Class UniformPlusMinusOp

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

Inheritance

Object

UniformPlusMinusOp

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), "UniformPlusMinus", new Type[]{typeof(double)})]
[Quality(QualityBand.Experimental)]
public static class UniformPlusMinusOp

Methods

LogEvidenceRatio(Double, Pareto)

Evidence message for EP.

Declaration

public static double LogEvidenceRatio(double sample, Pareto upperBound)

Parameters

Type	Name	Description
Double	sample	Constant value for sample.
Pareto	upperBound	Incoming message from upperBound.

Returns

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

Remarks

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

UpperBoundAverageConditional(Double)

EP message to upperBound.

Declaration

public static Pareto UpperBoundAverageConditional(double sample)

Parameters

Type	Name	Description
Double	sample	Constant value for sample.

Returns

Type	Description
Pareto	The outgoing EP message to the upperBound argument.

Remarks

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

UpperBoundAverageLogarithm(Double)

VMP message to upperBound.

Declaration

public static Pareto UpperBoundAverageLogarithm(double sample)

Parameters

Type	Name	Description
Double	sample	Constant value for sample.

Returns

Type	Description
Pareto	The outgoing VMP message to the upperBound argument.

Remarks

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