README.md

Base

Base ndarray.

Usage

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

ns

Base ndarray.

var o = ns;
// returns {...}

• assign( arrays ): assign elements in an input ndarray to elements in an output ndarray.
• binaryLoopOrder( shape, stridesX, stridesY, stridesZ ): reorder ndarray dimensions and associated strides for loop interchange.
• binaryBlockSize( dtypeX, dtypeY, dtypeZ ): resolve a loop block size for multi-dimensional array tiled loops.
• bind2vind( shape, strides, offset, order, idx, mode ): convert a linear index in an underlying data buffer to a linear index in an array view.
• broadcastArray( arr, shape ): broadcast an ndarray to a specified shape.
• broadcastArrays( arrays ): broadcast ndarrays to a common shape.
• broadcastScalar( value, dtype, shape, order ): broadcast a scalar value to an ndarray having a specified shape.
• broadcastShapes( shapes ): broadcast array shapes to a single shape.
• bufferCtors( dtype ): ndarray data buffer constructors.
• bufferDataTypeEnum( buffer ): return the data type enumeration constant of an ndarray data buffer.
• bufferDataType( buffer ): return the data type of an ndarray data buffer.
• buffer( dtype, size ): create a zero-filled contiguous linear ndarray data buffer.
• bytesPerElement( dtype ): return the number of bytes per element provided an underlying array data type.
• char2dtype( [ch] ): return the data type string associated with a provided single letter character abbreviation.
• clampIndex( idx, max ): restrict an index to the interval [0,max].
• ndarray( dtype, buffer, shape, strides, offset, order ): create a multidimensional array.
• data( x ): return the underlying data buffer of a provided ndarray.
• dtypeChar( [dtype] ): return the single letter abbreviation for an underlying array data type.
• dtypeDesc( [dtype] ): return the description for a specified data type.
• dtypeEnum2Str( dtype ): return the data type string associated with an ndarray data type enumeration constant.
• dtypeResolveEnum( dtype ): return the enumeration constant associated with a supported ndarray data type value.
• dtypeResolveStr( dtype ): return the data type string associated with a supported ndarray data type value.
• dtypeStr2Enum( dtype ): return the enumeration constant associated with an ndarray data type string.
• dtype( x ): return the data type of a provided ndarray.
• dtype2c( dtype ): return the C data type associated with a provided data type value.
• dtypes2signatures( dtypes, nin, nout ): transform a list of array argument data types into a list of signatures.
• emptyLike( x ): create an uninitialized ndarray having the same shape and data type as a provided ndarray.
• empty( dtype, shape, order ): create an uninitialized ndarray having a specified shape and data type.
• expandDimensions( x, axis ): expand the shape of an array by inserting a new dimension of size one at a specified axis.
• flag( x, name ): return a specified flag for a provided ndarray.
• flags( x, copy ): return the flags of a provided ndarray.
• fliplr( x, writable ): return a view of an input ndarray in which the order of elements along the last dimension is reversed.
• flipud( x, writable ): return a view of an input ndarray in which the order of elements along the second-to-last dimension is reversed.
• scalar2ndarray( value, dtype, order ): convert a scalar value to a zero-dimensional ndarray.
• ind( idx, max, mode ): return an index given an index mode.
• ind2sub( shape, strides, offset, order, idx, mode ): convert a linear index to an array of subscripts.
• iterationOrder( strides ): given a stride array, determine array iteration order.
• maxViewBufferIndex( shape, strides, offset ): compute the maximum linear index in an underlying data buffer accessible to an array view.
• maybeBroadcastArray( arr, shape ): broadcast an ndarray to a specified shape if and only if the specified shape differs from the provided ndarray's shape.
• maybeBroadcastArrays( arrays ): broadcast ndarrays to a common shape.
• metaDataProps( meta, dtypes, obj ): define non-enumerable read-only properties which expose ndarray function meta data.
• minViewBufferIndex( shape, strides, offset ): compute the minimum linear index in an underlying data buffer accessible to an array view.
• minmaxViewBufferIndex( shape, strides, offset ): compute the minimum and maximum linear indices in an underlying data buffer which are accessible to an array view.
• ndarraylike2object( x ): convert an ndarray-like object to an object likely to have the same "shape".
• ndims( x ): return the number of ndarray dimensions.
• nextCartesianIndex( shape, order, idx, dim ): return the next Cartesian index (i.e., set of subscripts/dimension indices).
• nonsingletonDimensions( shape ): return the number of non-singleton dimensions.
• normalizeIndex( idx, max ): normalize an index to the interval [0,max].
• nullaryLoopOrder( shape, stridesX ): reorder ndarray dimensions and associated strides for loop interchange.
• nullaryBlockSize( dtypeX ): resolve a loop block size for multi-dimensional array tiled loops.
• nullary( arrays, fcn ): apply a nullary callback and assign results to elements in an output ndarray.
• numelDimension( x, dim ): return the size (i.e., number of elements) of a specified dimension for a provided ndarray.
• numel( shape ): return the number of elements in an array.
• offset( x ): return the index offset specifying the underlying buffer index of the first iterated ndarray element.
• order( x ): return the layout order of a provided ndarray.
• outputPolicyEnum2Str( policy ): return the policy string associated with an output ndarray data type policy enumeration constant.
• outputPolicyResolveEnum( policy ): return the enumeration constant associated with a supported ndarray data type policy value.
• outputPolicyResolveStr( dtype ): return the policy string associated with a supported ndarray data type policy value.
• outputPolicyStr2Enum( policy ): return the enumeration constant associated with an output ndarray data type policy string.
• prependSingletonDimensions( x, n ): prepend singleton dimensions.
• removeSingletonDimensions( x ): remove singleton dimensions.
• reverseDimension( x, dim, writable ): return a view of an input ndarray in which the order of elements along a specified dimension is reversed.
• reverse( x, writable ): return a view of an input ndarray in which the order of elements along each dimension is reversed.
• serializeMetaData( x ): serialize ndarray meta data.
• shape( x, copy ): return the shape of a provided ndarray.
• shape2strides( shape, order ): generate a stride array from an array shape.
• singletonDimensions( shape ): return the number of singleton dimensions.
• sliceAssign( x, y, slice, strict ): assign element values from a broadcasted input ndarray to corresponding elements in an output ndarray view.
• sliceDimensionFrom( x, dim, start, strict, writable ): return a shifted view of an input ndarray along a specified dimension.
• sliceDimensionTo( x, dim, stop, strict, writable ): return a truncated view of an input ndarray along a specified dimension.
• sliceDimension( x, dim, slice, strict, writable ): return a view of an input ndarray when sliced along a specified dimension.
• sliceFrom( x, start, strict, writable ): return a shifted view of an input ndarray.
• sliceTo( x, stop, strict, writable ): return a truncated view of an input ndarray.
• slice( x, slice, strict, writable ): return a view of an input ndarray.
• stride( x, dim ): return the stride along a specified dimension for a provided ndarray.
• strides( x, copy ): return the strides of a provided ndarray.
• strides2offset( shape, strides ): determine the index offset which specifies the location of the first indexed value in a multidimensional array based on a stride array.
• strides2order( strides ): determine the order of a multidimensional array based on a provided stride array.
• sub2ind( shape, strides, offset, ...subscripts, mode ): convert subscripts to a linear index.
• ndarray2array( buffer, shape, strides, offset, order ): convert an ndarray buffer to a generic array.
• transpose( x ): transpose a matrix (or a stack of matrices).
• unaryBy( arrays, fcn, clbk[, thisArg] ): apply a unary function to each element in an input ndarray according to a callback function and assign results to elements in an output ndarray.
• unaryLoopOrder( shape, stridesX, stridesY ): reorder ndarray dimensions and associated strides for loop interchange.
• unaryOutputDataType( dtype, policy ): resolve the output ndarray data type for a unary function.
• unaryBlockSize( dtypeX, dtypeY ): resolve a loop block size for multi-dimensional array tiled loops.
• unary( arrays, fcn ): apply a unary callback to elements in an input ndarray and assign results to elements in an output ndarray.
• vind2bind( shape, strides, offset, order, idx, mode ): convert a linear index in an array view to a linear index in an underlying data buffer.
• wrapIndex( idx, max ): wrap an index on the interval [0,max].
• zerosLike( x ): create a zero-filled ndarray having the same shape and data type as a provided ndarray.
• zeros( dtype, shape, order ): create a zero-filled ndarray having a specified shape and data type.

