matblock_math.h File Reference

Declarations for mathematical matrix block operations. More...

#include <jwsc/config.h>
#include <stdlib.h>
#include <inttypes.h>
#include "jwsc/base/error.h"
#include "jwsc/math/complex.h"
#include "jwsc/matblock/matblock.h"

Include dependency graph for matblock_math.h:

Go to the source code of this file.

Functions
add_scalar_to_matblock
Adds a scalar to a matrix block.
void	add_scalar_to_matblock_f (Matblock_f **m_out, const Matblock_f *m_in, float a)
	Adds a scalar to a single precision matrix block.
void	add_scalar_to_matblock_d (Matblock_d **m_out, const Matblock_d *m_in, double a)
	Adds a scalar to a double precision matrix block.
void	add_scalar_to_matblock_cf (Matblock_cf **m_out, const Matblock_cf *m_in, Complex_f z)
	Adds a scalar to a single precision complex matrix block.
void	add_scalar_to_matblock_cd (Matblock_cd **m_out, const Matblock_cd *m_in, Complex_d z)
	Adds a scalar to a double precision complex matrix block.
multiply_matblock_by_scalar
Multiplies a matrix block by a scalar.
void	multiply_matblock_by_scalar_f (Matblock_f **m_out, const Matblock_f *m_in, float a)
	Multiplies a single precision matrix block by a scalar.
void	multiply_matblock_by_scalar_d (Matblock_d **m_out, const Matblock_d *m_in, double a)
	Multiplies a double precision matrix block by a scalar.
void	multiply_matblock_by_scalar_cf (Matblock_cf **m_out, const Matblock_cf *m_in, Complex_f z)
	Multiplies a single precision complex matrix block by a scalar.
void	multiply_matblock_by_scalar_cd (Matblock_cd **m_out, const Matblock_cd *m_in, Complex_d z)
	Multiplies a double precision complex matrix block by a scalar.
multiply_matblocks_elt_wise
Multiplies a two matrix blocks element by element.
Error *	multiply_matblocks_elt_wise_f (Matblock_f **m_out, const Matblock_f *m_1, const Matblock_f *m_2)
	Multiples two single precision matrix blocks element by element.
Error *	multiply_matblocks_elt_wise_d (Matblock_d **m_out, const Matblock_d *m_1, const Matblock_d *m_2)
	Multiples two double precision matrix blocks element by element.
Error *	multiply_matblocks_elt_wise_cf (Matblock_cf **m_out, const Matblock_cf *m_1, const Matblock_cf *m_2)
	Multiples two single precision complex matrix blocks element by element.
Error *	multiply_matblocks_elt_wise_cd (Matblock_cd **m_out, const Matblock_cd *m_1, const Matblock_cd *m_2)
	Multiples two double precision complex matrix blocks element by element.
add_matblocks
Adds two matrix blocks element-wise.
Error *	add_matblocks_f (Matblock_f **m_out, const Matblock_f *m_1, const Matblock_f *m_2)
	Adds two single precision matrix blocks element-wise.
Error *	add_matblocks_d (Matblock_d **m_out, const Matblock_d *m_1, const Matblock_d *m_2)
	Adds two double precision matrix blocks element-wise.
subtract_matblocks
Subtracts two matrix blocks element-wise.
Error *	subtract_matblocks_f (Matblock_f **m_out, const Matblock_f *m_1, const Matblock_f *m_2)
	Subtracts two single precision matrix blocks element-wise.
Error *	subtract_matblocks_d (Matblock_d **m_out, const Matblock_d *m_1, const Matblock_d *m_2)
	Subtracts two double precision matrix blocks element-wise.
calc_matblock_asum
Calculates the absolute sum of a real matrix block.
void	calc_matblock_asum_f (float *sum_out, const Matblock_f *m)
	Calculates the absolute value sum of a single precision matrix block.
void	calc_matblock_asum_d (double *sum_out, const Matblock_d *m)
	Calculates the absolute value sum of a double precision matrix block.
normalize_matblock_sum
Normalizes a real matrix block by its sum.
void	normalize_matblock_sum_f (Matblock_f **m_out, const Matblock_f *m_in)
	Normalizes a single precision matrix block by its sum.
void	normalize_matblock_sum_d (Matblock_d **m_out, const Matblock_d *m_in)
	Normalizes a double precision matrix block by its sum.
abs_matblock
Takes the absolute value of all elements of a matrix.
void	abs_matblock_f (Matblock_f **m_out, const Matblock_f *m_in)
	Takes the absolute value of all elements of a single precision matrix.
void	abs_matblock_d (Matblock_d **m_out, const Matblock_d *m_in)
	Takes the absolute value of all elements of a double precision matrix.
threshold_matblock
Thresholds a matblock by setting all values below the threshold to zero.
void	threshold_matblock_f (Matblock_f **m_out, const Matblock_f *m_in, float threshold)
	Thresholds a single precision matrix block by setting all values below the threshold to zero.
void	threshold_matblock_d (Matblock_d **m_out, const Matblock_d *m_in, double threshold)
	Thresholds a double precision matrix block by setting all values below the threshold to zero.

