sdsdot
Calculate the dot product of two single-precision floating-point vectors with extended accumulation.
The dot product (or scalar product) is defined as
Usage
var sdsdot = require( '@stdlib/blas/base/sdsdot' );
sdsdot( N, scalar, x, strideX, y, strideY )
Calculates the dot product of vectors x
and y
with extended accumulation.
var Float32Array = require( '@stdlib/array/float32' );
var x = new Float32Array( [ 4.0, 2.0, -3.0, 5.0, -1.0 ] );
var y = new Float32Array( [ 2.0, 6.0, -1.0, -4.0, 8.0 ] );
var z = sdsdot( x.length, 0.0, x, 1, y, 1 );
// returns -5.0
The function has the following parameters:
- N: number of indexed elements.
- scalar: scalar constant added to the dot product.
- x: input
Float32Array
. - strideX: index increment for
x
. - y: input
Float32Array
. - strideY: index increment for
y
.
The N
and stride parameters determine which elements in the strided arrays are accessed at runtime. For example, to calculate the dot product of every other value in x
and the first N
elements of y
in reverse order,
var Float32Array = require( '@stdlib/array/float32' );
var x = new Float32Array( [ 1.0, 2.0, 3.0, 4.0, 5.0, 6.0 ] );
var y = new Float32Array( [ 1.0, 1.0, 1.0, 1.0, 1.0, 1.0 ] );
var z = sdsdot( 3, 0.0, x, 2, y, -1 );
// returns 9.0
Note that indexing is relative to the first index. To introduce an offset, use typed array
views.
var Float32Array = require( '@stdlib/array/float32' );
// Initial arrays...
var x0 = new Float32Array( [ 1.0, 2.0, 3.0, 4.0, 5.0, 6.0 ] );
var y0 = new Float32Array( [ 7.0, 8.0, 9.0, 10.0, 11.0, 12.0 ] );
// Create offset views...
var x1 = new Float32Array( x0.buffer, x0.BYTES_PER_ELEMENT*1 ); // start at 2nd element
var y1 = new Float32Array( y0.buffer, y0.BYTES_PER_ELEMENT*3 ); // start at 4th element
var z = sdsdot( 3, 0.0, x1, -2, y1, 1 );
// returns 128.0
sdsdot.ndarray( N, x, strideX, offsetX, y, strideY, offsetY )
Calculates the dot product of vectors x
and y
with extended accumulation and using alternative indexing semantics.
var Float32Array = require( '@stdlib/array/float32' );
var x = new Float32Array( [ 4.0, 2.0, -3.0, 5.0, -1.0 ] );
var y = new Float32Array( [ 2.0, 6.0, -1.0, -4.0, 8.0 ] );
var z = sdsdot.ndarray( x.length, 0.0, x, 1, 0, y, 1, 0 );
// returns -5.0
The function has the following additional parameters:
- offsetX: starting index for
x
. - offsetY: starting index for
y
.
While typed array
views mandate a view offset based on the underlying buffer, the offset parameters support indexing semantics based on starting indices. For example, to calculate the dot product of every other value in x
starting from the second value with the last 3 elements in y
in reverse order
var Float32Array = require( '@stdlib/array/float32' );
var x = new Float32Array( [ 1.0, 2.0, 3.0, 4.0, 5.0, 6.0 ] );
var y = new Float32Array( [ 7.0, 8.0, 9.0, 10.0, 11.0, 12.0 ] );
var z = sdsdot.ndarray( 3, 0.0, x, 2, 1, y, -1, y.length-1 );
// returns 128.0
Notes
Examples
var discreteUniform = require( '@stdlib/random/array/discrete-uniform' );
var sdsdot = require( '@stdlib/blas/base/sdsdot' );
var opts = {
'dtype': 'float32'
};
var x = discreteUniform( 10, 0, 100, opts );
console.log( x );
var y = discreteUniform( x.length, 0, 10, opts );
console.log( y );
var out = sdsdot( x.length, 0.0, x, 1, y, -1 );
console.log( out );
C APIs
Usage
#include "stdlib/blas/base/sdsdot.h"
c_sdsdot( N, scalar, *X, strideX, *Y, strideY )
Calculates the dot product of vectors x
and y
with extended accumulation.
