Class LogOp_EP

Provides outgoing messages for Log(Double), given random arguments to the function.

Inheritance

Object

LogOp_EP

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(Math), "Log", new Type[]{typeof(double)})]
[Quality(QualityBand.Preview)]
public class LogOp_EP

Methods

AverageLogFactor(Double, Double)

Evidence message for VMP.

Declaration

public static double AverageLogFactor(double log, double d)

Parameters

Type	Name	Description
Double	log	Constant value for log.
Double	d	Constant value for d.

Returns

Type	Description
Double	Zero.

Remarks

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

DAverageConditional(Gaussian, Gamma, Gaussian)

EP message to d.

Declaration

public static Gamma DAverageConditional(Gaussian log, Gamma d, Gaussian to_log)

Parameters

Type	Name	Description
Gaussian	log	Incoming message from log. Must be a proper distribution. If uniform, the result will be uniform.
Gamma	d	Incoming message from d.
Gaussian	to_log	Previous outgoing message to log.

Returns

Type	Description
Gamma	The outgoing EP message to the d argument.

Remarks

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

Exceptions

Type	Condition
ImproperMessageException	log is not a proper distribution.

DAverageConditional(Double)

EP message to d.

Declaration

public static Gamma DAverageConditional(double log)

Parameters

Type	Name	Description
Double	log	Constant value for log.

Returns

Type	Description
Gamma	The outgoing EP message to the d argument.

Remarks

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

LogAverageConditional(Gaussian, Gamma, Gaussian)

EP message to log.

Declaration

public static Gaussian LogAverageConditional(Gaussian log, Gamma d, Gaussian result)

Parameters

Type	Name	Description
Gaussian	log	Incoming message from log. Must be a proper distribution. If uniform, the result will be uniform.
Gamma	d	Incoming message from d. Must be a proper distribution. If uniform, the result will be uniform.
Gaussian	result	Modified to contain the outgoing message.

Returns

Type	Description
Gaussian	result

Remarks

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

Exceptions

Type	Condition
ImproperMessageException	log is not a proper distribution.
ImproperMessageException	d is not a proper distribution.

LogAverageConditional(Double)

EP message to log.

Declaration

public static Gaussian LogAverageConditional(double d)

Parameters

Type	Name	Description
Double	d	Constant value for d.

Returns

Type	Description
Gaussian	The outgoing EP message to the log argument.

Remarks

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

LogAverageConditionalInit()

Declaration

public static Gaussian LogAverageConditionalInit()

Returns

Type	Description
Gaussian

Remarks

LogAverageFactor(Gaussian, Gamma, Gaussian)

Evidence message for EP.

Declaration

public static double LogAverageFactor(Gaussian log, Gamma d, Gaussian to_log)

Parameters

Type	Name	Description
Gaussian	log	Incoming message from log.
Gamma	d	Incoming message from d.
Gaussian	to_log	Outgoing message to log.

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_(log,d) p(log,d) factor(log,d)).

LogAverageFactor(Gaussian, Double)

Evidence message for EP.

Declaration

public static double LogAverageFactor(Gaussian log, double d)

Parameters

Type	Name	Description
Gaussian	log	Incoming message from log.
Double	d	Constant value for d.

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_(log) p(log) factor(log,d)).

LogAverageFactor(Double, Gamma)

Evidence message for EP.

Declaration

public static double LogAverageFactor(double log, Gamma d)

Parameters

Type	Name	Description
Double	log	Constant value for log.
Gamma	d	Incoming message from d.

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_(d) p(d) factor(log,d)).

LogAverageFactor(Double, Double)

Evidence message for EP.

Declaration

public static double LogAverageFactor(double log, double d)

Parameters

Type	Name	Description
Double	log	Constant value for log.
Double	d	Constant value for d.

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(log,d)).

LogEvidenceRatio(Gaussian, Gamma, Gaussian)

Evidence message for EP.

Declaration

public static double LogEvidenceRatio(Gaussian log, Gamma d, Gaussian to_log)

Parameters

Type	Name	Description
Gaussian	log	Incoming message from log.
Gamma	d	Incoming message from d.
Gaussian	to_log	Outgoing message to log.

Returns

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

Remarks

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

LogEvidenceRatio(Double, Gamma)

Evidence message for EP.

Declaration

public static double LogEvidenceRatio(double log, Gamma d)

Parameters

Type	Name	Description
Double	log	Constant value for log.
Gamma	d	Incoming message from d.

Returns

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

Remarks

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

LogEvidenceRatio(Double, Double)

Evidence message for EP.

Declaration

public static double LogEvidenceRatio(double log, double d)

Parameters

Type	Name	Description
Double	log	Constant value for log.
Double	d	Constant value for d.

Returns

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

Remarks

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