at.tuwien.ifs.somtoolbox.models

Class GrowingCellStructures

java.lang.Object

at.tuwien.ifs.somtoolbox.models.AbstractNetworkModel

at.tuwien.ifs.somtoolbox.models.GrowingCellStructures

All Implemented Interfaces:

NetworkModel

public class GrowingCellStructures
extends AbstractNetworkModel

This class implements the Growing Cell Structures. It is basically a wrapper for the GrowingCellLayer and mainly handles command line execution and parameters. It implements the NetworkModel interface which is currently not used, but may be used in the future.

Version:

$Id: GrowingCellStructures.java 3589 2010-05-21 10:42:01Z mayer $

Author:

Johannes Inführ, Andreas Zweng

Field Summary

Fields

Modifier and Type	Field and Description
private static float	alpha
private static float	epsilonB
private static float	epsilonN
private static float	eta
private static int	lamda
protected GrowingCellLayer	layer

Fields inherited from class at.tuwien.ifs.somtoolbox.models.AbstractNetworkModel

contentType, DEFAULT_LABEL_COUNT, labelled, sharedInputObjects, trainingStart

Constructor Summary

Constructors

Constructor and Description
GrowingCellStructures(int dim, boolean normalize, SOMProperties props, InputData data) Constructs a new GrowingCellStructures with dim-dimensional weight vectors.

Method Summary

Modifier and Type	Method and Description
GrowingCellLayer	getLayer()
static void	main(java.lang.String[] args) Method for stand-alone execution of map training.
private void	train(InputData data, SOMProperties props) Trains a GrowingCellStructures

Methods inherited from class at.tuwien.ifs.somtoolbox.models.AbstractNetworkModel

getDataContentType, getInputData, getSharedInputObjects, isLabelled, printTrainingTime, setLabelled, setSharedInputObjects

Methods inherited from class java.lang.Object

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

Field Detail

layer

protected GrowingCellLayer layer

epsilonB

private static float epsilonB

epsilonN

private static float epsilonN

alpha

private static float alpha

lamda

private static int lamda

eta

private static float eta

Constructor Detail

GrowingCellStructures

public GrowingCellStructures(int dim,
                             boolean normalize,
                             SOMProperties props,
                             InputData data)

Constructs a new GrowingCellStructures with dim-dimensional weight vectors. Argument norm determines whether the randomly initialized weight vectors should be normalized to unit length or not.

Parameters:

dim - the dimensionality of the weight vectors.

normalize - specifies if the weight vectors are to be normalized to unit length.

props - the network properties.

Method Detail

main

public static void main(java.lang.String[] args)

Method for stand-alone execution of map training. Options are:

• --noDWM switch to not write the data winner mapping file
• properties name of properties file, mandatory

Parameters:

args - the execution arguments as stated above.

train

private void train(InputData data,
                   SOMProperties props)

Trains a GrowingCellStructures

Parameters:

data - inputdata used for training

props - properties for training

getLayer

public GrowingCellLayer getLayer()