J. (See LaIndex class.) These indices specify start, increment, and ending offsets, similar to triplet notation of Matlab or Fortran 90. For example, if B is a 10 x 10 matrix, I is (0:2:2) and J is (3:1:4), then B(I,J) denotes the 2 x 2 matrix

$\left(\begin{array}{cc} b_{0,3} & b_{2,3} \\ b_{0,4} & b_{4,4} \end{array}\right)$

LaGenMatDouble LaGenMatDouble::operator()	(	const LaIndex &	I,
		const LaIndex &	J
	)			const

Return a submatrix view specified by the indices I and J. (See LaIndex class.) These indices specify start, increment, and ending offsets, similar to triplet notation of Matlab or Fortran 90. For example, if B is a 10 x 10 matrix, I is (0:2:2) and J is (3:1:4), then B(I,J) denotes the 2 x 2 matrix

$\left(\begin{array}{cc} b_{0,3} & b_{2,3} \\ b_{0,4} & b_{4,4} \end{array}\right)$

LaGenMatDouble LaGenMatDouble::row ( int k )

Returns a submatrix view for the specified row k of this matrix.

The returned object references still the same memory as this object, so if you modify elements, they will appear modified in both objects. (New in lapackpp-2.4.6)

LaGenMatDouble LaGenMatDouble::row ( int k ) const

Returns a submatrix view for the specified row k of this matrix.

The returned object references still the same memory as this object, so if you modify elements, they will appear modified in both objects. (New in lapackpp-2.4.6)

LaGenMatDouble LaGenMatDouble::col ( int k )

Returns a submatrix view for the specified column k of this matrix.

The returned object references still the same memory as this object, so if you modify elements, they will appear modified in both objects. (New in lapackpp-2.4.6)

LaGenMatDouble LaGenMatDouble::col ( int k ) const

Returns a submatrix view for the specified column k of this matrix.

The returned object references still the same memory as this object, so if you modify elements, they will appear modified in both objects. (New in lapackpp-2.4.6)

LaGenMatDouble& LaGenMatDouble::operator= ( double s )

Set elements of left-hand side to the scalar value s. No new matrix is created, so that if there are other matrices that reference this memory space, they will also be affected.

Reimplemented in LaColVectorDouble, LaRowVectorDouble, and LaVectorDouble.

LaGenMatDouble& LaGenMatDouble::operator= ( const LaGenMatDouble & s )

Release left-hand side (reclaiming memory space if possible) and copy elements of elements of s. Unline inject(), it does not require conformity, and previous references of left-hand side are unaffected.

This is an alias for copy().

Watch out! Due to the C++ "named return value optimization" you cannot use this as an alias for copy() when declaring a variable if the right-side is a return value of operator(). More precisely, you cannot write the following:

       LaGenMatDouble x = y(LaIndex(),LaIndex()); // erroneous reference copy!

Instead, if the initialization should create a new copy of the right-side matrix, you have to write it this way:

       LaGenMatDouble x = y(LaIndex(),LaIndex()).copy(); // correct deep-copy

Or this way:

       LaGenMatDouble x;
       x = y(LaIndex(),LaIndex()); // correct deep-copy

Note: The manual for lapack++-1.1 claimed that this operator would be an alias for ref(), not for copy(), i.e. this operator creates a reference instead of a deep copy. However, since that confused many people, the behaviour was changed so that B=A will now create B as a deep copy instead of a reference. If you want a reference, please write B.ref(A) explicitly.

Reimplemented in LaColVectorDouble, LaRowVectorDouble, and LaVectorDouble.

LaGenMatDouble& LaGenMatDouble::operator+= ( double s )

Add the scalar value s to elements of left-hand side. No new matrix is created, so that if there are other matrices that reference this memory space, they will also be affected.

Note:: This method is rather slow. In many cases, it can be much faster to use Blas_Mat_Mult() with a Ones-Matrix instead.

LaGenMatDouble& LaGenMatDouble::add ( double s )

Add the scalar value s to elements of left-hand side. No new matrix is created, so that if there are other matrices that reference this memory space, they will also be affected. (New in lapackpp-2.4.7.)

LaGenMatDouble& LaGenMatDouble::operator *= ( double s )

Scale the left-hand side matrix by the given scalar value. No new matrix is created, so that if there are other matrices that reference this memory space, they will also be affected. (New in lapackpp-2.4.7.)

LaGenMatDouble& LaGenMatDouble::scale ( double s )

Scale the left-hand side matrix by the given scalar value. No new matrix is created, so that if there are other matrices that reference this memory space, they will also be affected. (New in lapackpp-2.4.7.)

