Class DiscreteFromDirichletOp

Provides outgoing messages for the following factors:

, given random arguments to the function.

Inheritance

Object

DiscreteFromDirichletOp

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(Discrete), "Sample", new Type[]{typeof(Vector)})]
[FactorMethod(new string[]{"sample", "probs"}, typeof(Factor), "Discrete", new Type[]{typeof(Vector)})]
[FactorMethod(typeof(EnumSupport), "DiscreteEnum<>", new Type[]{})]
[Quality(QualityBand.Mature)]
public static class DiscreteFromDirichletOp

Methods

AverageLogFactor(Discrete, Dirichlet)

Evidence message for VMP.

Declaration

public static double AverageLogFactor(Discrete sample, Dirichlet probs)

Parameters

Type	Name	Description
Discrete	sample	Incoming message from `sample`.
Dirichlet	probs	Incoming message from `probs`. Must be a proper distribution. If any element is uniform, the result will be uniform.

Returns

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

Remarks

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

Exceptions

Type	Condition
ImproperMessageException	`probs` is not a proper distribution.

AverageLogFactor(Discrete, Vector)

Evidence message for VMP.

Declaration

public static double AverageLogFactor(Discrete sample, Vector probs)

Parameters

Type	Name	Description
Discrete	sample	Incoming message from `sample`.
Vector	probs	Constant value for `probs`.

Returns

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

Remarks

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

AverageLogFactor(Int32, Dirichlet)

Evidence message for VMP.

Declaration

public static double AverageLogFactor(int sample, Dirichlet probs)

Parameters

Type	Name	Description
Int32	sample	Constant value for `sample`.
Dirichlet	probs	Incoming message from `probs`. Must be a proper distribution. If any element is uniform, the result will be uniform.

Returns

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

Remarks

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

Exceptions

Type	Condition
ImproperMessageException	`probs` is not a proper distribution.

AverageLogFactor(Int32, Vector)

Evidence message for VMP.

Declaration

public static double AverageLogFactor(int sample, Vector probs)

Parameters

Type	Name	Description
Int32	sample	Constant value for `sample`.
Vector	probs	Constant value for `probs`.

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(sample,probs)). Adding up these values across all factors and variables gives the log-evidence estimate for VMP.

LogAverageFactor(Discrete, Discrete)

Evidence message for EP.

Declaration

public static double LogAverageFactor(Discrete sample, Discrete to_sample)

Parameters

Type	Name	Description
Discrete	sample	Incoming message from `sample`.
Discrete	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_(sample) p(sample) factor(sample,probs)).

LogAverageFactor(Int32, Dirichlet)

Evidence message for EP.

Declaration

public static double LogAverageFactor(int sample, Dirichlet probs)

Parameters

Type	Name	Description
Int32	sample	Constant value for `sample`.
Dirichlet	probs	Incoming message from `probs`.

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_(probs) p(probs) factor(sample,probs)).

LogAverageFactor(Int32, Vector)

Evidence message for EP.

Declaration

public static double LogAverageFactor(int sample, Vector probs)

Parameters

Type	Name	Description
Int32	sample	Constant value for `sample`.
Vector	probs	Constant value for `probs`.

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(sample,probs)).

LogEvidenceRatio(Discrete, Dirichlet)

Evidence message for EP.

Declaration

public static double LogEvidenceRatio(Discrete sample, Dirichlet probs)

Parameters

Type	Name	Description
Discrete	sample	Incoming message from `sample`.
Dirichlet	probs	Incoming message from `probs`.

Returns

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

Remarks

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

LogEvidenceRatio(Discrete, Vector)

Evidence message for EP.

Declaration

public static double LogEvidenceRatio(Discrete sample, Vector probs)

Parameters

Type	Name	Description
Discrete	sample	Incoming message from `sample`.
Vector	probs	Constant value for `probs`.

Returns

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

Remarks

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

LogEvidenceRatio(Int32, Dirichlet)

Evidence message for EP.

Declaration

public static double LogEvidenceRatio(int sample, Dirichlet probs)

Parameters

Type	Name	Description
Int32	sample	Constant value for `sample`.
Dirichlet	probs	Incoming message from `probs`.

Returns

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

Remarks

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

LogEvidenceRatio(Int32, Vector)

Evidence message for EP.

Declaration

public static double LogEvidenceRatio(int sample, Vector probs)

Parameters

Type	Name	Description
Int32	sample	Constant value for `sample`.
Vector	probs	Constant value for `probs`.

Returns

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

Remarks

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

ProbsAverageConditional(Discrete, Dirichlet, Dirichlet)

EP message to probs.

Declaration

public static Dirichlet ProbsAverageConditional(Discrete sample, Dirichlet probs, Dirichlet result)

Parameters

Type	Name	Description
Discrete	sample	Incoming message from `sample`. Must be a proper distribution. If uniform, the result will be uniform.
Dirichlet	probs	Incoming message from `probs`.
Dirichlet	result	Modified to contain the outgoing message.

Returns

Type	Description
Dirichlet	`result`

Remarks

The outgoing message is a distribution matching the moments of probs as the random arguments are varied. The formula is proj[p(probs) sum_(sample) p(sample) factor(sample,probs)]/p(probs).

Exceptions

Type	Condition
ImproperMessageException	`sample` is not a proper distribution.

ProbsAverageConditional(Discrete, Vector, Dirichlet)

EP message to probs.

Declaration

public static Dirichlet ProbsAverageConditional(Discrete sample, Vector probs, Dirichlet result)

Parameters

Type	Name	Description
Discrete	sample	Incoming message from `sample`. Must be a proper distribution. If uniform, the result will be uniform.
Vector	probs	Constant value for `probs`.
Dirichlet	result	Modified to contain the outgoing message.

