Class StringsAreEqualOp

Provides outgoing messages for AreEqual(String, String), given random arguments to the function.

Inheritance

Object

StringsAreEqualOp

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), "AreEqual", new Type[]{typeof(string), typeof(string)})]
[Quality(QualityBand.Experimental)]
public static class StringsAreEqualOp

Methods

AreEqualAverageConditional(StringDistribution, StringDistribution)

EP message to areEqual.

Declaration

public static Bernoulli AreEqualAverageConditional(StringDistribution str1, StringDistribution str2)

Parameters

Type	Name	Description
StringDistribution	str1	Incoming message from `str1`.
StringDistribution	str2	Incoming message from `str2`.

Returns

Type	Description
Bernoulli	The outgoing EP message to the `areEqual` argument.

Remarks

The outgoing message is a distribution matching the moments of areEqual as the random arguments are varied. The formula is proj[p(areEqual) sum_(str1,str2) p(str1,str2) factor(areEqual,str1,str2)]/p(areEqual).

LogAverageFactor(Boolean, StringDistribution, StringDistribution)

Evidence message for EP.

Declaration

public static double LogAverageFactor(bool areEqual, StringDistribution str1, StringDistribution str2)

Parameters

Type	Name	Description
Boolean	areEqual	Constant value for `areEqual`.
StringDistribution	str1	Incoming message from `str1`.
StringDistribution	str2	Incoming message from `str2`.

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_(str1,str2) p(str1,str2) factor(areEqual,str1,str2)).

LogEvidenceRatio(Bernoulli)

Evidence message for EP.

Declaration

public static double LogEvidenceRatio(Bernoulli areEqual)

Parameters

Type	Name	Description
Bernoulli	areEqual	Incoming message from `areEqual`.

Returns

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

Remarks

The formula for the result is log(sum_(areEqual) p(areEqual) factor(areEqual,str1,str2) / sum_areEqual p(areEqual) messageTo(areEqual)). Adding up these values across all factors and variables gives the log-evidence estimate for EP.

LogEvidenceRatio(Boolean, StringDistribution, StringDistribution)

Evidence message for EP.

Declaration

public static double LogEvidenceRatio(bool areEqual, StringDistribution str1, StringDistribution str2)

Parameters

Type	Name	Description
Boolean	areEqual	Constant value for `areEqual`.
StringDistribution	str1	Incoming message from `str1`.
StringDistribution	str2	Incoming message from `str2`.

Returns

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

Remarks

The formula for the result is log(sum_(str1,str2) p(str1,str2) factor(areEqual,str1,str2)). Adding up these values across all factors and variables gives the log-evidence estimate for EP.

Str1AverageConditional(StringDistribution, Bernoulli, StringDistribution)

EP message to str1.

Declaration

public static StringDistribution Str1AverageConditional(StringDistribution str2, Bernoulli areEqual, StringDistribution result)

Parameters

Type	Name	Description
StringDistribution	str2	Incoming message from `str2`.
Bernoulli	areEqual	Incoming message from `areEqual`. Must be a proper distribution. If uniform, the result will be uniform.
StringDistribution	result	Modified to contain the outgoing message.

Returns

Type	Description
StringDistribution	`result`

Remarks

The outgoing message is a distribution matching the moments of str1 as the random arguments are varied. The formula is proj[p(str1) sum_(str2,areEqual) p(str2,areEqual) factor(areEqual,str1,str2)]/p(str1).

Exceptions

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

Str2AverageConditional(StringDistribution, Bernoulli, StringDistribution)

EP message to str2.

Declaration

public static StringDistribution Str2AverageConditional(StringDistribution str1, Bernoulli areEqual, StringDistribution result)

Parameters

Type	Name	Description
StringDistribution	str1	Incoming message from `str1`.
Bernoulli	areEqual	Incoming message from `areEqual`. Must be a proper distribution. If uniform, the result will be uniform.
StringDistribution	result	Modified to contain the outgoing message.

Returns

Type	Description
StringDistribution	`result`

Remarks

The outgoing message is a distribution matching the moments of str2 as the random arguments are varied. The formula is proj[p(str2) sum_(str1,areEqual) p(str1,areEqual) factor(areEqual,str1,str2)]/p(str2).

Exceptions

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