---

Detailed Description

Declarations for mathematical matrix block operations.

Author:

Joseph Schlecht

License:

Creative Commons BY-NC-SA 3.0

Many of the functions are accelerated if linked against an optimized BLAS library.

Definition in file matblock_math.h.

---

Function Documentation

void add_scalar_to_matblock_f	(	Matblock_f **	m_out,
		const Matblock_f *	m_in,
		float	a
	)

Adds a scalar to a single precision matrix block.

Parameters:

m_out	Result parameter. If *m_out is NULL, a matrix block is allocated; otherwise its space is re-used.
m_in	Matrix block to add a to.
a	Scalar to add to the values of m_in.

Note:

If *m_out == m_in, then m_in is overwritten.

Definition at line 79 of file matblock_math.c.

void add_scalar_to_matblock_d	(	Matblock_d **	m_out,
		const Matblock_d *	m_in,
		double	a
	)

Adds a scalar to a double precision matrix block.

Parameters:

m_out	Result parameter. If *m_out is NULL, a matrix block is allocated; otherwise its space is re-used.
m_in	Matrix block to add a to.
a	Scalar to add to the values of m_in.

Note:

If *m_out == m_in, then m_in is overwritten.

Definition at line 115 of file matblock_math.c.

void add_scalar_to_matblock_cf	(	Matblock_cf **	m_out,
		const Matblock_cf *	m_in,
		Complex_f	z
	)

Adds a scalar to a single precision complex matrix block.

Parameters:

m_out	Result parameter. If *m_out is NULL, a matrix block is allocated; otherwise its space is re-used.
m_in	Matrix block to add a to.
z	Scalar to add to the values of m_in.

Note:

If *m_out == m_in, then m_in is overwritten.

Definition at line 151 of file matblock_math.c.

void add_scalar_to_matblock_cd	(	Matblock_cd **	m_out,
		const Matblock_cd *	m_in,
		Complex_d	z
	)

Adds a scalar to a double precision complex matrix block.

Parameters:

m_out	Result parameter. If *m_out is NULL, a matrix block is allocated; otherwise its space is re-used.
m_in	Matrix block to add a to.
z	Scalar to add to the values of m_in.

Note:

If *m_out == m_in, then m_in is overwritten.

Definition at line 188 of file matblock_math.c.

void multiply_matblock_by_scalar_f	(	Matblock_f **	m_out,
		const Matblock_f *	m_in,
		float	a
	)

Multiplies a single precision matrix block by a scalar.

Parameters:

m_out	Result parameter. If *m_out is NULL, a matrix block is allocated; otherwise its space is re-used.
m_in	Matrix block to multiply scalar with.
a	Scalar to multiply by the values of m_in.

Note:

If *m_out == m_in, then m_in is overwritten.

Definition at line 237 of file matblock_math.c.

void multiply_matblock_by_scalar_d	(	Matblock_d **	m_out,
		const Matblock_d *	m_in,
		double	a
	)

Multiplies a double precision matrix block by a scalar.

Parameters:

m_out	Result parameter. If *m_out is NULL, a matrix block is allocated; otherwise its space is re-used.
m_in	Matrix block to multiply scalar with.
a	Scalar to multiply by the values of m_in.

Note:

If *m_out == m_in, then m_in is overwritten.

Definition at line 286 of file matblock_math.c.

void multiply_matblock_by_scalar_cf	(	Matblock_cf **	m_out,
		const Matblock_cf *	m_in,
		Complex_f	z
	)

Multiplies a single precision complex matrix block by a scalar.

Parameters:

m_out	Result parameter. If *m_out is NULL, a matrix block is allocated; otherwise its space is re-used.
m_in	Matrix block to multiply scalar with.
z	Scalar to multiply by the values of m_in.

