Class GateEnterOneOp<T>

Provides outgoing messages for EnterOne<T>(Int32, T, Int32), given random arguments to the function.

Inheritance

Object

GateEnterOneOp<T>

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(Gate), "EnterOne<>", new Type[]{null, typeof(int), null, typeof(int)})]
[Quality(QualityBand.Mature)]
public static class GateEnterOneOp<T>

Type Parameters

Name	Description
T	The type of the variable entering the gate.

Methods

AverageLogFactor()

Evidence message for VMP.

Declaration

public static double AverageLogFactor()

Returns

Type	Description
Double	Zero.

Remarks

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

EnterOneAverageConditional<TValue>(TValue)

EP message to enterOne.

Declaration

public static TValue EnterOneAverageConditional<TValue>(TValue value)

Parameters

Type	Name	Description
TValue	value	Incoming message from `value`.

Returns

Type	Description
TValue	The outgoing EP message to the `enterOne` argument.

Type Parameters

Name	Description
TValue	The type of the incoming message from `value`.

Remarks

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

EnterOneAverageLogarithm<TValue>(TValue)

VMP message to enterOne.

Declaration

public static TValue EnterOneAverageLogarithm<TValue>(TValue value)

Parameters

Type	Name	Description
TValue	value	Incoming message from `value`.

Returns

Type	Description
TValue	The outgoing VMP message to the `enterOne` argument.

Type Parameters

Name	Description
TValue	The type of the incoming message from `value`.

Remarks

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

LogAverageFactor()

Evidence message for EP.

Declaration

public static double LogAverageFactor()

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(enterOne,selector,value,index)).

LogEvidenceRatio<TDist>(TDist)

Evidence message for EP.

Declaration

public static double LogEvidenceRatio<TDist>(TDist enterOne)
    where TDist : IDistribution<T>

Parameters

Type	Name	Description
TDist	enterOne	Incoming message from `enterOne`.

Returns

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

Type Parameters

Name	Description
TDist

Remarks

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

SelectorAverageConditional(Discrete)

EP message to selector.

Declaration

public static Discrete SelectorAverageConditional(Discrete result)

Parameters

Type	Name	Description
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 selector conditioned on the given values.

SelectorAverageLogarithm(Discrete)

VMP message to selector.

Declaration

public static Discrete SelectorAverageLogarithm(Discrete result)

Parameters

Type	Name	Description
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 selector conditioned on the given values.

ValueAverageConditional<TDist>(TDist, Discrete, TDist, Int32, TDist)

EP message to value.

Declaration

public static TDist ValueAverageConditional<TDist>([SkipIfAllUniform] TDist enterOne, Discrete selector, TDist value, int index, TDist result)
    where TDist : IDistribution<T>, SettableToProduct<TDist>, SettableToRatio<TDist>, SettableToWeightedSum<TDist>, SettableTo<TDist>, SettableToPower<TDist>

Parameters

Type	Name	Description
TDist	enterOne	Incoming message from `enterOne`. Must be a proper distribution. If uniform, the result will be uniform.
Discrete	selector	Incoming message from `selector`.
TDist	value	Incoming message from `value`. Must be a proper distribution. If uniform, the result will be uniform.
Int32	index	Constant value for `index`.
TDist	result	Modified to contain the outgoing message.

Returns

Type	Description
TDist	`result`

Type Parameters

Name	Description
TDist	The type of the distribution over the variable entering the gate.

Remarks

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

Exceptions

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

ValueAverageConditional<TDist>(TDist, Int32, Int32, TDist)

EP message to value.

Declaration

public static TDist ValueAverageConditional<TDist>([SkipIfAllUniform] TDist enterOne, int selector, int index, TDist result)
    where TDist : IDistribution<T>, SettableTo<TDist>

Parameters

Type	Name	Description
TDist	enterOne	Incoming message from `enterOne`. Must be a proper distribution. If uniform, the result will be uniform.
Int32	selector	Constant value for `selector`.
Int32	index	Constant value for `index`.
TDist	result	Modified to contain the outgoing message.

Returns

Type	Description
TDist	`result`

Type Parameters

Name	Description
TDist	The type of the distribution over the variable entering the gate.

Remarks

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

Exceptions

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

ValueAverageLogarithm<TDist>(TDist, Discrete, Int32, TDist)

VMP message to value.

Declaration

public static TDist ValueAverageLogarithm<TDist>(TDist enterOne, Discrete selector, int index, TDist result)
    where TDist : SettableToPower<TDist>

Parameters

Type	Name	Description
TDist	enterOne	Incoming message from `enterOne`. Must be a proper distribution. If uniform, the result will be uniform.
Discrete	selector	Incoming message from `selector`.
Int32	index	Constant value for `index`.
TDist	result	Modified to contain the outgoing message.

Returns

Type	Description
TDist	`result`

Type Parameters

Name	Description
TDist	The type of the distribution over the variable entering the gate.

Remarks

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

Exceptions

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