smile.math.distance

Class SparseMinkowskiDistance

java.lang.Object

smile.math.distance.SparseMinkowskiDistance

All Implemented Interfaces:

Distance<SparseArray>, Metric<SparseArray>

public class SparseMinkowskiDistance
extends Object
implements Metric<SparseArray>

Minkowski distance of order p or L_p-norm, is a generalization of Euclidean distance that is actually L₂-norm. You may also provide a specified weight vector. For float or double arrays, missing values (i.e. NaN) are also handled. Also support sparse arrays of which zeros are excluded to save space.

Author:

Haifeng Li

Constructor Summary

Constructors

Constructor and Description
SparseMinkowskiDistance(int p) Constructor.
SparseMinkowskiDistance(int p, double[] weight) Constructor.

Method Summary

Methods

Modifier and Type	Method and Description
double	d(SparseArray x, SparseArray y) Returns the distance measure between two objects.
String	toString()

Methods inherited from class java.lang.Object

clone, equals, finalize, getClass, hashCode, notify, notifyAll, wait, wait, wait

Constructor Detail

SparseMinkowskiDistance

public SparseMinkowskiDistance(int p)

Constructor.

SparseMinkowskiDistance

public SparseMinkowskiDistance(int p,
                       double[] weight)

Constructor.

Parameters:

weight - the weight vector.

Method Detail

toString

public String toString()

Overrides:

toString in class Object

d

public double d(SparseArray x,
       SparseArray y)

Description copied from interface: Distance

Returns the distance measure between two objects.

Specified by:

d in interface Distance<SparseArray>