jsat.classifiers.neuralnetwork

Class SOM

java.lang.Object

jsat.classifiers.neuralnetwork.SOM

All Implemented Interfaces:

Serializable, Cloneable, Classifier, Parameterized

public class SOM
extends Object
implements Classifier, Parameterized

An implementation of a Self Organizing Map, also called a Kohonen Map. It is linked to many other algorithms, and is an unsupervised learning algorithm that can perform classification.

The SOM is useful for visualizing data sets, though this is not yet implemented.

Author:

Edward Raff

See Also:

Serialized Form

Field Summary

Fields

Modifier and Type	Field and Description
static KernelFunction	DEFAULT_KF
static DecayRate	DEFAULT_LEARNING_DECAY
static double	DEFAULT_LEARNING_RATE
static int	DEFAULT_MAX_ITERS
static DecayRate	DEFAULT_NEIGHBOR_DECAY

Constructor Summary

Constructors

Constructor and Description
SOM(DistanceMetric dm, int somHeight, int somWeight) Creates a new SOM using the given parameters
SOM(DistanceMetric dm, int somHeight, int somWeight, VectorCollectionFactory<VecPaired<Vec,Integer>> vcFactory) Creates a new SOM using the given parameters
SOM(int somHeight, int somWeight) Creates a new SOM using the given parameters using the EuclideanDistance

Method Summary

Modifier and Type	Method and Description
CategoricalResults	classify(DataPoint data) Performs classification on the given data point.
SOM	clone()
double	getInitialLearningRate() Returns the rate at which input is incorporated at each iteration of the SOM
DecayRate	getLearningDecay() The rate the SOM learns decays over each iteration, and this defines the way in which the rate decays.
int	getMaxIterations() Returns the maximum number of iterations that will be used to converge
DecayRate	getNeighborDecay() The range of effect each data point has decays with each iteration, and this defines the way in which the rate decays.
Parameter	getParameter(String paramName) Returns the parameter with the given name.
List<Parameter>	getParameters() Returns the list of parameters that can be altered for this learner.
int	getSomHeight() Returns the height of the SOM lattice to create
int	getSomWidth() Returns the width of the SOM lattice to create
void	setInitialLearningRate(double initialLearningRate) Sets the rate at which input is incorporated at each iteration of the SOM algorithm
void	setLearningDecay(DecayRate learningDecay) The rate the SOM learns decays over each iteration, and this defines the way in which the rate decays.
void	setMaxIterations(int maxIters) Sets the maximum number of iterations that will be used to converge
void	setNeighborDecay(DecayRate neighborDecay) The range of effect each data point has decays with each iteration, and this defines the way in which the rate decays.
void	setSomHeight(int somHeight) Sets the height of the SOM lattice to create
void	setSomWidth(int somWidth) Sets the width of the SOM lattice to create
boolean	supportsWeightedData() Indicates whether the model knows how to train using weighted data points.
void	trainC(ClassificationDataSet dataSet) Trains the classifier and constructs a model for classification using the given data set.
void	trainC(ClassificationDataSet dataSet, ExecutorService threadPool) Trains the classifier and constructs a model for classification using the given data set.

Methods inherited from class java.lang.Object

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

Field Detail

DEFAULT_MAX_ITERS

public static final int DEFAULT_MAX_ITERS

See Also:

Constant Field Values

DEFAULT_KF

public static final KernelFunction DEFAULT_KF

DEFAULT_LEARNING_RATE

public static final double DEFAULT_LEARNING_RATE

See Also:

Constant Field Values

DEFAULT_LEARNING_DECAY

public static final DecayRate DEFAULT_LEARNING_DECAY

DEFAULT_NEIGHBOR_DECAY

public static final DecayRate DEFAULT_NEIGHBOR_DECAY

Constructor Detail

SOM

public SOM(int somHeight,
           int somWeight)

Creates a new SOM using the given parameters using the EuclideanDistance

Parameters:

somHeight - the height of the SOM lattice

somWeight - the weight of the SOM lattice

SOM

public SOM(DistanceMetric dm,
           int somHeight,
           int somWeight)

Creates a new SOM using the given parameters

Parameters:

dm - the distance metric to use when comparing points

somHeight - the height of the SOM lattice

somWeight - the weight of the SOM lattice

SOM

public SOM(DistanceMetric dm,
           int somHeight,
           int somWeight,
           VectorCollectionFactory<VecPaired<Vec,Integer>> vcFactory)

