Class PlusWrappedGaussianOp

Provides outgoing messages for the following factors:

, given random arguments to the function.

Inheritance

Object

PlusWrappedGaussianOp

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), "Plus", new Type[]{typeof(double), typeof(double)}, Default = true)]
[FactorMethod(new string[]{"A", "Sum", "B"}, typeof(Factor), "Difference", new Type[]{typeof(double), typeof(double)}, Default = true)]
[Quality(QualityBand.Mature)]
public static class PlusWrappedGaussianOp

Methods

AAverageConditional(WrappedGaussian, WrappedGaussian)

EP message to difference.

Declaration

public static WrappedGaussian AAverageConditional(WrappedGaussian sum, WrappedGaussian b)

Parameters

Type	Name	Description
WrappedGaussian	sum	Incoming message from `a`. Must be a proper distribution. If uniform, the result will be uniform.
WrappedGaussian	b	Incoming message from `b`. Must be a proper distribution. If uniform, the result will be uniform.

Returns

Type	Description
WrappedGaussian	The outgoing EP message to the `difference` argument.

Remarks

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

Exceptions

Type	Condition
ImproperMessageException	`sum` is not a proper distribution.
ImproperMessageException	`b` is not a proper distribution.

AAverageConditional(WrappedGaussian, Double)

EP message to difference.

Declaration

public static WrappedGaussian AAverageConditional(WrappedGaussian sum, double b)

Parameters

Type	Name	Description
WrappedGaussian	sum	Incoming message from `a`. Must be a proper distribution. If uniform, the result will be uniform.
Double	b	Constant value for `b`.

Returns

Type	Description
WrappedGaussian	The outgoing EP message to the `difference` argument.

Remarks

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

Exceptions

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

AAverageLogarithm(WrappedGaussian, WrappedGaussian)

VMP message to difference.

Declaration

public static WrappedGaussian AAverageLogarithm(WrappedGaussian sum, WrappedGaussian b)

Parameters

Type	Name	Description
WrappedGaussian	sum	Incoming message from `a`. Must be a proper distribution. If uniform, the result will be uniform.
WrappedGaussian	b	Incoming message from `b`. Must be a proper distribution. If uniform, the result will be uniform.

Returns

Type	Description
WrappedGaussian	The outgoing VMP message to the `difference` argument.

Remarks

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

Exceptions

Type	Condition
ImproperMessageException	`sum` is not a proper distribution.
ImproperMessageException	`b` is not a proper distribution.

AAverageLogarithm(WrappedGaussian, Double)

VMP message to difference.

Declaration

public static WrappedGaussian AAverageLogarithm(WrappedGaussian sum, double b)

Parameters

Type	Name	Description
WrappedGaussian	sum	Incoming message from `a`. Must be a proper distribution. If uniform, the result will be uniform.
Double	b	Constant value for `b`.

Returns

Type	Description
WrappedGaussian	The outgoing VMP message to the `difference` argument.

Remarks

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

Exceptions

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

AverageLogFactor(WrappedGaussian)

Evidence message for VMP.

Declaration

public static double AverageLogFactor(WrappedGaussian sum)

Parameters

Type	Name	Description
WrappedGaussian	sum	Incoming message from `a`.

Returns

Type	Description
Double	Zero.

Remarks

In Variational Message Passing, the evidence contribution of a deterministic factor is zero. Adding up these values across all factors and variables gives the log-evidence estimate for VMP.

BAverageConditional(WrappedGaussian, WrappedGaussian)

EP message to b.

Declaration

public static WrappedGaussian BAverageConditional(WrappedGaussian sum, WrappedGaussian a)

Parameters

Type	Name	Description
WrappedGaussian	sum	Incoming message from `a`. Must be a proper distribution. If uniform, the result will be uniform.
WrappedGaussian	a	Incoming message from `difference`. Must be a proper distribution. If uniform, the result will be uniform.

Returns

Type	Description
WrappedGaussian	The outgoing EP message to the `b` argument.

Remarks

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

Exceptions

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

BAverageConditional(WrappedGaussian, Double)

EP message to b.

Declaration

public static WrappedGaussian BAverageConditional(WrappedGaussian sum, double a)

Parameters

Type	Name	Description
WrappedGaussian	sum	Incoming message from `a`. Must be a proper distribution. If uniform, the result will be uniform.
Double	a	Constant value for `difference`.

Returns

Type	Description
WrappedGaussian	The outgoing EP message to the `b` argument.

Remarks

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

Exceptions

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

BAverageLogarithm(WrappedGaussian, WrappedGaussian)

VMP message to b.

