time-to-botec

Benchmark sampling in different programming languages
Log | Files | Refs | README

README.md (6703B)


      1 <!--
      2 
      3 @license Apache-2.0
      4 
      5 Copyright (c) 2020 The Stdlib Authors.
      6 
      7 Licensed under the Apache License, Version 2.0 (the "License");
      8 you may not use this file except in compliance with the License.
      9 You may obtain a copy of the License at
     10 
     11    http://www.apache.org/licenses/LICENSE-2.0
     12 
     13 Unless required by applicable law or agreed to in writing, software
     14 distributed under the License is distributed on an "AS IS" BASIS,
     15 WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
     16 See the License for the specific language governing permissions and
     17 limitations under the License.
     18 
     19 -->
     20 
     21 # sdsdot
     22 
     23 > Calculate the dot product of two single-precision floating-point vectors with extended accumulation.
     24 
     25 <section class="intro">
     26 
     27 The [dot product][dot-product] (or scalar product) is defined as
     28 
     29 <!-- <equation class="equation" label="eq:dot_product" align="center" raw="\mathbf{x}\cdot\mathbf{y} = \sum_{i=0}^{N-1} x_i y_i = x_0 y_0 + x_1 y_1 + \ldots + x_{N-1} y_{N-1}" alt="Dot product definition."> -->
     30 
     31 <div class="equation" align="center" data-raw-text="\mathbf{x}\cdot\mathbf{y} = \sum_{i=0}^{N-1} x_i y_i = x_0 y_0 + x_1 y_1 + \ldots + x_{N-1} y_{N-1}" data-equation="eq:dot_product">
     32     <img src="https://cdn.jsdelivr.net/gh/stdlib-js/stdlib@37cdbad9473fd050c8f7185c122a33d0f680b6c6/lib/node_modules/@stdlib/blas/base/sdsdot/docs/img/equation_dot_product.svg" alt="Dot product definition.">
     33     <br>
     34 </div>
     35 
     36 <!-- </equation> -->
     37 
     38 </section>
     39 
     40 <!-- /.intro -->
     41 
     42 <section class="usage">
     43 
     44 ## Usage
     45 
     46 ```javascript
     47 var sdsdot = require( '@stdlib/blas/base/sdsdot' );
     48 ```
     49 
     50 #### sdsdot( N, scalar, x, strideX, y, strideY )
     51 
     52 Calculates the dot product of vectors `x` and `y` with extended accumulation.
     53 
     54 ```javascript
     55 var Float32Array = require( '@stdlib/array/float32' );
     56 
     57 var x = new Float32Array( [ 4.0, 2.0, -3.0, 5.0, -1.0 ] );
     58 var y = new Float32Array( [ 2.0, 6.0, -1.0, -4.0, 8.0 ] );
     59 
     60 var z = sdsdot( x.length, 0.0, x, 1, y, 1 );
     61 // returns -5.0
     62 ```
     63 
     64 The function has the following parameters:
     65 
     66 -   **N**: number of indexed elements.
     67 -   **scalar**: scalar constant added to the dot product.
     68 -   **x**: input [`Float32Array`][@stdlib/array/float32].
     69 -   **strideX**: index increment for `x`.
     70 -   **y**: input [`Float32Array`][@stdlib/array/float32].
     71 -   **strideY**: index increment for `y`.
     72 
     73 The `N` and `stride` parameters determine which elements in `x` and `y` 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,
     74 
     75 ```javascript
     76 var Float32Array = require( '@stdlib/array/float32' );
     77 var floor = require( '@stdlib/math/base/special/floor' );
     78 
     79 var x = new Float32Array( [ 1.0, 2.0, 3.0, 4.0, 5.0, 6.0 ] );
     80 var y = new Float32Array( [ 1.0, 1.0, 1.0, 1.0, 1.0, 1.0 ] );
     81 
     82 var N = floor( x.length / 2 );
     83 
     84 var z = sdsdot( N, 0.0, x, 2, y, -1 );
     85 // returns 9.0
     86 ```
     87 
     88 Note that indexing is relative to the first index. To introduce an offset, use [`typed array`][mdn-typed-array] views.
     89 
     90 <!-- eslint-disable stdlib/capitalized-comments -->
     91 
     92 ```javascript
     93 var Float32Array = require( '@stdlib/array/float32' );
     94 var floor = require( '@stdlib/math/base/special/floor' );
     95 
     96 // Initial arrays...
     97 var x0 = new Float32Array( [ 1.0, 2.0, 3.0, 4.0, 5.0, 6.0 ] );