Creates a new SOM using the given parameters

Parameters:

dm - the distance metric to use when comparing points

somHeight - the height of the SOM lattice

somWeight - the weight of the SOM lattice

vcFactory - the vector collection factory to use for containing points

Method Detail

setMaxIterations

public void setMaxIterations(int maxIters)

Sets the maximum number of iterations that will be used to converge

Parameters:

maxIters - the max iterations of the algorithm

getMaxIterations

public int getMaxIterations()

Returns the maximum number of iterations that will be used to converge

Returns:

the max iterations of the algorithm

setSomWidth

public void setSomWidth(int somWidth)

Sets the width of the SOM lattice to create

Parameters:

somWidth - the width of the lattice

setSomHeight

public void setSomHeight(int somHeight)

Sets the height of the SOM lattice to create

Parameters:

somHeight - the height of the lattice

getSomHeight

public int getSomHeight()

Returns the height of the SOM lattice to create

Returns:

the height of the lattice

getSomWidth

public int getSomWidth()

Returns the width of the SOM lattice to create

Returns:

the width of the lattice

setInitialLearningRate

public void setInitialLearningRate(double initialLearningRate)

Sets the rate at which input is incorporated at each iteration of the SOM algorithm

Parameters:

initialLearningRate - the rate the SOM learns at

getInitialLearningRate

public double getInitialLearningRate()

Returns the rate at which input is incorporated at each iteration of the SOM

Returns:

the rate the SOM learns at

setLearningDecay

public void setLearningDecay(DecayRate learningDecay)

The rate the SOM learns decays over each iteration, and this defines the way in which the rate decays.

Parameters:

learningDecay - the decay for the learning rate

getLearningDecay

public DecayRate getLearningDecay()

The rate the SOM learns decays over each iteration, and this defines the way in which the rate decays.

Returns:

the decay for the learning rate

setNeighborDecay

public void setNeighborDecay(DecayRate neighborDecay)

The range of effect each data point has decays with each iteration, and this defines the way in which the rate decays.

Parameters:

neighborDecay - the decay for the neighbor range.

getNeighborDecay

public DecayRate getNeighborDecay()

The range of effect each data point has decays with each iteration, and this defines the way in which the rate decays.

Returns:

the decay for the neighbor range.

getParameters

public List<Parameter> getParameters()

Description copied from interface: Parameterized

Returns the list of parameters that can be altered for this learner.

Specified by:

getParameters in interface Parameterized

Returns:

the list of parameters that can be altered for this learner.

getParameter

public Parameter getParameter(String paramName)

Description copied from interface: Parameterized

Returns the parameter with the given name. Two different strings may map to a single Parameter object. An ASCII only string, and a Unicode style string.

Specified by:

getParameter in interface Parameterized

Parameters:

paramName - the name of the parameter to obtain

Returns:

the Parameter in question, or null if no such named Parameter exists.

classify

public CategoricalResults classify(DataPoint data)

Description copied from interface: Classifier

Performs classification on the given data point.

Specified by:

classify in interface Classifier

Parameters:

data - the data point to classify

Returns:

the results of the classification.

trainC

public void trainC(ClassificationDataSet dataSet,
                   ExecutorService threadPool)

Description copied from interface: Classifier

Trains the classifier and constructs a model for classification using the given data set. If the training method knows how, it will used the threadPool to conduct training in parallel. This method will block until the training has completed.

Specified by:

trainC in interface Classifier

Parameters:

dataSet - the data set to train on

threadPool - the source of threads to use.

trainC

public void trainC(ClassificationDataSet dataSet)

Description copied from interface: Classifier

Trains the classifier and constructs a model for classification using the given data set.

Specified by:

trainC in interface Classifier

Parameters:

dataSet - the data set to train on

supportsWeightedData

public boolean supportsWeightedData()

Description copied from interface: Classifier

Indicates whether the model knows how to train using weighted data points. If it does, the model will train assuming the weights. The values returned by this method may change depending on the parameters set for the model.

Specified by:

supportsWeightedData in interface Classifier

Returns:

true if the model supports weighted data, false otherwise

clone

public SOM clone()

Specified by:

clone in interface Classifier

Overrides:

clone in class Object