Modifier and Type	Class and Description
private static class	MapUtils.PairNeuronDouble Helper data structure holding a (Neuron, double) pair.

Access	Constructor and Description
private	MapUtils() Class contains only static methods.

Method Summary

Modifier and Type	Method and Description
public static int[][]	Returns: the number of hits for each neuron in the map. computeHitHistogram(Iterable<double[]> Feature vectors. data, NeuronSquareMesh2D Network. map, DistanceMeasure Function to use for determining the best matching unit. distance) Computes the "hit" histogram of a two-dimensional map.
public static double	Returns: the error. computeQuantizationError(Iterable<double[]> Feature vectors. data, Iterable<Neuron> List of neurons to scan. neurons, DistanceMeasure Distance function. distance) Computes the quantization error.
public static double	Returns: the error. computeTopographicError(Iterable<double[]> Feature vectors. data, Network Network. net, DistanceMeasure Distance function. distance) Computes the topographic error.
public static double[][]	Returns: the matrix of average distances. computeU(NeuronSquareMesh2D Network. map, DistanceMeasure Function to use for computing the average distance from a neuron to its neighbours. distance) Computes the U-matrix of a two-dimensional map.
public static Neuron	Returns: the neuron whose features are closest to the given data. findBest(double[] Data. features, Iterable<Neuron> List of neurons to scan. If the list is empty `null` will be returned. neurons, DistanceMeasure Distance function. The neuron's features are passed as the first argument to `DistanceMeasure#compute(double[], double[])`. distance) Finds the neuron that best matches the given features.
public static Pair<Neuron, Neuron>	Returns: the two neurons whose features are closest to the given data. findBestAndSecondBest(double[] Data. features, Iterable<Neuron> List of neurons to scan. If the list is empty `null` will be returned. neurons, DistanceMeasure Distance function. The neuron's features are passed as the first argument to `DistanceMeasure#compute(double[], double[])`. distance) Finds the two neurons that best match the given features.
public static Neuron[]	Returns: the neurons, sorted in increasing order of distance in data space. sort(double[] Data. features, Iterable<Neuron> List of neurons to scan. If it is empty, an empty array will be returned. neurons, DistanceMeasure Distance function. distance) Creates a list of neurons sorted in increased order of the distance to the given `features`.

Inherited from java.lang.Object:: clone equals finalize getClass hashCode notify notifyAll toString wait wait wait

Constructor Detail

MapUtils	back to summary
private MapUtils() Class contains only static methods.

Method Detail

computeHitHistogram	back to summary
public static int[][] computeHitHistogram(Iterable<double[]> data, NeuronSquareMesh2D map, DistanceMeasure distance) Computes the "hit" histogram of a two-dimensional map. Parameters data:Iterable<double[]> Feature vectors. map:NeuronSquareMesh2D Network. distance:DistanceMeasure Function to use for determining the best matching unit. Returns:int[][] the number of hits for each neuron in the map.

computeHitHistogram

back to summary

public static int[][] computeHitHistogram(Iterable<double[]> data, NeuronSquareMesh2D map, DistanceMeasure distance)

Computes the "hit" histogram of a two-dimensional map.

Parameters

data:Iterable<double[]>: Feature vectors.
map:NeuronSquareMesh2D: Network.
distance:DistanceMeasure: Function to use for determining the best matching unit.

Returns:int[][]

the number of hits for each neuron in the map.

computeQuantizationError back to summary

computeQuantizationError	back to summary
public static double computeQuantizationError(Iterable<double[]> data, Iterable<Neuron> neurons, DistanceMeasure distance) Computes the quantization error. The quantization error is the average distance between a feature vector and its "best matching unit" (closest neuron). Parameters data:Iterable<double[]> Feature vectors. neurons:Iterable<Neuron> List of neurons to scan. distance:DistanceMeasure Distance function. Returns:double the error. Exceptions NoDataException: if `data` is empty.

public static double computeQuantizationError(Iterable<double[]> data, Iterable<Neuron> neurons, DistanceMeasure distance)

Computes the quantization error. The quantization error is the average distance between a feature vector and its "best matching unit" (closest neuron).

Parameters

data:Iterable<double[]>: Feature vectors.
neurons:Iterable<Neuron>: List of neurons to scan.
distance:DistanceMeasure: Distance function.

Returns:double

the error.

Exceptions

NoDataException:: if data is empty.

computeTopographicError back to summary

computeTopographicError	back to summary
public static double computeTopographicError(Iterable<double[]> data, Network net, DistanceMeasure distance) Computes the topographic error. The topographic error is the proportion of data for which first and second best matching units are not adjacent in the map. Parameters data:Iterable<double[]> Feature vectors. net:Network Network. distance:DistanceMeasure Distance function. Returns:double the error. Exceptions NoDataException: if `data` is empty.

public static double computeTopographicError(Iterable<double[]> data, Network net, DistanceMeasure distance)

Computes the topographic error. The topographic error is the proportion of data for which first and second best matching units are not adjacent in the map.

Parameters

data:Iterable<double[]>: Feature vectors.
net:Network: Network.
distance:DistanceMeasure: Distance function.

Returns:double

the error.

Exceptions

NoDataException:: if data is empty.

computeU	back to summary
public static double[][] computeU(NeuronSquareMesh2D map, DistanceMeasure distance) Computes the U-matrix of a two-dimensional map. Parameters map:NeuronSquareMesh2D Network. distance:DistanceMeasure Function to use for computing the average distance from a neuron to its neighbours. Returns:double[][] the matrix of average distances.

computeU

back to summary

public static double[][] computeU(NeuronSquareMesh2D map, DistanceMeasure distance)

Computes the U-matrix of a two-dimensional map.

