The SUMO toolbox  2018a
 All Classes Namespaces Files Functions Variables Typedefs Enumerations Enumerator Pages
Functions
buildFractionalDistanceMatrix.m File Reference

Functions

function buildFractionalDistanceMatrix (var samples, var targets, var d)
 Both `samples' and `targets' give an array of d-dimensional points. More...
 

Detailed Description

Authors
SUMO Lab Team
Version
2018a
Date
Copyright 2006-2018

This file is part of the Surrogate Modeling Toolbox ("SUMO Toolbox") and you can redistribute it and/or modify it under the terms of the GNU Affero General Public License version 3 as published by the Free Software Foundation. With the additional provision that a commercial license must be purchased if the SUMO Toolbox is used, modified, or extended in a commercial setting. For details see the included LICENSE.txt file. When referring to the SUMO Toolbox please make reference to the corresponding publication:

Contact : sumo@.nosp@m.sumo.nosp@m..inte.nosp@m.c.ug.nosp@m.ent.b.nosp@m.e - http://sumo.intec.ugent.be Signature [ distances ] = buildFractionalDistanceMatrix( samples, targets, d )

Function Documentation

function buildFractionalDistanceMatrix ( var  samples,
var  targets,
var  d 
)

Both `samples' and `targets' give an array of d-dimensional points.

If `samples' is N x d and `targets' is M x d, then this function returns an N x M matrix, where the i,j element is the carthesian distance between samples_i and targets_j If `targets' is omitted or empty, it is assumed to be equal to `samples', in that case the returned matrix is square and symmetrical, and has zeros on its diagonal.

The parameter fraction determines the order of the fractional pseudonorm, if it is omitted it is chosen as floor(log(d+1)), (the log is used to lower the order for very high dimensional problems, as this results in potentially unstable computation).

Example:

buildFractionalDistanceMatrix( [ 0 0 ; 1 1 ; 2 2 ], [ 1 0 ; 0 1 ] )

ans = 1 1 1 1 3 3