Note:

If *m_out == m_in, then m_in is overwritten.

Definition at line 335 of file matblock_math.c.

void multiply_matblock_by_scalar_cd	(	Matblock_cd **	m_out,
		const Matblock_cd *	m_in,
		Complex_d	z
	)

Multiplies a double precision complex matrix block by a scalar.

Parameters:

m_out	Result parameter. If *m_out is NULL, a matrix block is allocated; otherwise its space is re-used.
m_in	Matrix block to multiply scalar with.
z	Scalar to multiply by the values of m_in.

Note:

If *m_out == m_in, then m_in is overwritten.

Definition at line 389 of file matblock_math.c.

Error* multiply_matblocks_elt_wise_f	(	Matblock_f **	m_out,
		const Matblock_f *	m_1,
		const Matblock_f *	m_2
	)

Multiples two single precision matrix blocks element by element.

The matrix blocks must have the same dimensions.

Parameters:

m_out	Result parameter. If *m_out is NULL, a matrix block is allocated; otherwise its space is re-used.
m_1	First (left) matrix block in the multiply operation.
m_2	Second (right) matrix block in the multiply operation.

Returns:

On success, NULL is returned; otherwise an Error is returned and *m_out is freed and set to NULL (if it is not an input matrix block).

• ERROR_INV_ARG Matrix blocks m_1 and m_2 do not have the same dimensions.

Note:

If *m_out == m_[12], it is overwritten.

Definition at line 462 of file matblock_math.c.

Error* multiply_matblocks_elt_wise_d	(	Matblock_d **	m_out,
		const Matblock_d *	m_1,
		const Matblock_d *	m_2
	)

Multiples two double precision matrix blocks element by element.

Definition at line 504 of file matblock_math.c.

Error* multiply_matblocks_elt_wise_cf	(	Matblock_cf **	m_out,
		const Matblock_cf *	m_1,
		const Matblock_cf *	m_2
	)

Multiples two single precision complex matrix blocks element by element.

Definition at line 546 of file matblock_math.c.

Error* multiply_matblocks_elt_wise_cd	(	Matblock_cd **	m_out,
		const Matblock_cd *	m_1,
		const Matblock_cd *	m_2
	)

Multiples two double precision complex matrix blocks element by element.

Definition at line 595 of file matblock_math.c.

Error* add_matblocks_f	(	Matblock_f **	m_out,
		const Matblock_f *	m_1,
		const Matblock_f *	m_2
	)

Adds two single precision matrix blocks element-wise.

The matrix block must have equal dimensions.

Parameters:

m_out	Result parameter. If *m_out is NULL, a matrix block is allocated; otherwise its space is re-used.
m_1	Matrix block to add.
m_2	Matrix block to add.

Returns:

On success, NULL is returned. On error, an Error is returned and *m_out is freed and set to NULL (if it is not m_1 or m_2).

• ERROR_INV_ARG Matrix blocks m_1 and m_2 do not have equal dimensions.

Note:

If *m_out == m_[12], it is overwritten.

Definition at line 671 of file matblock_math.c.

Error* add_matblocks_d	(	Matblock_d **	m_out,
		const Matblock_d *	m_1,
		const Matblock_d *	m_2
	)

Adds two double precision matrix blocks element-wise.

The matrix block must have equal dimensions.

Parameters:

m_out	Result parameter. If *m_out is NULL, a matrix block is allocated; otherwise its space is re-used.
m_1	Matrix block to add.
m_2	Matrix block to add.

Returns:

On success, NULL is returned. On error, an Error is returned and *m_out is freed and set to NULL (if it is not m_1 or m_2).

• ERROR_INV_ARG Matrix blocks m_1 and m_2 do not have equal dimensions.

Note:

If *m_out == m_[12], it is overwritten.

Definition at line 761 of file matblock_math.c.

Error* subtract_matblocks_f	(	Matblock_f **	m_out,
		const Matblock_f *	m_1,
		const Matblock_f *	m_2
	)

Subtracts two single precision matrix blocks element-wise.

The matrix block must have equal dimensions.

Parameters:

m_out	Result parameter. If *m_out is NULL, a matrix block is allocated; otherwise its space is re-used.
m_1	Matrix block to subtract.
m_2	Matrix block to subtract.

Returns:

On success, NULL is returned. On error, an Error is returned and *m_out is freed and set to NULL (if it is not m_1 or m_2).

