time-to-botec

4d_blocked.js (5387B)

---

 /**
 * @license Apache-2.0
 *
 * Copyright (c) 2021 The Stdlib Authors.
 *
 * Licensed under the Apache License, Version 2.0 (the "License");
 * you may not use this file except in compliance with the License.
 * You may obtain a copy of the License at
 *
 *    http://www.apache.org/licenses/LICENSE-2.0
 *
 * Unless required by applicable law or agreed to in writing, software
 * distributed under the License is distributed on an "AS IS" BASIS,
 * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
 * See the License for the specific language governing permissions and
 * limitations under the License.
 */
 
 /* eslint-disable max-depth */
 
 'use strict';
 
 // MODULES //
 
 var init = require( './init.js' );
 
 
 // MAIN //
 
 /**
 * Applies a unary callback to elements in a four-dimensional input ndarray and assigns results to elements in an equivalently shaped output ndarray via loop blocking.
 *
 * @private
 * @param {Object} x - object containing input ndarray meta data
 * @param {string} x.dtype - data type
 * @param {Collection} x.data - data buffer
 * @param {NonNegativeIntegerArray} x.shape - dimensions
 * @param {IntegerArray} x.strides - stride lengths
 * @param {NonNegativeInteger} x.offset - index offset
 * @param {string} x.order - specifies whether `x` is row-major (C-style) or column-major (Fortran-style)
 * @param {Object} y - object containing output ndarray meta data
 * @param {string} y.dtype - data type
 * @param {Collection} y.data - data buffer
 * @param {NonNegativeIntegerArray} y.shape - dimensions
 * @param {IntegerArray} y.strides - stride lengths
 * @param {NonNegativeInteger} y.offset - index offset
 * @param {string} y.order - specifies whether `y` is row-major (C-style) or column-major (Fortran-style)
 * @param {Callback} fcn - unary callback
 *
 * @example
 * var Float64Array = require( '@stdlib/array/float64' );
 *
 * function scale( x ) {
 *     return x * 10.0;
 * }
 *
 * // Create data buffers:
 * var xbuf = new Float64Array( [ 1.0, 2.0, 3.0, 4.0, 5.0, 6.0, 7.0, 8.0, 9.0, 10.0, 11.0, 12.0 ] );
 * var ybuf = new Float64Array( 6 );
 *
 * // Define the shape of the input and output arrays:
 * var shape = [ 1, 3, 1, 2 ];
 *
 * // Define the array strides:
 * var sx = [ 12, 4, 4, 1 ];
 * var sy = [ 6, 2, 2, 1 ];
 *
 * // Define the index offsets:
 * var ox = 1;
 * var oy = 0;
 *
 * // Create the input and output ndarray-like objects:
 * var x = {
 *     'dtype': 'float64',
 *     'data': xbuf,
 *     'shape': shape,
 *     'strides': sx,
 *     'offset': ox,
 *     'order': 'row-major'
 * };
 * var y = {
 *     'dtype': 'float64',
 *     'data': ybuf,
 *     'shape': shape,
 *     'strides': sy,
 *     'offset': oy,
 *     'order': 'row-major'
 * };
 *
 * // Apply the unary function:
 * blockedunary4d( x, y, scale );
 *
 * console.log( y.data );
 * // => <Float64Array>[ 20.0, 30.0, 60.0, 70.0, 100.0, 110.0 ]
 */
 function blockedunary4d( x, y, fcn ) {
 	var bsize;
 	var xbuf;
 	var ybuf;
 	var dx0;
 	var dx1;
 	var dx2;
 	var dx3;
 	var dy0;
 	var dy1;
 	var dy2;
 	var dy3;
 	var ox1;
 	var ox2;
 	var ox3;
 	var oy1;
 	var oy2;
 	var oy3;
 	var sh;
 	var s0;
 	var s1;
 	var s2;
 	var s3;
 	var sx;
 	var sy;
 	var ox;
 	var oy;
 	var ix;
 	var iy;
 	var i0;
 	var i1;
 	var i2;
 	var i3;
 	var j0;
 	var j1;
 	var j2;
 	var j3;
 	var o;
 
 	// Note on variable naming convention: s#, dx#, dy#, i#, j# where # corresponds to the loop number, with `0` being the innermost loop...
 
 	// Initialize and unpack block data:
 	o = init( x, y );
 	sh = o.sh;
 	sx = o.sx;
 	sy = o.sy;
 	bsize = o.bsize;
 
 	// Cache the indices of the first indexed elements in the respective ndarrays...
 	ox = x.offset;
 	oy = y.offset;
 
 	// Cache references to the input and output ndarray buffers...
 	xbuf = x.data;
 	ybuf = y.data;
 
 	// Cache offset increments for the innermost loop...
 	dx0 = sx[0];
 	dy0 = sy[0];
 
 	// Iterate over blocks...
 	for ( j3 = sh[3]; j3 > 0; ) {
 		if ( j3 < bsize ) {
 			s3 = j3;
 			j3 = 0;
 		} else {
 			s3 = bsize;
 			j3 -= bsize;
 		}
 		ox3 = ox + ( j3*sx[3] );
 		oy3 = oy + ( j3*sy[3] );
 		for ( j2 = sh[2]; j2 > 0; ) {
 			if ( j2 < bsize ) {
 				s2 = j2;
 				j2 = 0;
 			} else {
 				s2 = bsize;
 				j2 -= bsize;
 			}
 			dx3 = sx[3] - ( s2*sx[2] );
 			dy3 = sy[3] - ( s2*sy[2] );
 			ox2 = ox3 + ( j2*sx[2] );
 			oy2 = oy3 + ( j2*sy[2] );
 			for ( j1 = sh[1]; j1 > 0; ) {
 				if ( j1 < bsize ) {
 					s1 = j1;
 					j1 = 0;
 				} else {
 					s1 = bsize;
 					j1 -= bsize;
 				}
 				dx2 = sx[2] - ( s1*sx[1] );
 				dy2 = sy[2] - ( s1*sy[1] );
 				ox1 = ox2 + ( j1*sx[1] );
 				oy1 = oy2 + ( j1*sy[1] );
 				for ( j0 = sh[0]; j0 > 0; ) {
 					if ( j0 < bsize ) {
 						s0 = j0;
 						j0 = 0;
 					} else {
 						s0 = bsize;
 						j0 -= bsize;
 					}
 					// Compute index offsets for the first input and output ndarray elements in the current block...
 					ix = ox1 + ( j0*sx[0] );
 					iy = oy1 + ( j0*sy[0] );
 
 					// Compute loop offset increments...
 					dx1 = sx[1] - ( s0*sx[0] );
 					dy1 = sy[1] - ( s0*sy[0] );
 
 					// Iterate over the ndarray dimensions...
 					for ( i3 = 0; i3 < s3; i3++ ) {
 						for ( i2 = 0; i2 < s2; i2++ ) {
 							for ( i1 = 0; i1 < s1; i1++ ) {
 								for ( i0 = 0; i0 < s0; i0++ ) {
 									ybuf[ iy ] = fcn( xbuf[ ix ] );
 									ix += dx0;
 									iy += dy0;
 								}
 								ix += dx1;
 								iy += dy1;
 							}
 							ix += dx2;
 							iy += dy2;
 						}
 						ix += dx3;
 						iy += dy3;
 					}
 				}
 			}
 		}
 	}
 }
 
 
 // EXPORTS //
 
 module.exports = blockedunary4d;