org.apache.flink.ml.metrics.distances

Class MinkowskiDistanceMetric

java.lang.Object

org.apache.flink.ml.metrics.distances.MinkowskiDistanceMetric

All Implemented Interfaces:

Serializable, DistanceMetric

public class MinkowskiDistanceMetric
extends Object
implements DistanceMetric

This class implements a Minkowski distance metric. The metric is a generalization of L(p) distances: Euclidean distance and Manhattan distance. If you need for a special case of p = 1 or p = 2, use ManhattanDistanceMetric, EuclideanDistanceMetric. This class is useful for high exponents.

See Also:

http://en.wikipedia.org/wiki/Minkowski_distance, Serialized Form

Constructor Summary

Constructors

Constructor and Description
MinkowskiDistanceMetric(double p)

Method Summary

Modifier and Type	Method and Description
static MinkowskiDistanceMetric	apply(double p)
double	distance(Vector a, Vector b) Returns the distance between the arguments.
double	p()

Methods inherited from class java.lang.Object

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

Methods inherited from interface org.apache.flink.ml.metrics.distances.DistanceMetric

checkValidArguments

Constructor Detail

MinkowskiDistanceMetric

public MinkowskiDistanceMetric(double p)

Method Detail

apply

public static MinkowskiDistanceMetric apply(double p)

p

public double p()

distance

public double distance(Vector a,
                       Vector b)

Description copied from interface: DistanceMetric

Returns the distance between the arguments.

Specified by:

distance in interface DistanceMetric

Parameters:

a - a Vector defining a multi-dimensional point in some space

b - a Vector defining a multi-dimensional point in some space

Returns:

a scalar double of the distance