Skip to content

Latest commit

 

History

History
781 lines (527 loc) · 22 KB

README.md

File metadata and controls

781 lines (527 loc) · 22 KB
About stdlib...

We believe in a future in which the web is a preferred environment for numerical computation. To help realize this future, we've built stdlib. stdlib is a standard library, with an emphasis on numerical and scientific computation, written in JavaScript (and C) for execution in browsers and in Node.js.

The library is fully decomposable, being architected in such a way that you can swap out and mix and match APIs and functionality to cater to your exact preferences and use cases.

When you use stdlib, you can be absolutely certain that you are using the most thorough, rigorous, well-written, studied, documented, tested, measured, and high-quality code out there.

To join us in bringing numerical computing to the web, get started by checking us out on GitHub, and please consider financially supporting stdlib. We greatly appreciate your continued support!

ndarray

NPM version Build Status Coverage Status

Create a multidimensional array.

Installation

npm install @stdlib/ndarray-base-ctor

Alternatively,

  • To load the package in a website via a script tag without installation and bundlers, use the ES Module available on the esm branch (see README).
  • If you are using Deno, visit the deno branch (see README for usage intructions).
  • For use in Observable, or in browser/node environments, use the Universal Module Definition (UMD) build available on the umd branch (see README).

The branches.md file summarizes the available branches and displays a diagram illustrating their relationships.

To view installation and usage instructions specific to each branch build, be sure to explicitly navigate to the respective README files on each branch, as linked to above.

Usage

var ndarray = require( '@stdlib/ndarray-base-ctor' );

ndarray( dtype, buffer, shape, strides, offset, order )

Returns an ndarray instance.

// Specify the array configuration:
var buffer = [ 1, 2, 3, 4 ];
var shape = [ 2, 2 ];
var order = 'row-major';
var strides = [ 2, 1 ];
var offset = 0;

// Create a new ndarray:
var arr = ndarray( 'generic', buffer, shape, strides, offset, order );
// returns <ndarray>

The constructor has the following parameters:

  • dtype: underlying data type.
  • buffer: data buffer.
  • shape: array shape (dimensions).
  • strides: array strides which are index offsets specifying how to access along corresponding dimensions.
  • offset: index offset specifying the location of the first indexed element in the data buffer.
  • order: array order, which is either row-major (C-style) or column-major (Fortran-style).

To create a zero-dimensional array, provide an empty shape and a single strides element equal to 0. The order can be either row-major or column-major and has no effect on data storage or access.

var buffer = [ 1 ];
var shape = [];
var order = 'row-major';
var strides = [ 0 ];
var offset = 0;

// Create a new zero-dimensional array:
var arr = ndarray( 'generic', buffer, shape, strides, offset, order );
// returns <ndarray>

Properties

ndarray.name

String value of the ndarray constructor name.

var str = ndarray.name;
// returns 'ndarray'

ndarray.prototype.byteLength

Size (in bytes) of the array (if known).

var Float64Array = require( '@stdlib/array-float64' );

// Specify the array configuration:
var buffer = new Float64Array( [ 1, 2, 3, 4 ] );
var shape = [ 2, 2 ];
var order = 'row-major';
var strides = [ 2, 1 ];
var offset = 0;

// Create a new ndarray:
var arr = ndarray( 'float64', buffer, shape, strides, offset, order );

// Get the byte length:
var nbytes = arr.byteLength;
// returns 32

If unable to determine the size of the array, the property value is null.

// Specify the array configuration:
var buffer = [ 1, 2, 3, 4 ];
var shape = [ 2, 2 ];
var order = 'row-major';
var strides = [ 2, 1 ];
var offset = 0;

// Create a new ndarray:
var arr = ndarray( 'generic', buffer, shape, strides, offset, order );

// Get the byte length:
var nbytes = arr.byteLength;
// returns null

ndarray.prototype.BYTES_PER_ELEMENT

Size (in bytes) of each array element (if known).

var Float32Array = require( '@stdlib/array-float32' );

// Specify the array configuration:
var buffer = new Float32Array( [ 1, 2, 3, 4 ] );
var shape = [ 2, 2 ];
var order = 'row-major';
var strides = [ 2, 1 ];
var offset = 0;

// Create a new ndarray:
var arr = ndarray( 'float32', buffer, shape, strides, offset, order );

// Get the number of bytes per element:
var nbytes = arr.BYTES_PER_ELEMENT;
// returns 4

If size of each array element is unknown, the property value is null.