Returns

Type	Description
Dirichlet	`result`

Remarks

The outgoing message is a distribution matching the moments of probs as the random arguments are varied. The formula is proj[p(probs) sum_(sample) p(sample) factor(sample,probs)]/p(probs).

Exceptions

Type	Condition
ImproperMessageException	`sample` is not a proper distribution.

ProbsAverageConditional(Int32, Dirichlet)

EP message to probs.

Declaration

public static Dirichlet ProbsAverageConditional(int sample, Dirichlet result)

Parameters

Type	Name	Description
Int32	sample	Constant value for `sample`.
Dirichlet	result	Modified to contain the outgoing message.

Returns

Type	Description
Dirichlet	`result`

Remarks

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

ProbsAverageLogarithm(Discrete, Dirichlet)

VMP message to probs.

Declaration

public static Dirichlet ProbsAverageLogarithm(Discrete sample, Dirichlet result)

Parameters

Type	Name	Description
Discrete	sample	Incoming message from `sample`.
Dirichlet	result	Modified to contain the outgoing message.

Returns

Type	Description
Dirichlet	`result`

Remarks

The outgoing message is the exponential of the average log-factor value, where the average is over all arguments except probs. The formula is exp(sum_(sample) p(sample) log(factor(sample,probs))).

ProbsAverageLogarithm(Int32, Dirichlet)

VMP message to probs.

Declaration

public static Dirichlet ProbsAverageLogarithm(int sample, Dirichlet result)

Parameters

Type	Name	Description
Int32	sample	Constant value for `sample`.
Dirichlet	result	Modified to contain the outgoing message.

Returns

Type	Description
Dirichlet	`result`

Remarks

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

ProbsConditional(Int32, Dirichlet)

Gibbs message to probs.

Declaration

public static Dirichlet ProbsConditional(int sample, Dirichlet result)

Parameters

Type	Name	Description
Int32	sample	Constant value for `sample`.
Dirichlet	result	Modified to contain the outgoing message.

Returns

Type	Description
Dirichlet	`result`

Remarks

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

SampleAverageConditional(Dirichlet, Discrete)

EP message to sample.

Declaration

public static Discrete SampleAverageConditional(Dirichlet probs, Discrete result)

Parameters

Type	Name	Description
Dirichlet	probs	Incoming message from `probs`. Must be a proper distribution. If any element is uniform, the result will be uniform.
Discrete	result	Modified to contain the outgoing message.

Returns

Type	Description
Discrete	`result`

Remarks

The outgoing message is a distribution matching the moments of sample as the random arguments are varied. The formula is proj[p(sample) sum_(probs) p(probs) factor(sample,probs)]/p(sample).

Exceptions

Type	Condition
ImproperMessageException	`probs` is not a proper distribution.

SampleAverageConditional(Vector, Discrete)

EP message to sample.

Declaration

public static Discrete SampleAverageConditional(Vector probs, Discrete result)

Parameters

Type	Name	Description
Vector	probs	Constant value for `probs`.
Discrete	result	Modified to contain the outgoing message.

Returns

Type	Description
Discrete	`result`

Remarks

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

SampleAverageConditionalInit(Dirichlet)

Declaration

public static Discrete SampleAverageConditionalInit(Dirichlet probs)

Parameters

Type	Name	Description
Dirichlet	probs	Incoming message from `probs`.

Returns

Type	Description
Discrete

Remarks

SampleAverageConditionalInit(Vector)

Declaration

public static Discrete SampleAverageConditionalInit(Vector probs)

Parameters

Type	Name	Description
Vector	probs	Constant value for `probs`.

Returns

Type	Description
Discrete

Remarks

SampleAverageLogarithm(Dirichlet, Discrete)

VMP message to sample.

Declaration

public static Discrete SampleAverageLogarithm(Dirichlet probs, Discrete result)

Parameters

Type	Name	Description
Dirichlet	probs	Incoming message from `probs`. Must be a proper distribution. If any element is uniform, the result will be uniform.
Discrete	result	Modified to contain the outgoing message.

Returns

Type	Description
Discrete	`result`

Remarks

The outgoing message is the exponential of the average log-factor value, where the average is over all arguments except sample. The formula is exp(sum_(probs) p(probs) log(factor(sample,probs))).

Exceptions

Type	Condition
ImproperMessageException	`probs` is not a proper distribution.

SampleAverageLogarithm(Vector, Discrete)

VMP message to sample.

Declaration

public static Discrete SampleAverageLogarithm(Vector probs, Discrete result)

Parameters

Type	Name	Description
Vector	probs	Constant value for `probs`.
Discrete	result	Modified to contain the outgoing message.

Returns

Type	Description
Discrete	`result`

Remarks

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

SampleAverageLogarithmInit(Dirichlet)

Declaration

public static Discrete SampleAverageLogarithmInit(Dirichlet probs)

Parameters

Type	Name	Description
Dirichlet	probs	Incoming message from `probs`.

Returns

Type	Description
Discrete

Remarks

SampleAverageLogarithmInit(Vector)

Declaration

public static Discrete SampleAverageLogarithmInit(Vector probs)

Parameters

Type	Name	Description
Vector	probs	Constant value for `probs`.

Returns

Type	Description
Discrete

Remarks

SampleConditional(Vector, Discrete)

Gibbs message to sample.

Declaration

public static Discrete SampleConditional(Vector probs, Discrete result)

Parameters

Type	Name	Description
Vector	probs	Constant value for `probs`.
Discrete	result	Modified to contain the outgoing message.

Returns

Type	Description
Discrete	`result`

Remarks

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