Class DoubleOp

Provides outgoing messages for Double(Int32), given random arguments to the function.

Inheritance

Object

DoubleOp

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(Factor), "Double", new Type[]{typeof(int)})]
[Quality(QualityBand.Preview)]
public static class DoubleOp

Fields

ForceProper

Declaration

public static bool ForceProper

Field Value

Type	Description
Boolean

Methods

DoubleAverageConditional(Gaussian, Discrete)

EP message to double.

Declaration

public static Gaussian DoubleAverageConditional(Gaussian Double, Discrete Integer)

Parameters

Type	Name	Description
Gaussian	Double	Incoming message from double.
Discrete	Integer	Incoming message from integer.

Returns

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

Remarks

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

IntegerAverageConditional(Gaussian, Discrete)

EP message to integer.

Declaration

public static Discrete IntegerAverageConditional(Gaussian Double, Discrete result)

Parameters

Type	Name	Description
Gaussian	Double	Incoming message from double.
Discrete	result	Modified to contain the outgoing message.

Returns

Type	Description
Discrete	result

Remarks

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

LogAverageFactor(Gaussian, Discrete)

Evidence message for EP.

Declaration

public static double LogAverageFactor(Gaussian Double, Discrete Integer)

Parameters

Type	Name	Description
Gaussian	Double	Incoming message from double.
Discrete	Integer	Incoming message from integer.

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_(double,integer) p(double,integer) factor(double,integer)).

LogEvidenceRatio(Gaussian, Discrete, Gaussian)

Evidence message for EP.

Declaration

public static double LogEvidenceRatio(Gaussian Double, Discrete Integer, Gaussian to_double)

Parameters

Type	Name	Description
Gaussian	Double	Incoming message from double.
Discrete	Integer	Incoming message from integer.
Gaussian	to_double	Previous outgoing message to double.

Returns

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

Remarks

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

LogEvidenceRatio(Gaussian, Int32)

Evidence message for EP.

Declaration

public static double LogEvidenceRatio(Gaussian Double, int Integer)

Parameters

Type	Name	Description
Gaussian	Double	Incoming message from double.
Int32	Integer	Constant value for integer.

Returns

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

Remarks

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

LogEvidenceRatio(Double, Discrete)

Evidence message for EP.

Declaration

public static double LogEvidenceRatio(double Double, Discrete Integer)

Parameters

Type	Name	Description
Double	Double	Constant value for double.
Discrete	Integer	Incoming message from integer.

Returns

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

Remarks

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

LogEvidenceRatio(Double, Int32)

Evidence message for EP.

Declaration

public static double LogEvidenceRatio(double Double, int Integer)

Parameters

Type	Name	Description
Double	Double	Constant value for double.
Int32	Integer	Constant value for integer.

Returns

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

Remarks

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