uk.ac.soton.simulation.jsit.core

Class DistCustomCategorical<C extends java.lang.Enum<C>>

java.lang.Object

uk.ac.soton.simulation.jsit.core.AbstractStochasticItem

uk.ac.soton.simulation.jsit.core.Distribution

uk.ac.soton.simulation.jsit.core.DistributionDiscrete

uk.ac.soton.simulation.jsit.core.DistributionCategorical<C>

uk.ac.soton.simulation.jsit.core.DistCustomCategorical<C>

All Implemented Interfaces:

java.io.Serializable, ContinuousSampler

public class DistCustomCategorical<C extends java.lang.Enum<C>>
extends DistributionCategorical<C>
implements java.io.Serializable

JSIT representation of a custom categorical distribution.

Since:

0.1

Author:

Stuart Rossiter

See Also:

Serialized Form

Nested Class Summary

Nested classes/interfaces inherited from class uk.ac.soton.simulation.jsit.core.DistributionCategorical

DistributionCategorical.Range

Field Summary

Fields inherited from class uk.ac.soton.simulation.jsit.core.DistributionDiscrete

CUMULATIVE_PROB_TOLERANCE

Constructor Summary

Constructors

Constructor and Description
DistCustomCategorical(java.lang.Class<C> categoryEnumType, double[] pmf) Create an instance where outcomes map to an Java enum category.
DistCustomCategorical(double[] pmf, boolean keepLocked) Create an instance where outcomes map to integers 1-K.

Method Summary

Modifier and Type	Method and Description
Distribution	createUnregisteredCopy() Create an unregistered copy of this distribution.
void	setProbability(C categoryValue, double prob) Set (change) a probability value linked to a categorical outcome.
void	setProbability(int outcomeNum, double prob) Set (change) a probability value linked to a numeric (integer) outcome.
java.lang.String	toString() Custom toString implementation.
void	unlock() Unlock the distribution (allowing it to be sampled) after changes to probabilities are completed.

Methods inherited from class uk.ac.soton.simulation.jsit.core.DistributionCategorical

addRange, clearRanges, getK, isLocked, lock, sampleCategory, usesEnumCategory, usesMappedRanges

Methods inherited from class uk.ac.soton.simulation.jsit.core.DistributionDiscrete

sampleDouble, sampleInt

Methods inherited from class uk.ac.soton.simulation.jsit.core.Distribution

createRegisteredCopy

Methods inherited from class uk.ac.soton.simulation.jsit.core.AbstractStochasticItem

getSampler, registerAccessInfo, registerSampler

Methods inherited from class java.lang.Object

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

Constructor Detail

DistCustomCategorical

public DistCustomCategorical(java.lang.Class<C> categoryEnumType,
                             double[] pmf)

Create an instance where outcomes map to an Java enum category.

Parameters:

categoryEnumType - The enum which defines your categories.

pmf - The probability mass function (PMF) as an array of probabilities.

Since:

0.1

DistCustomCategorical

public DistCustomCategorical(double[] pmf,
                             boolean keepLocked)

Create an instance where outcomes map to integers 1-K. You may wish to add ranges so that one of a particular set of integers is returned, in which case you should keep the distribution locked until all ranges have been added (and then unlock it).

Parameters:

pmf - The probability mass function (PMF) as an array of probabilities.

keepLocked - Whether to keep the distribution locked (see above).

Since:

0.2

Method Detail

unlock

public void unlock()

Unlock the distribution (allowing it to be sampled) after changes to probabilities are completed. The probabilities will need to sum to 1 at this point.

Overrides:

unlock in class DistributionCategorical<C extends java.lang.Enum<C>>

Since:

0.2

setProbability

public void setProbability(C categoryValue,
                           double prob)

Set (change) a probability value linked to a categorical outcome. The distribution will be locked automatically when you change the first probability; you should unlock it after the last change is done.

Parameters:

categoryValue - The enum value (category) being changed.

prob - The new probability.

Since:

0.1

setProbability

public void setProbability(int outcomeNum,
                           double prob)

Set (change) a probability value linked to a numeric (integer) outcome. The distribution will be locked automatically when you change the first probability; you should unlock it after the last change is done.

Parameters:

outcomeNum - The number ('position') of the outcome to change (1-K).

prob - The new probability.

Since:

0.2

toString

public java.lang.String toString()

Custom toString implementation.

Specified by:

toString in class DistributionCategorical<C extends java.lang.Enum<C>>

Returns:

A string representation of the distribution; e.g., "CustomPMF[0.1,0.2,0.7]".

Since:

0.1

createUnregisteredCopy

public Distribution createUnregisteredCopy()

Create an unregistered copy of this distribution.

Specified by:

createUnregisteredCopy in class Distribution

Returns:

The unregistered copy. (This will need casting as appropriate.)

Since:

0.2