org.encog.ml.bayesian.query.sample

Class SamplingQuery

java.lang.Object

org.encog.ml.bayesian.query.BasicQuery

org.encog.ml.bayesian.query.sample.SamplingQuery

All Implemented Interfaces:

java.io.Serializable, java.lang.Cloneable, BayesianQuery

public class SamplingQuery
extends BasicQuery
implements java.io.Serializable

A sampling query allows probabilistic queries on a Bayesian network. Sampling works by actually simulating the probabilities using a random number generator. A sample size must be specified. The higher the sample size, the more accurate the probability will be. However, the higher the sampling size, the longer it takes to run the query. An enumeration query is more precise than the sampling query. However, the enumeration query will become slow as the size of the Bayesian network grows. Sampling can often be used for a quick estimation of a probability.

See Also:

Serialized Form

Field Summary

Fields

Modifier and Type	Field and Description
static int	DEFAULT_SAMPLE_SIZE The default sample size.

Constructor Summary

Constructors

Constructor and Description
SamplingQuery(BayesianNetwork theNetwork) Construct a sampling query.

Method Summary

Modifier and Type	Method and Description
SamplingQuery	clone()
java.lang.String	dumpCurrentState()
void	execute() Execute the query.
double	getProbability()
int	getSampleSize()
void	setSampleSize(int sampleSize)
java.lang.String	toString()

Methods inherited from class org.encog.ml.bayesian.query.BasicQuery

defineEventType, finalizeStructure, getEvents, getEventState, getEventType, getEvidenceEvents, getNetwork, getOutcomeEvents, getProblem, isNeededEvidence, locateEventTypes, reset, satisfiesDesiredOutcome, setEventValue, setEventValue

Methods inherited from class java.lang.Object

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

Field Detail

DEFAULT_SAMPLE_SIZE

public static final int DEFAULT_SAMPLE_SIZE

The default sample size.

See Also:

Constant Field Values

Constructor Detail

SamplingQuery

public SamplingQuery(BayesianNetwork theNetwork)

Construct a sampling query.

Parameters:

theNetwork - The network that will be queried.

Method Detail

getSampleSize

public int getSampleSize()

Returns:

the sampleSize

setSampleSize

public void setSampleSize(int sampleSize)

Parameters:

sampleSize - the sampleSize to set

execute

public void execute()

Execute the query.

Specified by:

execute in interface BayesianQuery

getProbability

public double getProbability()

Specified by:

getProbability in interface BayesianQuery

Returns:

Obtains the probability after execute has been called.

dumpCurrentState

public java.lang.String dumpCurrentState()

Returns:

The current state as a string.

clone

public SamplingQuery clone()

Specified by:

clone in interface BayesianQuery

Overrides:

clone in class BasicQuery

toString

public java.lang.String toString()

Overrides:

toString in class java.lang.Object