gov.sandia.cognition.learning.function.distance

Class MinkowskiDistanceMetric

java.lang.Object

gov.sandia.cognition.util.AbstractCloneableSerializable

gov.sandia.cognition.learning.function.distance.MinkowskiDistanceMetric

All Implemented Interfaces:

DivergenceFunction<Vectorizable,Vectorizable>, Metric<Vectorizable>, Semimetric<Vectorizable>, CloneableSerializable, java.io.Serializable, java.lang.Cloneable

@PublicationReference(title="Minkowski Distance",
                      author="Wikipedia",
                      year=2011,
                      type=WebPage,
                      url="http://en.wikipedia.org/wiki/Minkowski_distance")
public class MinkowskiDistanceMetric
extends AbstractCloneableSerializable
implements Metric<Vectorizable>

An implementation of the Minkowski distance metric. The metric is a generalization of the 2-norm (Euclidean) and 1-norm (Manhattan) distances to an arbitrary p-norm (p > 0). It is defined as: d(x, y) = (||x - y||_p)^(1/p) To support a power of infinity, see ChebyshevDistanceMetric.

Since:

3.3.3

Author:

Justin Basilico

See Also:

Serialized Form

Field Summary

Fields

Modifier and Type	Field and Description
static double	DEFAULT_POWER The default power is 2.0.
protected double	power The power that the distance is computed to.

Constructor Summary

Constructors

Constructor and Description
MinkowskiDistanceMetric() Creates a new MinkowskiDistanceMetric with the default power of 2.0.
MinkowskiDistanceMetric(double power) Creates a new MinkowskiDistanceMetric with the given power.

Method Summary

Modifier and Type	Method and Description
double	evaluate(Vectorizable first, Vectorizable second) Evaluates the divergence between the two given objects.
double	getPower() Gets the power used for the distance.
void	setPower(double power) Sets the power used for the distance.

Methods inherited from class gov.sandia.cognition.util.AbstractCloneableSerializable

clone

Methods inherited from class java.lang.Object

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

Field Detail

DEFAULT_POWER

public static final double DEFAULT_POWER

The default power is 2.0.

See Also:

Constant Field Values

power

protected double power

The power that the distance is computed to.

Constructor Detail

MinkowskiDistanceMetric

public MinkowskiDistanceMetric()

Creates a new MinkowskiDistanceMetric with the default power of 2.0.

MinkowskiDistanceMetric

public MinkowskiDistanceMetric(double power)

Creates a new MinkowskiDistanceMetric with the given power.

Parameters:

power - The power for the distance metric. Must be positive.

Method Detail

evaluate

public double evaluate(Vectorizable first,
                       Vectorizable second)

Description copied from interface: DivergenceFunction

Evaluates the divergence between the two given objects.

Specified by:

evaluate in interface DivergenceFunction<Vectorizable,Vectorizable>

Parameters:

first - The first object.

second - The second object.

Returns:

The divergence between the objects.

getPower

public double getPower()

Gets the power used for the distance.

Returns:

The power used for the distance. Must be positive.

setPower

public void setPower(double power)

Sets the power used for the distance.

Parameters:

power - The power used for the distance. Must be positive.