// Specify the array configuration:
var buffer = [ 1, 2, 3, 4 ];
var shape = [ 2, 2 ];
var order = 'row-major';
var strides = [ 2, 1 ];
var offset = 0;

// Create a new ndarray:
var arr = ndarray( 'generic', buffer, shape, strides, offset, order );

// Get the number of bytes per element:
var nbytes = arr.BYTES_PER_ELEMENT;
// returns null

ndarray.prototype.data

A reference to the underlying data buffer.

var Int8Array = require( '@stdlib/array-int8' );

// Specify the array configuration:
var buffer = new Int8Array( [ 1, 2, 3, 4 ] );
var shape = [ 2, 2 ];
var order = 'row-major';
var strides = [ 2, 1 ];
var offset = 0;

// Create a new ndarray:
var arr = ndarray( 'int8', buffer, shape, strides, offset, order );

// Get the buffer reference:
var d = arr.data;
// returns <Int8Array>[ 1, 2, 3, 4 ]

var bool = ( d === buffer );
// returns true

ndarray.prototype.dtype

Underlying data type.

var Uint8Array = require( '@stdlib/array-uint8' );

// Specify the array configuration:
var buffer = new Uint8Array( [ 1, 2, 3, 4 ] );
var shape = [ 2, 2 ];
var order = 'row-major';
var strides = [ -2, 1 ];
var offset = 2;

// Create a new ndarray:
var arr = ndarray( 'uint8', buffer, shape, strides, offset, order );

// Get the underlying data type:
var dtype = arr.dtype;
// returns 'uint8'

ndarray.prototype.flags

Meta information, such as information concerning the memory layout of the array. The returned object has the following properties:

  • ROW_MAJOR_CONTIGUOUS: boolean indicating if an array is row-major contiguous.
  • COLUMN_MAJOR_CONTIGUOUS: boolean indicating if an array is column-major contiguous.
  • READONLY: boolean indicating if an array is read-only.

An array is contiguous if (1) an array is compatible with being stored in a single memory segment and (2) each array element is adjacent to the next array element. Note that an array can be both row-major contiguous and column-major contiguous at the same time (e.g., if an array is a 1-dimensional ndarray with strides = [1]).

var Int32Array = require( '@stdlib/array-int32' );

// Specify the array configuration:
var buffer = new Int32Array( [ 1, 2, 3, 4 ] );
var shape = [ 2, 2 ];
var order = 'column-major';
var strides = [ 1, 2 ];
var offset = 0;

// Create a new ndarray:
var arr = ndarray( 'int32', buffer, shape, strides, offset, order );

// Get the array flags:
var flg = arr.flags;
// returns {...}

ndarray.prototype.length

Number of array elements.

var Uint16Array = require( '@stdlib/array-uint16' );

// Specify the array configuration:
var buffer = new Uint16Array( [ 1, 2, 3, 4 ] );
var shape = [ 2, 2 ];
var order = 'column-major';
var strides = [ -1, -2 ];
var offset = 3;

// Create a new ndarray:
var arr = ndarray( 'uint16', buffer, shape, strides, offset, order );

// Get the array length:
var len = arr.length;
// returns 4

ndarray.prototype.ndims

Number of dimensions.

var Uint8ClampedArray = require( '@stdlib/array-uint8c' );

// Specify the array configuration:
var buffer = new Uint8ClampedArray( [ 1, 2, 3, 4 ] );
var shape = [ 2, 2 ];
var order = 'row-major';
var strides = [ -2, -1 ];
var offset = 3;

// Create a new ndarray:
var arr = ndarray( 'uint8c', buffer, shape, strides, offset, order );

// Get the number of dimensions:
var ndims = arr.ndims;
// returns 2

ndarray.prototype.offset

Index offset which specifies the buffer index at which to start iterating over array elements.

var Int16Array = require( '@stdlib/array-int16' );