     98 var y0 = new Float32Array( [ 7.0, 8.0, 9.0, 10.0, 11.0, 12.0 ] );
     99 
    100 // Create offset views...
    101 var x1 = new Float32Array( x0.buffer, x0.BYTES_PER_ELEMENT*1 ); // start at 2nd element
    102 var y1 = new Float32Array( y0.buffer, y0.BYTES_PER_ELEMENT*3 ); // start at 4th element
    103 
    104 var N = floor( x0.length / 2 );
    105 
    106 var z = sdsdot( N, 0.0, x1, -2, y1, 1 );
    107 // returns 128.0
    108 ```
    109 
    110 #### sdsdot.ndarray( N, x, strideX, offsetX, y, strideY, offsetY )
    111 
    112 Calculates the dot product of vectors `x` and `y` with extended accumulation and using alternative indexing semantics.
    113 
    114 ```javascript
    115 var Float32Array = require( '@stdlib/array/float32' );
    116 
    117 var x = new Float32Array( [ 4.0, 2.0, -3.0, 5.0, -1.0 ] );
    118 var y = new Float32Array( [ 2.0, 6.0, -1.0, -4.0, 8.0 ] );
    119 
    120 var z = sdsdot.ndarray( x.length, 0.0, x, 1, 0, y, 1, 0 );
    121 // returns -5.0
    122 ```
    123 
    124 The function has the following additional parameters:
    125 
    126 -   **offsetX**: starting index for `x`.
    127 -   **offsetY**: starting index for `y`.
    128 
    129 While [`typed array`][mdn-typed-array] views mandate a view offset based on the underlying `buffer`, the `offsetX` and `offsetY` 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
    130 
    131 ```javascript
    132 var Float32Array = require( '@stdlib/array/float32' );
    133 var floor = require( '@stdlib/math/base/special/floor' );
    134 
    135 var x = new Float32Array( [ 1.0, 2.0, 3.0, 4.0, 5.0, 6.0 ] );
    136 var y = new Float32Array( [ 7.0, 8.0, 9.0, 10.0, 11.0, 12.0 ] );
    137 
    138 var N = floor( x.length / 2 );
    139 
    140 var z = sdsdot.ndarray( N, 0.0, x, 2, 1, y, -1, y.length-1 );
    141 // returns 128.0
    142 ```
    143 
    144 </section>
    145 
    146 <!-- /.usage -->
    147 
    148 <section class="notes">
    149 
    150 ## Notes
    151 
    152 -   If `N <= 0`, both functions return `scalar`.
    153 -   `sdsdot()` corresponds to the [BLAS][blas] level 1 function [`sdsdot`][sdsdot].
    154 
    155 </section>
    156 
    157 <!-- /.notes -->
    158 
    159 <section class="examples">
    160 
    161 ## Examples
    162 
    163 <!-- eslint no-undef: "error" -->
    164 
    165 ```javascript
    166 var randu = require( '@stdlib/random/base/randu' );
    167 var round = require( '@stdlib/math/base/special/round' );
    168 var Float32Array = require( '@stdlib/array/float32' );
    169 var sdsdot = require( '@stdlib/blas/base/sdsdot' );
    170 
    171 var x;
    172 var y;
    173 var i;
    174 
    175 x = new Float32Array( 10 );
    176 y = new Float32Array( 10 );
    177 for ( i = 0; i < x.length; i++ ) {
    178     x[ i ] = round( randu() * 100.0 );
    179     y[ i ] = round( randu() * 10.0 );
    180 }
    181 console.log( x );
    182 console.log( y );
    183 
    184 var z = sdsdot( x.length, 0.0, x, 1, y, -1 );
    185 console.log( z );
    186 ```
    187 
    188 </section>
    189 
    190 <!-- /.examples -->
    191 
    192 * * *
    193 
    194 <section class="references">
    195 
    196 ## References
    197 
    198 -   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][@lawson:1979a].
    199 
    200 </section>
    201 
    202 <!-- /.references -->
    203 
    204 <section class="links">
    205 
    206 [dot-product]: https://en.wikipedia.org/wiki/Dot_product
    207 
    208 [blas]: http://www.netlib.org/blas
    209 
    210 [sdsdot]: http://www.netlib.org/lapack/explore-html/df/d28/group__single__blas__level1.html
    211 
    212 [@stdlib/array/float32]: https://www.npmjs.com/package/@stdlib/array-float32
    213 
    214 [mdn-typed-array]: https://developer.mozilla.org/en-US/docs/Web/JavaScript/Reference/Global_Objects/TypedArray
    215 
    216 [@lawson:1979a]: https://doi.org/10.1145/355841.355848
    217 
    218 </section>
    219 
    220 <!-- /.links -->