com.ibm.bluej.consistency.learning

Class WorldWeightParam

java.lang.Object

com.ibm.bluej.consistency.learning.WorldWeightParam

All Implemented Interfaces:

IWorldWeight, java.io.Serializable, java.lang.Cloneable

public class WorldWeightParam
extends java.lang.Object
implements IWorldWeight, java.lang.Cloneable

See Also:

Serialized Form

Constructor Summary

Constructors

Constructor and Description
WorldWeightParam()

Method Summary

Methods

Modifier and Type	Method and Description
void	badDelta(LearningState learningState) This SGWeight represents the difference between two states: this = weight(s2) - weight(s1).
void	clear()
WorldWeightParam	clone()
static WorldWeightParam	difference(WorldWeightParam w1, WorldWeightParam w2)
boolean	equals(java.lang.Object o)
double	getWeight()
int	hashCode()
boolean	hasParam()
void	merge(WorldWeightParam delta) Expert
void	scaleEveryPart(double scale) Expert usage
java.lang.String	toOptimizeString()
java.lang.String	toString()
void	update(double v)
void	update(Learnable p, double v)

Methods inherited from class java.lang.Object

getClass, notify, notifyAll, wait, wait, wait

Constructor Detail

WorldWeightParam

public WorldWeightParam()

Method Detail

toOptimizeString

public java.lang.String toOptimizeString()

scaleEveryPart

public void scaleEveryPart(double scale)

Expert usage

Parameters:

scale -

hasParam

public boolean hasParam()

toString

public java.lang.String toString()

Overrides:

toString in class java.lang.Object

clone

public WorldWeightParam clone()

Overrides:

clone in class java.lang.Object

difference

public static WorldWeightParam difference(WorldWeightParam w1,
                          WorldWeightParam w2)

merge

public void merge(WorldWeightParam delta)

Expert

Parameters:

delta -

getWeight

public double getWeight()

Specified by:

getWeight in interface IWorldWeight

clear

public void clear()

update

public void update(double v)

update

public void update(Learnable p,
          double v)

badDelta

public void badDelta(LearningState learningState)

This SGWeight represents the difference between two states: this = weight(s2) - weight(s1). gold(s1) > gold(s2). |weight(s1)| =< |weight(s2)|. So |this| is non-negative. NOTE: after any learning update recompute() will have to be called on all SGWeights

Parameters:

timestep - The number of SampleRanks so far, beginning at 1

hashCode

public int hashCode()

Overrides:

hashCode in class java.lang.Object

equals

public boolean equals(java.lang.Object o)

Overrides:

equals in class java.lang.Object