Class DirichletFromPseudoCountsOp

Provides outgoing messages for SampleFromPseudoCounts(Vector), given random arguments to the function.

Inheritance

Object

DirichletFromPseudoCountsOp

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(new string[]{"sample", "pseudoCounts"}, typeof(Dirichlet), "SampleFromPseudoCounts", new Type[]{})]
[Quality(QualityBand.Stable)]
public static class DirichletFromPseudoCountsOp

Methods

AverageLogFactor(Dirichlet, Vector, Dirichlet)

Evidence message for VMP.

Declaration

public static double AverageLogFactor(Dirichlet sample, Vector pseudoCounts, Dirichlet to_sample)

Parameters

Type	Name	Description
Dirichlet	sample	Incoming message from sampleFromPseudoCounts.
Vector	pseudoCounts	Constant value for pseudoCount.
Dirichlet	to_sample	Outgoing message to sample.

Returns

Type	Description
Double	Average of the factor's log-value across the given argument distributions.

Remarks

The formula for the result is sum_(sampleFromPseudoCounts) p(sampleFromPseudoCounts) log(factor(sampleFromPseudoCounts,pseudoCount)). Adding up these values across all factors and variables gives the log-evidence estimate for VMP.

AverageLogFactor(Vector, Vector)

Evidence message for VMP.

Declaration

public static double AverageLogFactor(Vector sample, Vector pseudoCounts)

Parameters

Type	Name	Description
Vector	sample	Constant value for sampleFromPseudoCounts.
Vector	pseudoCounts	Constant value for pseudoCount.

Returns

Type	Description
Double	Average of the factor's log-value across the given argument distributions.

Remarks

The formula for the result is log(factor(sampleFromPseudoCounts,pseudoCount)). Adding up these values across all factors and variables gives the log-evidence estimate for VMP.

LogAverageFactor(Dirichlet, Vector, Dirichlet)

Evidence message for EP.

Declaration

public static double LogAverageFactor(Dirichlet sample, Vector pseudoCounts, Dirichlet to_sample)

Parameters

Type	Name	Description
Dirichlet	sample	Incoming message from sampleFromPseudoCounts.
Vector	pseudoCounts	Constant value for pseudoCount.
Dirichlet	to_sample	Outgoing message to sample.

Returns

Type	Description
Double	Logarithm of the factor's average value across the given argument distributions.

Remarks

The formula for the result is log(sum_(sampleFromPseudoCounts) p(sampleFromPseudoCounts) factor(sampleFromPseudoCounts,pseudoCount)).

LogAverageFactor(Vector, Vector)

Evidence message for EP.

Declaration

public static double LogAverageFactor(Vector sample, Vector pseudoCounts)

Parameters

Type	Name	Description
Vector	sample	Constant value for sampleFromPseudoCounts.
Vector	pseudoCounts	Constant value for pseudoCount.

Returns

Type	Description
Double	Logarithm of the factor's average value across the given argument distributions.

Remarks

The formula for the result is log(factor(sampleFromPseudoCounts,pseudoCount)).

LogEvidenceRatio(Dirichlet, Vector)

Evidence message for EP.

Declaration

public static double LogEvidenceRatio(Dirichlet sample, Vector pseudoCounts)

Parameters

Type	Name	Description
Dirichlet	sample	Incoming message from sampleFromPseudoCounts.
Vector	pseudoCounts	Constant value for pseudoCount.

Returns

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

Remarks

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

LogEvidenceRatio(Vector, Vector)

Evidence message for EP.

Declaration

public static double LogEvidenceRatio(Vector sample, Vector pseudoCounts)

Parameters

Type	Name	Description
Vector	sample	Constant value for sampleFromPseudoCounts.
Vector	pseudoCounts	Constant value for pseudoCount.

Returns

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

Remarks

The formula for the result is log(factor(sampleFromPseudoCounts,pseudoCount)). Adding up these values across all factors and variables gives the log-evidence estimate for EP.

SampleAverageConditional(Vector)

EP message to sampleFromPseudoCounts.

Declaration

public static Dirichlet SampleAverageConditional(Vector pseudoCounts)

Parameters

Type	Name	Description
Vector	pseudoCounts	Constant value for pseudoCount.

Returns

Type	Description
Dirichlet	The outgoing EP message to the sampleFromPseudoCounts argument.

Remarks

The outgoing message is the factor viewed as a function of sampleFromPseudoCounts conditioned on the given values.

SampleAverageLogarithm(Vector)

VMP message to sampleFromPseudoCounts.

Declaration

public static Dirichlet SampleAverageLogarithm(Vector pseudoCounts)

Parameters

Type	Name	Description
Vector	pseudoCounts	Constant value for pseudoCount.

Returns

Type	Description
Dirichlet	The outgoing VMP message to the sampleFromPseudoCounts argument.

Remarks

The outgoing message is the factor viewed as a function of sampleFromPseudoCounts conditioned on the given values.