// Specify the array configuration:
var buffer = new Int16Array( [ 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12 ] );
var shape = [ 2, 2 ];
var order = 'row-major';
var strides = [ -2, -1 ];
var offset = 10;

// Create a new ndarray:
var arr = ndarray( 'int16', buffer, shape, strides, offset, order );

// Get the index offset:
var o = arr.offset;
// returns 10

ndarray.prototype.order

Array order. The array order is either row-major (C-style) or column-major (Fortran-style).

var Uint32Array = require( '@stdlib/array-uint32' );

// Specify the array configuration:
var buffer = new Uint32Array( [ 1, 2, 3, 4 ] );
var shape = [ 2, 2 ];
var order = 'row-major';
var strides = [ 2, 1 ];
var offset = 0;

// Create a new ndarray:
var arr = ndarray( 'uint32', buffer, shape, strides, offset, order );

// Get the array order:
var ord = arr.order;
// returns 'row-major'

ndarray.prototype.shape

Returns a copy of the array shape.

// Specify the array configuration:
var buffer = [ 1, 2, 3, 4, 5, 6 ];
var shape = [ 2, 2 ];
var order = 'row-major';
var strides = [ 2, 1 ];
var offset = 2;

// Create a new ndarray:
var arr = ndarray( 'generic', buffer, shape, strides, offset, order );

// Get the array shape:
var dims = arr.shape;
// returns [ 2, 2 ]

ndarray.prototype.strides

Returns a copy of the array strides which specify how to access data along corresponding array dimensions.

// Specify the array configuration:
var buffer = [ 1, 2, 3, 4 ];
var shape = [ 2, 2 ];
var order = 'column-major';
var strides = [ -1, 2 ];
var offset = 1;

// Create a new ndarray:
var arr = ndarray( 'generic', buffer, shape, strides, offset, order );

// Get the array strides:
var s = arr.strides;
// returns [ -1, 2 ]

Methods

ndarray.prototype.get( i, j, k, ... )

Returns an array element specified according to provided subscripts. The number of provided subscripts should equal the number of dimensions.

// Specify the array configuration:
var buffer = [ 1, 2, 3, 4, 5, 6, 7, 8 ];
var shape = [ 2, 2 ];
var order = 'row-major';
var strides = [ 2, 1 ];
var offset = 2;

// Create a new ndarray:
var arr = ndarray( 'generic', buffer, shape, strides, offset, order );

// Get the element located at (1,1):
var v = arr.get( 1, 1 );
// returns 6

ndarray.prototype.iget( idx )

Returns an array element located at a specified linear index.

// Specify the array configuration:
var buffer = [ 1, 2, 3, 4, 5, 6, 7, 8 ];
var shape = [ 2, 2 ];
var order = 'row-major';
var strides = [ 2, 1 ];
var offset = 2;

// Create a new ndarray:
var arr = ndarray( 'generic', buffer, shape, strides, offset, order );

// Get the element located at index 3:
var v = arr.iget( 3 );
// returns 6

For zero-dimensional arrays, the input argument is ignored and, for clarity, should not be provided.

ndarray.prototype.set( i, j, k, ..., v )

Sets an array element specified according to provided subscripts. The number of provided subscripts should equal the number of dimensions.

// Specify the array configuration:
var buffer = [ 1, 2, 3, 4 ];
var shape = [ 2, 2 ];
var order = 'row-major';
var strides = [ 2, 1 ];
var offset = 0;

// Create a new ndarray:
var arr = ndarray( 'generic', buffer, shape, strides, offset, order );

// Set the element located at (1,1):
arr.set( 1, 1, 40 );
var v = arr.get( 1, 1 );
// returns 40

// Get the underlying buffer:
var d = arr.data;
// returns [ 1, 2, 3, 40 ]

The method returns the ndarray instance.

ndarray.prototype.iset( idx, v )

Sets an array element located at a specified linear index.