const float x[] = { 1.0f, -2.0f, 3.0f, -4.0f, 5.0f, -6.0f, 7.0f, -8.0f };
const float y[] = { 1.0f, -2.0f, 3.0f, -4.0f, 5.0f, -6.0f, 7.0f, -8.0f };
float v = c_sdsdot( 5, 0.0f, x, 1, y, -1 );
// returns -120.0f
The function accepts the following arguments:
- N:
[in] CBLAS_INT
number of indexed elements. - scalar:
[in] float
scalar constant to add to dot product. - X:
[in] float*
first input array. - strideX:
[in] CBLAS_INT
index increment forX
. - Y:
[in] float*
second input array. - strideY:
[in] CBLAS_INT
index increment forY
.
float c_sdsdot( const CBLAS_INT N, const float scalar, const float *X, const CBLAS_INT strideX, const float *Y, const CBLAS_INT strideY );
c_sdsdot_ndarray( N, scalar, *X, strideX, offsetX, *Y, strideY, offsetY )
Calculates the dot product of vectors x
and y
with extended accumulation using alternative indexing semantics.
const float x[] = { 1.0f, -2.0f, 3.0f, -4.0f, 5.0f, -6.0f, 7.0f, -8.0f };
const float y[] = { 1.0f, -2.0f, 3.0f, -4.0f, 5.0f, -6.0f, 7.0f, -8.0f };
float v = c_sdsdot_ndarray( 5, 0.0f, x, 1, 0, y, -1, 7 );
// returns -80.0f
The function accepts the following arguments:
- N:
[in] CBLAS_INT
number of indexed elements. - scalar:
[in] float
scalar constant to add to dot product. - X:
[in] float*
first input array. - strideX:
[in] CBLAS_INT
index increment forX
. - offsetX:
[in] CBLAS_INT
starting index forX
. - Y:
[in] float*
second input array. - strideY:
[in] CBLAS_INT
index increment forY
. - offsetY:
[in] CBLAS_INT
starting index forY
.
float c_sdsdot_ndarray( const CBLAS_INT N, const float scalar, const float *X, const CBLAS_INT strideX, const CBLAS_INT offsetX, const float *Y, const CBLAS_INT strideY, const CBLAS_INT offsetY );
Examples
#include "stdlib/blas/base/sdsdot.h"
#include <stdio.h>
int main( void ) {
// Create strided arrays:
const float x[] = { 1.0f, -2.0f, 3.0f, -4.0f, 5.0f, -6.0f, 7.0f, -8.0f };
const float y[] = { 1.0f, -2.0f, 3.0f, -4.0f, 5.0f, -6.0f, 7.0f, -8.0f };
// Specify the number of indexed elements:
const int N = 8;
// Specify strides:
const int strideX = 1;
const int strideY = -1;
// Compute the dot product:
float d = c_sdsdot( N, 0.0f, x, strideX, y, strideY );
// Print the result:
printf( "dot product: %f\n", d );
// Compute the dot product:
d = c_sdsdot_ndarray( N, 0.0f, x, strideX, 0, y, strideY, 7 );
// Print the result:
printf( "dot product: %f\n", d );
}
References
- Lawson, Charles L., Richard J. Hanson, Fred T. Krogh, and David Ronald Kincaid. 1979. "Algorithm 539: Basic Linear Algebra Subprograms for Fortran Usage [F1]." ACM Transactions on Mathematical Software 5 (3). New York, NY, USA: Association for Computing Machinery: 324–25. doi:10.1145/355841.355848.