Class CountTrueOp

Provides outgoing messages for CountTrue(Boolean[]), given random arguments to the function.

Inheritance

Object

CountTrueOp

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), "CountTrue", new Type[]{})]
[Quality(QualityBand.Preview)]
[Buffers(new string[]{"PoissonBinomialTable"})]
public static class CountTrueOp

Methods

ArrayAverageConditional<TBernoulliArray>(TBernoulliArray, Discrete, Double[,], TBernoulliArray)

EP message to array.

Declaration

public static TBernoulliArray ArrayAverageConditional<TBernoulliArray>(TBernoulliArray array, Discrete count, double[, ] poissonBinomialTable, TBernoulliArray result)
    where TBernoulliArray : IList<Bernoulli>

Parameters

Type	Name	Description
TBernoulliArray	array	Incoming message from array.
Discrete	count	Incoming message from count. Must be a proper distribution. If uniform, the result will be uniform.
Double[,]	poissonBinomialTable	Buffer poissonBinomialTable.
TBernoulliArray	result	Modified to contain the outgoing message.

Returns

Type	Description
TBernoulliArray	result

Type Parameters

Name	Description
TBernoulliArray	The type of messages from/to 'array'.

Remarks

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

Exceptions

Type	Condition
ImproperMessageException	count is not a proper distribution.

ArrayAverageConditional<TBernoulliArray>(TBernoulliArray, Int32, Double[,], TBernoulliArray)

EP message to array.

Declaration

public static TBernoulliArray ArrayAverageConditional<TBernoulliArray>(TBernoulliArray array, int count, double[, ] poissonBinomialTable, TBernoulliArray result)
    where TBernoulliArray : IList<Bernoulli>

Parameters

Type	Name	Description
TBernoulliArray	array	Incoming message from array.
Int32	count	Constant value for count.
Double[,]	poissonBinomialTable	Buffer poissonBinomialTable.
TBernoulliArray	result	Modified to contain the outgoing message.

Returns

Type	Description
TBernoulliArray	result

Type Parameters

Name	Description
TBernoulliArray	The type of messages from/to 'array'.

Remarks

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

CountAverageConditional(Double[,])

EP message to count.

Declaration

public static Discrete CountAverageConditional(double[, ] poissonBinomialTable)

Parameters

Type	Name	Description
Double[,]	poissonBinomialTable	Buffer poissonBinomialTable.

Returns

Type	Description
Discrete	The outgoing EP message to the count argument.

Remarks

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

LogEvidenceRatio(Discrete)

Evidence message for EP.

Declaration

public static double LogEvidenceRatio(Discrete count)

Parameters

Type	Name	Description
Discrete	count	Incoming message from count. Must be a proper distribution. If uniform, the result will be uniform.

Returns

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

Remarks

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

Exceptions

Type	Condition
ImproperMessageException	count is not a proper distribution.

LogEvidenceRatio(Int32, Boolean[])

Evidence message for EP.

Declaration

public static double LogEvidenceRatio(int count, bool[] array)

Parameters

Type	Name	Description
Int32	count	Constant value for count.
Boolean[]	array	Constant value for array.

Returns

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

Remarks

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

LogEvidenceRatio(Int32, Double[,])

Evidence message for EP.

Declaration

public static double LogEvidenceRatio(int count, double[, ] poissonBinomialTable)

Parameters

Type	Name	Description
Int32	count	Constant value for count.
Double[,]	poissonBinomialTable	Buffer poissonBinomialTable.

Returns

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

Remarks

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

PoissonBinomialTable(IList<Bernoulli>)

Update the buffer PoissonBinomialTable.

Declaration

public static double[, ] PoissonBinomialTable(IList<Bernoulli> array)

Parameters

Type	Name	Description
IList<Bernoulli>	array	Incoming message from array.

Returns

Type	Description
Double[,]	New value of buffer PoissonBinomialTable.

Remarks

PoissonBinomialTableInit(IList<Bernoulli>)

Initialize the buffer PoissonBinomialTable.

Declaration

public static double[, ] PoissonBinomialTableInit(IList<Bernoulli> array)

Parameters

Type	Name	Description
IList<Bernoulli>	array	Incoming message from array.

Returns

Type	Description
Double[,]	Initial value of buffer PoissonBinomialTable.

Remarks