org.joone.util

Class ToBinaryPlugin

java.lang.Object

org.joone.util.AbstractConverterPlugIn

org.joone.util.ConverterPlugIn

org.joone.util.ToBinaryPlugin

All Implemented Interfaces:

java.io.Serializable, PlugInListener

public class ToBinaryPlugin
extends ConverterPlugIn

This plug-in converts 10-base data to binary format. The plug-in ingnores the broken part (the part after the . ) for non-integer numbers. It works correct for positive as well as for negative numbers.

See Also:

Serialized Form

Constructor Summary

Constructors

Constructor and Description
ToBinaryPlugin() Creates a new instance of ToBinaryPlugin
ToBinaryPlugin(java.lang.String anAdvancedSerieSelector) Creates a new instance of ToBinaryPlugin

Method Summary

Modifier and Type	Method and Description
double	getLowerBit() Gets the value used for the lower bit.
double	getUpperBit() Gets the value used for the upper bit.
void	setLowerBit(double aValue) Sets the value for the lower bit.
void	setUpperBit(double aValue) Sets the value for the upper bit.

Methods inherited from class org.joone.util.ConverterPlugIn

isApplyEveryCycle, newCycle, setApplyEveryCycle

Methods inherited from class org.joone.util.AbstractConverterPlugIn

addPlugIn, addPlugInListener, check, convertPatterns, dataChanged, getAdvancedSerieSelector, getName, getNextPlugIn, isConnected, removeAllPlugIns, removePlugInListener, setAdvancedSerieSelector, setConnected, setInputVector, setName, setNextPlugin, setNextPlugIn

Methods inherited from class java.lang.Object

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

Constructor Detail

ToBinaryPlugin

public ToBinaryPlugin()

Creates a new instance of ToBinaryPlugin

ToBinaryPlugin

public ToBinaryPlugin(java.lang.String anAdvancedSerieSelector)

Creates a new instance of ToBinaryPlugin

Parameters:

anAdvancedSerieSelector - the advanced serie selector to use.

See Also:

setAdvancedSerieSelector()

Method Detail

setUpperBit

public void setUpperBit(double aValue)

Sets the value for the upper bit. In binary problems it is often better to use 0.3 and 0.7 or -0.7 and 0.7 as target instead of 0 and 1 or -1 and 1, because the asymptotes (0 and 1) tend to take a long time to train, worsen generalization and drive the weights to very large values. By using this function you can set a different value for the upper bit.

Parameters:

aValue - the value to use for the upper bit.

getUpperBit

public double getUpperBit()

Gets the value used for the upper bit.

setLowerBit

public void setLowerBit(double aValue)

Sets the value for the lower bit.

Parameters:

aValue - the value to use for the lower bit.

getLowerBit

public double getLowerBit()

Gets the value used for the lower bit.