LaGenMatDouble& LaGenMatDouble::inject ( const LaGenMatDouble & s )

Copy elements of s into the memory space referenced by the left-hand side, without first releasing it. The effect is that if other matrices share memory with left-hand side, they too will be affected. Note that the size of s must be the same as that of the left-hand side matrix.

Note:: If you rather wanted to create a new copy of s, you should use copy() instead.

Reimplemented in LaVectorDouble.

LaGenMatDouble& LaGenMatDouble::copy ( const LaGenMatDouble & s )

Release left-hand side (reclaiming memory space if possible) and copy elements of elements of s. Unline inject(), it does not require conformity, and previous references of left-hand side are unaffected.

Reimplemented in LaVectorDouble.

LaGenMatDouble LaGenMatDouble::copy ( ) const

Returns a newly allocated matrix that is an element-by-element copy of this matrix.

New in lapackpp-2.5.2

LaGenMatDouble & LaGenMatDouble::shallow_assign ( ) [inline]

This is an optimization for returning temporary matrices from functions, without copying. The shallow_assign() function essentially sets an internal flag which instructs the X::X(&X) copy constructor to avoid the copying.

LaGenMatDouble& LaGenMatDouble::ref ( const LaGenMatDouble & s )

Let this matrix reference the given matrix s, so that the given matrix memory s is now referenced by multiple objects (by the given object s and now also by this object). Handle this with care!

This function releases any previously referenced memory of this object.

Reimplemented in LaVectorDouble.

LaGenMatDouble LaGenMatDouble::repmat	(	int	M,
		int	N
	)			const

Returns a newly allocated large matrix that consists of M-by-N copies of the given matrix. (New in lapackpp-2.4.5.)

value_type LaGenMatDouble::trace ( ) const

Returns the trace, i.e. the sum of all diagonal elements of the matrix. (New in lapackpp-2.4.5)

LaGenMatDouble LaGenMatDouble::diag ( ) const

Returns a newly allocated column vector of dimension Nx1 that contains the diagonal of the given matrix. (New in lapackpp-2.4.5)

int LaGenMatDouble::shallow ( ) const [inline]

Returns global shallow flag

int LaGenMatDouble::debug ( ) const [inline]

Returns global debug flag

int LaGenMatDouble::debug ( int d ) [inline]

Set global debug flag

const LaGenMatDouble& LaGenMatDouble::info ( ) const [inline]

use as in

std::cout << B.info() << std::endl;

this *info_ member is unique in that it really isn't part of the matrix info, just a flag as to how to print it. We've included in this beta release as part of our testing, but we do not expect it to be user accessable.

std::ostream& LaGenMatDouble::Info ( std::ostream & s ) const [inline]

Print the matrix info (not the actual elements) to the given ostream.

LaGenMatComplex LaGenMatDouble::to_LaGenMatComplex ( ) const

Convert this matrix to a complex matrix with imaginary part zero.

LaGenMatFloat LaGenMatDouble::to_LaGenMatFloat ( ) const

Convert this matrix to a float (floating-point single precision) matrix.

LaGenMatInt LaGenMatDouble::to_LaGenMatInt ( ) const

Convert this matrix to an int matrix.

LaGenMatLongInt LaGenMatDouble::to_LaGenMatLongInt ( ) const

Convert this matrix to a long int matrix.

static LaGenMatDouble LaGenMatDouble::zeros	(	int	N,
		int	M = `0`
	)			`[static]`

Returns a newly allocated all-zero matrix of dimension NxN, if M is not given, or NxM if M is given. (New in lapackpp-2.4.5)

static LaGenMatDouble LaGenMatDouble::ones	(	int	N,
		int	M = `0`
	)			`[static]`

Returns a newly allocated all-one matrix of dimension NxN, if M is not given, or NxM if M is given. (New in lapackpp-2.4.5)

static LaGenMatDouble LaGenMatDouble::eye	(	int	N,
		int	M = `0`
	)			`[static]`

Returns a newly allocated identity matrix of dimension NxN, if M is not given, or a rectangular matrix NxM if M is given. (New in lapackpp-2.4.5)

static LaGenMatDouble LaGenMatDouble::rand	(	int	N,
		int	M,
		value_type	low = `0`,
		value_type	high = `1`
	)			`[static]`

Returns a newly allocated matrix of dimension NxM with pseudo-random values. The values are uniformly distributed in the interval (0,1) or, if specified, (low,high). (New in lapackpp-2.4.5)

