Class ReplicateGibbsOp<T>

Provides outgoing messages for the following factors:

• Replicate<T>(T, Int32)
• ReplicateWithMarginal<T>(T, Int32, out T)

, given random arguments to the function.

Inheritance

Object

ReplicateGibbsOp<T>

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), "Replicate<>", new Type[]{}, Default = true)]
[FactorMethod(typeof(Clone), "ReplicateWithMarginal<>", new Type[]{}, Default = true)]
[Quality(QualityBand.Mature)]
public static class ReplicateGibbsOp<T>

Type Parameters

Name	Description
T	The type of the variable being replicated.

Methods

DefGibbs<TDist>(TDist[], TDist)

Declaration

public static TDist DefGibbs<TDist>([SkipIfAllUniform] TDist[] Uses, TDist result)
    where TDist : IDistribution<T>, Sampleable<T>, SettableToProduct<TDist>, SettableTo<TDist>

Parameters

Type	Name	Description
TDist[]	Uses	Incoming message from Uses. Must be a proper distribution. If all elements are uniform, the result will be uniform.
TDist	result	Modified to contain the outgoing message.

Returns

Type	Description
TDist	result

Type Parameters

Name	Description
TDist	The type of the distribution over the replicated variable.

Remarks

Exceptions

Type	Condition
ImproperMessageException	Uses is not a proper distribution.

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

Declaration

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

Parameters

Type	Name	Description
GibbsMarginal<TDist, T>	to_marginal
T	result	Modified to contain the outgoing message.

Returns

Type	Description
T	result

Type Parameters

Name	Description
TDist	The type of the distribution over the replicated variable.

Remarks

DefGibbs<TDist>(IReadOnlyList<TDist>, TDist)

Declaration

public static TDist DefGibbs<TDist>([SkipIfAllUniform] IReadOnlyList<TDist> Uses, TDist result)
    where TDist : IDistribution<T>, Sampleable<T>, SettableTo<TDist>, SettableToProduct<TDist>, SettableToRatio<TDist>

Parameters

Type	Name	Description
IReadOnlyList<TDist>	Uses
TDist	result

Returns

Type	Description
TDist

Type Parameters

Name	Description
TDist	The type of the distribution over the replicated variable.

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

Declaration

public static double GibbsEvidence<TDist>(IList<TDist> Uses, T Def)
    where TDist : IDistribution<T>

Parameters

Type	Name	Description
IList<TDist>	Uses	Incoming message from Uses.
T	Def	Incoming message from Def.

Returns

Type	Description
Double

Type Parameters

Name	Description
TDist	The type of the distribution over the replicated variable.

Remarks

LogEvidenceRatio<TDist>(IList<TDist>, T)

Evidence message for EP.

Declaration

public static double LogEvidenceRatio<TDist>([SkipIfAllUniform] IList<TDist> Uses, T Def)
    where TDist : IDistribution<T>

Parameters

Type	Name	Description
IList<TDist>	Uses	Incoming message from Uses. Must be a proper distribution. If all elements are uniform, the result will be uniform.
T	Def	Incoming message from Def.

Returns

Type	Description
Double	Logarithm of the factor's contribution the EP model evidence.

Type Parameters

Name	Description
TDist	The type of the distribution over the replicated variable.

Remarks

The formula for the result is log(sum_(Uses,Def) p(Uses,Def) factor(Uses,Def,Count) / sum_Uses p(Uses) messageTo(Uses)). Adding up these values across all factors and variables gives the log-evidence estimate for EP.

Exceptions

Type	Condition
ImproperMessageException	Uses is not a proper distribution.

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

Declaration

[SkipIfAllUniform]
public static GibbsMarginal<TDist, T> MarginalGibbs<TDist>(IReadOnlyList<TDist> Uses, TDist Def, GibbsMarginal<TDist, T> to_marginal)
    where TDist : IDistribution<T>, SettableToProduct<TDist>, SettableToRatio<TDist>, SettableTo<TDist>, Sampleable<T>

Parameters

Type	Name	Description
IReadOnlyList<TDist>	Uses
TDist	Def
GibbsMarginal<TDist, T>	to_marginal

Returns

Type	Description
GibbsMarginal<TDist, T>

Type Parameters

Name	Description
TDist	The type of the distribution over the replicated variable.

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

Declaration

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

Parameters

Type	Name	Description
TDomain	Def
GibbsMarginal<TDist, TDomain>	to_marginal

Returns

Type	Description
GibbsMarginal<TDist, TDomain>

Type Parameters

Name	Description
TDist	The type of the distribution over the replicated variable.
TDomain	The domain type of TDist.

MarginalGibbs2<TDist>(T[], GibbsMarginal<TDist, T>)

Declaration

public static GibbsMarginal<TDist, T> MarginalGibbs2<TDist>(T[] Uses, GibbsMarginal<TDist, T> to_marginal)
    where TDist : IDistribution<T>, Sampleable<T>

Parameters

Type	Name	Description
T[]	Uses
GibbsMarginal<TDist, T>	to_marginal

Returns

Type	Description
GibbsMarginal<TDist, T>

Type Parameters

Name	Description
TDist	The type of the distribution over the replicated variable.

MarginalGibbsInit<TDist>(TDist)

Declaration

public static GibbsMarginal<TDist, T> MarginalGibbsInit<TDist>(TDist def)
    where TDist : IDistribution<T>, Sampleable<T>

Parameters

Type	Name	Description
TDist	def	Incoming message from Def.

Returns

Type	Description
GibbsMarginal<TDist, T>

Type Parameters

Name	Description
TDist	The type of the distribution over the replicated variable.

Remarks

UsesGibbs(T, Int32, T)

Declaration

public static T UsesGibbs(T def, int resultIndex, T result)

Parameters

Type	Name	Description
T	def	Incoming message from Def.
Int32	resultIndex	Index of the Uses for which a message is desired.
T	result	Modified to contain the outgoing message.

Returns

Type	Description
T	result

Remarks

UsesGibbs<TDist>(TDist, Int32, TDist)

Declaration

public static TDist UsesGibbs<TDist>(TDist def, int resultIndex, TDist result)
    where TDist : IDistribution<T>

Parameters

Type	Name	Description
TDist	def	Incoming message from Def.
Int32	resultIndex	Index of the Uses for which a message is desired.
TDist	result	Modified to contain the outgoing message.

Returns

Type	Description
TDist	result

Type Parameters

Name	Description
TDist	The type of the distribution over the replicated variable.

Remarks

UsesGibbs<TDist>(GibbsMarginal<TDist, T>, TDist, Int32, T)

Declaration

public static T UsesGibbs<TDist>(GibbsMarginal<TDist, T> to_marginal, TDist def, int resultIndex, T result)
    where TDist : IDistribution<T>, Sampleable<T>

Parameters

Type	Name	Description
GibbsMarginal<TDist, T>	to_marginal
TDist	def	Incoming message from Def.
Int32	resultIndex	Index of the Uses for which a message is desired.
T	result	Modified to contain the outgoing message.

Returns

Type	Description
T	result

Type Parameters

Name	Description
TDist	The type of the distribution over the replicated variable.

Remarks