Parameters

map:NeuronSquareMesh2D: Network.
distance:DistanceMeasure: Function to use for computing the average distance from a neuron to its neighbours.

Returns:double[][]

the matrix of average distances.

findBest back to summary

findBest	back to summary
public static Neuron findBest(double[] features, Iterable<Neuron> neurons, DistanceMeasure distance) Finds the neuron that best matches the given features. Parameters features:double[] Data. neurons:Iterable<Neuron> List of neurons to scan. If the list is empty `null` will be returned. distance:DistanceMeasure Distance function. The neuron's features are passed as the first argument to `DistanceMeasure#compute(double[], double[])`. Returns:Neuron the neuron whose features are closest to the given data. Exceptions DimensionMismatchException: if the size of the input is not compatible with the neurons features size.

public static Neuron findBest(double[] features, Iterable<Neuron> neurons, DistanceMeasure distance)

Finds the neuron that best matches the given features.

Parameters

features:double[]: Data.
neurons:Iterable<Neuron>: List of neurons to scan. If the list is empty null will be returned.
distance:DistanceMeasure: Distance function. The neuron's features are passed as the first argument to DistanceMeasure#compute(double[], double[]).

Returns:Neuron

the neuron whose features are closest to the given data.

Exceptions

DimensionMismatchException:: if the size of the input is not compatible with the neurons features size.

findBestAndSecondBest back to summary

findBestAndSecondBest	back to summary
public static Pair<Neuron, Neuron> findBestAndSecondBest(double[] features, Iterable<Neuron> neurons, DistanceMeasure distance) Finds the two neurons that best match the given features. Parameters features:double[] Data. neurons:Iterable<Neuron> List of neurons to scan. If the list is empty `null` will be returned. distance:DistanceMeasure Distance function. The neuron's features are passed as the first argument to `DistanceMeasure#compute(double[], double[])`. Returns:Pair<Neuron, Neuron> the two neurons whose features are closest to the given data. Exceptions DimensionMismatchException: if the size of the input is not compatible with the neurons features size.

public static Pair<Neuron, Neuron> findBestAndSecondBest(double[] features, Iterable<Neuron> neurons, DistanceMeasure distance)

Finds the two neurons that best match the given features.

Parameters

features:double[]: Data.
neurons:Iterable<Neuron>: List of neurons to scan. If the list is empty null will be returned.
distance:DistanceMeasure: Distance function. The neuron's features are passed as the first argument to DistanceMeasure#compute(double[], double[]).

Returns:Pair<Neuron, Neuron>

the two neurons whose features are closest to the given data.

Exceptions

DimensionMismatchException:: if the size of the input is not compatible with the neurons features size.

sort back to summary

sort	back to summary
public static Neuron[] sort(double[] features, Iterable<Neuron> neurons, DistanceMeasure distance) Creates a list of neurons sorted in increased order of the distance to the given `features`. Parameters features:double[] Data. neurons:Iterable<Neuron> List of neurons to scan. If it is empty, an empty array will be returned. distance:DistanceMeasure Distance function. Returns:Neuron[] the neurons, sorted in increasing order of distance in data space. Exceptions DimensionMismatchException: if the size of the input is not compatible with the neurons features size. Since 3.6 See Also `findBest(double[], Iterable, DistanceMeasure)`, `findBestAndSecondBest(double[], Iterable, DistanceMeasure)`

public static Neuron[] sort(double[] features, Iterable<Neuron> neurons, DistanceMeasure distance)

Creates a list of neurons sorted in increased order of the distance to the given features.

Parameters

features:double[]: Data.
neurons:Iterable<Neuron>: List of neurons to scan. If it is empty, an empty array will be returned.
distance:DistanceMeasure: Distance function.

Returns:Neuron[]

the neurons, sorted in increasing order of distance in data space.

Exceptions

DimensionMismatchException:: if the size of the input is not compatible with the neurons features size.

Since

3.6

org.apache.commons.math3.ml.neuralnet back to summary

private Class MapUtils.PairNeuronDouble

Class Inheritance

extends Object

java.lang.Object
org.apache.commons.math3.ml.neuralnet.MapUtils.PairNeuronDouble

Helper data structure holding a (Neuron, double) pair.

Field Summary

Modifier and Type	Field and Description
pack-priv static final Comparator<MapUtils.PairNeuronDouble>	COMPARATOR Comparator.
private final Neuron	neuron Key.
private final double	value Value.

Constructor Summary

Access	Constructor and Description
pack-priv	PairNeuronDouble(Neuron Neuron. neuron, double Value. value)

Method Summary

Modifier and Type	Method and Description
public Neuron	Returns: the neuron. getNeuron()

Inherited from java.lang.Object:: clone equals finalize getClass hashCode notify notifyAll toString wait wait wait

Field Detail

COMPARATOR	back to summary
pack-priv static final Comparator<MapUtils.PairNeuronDouble> COMPARATOR Comparator.

neuron	back to summary
private final Neuron neuron Key.

value	back to summary
private final double value Value.

Constructor Detail

PairNeuronDouble	back to summary
pack-priv PairNeuronDouble(Neuron neuron, double value) Parameters neuron:Neuron Neuron. value:double Value.

Method Detail

getNeuron	back to summary
public Neuron getNeuron() Returns:Neuron the neuron.

public Class MapUtils

Nested and Inner Type Summary

Constructor Summary

Method Summary

Constructor Detail

Method Detail

private Class MapUtils.PairNeuronDouble

Field Summary

Constructor Summary

Method Summary

Field Detail

Constructor Detail

Method Detail