Basic linear algebra functions. More...

This graph shows which files directly or indirectly include this file:

Functions
void	blas_gemm (const int TA, const int TB, const int M, const int N, const int K, const double ALPHA, const double A, const int lda, const double B, const int ldb, const double BETA, double *C, const int ldc)
	Performs the matrix-matrix multiplication: \( C = \alpha \mbox{op}(A) \mbox{op}(B) + \beta C \). More...

void	blas_axpy (const int N, const double ALPHA, const double X, const int INCX, double Y, const int INCY)
	Multiplies vector X by the scalar ALPHA and adds it to the vector Y. More...

void	blas_mul (const int N, const double X, const int INCX, double Y, const int INCY)
	Multiplies vector X by the vector Y and stores the result in vector Y. More...

void	blas_scal (const int N, const double ALPHA, double *X, const int INCX)
	Scales vector X by the scalar ALPHA and overwrites it with the result. More...

void	blas_fill (const int N, const double ALPHA, double *X, const int INCX)
	Fills the vector X with the value ALPHA. More...

double	blas_dot (const int N, const double X, const int INCX, const double Y, const int INCY)
	Computes the dot product of two vectors. More...

double	blas_sum (const double *X, const int N)
	Returns the sum of the vector X. More...

Detailed Description

Basic linear algebra functions.

Author: Richard Preen rpree.nosp@m.n@gm.nosp@m.ail.c.nosp@m.om

Copyright: The Authors.

Date: 2020.

Definition in file blas.h.

Function Documentation

◆ blas_axpy()

void blas_axpy	(	const int	N,
		const double	ALPHA,
		const double *	X,
		const int	INCX,
		double *	Y,
		const int	INCY
	)

Multiplies vector X by the scalar ALPHA and adds it to the vector Y.

Parameters

[in]	N	The number of elements in vectors X and Y.
[in]	ALPHA	Scalar used for multiplication.
[in]	X	Vector with N elements.
[in]	INCX	Stride between consecutive elements of X.
[in,out]	Y	Vector with N elements.
[in]	INCY	Stride between consecutive elements of Y.

Definition at line 138 of file blas.c.

Referenced by neural_layer_connected_backward(), neural_layer_connected_update(), neural_layer_convolutional_update(), neural_layer_lstm_backward(), neural_layer_lstm_forward(), neural_layer_recurrent_forward(), pred_nlms_update(), and pred_rls_update().

Here is the caller graph for this function:

◆ blas_dot()

double blas_dot	(	const int	N,
		const double *	X,
		const int	INCX,
		const double *	Y,
		const int	INCY
	)

Computes the dot product of two vectors.

Parameters

[in]	N	The number of elements in vectors X and Y.
[in]	X	Vector with N elements.
[in]	INCX	Stride between consecutive elements of X.
[in]	Y	Vector with N elements.
[in]	INCY	Stride between consecutive elements of Y.

Returns: The resulting dot product.

Definition at line 198 of file blas.c.

Referenced by pred_nlms_compute(), pred_nlms_update(), pred_rls_compute(), and pred_rls_update().

Here is the caller graph for this function:

◆ blas_fill()

void blas_fill	(	const int	N,
		const double	ALPHA,
		double *	X,
		const int	INCX
	)

Fills the vector X with the value ALPHA.

Parameters

[in]	N	The number of elements in vector X.
[in]	ALPHA	The value to fill the vector.
[out]	X	Vector with N elements.
[in]	INCX	Stride between consecutive elements of X.

Definition at line 181 of file blas.c.

Referenced by pred_nlms_init(), and pred_rls_init().

Here is the caller graph for this function:

◆ blas_gemm()

void blas_gemm	(	const int	TA,
		const int	TB,
		const int	M,
		const int	N,
		const int	K,
		const double	ALPHA,
		const double *	A,
		const int	lda,
		const double *	B,
		const int	ldb,
		const double	BETA,
		double *	C,
		const int	ldc
	)

Performs the matrix-matrix multiplication: \( C = \alpha \mbox{op}(A) \mbox{op}(B) + \beta C \).

Parameters

[in]	TA	Operation op(A) that is non- or (conj.) transpose.
[in]	TB	Operation op(B) that is non- or (conj.) transpose.
[in]	M	Number of rows of matrix op(A) and C.
[in]	N	Number of columns of matrix op(B) and C.
[in]	K	Number of columns of op(A) and rows of op(B).
[in]	ALPHA	Scalar used for multiplication.
[in]	A	Array of dimension lda × K with lda >= max(1,M) if TA=0 and lda × M with lda >= max(1,K) otherwise.
[in]	lda	Leading dimension of a 2-D array used to store the matrix A.
[in]	B	Array of dimension ldb × N with ldb >= max(1,K) if TB=0 and ldb × K with ldb >= max(1,N) otherwise.
[in]	ldb	Leading dimension of a 2-D array used to store the matrix B.
[in]	BETA	Scalar used for multiplication.
[in,out]	C	Array of dimension ldc × N with ldc >= max(1,M).
[in]	ldc	Leading dimension of a 2-D array used to store the matrix C.

Definition at line 108 of file blas.c.

References gemm_nn(), gemm_nt(), gemm_tn(), and gemm_tt().

Referenced by neural_layer_connected_backward(), neural_layer_connected_forward(), neural_layer_convolutional_backward(), neural_layer_convolutional_forward(), and pred_rls_update().

Here is the call graph for this function:

Here is the caller graph for this function:

◆ blas_mul()

void blas_mul	(	const int	N,
		const double *	X,
		const int	INCX,
		double *	Y,
		const int	INCY
	)

Multiplies vector X by the vector Y and stores the result in vector Y.

Parameters

[in]	N	The number of elements in vectors X and Y.
[in]	X	Vector with N elements.
[in]	INCX	Stride between consecutive elements of X.
[in,out]	Y	Vector with N elements.
[in]	INCY	Stride between consecutive elements of Y.

Definition at line 217 of file blas.c.

Referenced by neural_layer_lstm_backward(), and neural_layer_lstm_forward().

Here is the caller graph for this function:

◆ blas_scal()

void blas_scal	(	const int	N,
		const double	ALPHA,
		double *	X,
		const int	INCX
	)

Scales vector X by the scalar ALPHA and overwrites it with the result.

Parameters

[in]	N	The number of elements in vector X.
[in]	ALPHA	Scalar used for multiplication.
[in,out]	X	Vector with N elements.
[in]	INCX	Stride between consecutive elements of X.

Definition at line 160 of file blas.c.

Referenced by neural_layer_connected_update(), neural_layer_convolutional_update(), and pred_rls_update().

Here is the caller graph for this function:

◆ blas_sum()

double blas_sum	(	const double *	X,
		const int	N
	)

Returns the sum of the vector X.

Parameters

[in]	X	Vector with N elements.
[in]	N	The number of elements in vector X.

Returns: The resulting sum.

Definition at line 232 of file blas.c.

Referenced by neural_layer_convolutional_backward().

Here is the caller graph for this function:

Functions

Detailed Description

Function Documentation

◆ blas_axpy()

◆ blas_dot()

◆ blas_fill()

◆ blas_gemm()

◆ blas_mul()

◆ blas_scal()

◆ blas_sum()