Package org.apache.commons.math4.neuralnet

Class MapRanking

java.lang.Object

org.apache.commons.math4.neuralnet.MapRanking

public class MapRanking
extends Object

Utility for ranking the units (neurons) of a network.

Since:

4.0

Constructor Summary

Constructors

Constructor	Description
MapRanking(Iterable<Neuron> neurons, DistanceMeasure distance)

Method Summary

Modifier and Type	Method	Description
List<Neuron>	rank(double[] features)	Creates a list of the neurons whose features best correspond to the given features.
List<Neuron>	rank(double[] features, int max)	Creates a list of the neurons whose features best correspond to the given features.

Methods inherited from class java.lang.Object

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

Constructor Detail

MapRanking

public MapRanking(Iterable<Neuron> neurons,
                  DistanceMeasure distance)

Parameters:

neurons - List to be ranked. No defensive copy is performed. The created list of units will be sorted in increasing order of the distance.

distance - Distance function.

Method Detail

rank

public List<Neuron> rank(double[] features)

Creates a list of the neurons whose features best correspond to the given features.

Parameters:

features - Data.

Returns:

the list of neurons sorted in decreasing order of distance to the given data.

Throws:

IllegalArgumentException - if the size of the input is not compatible with the neurons features size.

rank

public List<Neuron> rank(double[] features,
                         int max)

Creates a list of the neurons whose features best correspond to the given features.

Parameters:

features - Data.

max - Maximum size of the returned list.

Returns:

the list of neurons sorted in decreasing order of distance to the given data.

Throws:

IllegalArgumentException - if the size of the input is not compatible with the neurons features size or max <= 0.