Declaration

public static WrappedGaussian BAverageLogarithm(WrappedGaussian sum, WrappedGaussian a)

Parameters

Type	Name	Description
WrappedGaussian	sum	Incoming message from `a`. Must be a proper distribution. If uniform, the result will be uniform.
WrappedGaussian	a	Incoming message from `difference`. Must be a proper distribution. If uniform, the result will be uniform.

Returns

Type	Description
WrappedGaussian	The outgoing VMP message to the `b` argument.

Remarks

The outgoing message is the exponential of the average log-factor value, where the average is over all arguments except b. Because the factor is deterministic, difference is integrated out before taking the logarithm. The formula is exp(sum_(a) p(a) log(sum_difference p(difference) factor(difference,a,b))).

Exceptions

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

BAverageLogarithm(WrappedGaussian, Double)

VMP message to b.

Declaration

public static WrappedGaussian BAverageLogarithm(WrappedGaussian sum, double a)

Parameters

Type	Name	Description
WrappedGaussian	sum	Incoming message from `a`. Must be a proper distribution. If uniform, the result will be uniform.
Double	a	Constant value for `difference`.

Returns

Type	Description
WrappedGaussian	The outgoing VMP message to the `b` argument.

Remarks

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

Exceptions

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

SumAverageConditional(WrappedGaussian, WrappedGaussian)

EP message to a.

Declaration

public static WrappedGaussian SumAverageConditional(WrappedGaussian a, WrappedGaussian b)

Parameters

Type	Name	Description
WrappedGaussian	a	Incoming message from `difference`. Must be a proper distribution. If uniform, the result will be uniform.
WrappedGaussian	b	Incoming message from `b`. Must be a proper distribution. If uniform, the result will be uniform.

Returns

Type	Description
WrappedGaussian	The outgoing EP message to the `a` argument.

Remarks

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

Exceptions

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

SumAverageConditional(WrappedGaussian, Double)

EP message to a.

Declaration

public static WrappedGaussian SumAverageConditional(WrappedGaussian a, double b)

Parameters

Type	Name	Description
WrappedGaussian	a	Incoming message from `difference`. Must be a proper distribution. If uniform, the result will be uniform.
Double	b	Constant value for `b`.

Returns

Type	Description
WrappedGaussian	The outgoing EP message to the `a` argument.

Remarks

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

Exceptions

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

SumAverageConditional(Double, WrappedGaussian)

EP message to a.

Declaration

public static WrappedGaussian SumAverageConditional(double a, WrappedGaussian b)

Parameters

Type	Name	Description
Double	a	Constant value for `difference`.
WrappedGaussian	b	Incoming message from `b`. Must be a proper distribution. If uniform, the result will be uniform.

Returns

Type	Description
WrappedGaussian	The outgoing EP message to the `a` argument.

Remarks

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

Exceptions

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

SumAverageLogarithm(WrappedGaussian, WrappedGaussian)

VMP message to a.

Declaration

public static WrappedGaussian SumAverageLogarithm(WrappedGaussian a, WrappedGaussian b)

Parameters

Type	Name	Description
WrappedGaussian	a	Incoming message from `difference`. Must be a proper distribution. If uniform, the result will be uniform.
WrappedGaussian	b	Incoming message from `b`. Must be a proper distribution. If uniform, the result will be uniform.

Returns

Type	Description
WrappedGaussian	The outgoing VMP message to the `a` argument.

Remarks

The outgoing message is the exponential of the average log-factor value, where the average is over all arguments except a. Because the factor is deterministic, difference is integrated out before taking the logarithm. The formula is exp(sum_(b) p(b) log(sum_difference p(difference) factor(difference,a,b))).

Exceptions

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

SumAverageLogarithm(WrappedGaussian, Double)

VMP message to a.

Declaration

public static WrappedGaussian SumAverageLogarithm(WrappedGaussian a, double b)

Parameters

Type	Name	Description
WrappedGaussian	a	Incoming message from `difference`. Must be a proper distribution. If uniform, the result will be uniform.
Double	b	Constant value for `b`.

Returns

Type	Description
WrappedGaussian	The outgoing VMP message to the `a` argument.

Remarks

The outgoing message is the factor viewed as a function of a with difference integrated out. The formula is sum_difference p(difference) factor(difference,a,b).

Exceptions

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

SumAverageLogarithm(Double, WrappedGaussian)

VMP message to a.

Declaration

public static WrappedGaussian SumAverageLogarithm(double a, WrappedGaussian b)

Parameters

Type	Name	Description
Double	a	Constant value for `difference`.
WrappedGaussian	b	Incoming message from `b`. Must be a proper distribution. If uniform, the result will be uniform.

Returns

Type	Description
WrappedGaussian	The outgoing VMP message to the `a` argument.

Remarks

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

Exceptions

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