// Specify the array configuration:
var buffer = [ 1, 2, 3, 4 ];
var shape = [ 2, 2 ];
var order = 'row-major';
var strides = [ 2, 1 ];
var offset = 0;

// Create a new ndarray:
var arr = ndarray( 'generic', buffer, shape, strides, offset, order );

// Set the element located at index 3:
arr.iset( 3, 40 );
var v = arr.iget( 3 );
// returns 40

// Get the underlying buffer:
var d = arr.data;
// returns [ 1, 2, 3, 40 ]

For zero-dimensional arrays, the first, and only, argument should be the value v to set. The method returns the ndarray instance.

ndarray.prototype.toString()

Serializes an ndarray as a string.

// Specify the array configuration:
var buffer = [ 1, 2, 3, 4, 5, 6, 7, 8 ];
var shape = [ 3, 2 ];
var order = 'row-major';
var strides = [ 2, 1 ];
var offset = 2;

// Create a new ndarray:
var arr = ndarray( 'generic', buffer, shape, strides, offset, order );

// Serialize to a string:
var str = arr.toString();
// returns "ndarray( 'generic', [ 3, 4, 5, 6, 7, 8 ], [ 3, 2 ], [ 2, 1 ], 0, 'row-major' )"

The method does not serialize data outside of the buffer region defined by the array configuration.

ndarray.prototype.toJSON()

Serializes an ndarray as a JSON object. JSON.stringify() implicitly calls this method when stringifying an ndarray instance.

// Specify the array configuration:
var buffer = [ 1, 2, 3, 4, 5, 6, 7, 8 ];
var shape = [ 3, 2 ];
var order = 'row-major';
var strides = [ 2, 1 ];
var offset = 2;

// Create a new ndarray:
var arr = ndarray( 'generic', buffer, shape, strides, offset, order );

// Serialize to JSON:
var o = arr.toJSON();
// returns { 'type': 'ndarray', 'dtype': 'generic', 'flags': {...}, 'offset': 0, 'order': 'row-major', 'shape': [ 3, 2 ], 'strides': [ 2, 1 ], 'data': [ 3, 4, 5, 6, 7, 8 ] }

The method does not serialize data outside of the buffer region defined by the array configuration.


Notes

  • A data buffer must be an array-like object (i.e., have a length property). For data buffers which are not indexed collections (i.e., collections which cannot support direct index access, such as buffer[ index ]; e.g., Complex64Array, Complex128Array, etc), a data buffer should provide #.get( idx ) and #.set( v[, idx] ) methods. Note that, for set methods, the value to set should be the first argument, followed by the linear index, similar to the native typed array set method.

Examples

var Float32Array = require( '@stdlib/array-float32' );
var ndarray = require( '@stdlib/ndarray-base-ctor' );

// Create a data buffer:
var buffer = new Float32Array( (3*3*3*3) + 100 );

// Specify the array shape:
var shape = [ 3, 3, 3, 3 ];

// Specify the array strides:
var strides = [ 27, 9, 3, 1 ];

// Specify the index offset:
var offset = 4;

// Specify the order:
var order = 'row-major'; // C-style

// Create a new ndarray:
var arr = ndarray( 'float32', buffer, shape, strides, offset, order );

// Retrieve an array value:
var v = arr.get( 1, 2, 1, 2 );
// returns 0.0

// Set an array value:
arr.set( 1, 2, 1, 2, 10.0 );

// Retrieve the array value:
v = arr.get( 1, 2, 1, 2 );
// returns 10.0

// Serialize the array as a string:
var str = arr.toString();
// returns "ndarray( 'float32', new Float32Array( [ 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 10, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0 ] ), [ 3, 3, 3, 3 ], [ 27, 9, 3, 1 ], 0, 'row-major' )"

// Serialize the array as JSON:
str = JSON.stringify( arr.toJSON() );
// e.g., returns '{"type":"ndarray","dtype":"float32","flags":{"READONLY":false},"order":"row-major","shape":[3,3,3,3],"strides":[27,9,3,1],"data":[0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,10,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0]}'

See Also


Notice

This package is part of stdlib, a standard library for JavaScript and Node.js, with an emphasis on numerical and scientific computing. The library provides a collection of robust, high performance libraries for mathematics, statistics, streams, utilities, and more.

For more information on the project, filing bug reports and feature requests, and guidance on how to develop stdlib, see the main project repository.

Community

Chat


License

See LICENSE.

Copyright

Copyright © 2016-2024. The Stdlib Authors.