The namespace contains the following sub-namespaces:

• assert: base ndarray assertion utilities.

Examples

var Float64Array = require( '@stdlib/array/float64' );
var linspace = require( '@stdlib/array/linspace' );
var ddot = require( '@stdlib/blas/base/ddot' );
var sin = require( '@stdlib/math/base/special/sin' );
var cos = require( '@stdlib/math/base/special/cos' );
var PI = require( '@stdlib/constants/float64/pi' );
var ns = require( '@stdlib/ndarray/base' );

var resultData;
var yStrides;
var xStrides;
var xShape;
var yData;
var xData;
var y;
var x;
var i;

xData = linspace( 0, 2 * PI, 100 );
yData = linspace( 0, PI, 100 );
x = ns.ndarray( 'float64', xData, [ 100 ], [ 1 ], 0, 'row-major' );
y = ns.ndarray( 'float64', yData, [ 100 ], [ 1 ], 0, 'row-major' );

// Apply sin function to each element of x and cos function to each element of y
for ( i = 0; i < x.shape[ 0 ]; i++ ) {
    x.set( i, sin( x.get( i ) ) );
    y.set( i, cos( y.get( i ) ) );
}

resultData = new Float64Array( 1 );

xShape = x.shape[ 0 ];
xStrides = x.strides[ 0 ];
yStrides = y.strides[ 0 ];

// Calculate dot product of x and y
resultData[ 0 ] = ddot( xShape, x.data, xStrides, y.data, yStrides );

// Logs 42.00632598024892
console.log( resultData[ 0 ] );