• ERROR_INV_ARG Matrix blocks m_1 and m_2 do not have equal dimensions.

Note:

If *m_out == m_[12], it is overwritten.

Definition at line 863 of file matblock_math.c.

Error* subtract_matblocks_d	(	Matblock_d **	m_out,
		const Matblock_d *	m_1,
		const Matblock_d *	m_2
	)

Subtracts two double precision matrix blocks element-wise.

The matrix block must have equal dimensions.

Parameters:

m_out	Result parameter. If *m_out is NULL, a matrix block is allocated; otherwise its space is re-used.
m_1	Matrix block to subtract.
m_2	Matrix block to subtract.

Returns:

On success, NULL is returned. On error, an Error is returned and *m_out is freed and set to NULL (if it is not m_1 or m_2).

• ERROR_INV_ARG Matrix blocks m_1 and m_2 do not have equal dimensions.

Note:

If *m_out == m_[12], it is overwritten.

Definition at line 957 of file matblock_math.c.

void calc_matblock_asum_f	(	float *	sum_out,
		const Matblock_f *	m
	)

Calculates the absolute value sum of a single precision matrix block.

Parameters:

sum_out	Result parameter. Sum of the absolute values of m.
m	Matrix block to calculate the sum of.

Definition at line 1051 of file matblock_math.c.

void calc_matblock_asum_d	(	double *	sum_out,
		const Matblock_d *	m
	)

Calculates the absolute value sum of a double precision matrix block.

Parameters:

sum_out	Result parameter. Sum of the absolute values of m.
m	Matrix block to calculate the sum of.

Definition at line 1075 of file matblock_math.c.

void normalize_matblock_sum_f	(	Matblock_f **	m_out,
		const Matblock_f *	m_in
	)

Normalizes a single precision matrix block by its sum.

Parameters:

m_out	Result parameter. If *m_out is NULL, a matrix block is allocated; otherwise its space is re-used.
m_in	Matrix block to normalize.

Note:

If *m_out == m_in, then m_in is overwritten.

Definition at line 1114 of file matblock_math.c.

void normalize_matblock_sum_d	(	Matblock_d **	m_out,
		const Matblock_d *	m_in
	)

Normalizes a double precision matrix block by its sum.

Parameters:

m_out	Result parameter. If *m_out is NULL, a matrix block is allocated; otherwise its space is re-used.
m_in	Matrix block to normalize.

Note:

If *m_out == m_in, then m_in is overwritten.

Definition at line 1137 of file matblock_math.c.

void abs_matblock_f	(	Matblock_f **	m_out,
		const Matblock_f *	m_in
	)

Takes the absolute value of all elements of a single precision matrix.

Parameters:

m_out	Result parameter. If *m_out is NULL, a matrix block is allocated; otherwise its space is re-used.
m_in	Matrix block to take the absolute value of.

Note:

If *m_out == m_in, then m_in is overwritten.

Definition at line 1170 of file matblock_math.c.

void abs_matblock_d	(	Matblock_d **	m_out,
		const Matblock_d *	m_in
	)

Takes the absolute value of all elements of a double precision matrix.

Parameters:

m_out	Result parameter. If *m_out is NULL, a matrix block is allocated; otherwise its space is re-used.
m_in	Matrix block to take the absolute value of.

Note:

If *m_out == m_in, then m_in is overwritten.

Definition at line 1201 of file matblock_math.c.

void threshold_matblock_f	(	Matblock_f **	m_out,
		const Matblock_f *	m_in,
		float	threshold
	)

Thresholds a single precision matrix block by setting all values below the threshold to zero.

Parameters:

m_out	Result parameter. If *m_out is NULL, a matrix block is allocated; otherwise its space is re-used.
m_in	Matrix block to threshold.
threshold	Elements below this value will be set to zero

Note:

If *m_out == m_in, then m_in is overwritten.

Definition at line 1243 of file matblock_math.c.

void threshold_matblock_d	(	Matblock_d **	m_out,
		const Matblock_d *	m_in,
		double	threshold
	)

Thresholds a double precision matrix block by setting all values below the threshold to zero.

Parameters:

m_out	Result parameter. If *m_out is NULL, a matrix block is allocated; otherwise its space is re-used.
m_in	Matrix block to threshold.
threshold	Elements below this value will be set to zero

Note:

If *m_out == m_in, then m_in is overwritten.

Definition at line 1277 of file matblock_math.c.