Class VariableGibbsOp

Provides outgoing messages for VariableGibbs<T>(T, out T), given random arguments to the function.

Inheritance

Object

VariableGibbsOp

Inherited Members

Object.Equals(Object)

Object.Equals(Object, Object)

Object.GetHashCode()

Object.GetType()

Object.MemberwiseClone()

Object.ReferenceEquals(Object, Object)

Object.ToString()

Namespace: Microsoft.ML.Probabilistic.Factors

Assembly: Microsoft.ML.Probabilistic.dll

Syntax

[FactorMethod(typeof(Clone), "VariableGibbs<>", new Type[]{})]
[Quality(QualityBand.Preview)]
public static class VariableGibbsOp

Methods

DefGibbs<TDist, T>(TDist, TDist)

Declaration

public static TDist DefGibbs<TDist, T>(TDist Use, TDist result)
    where TDist : IDistribution<T>, SettableTo<TDist>

Parameters

Type	Name	Description
TDist	Use
TDist	result

Returns

Type	Description
TDist

Type Parameters

Name	Description
TDist
T

DefGibbs<TDist, T>(GibbsMarginal<TDist, T>, T)

Gibbs sample message to 'Def'

Declaration

public static T DefGibbs<TDist, T>(GibbsMarginal<TDist, T> marginal, T result)
    where TDist : IDistribution<T>, Sampleable<T>

Parameters

Type	Name	Description
GibbsMarginal<TDist, T>	marginal	Incoming message from 'marginal'. Must be a proper distribution. If uniform, the result will be uniform.
T	result	Modified to contain the outgoing message

Returns

Type	Description
T	result

Type Parameters

Name	Description
TDist
T

Remarks

The outgoing message is the current Gibbs sample.

Exceptions

Type	Condition
ImproperMessageException	marginal is not a proper distribution

GibbsEvidence<TDist, T>(TDist, TDist, GibbsMarginal<TDist, T>)

Gibbs evidence

Declaration

public static double GibbsEvidence<TDist, T>(TDist Use, TDist Def, GibbsMarginal<TDist, T> marginal)
    where TDist : IDistribution<T>, Sampleable<T>, CanGetLogAverageOf<TDist>

Parameters

Type	Name	Description
TDist	Use
TDist	Def
GibbsMarginal<TDist, T>	marginal

Returns

Type	Description
Double

Type Parameters

Name	Description
TDist
T

MarginalGibbs<TDist, T>(TDist, TDist, GibbsMarginal<TDist, T>)

Gibbs message to 'Marginal'.

Declaration

public static GibbsMarginal<TDist, T> MarginalGibbs<TDist, T>(TDist Use, TDist Def, GibbsMarginal<TDist, T> to_marginal)
    where TDist : IDistribution<T>, SettableToProduct<TDist>, SettableTo<TDist>, Sampleable<T>

Parameters

Type	Name	Description
TDist	Use	Incoming message from 'use'.
TDist	Def	Incoming message from 'def'. Must be a proper distribution. If uniform, the result will be uniform.
GibbsMarginal<TDist, T>	to_marginal

Returns

Type	Description
GibbsMarginal<TDist, T>

Type Parameters

Name	Description
TDist
T

Remarks

The outgoing message is the product of 'Def' and 'Uses' messages.

Exceptions

Type	Condition
ImproperMessageException	Def is not a proper distribution

MarginalGibbs<TDist, T>(T, TDist, GibbsMarginal<TDist, T>)

Declaration

public static GibbsMarginal<TDist, T> MarginalGibbs<TDist, T>(T Use, TDist Def, GibbsMarginal<TDist, T> to_marginal)
    where TDist : IDistribution<T>, Sampleable<T>

Parameters

Type	Name	Description
T	Use
TDist	Def
GibbsMarginal<TDist, T>	to_marginal

Returns

Type	Description
GibbsMarginal<TDist, T>

Type Parameters

Name	Description
TDist
T

UseGibbs<TDist, T>(TDist, TDist)

Gibbs distribution message to 'Def'

Declaration

public static TDist UseGibbs<TDist, T>(TDist Def, TDist result)
    where TDist : SettableTo<TDist>

Parameters

Type	Name	Description
TDist	Def	Incoming message from 'def'. Must be a proper distribution. If uniform, the result will be uniform.
TDist	result	Modified to contain the outgoing message

Returns

Type	Description
TDist	result

Type Parameters

Name	Description
TDist
T

Remarks

The outgoing message is the product of the 'Def' message with all 'Uses' messages except the current

Exceptions

Type	Condition
ImproperMessageException	Def is not a proper distribution

UseGibbs<TDist, T>(GibbsMarginal<TDist, T>, T)

Gibbs sample message to 'Use'

Declaration

public static T UseGibbs<TDist, T>(GibbsMarginal<TDist, T> marginal, T result)
    where TDist : IDistribution<T>, Sampleable<T>

Parameters

Type	Name	Description
GibbsMarginal<TDist, T>	marginal	Incoming message from 'marginal'. Must be a proper distribution. If uniform, the result will be uniform.
T	result	Modified to contain the outgoing message

Returns

Type	Description
T	result

Type Parameters

Name	Description
TDist
T

Remarks

The outgoing message is the current Gibbs sample.

Exceptions

Type	Condition
ImproperMessageException	marginal is not a proper distribution