Note: Since this uses the system's rand() call, the randomness of the values might be questionable -- don't use this if you need really strong random numbers.

static LaGenMatDouble LaGenMatDouble::from_diag ( const LaGenMatDouble & vect ) [static]

Returns a newly allocated diagonal matrix of dimension NxN that has the vector vect of length N on the diagonal. (New in lapackpp-2.4.5)

static LaGenMatDouble LaGenMatDouble::linspace	(	value_type	start,
		value_type	end,
		int	nr_points
	)			`[static]`

Returns a newly allocated linarly spaced column vector with nr_points elements, between and including start and end. (New in lapackpp-2.4.5.)

std::ostream& operator<<	(	std::ostream &	,
		const LaGenMatDouble &
	)			`[friend]`

Print the matrix to the given output stream. If the matrix info flag is set, then this prints only the matrix info, see LaGenMatDouble::info(). Otherwise all matrix elements are printed.

See also:: LaPreferences::setPrintFormat()


Declaration
	LaGenMatDouble ()
	LaGenMatDouble (int m, int n)
	LaGenMatDouble (double *v, int m, int n, bool row_ordering=false)
	LaGenMatDouble (const LaGenMatDouble &)
	LaGenMatDouble (const LaGenMatFloat &)
LaGenMatDouble &	resize (int m, int n)
LaGenMatDouble &	resize (const LaGenMatDouble &s)
virtual	~LaGenMatDouble ()
Information Predicates
bool	is_zero () const
bool	is_submatrixview () const
bool	has_unitstride () const
bool	equal_to (const LaGenMatDouble &mat) const
Information
int	size (int d) const
int	cols () const
int	rows () const
int	inc (int d) const
int	gdim (int d) const
int	start (int d) const
int	end (int d) const
LaIndex	index (int d) const
int	ref_count () const
double *	addr () const
Access functions
double &	operator() (int i, int j)
double &	operator() (int i, int j) const
LaGenMatDouble	operator() (const LaIndex &I, const LaIndex &J)
LaGenMatDouble	operator() (const LaIndex &I, const LaIndex &J) const
LaGenMatDouble	row (int k)
LaGenMatDouble	row (int k) const
LaGenMatDouble	col (int k)
LaGenMatDouble	col (int k) const
Assignments
LaGenMatDouble &	operator= (double s)
LaGenMatDouble &	operator= (const LaGenMatDouble &s)
LaGenMatDouble &	operator+= (double s)
LaGenMatDouble &	add (double s)
LaGenMatDouble &	operator *= (double s)
LaGenMatDouble &	scale (double s)
LaGenMatDouble &	inject (const LaGenMatDouble &s)
LaGenMatDouble &	copy (const LaGenMatDouble &s)
LaGenMatDouble	copy () const
LaGenMatDouble &	shallow_assign ()
LaGenMatDouble &	ref (const LaGenMatDouble &s)
Expensive access functions
LaGenMatDouble	repmat (int M, int N) const
value_type	trace () const
LaGenMatDouble	diag () const
Debugging information
int	shallow () const
int	debug () const
int	debug (int d)
const LaGenMatDouble &	info () const
std::ostream &	Info (std::ostream &s) const
Matrix type conversions
LaGenMatComplex	to_LaGenMatComplex () const
LaGenMatFloat	to_LaGenMatFloat () const
LaGenMatInt	to_LaGenMatInt () const
LaGenMatLongInt	to_LaGenMatLongInt () const
Constructors for elementary matrices
static LaGenMatDouble	zeros (int N, int M=0)
static LaGenMatDouble	ones (int N, int M=0)
static LaGenMatDouble	eye (int N, int M=0)
static LaGenMatDouble	rand (int N, int M, value_type low=0, value_type high=1)
static LaGenMatDouble	from_diag (const LaGenMatDouble &vect)
static LaGenMatDouble	linspace (value_type start, value_type end, int nr_points)
Public Types
typedef double	value_type
typedef LaGenMatDouble	matrix_type
typedef VectorDouble	vec_type
Friends
std::ostream &	operator<< (std::ostream &, const LaGenMatDouble &)

LaGenMatDouble Class Reference

Declaration

Information Predicates

Information

Access functions

Assignments

Expensive access functions

Debugging information

Matrix type conversions

Constructors for elementary matrices

Public Types

Friends

Detailed Description

Member Typedef Documentation

Constructor & Destructor Documentation

Member Function Documentation

Friends And Related Function Documentation