Class BernoulliFromDiscreteOp

Provides outgoing messages for BernoulliFromDiscrete(Int32, Double[]), given random arguments to the function.

Inheritance

Object

BernoulliFromDiscreteOp

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(new string[]{"sample", "index", "probTrue"}, typeof(Factor), "BernoulliFromDiscrete", new Type[]{})]
public static class BernoulliFromDiscreteOp

Methods

IndexAverageConditional(Bernoulli, Double[], Discrete)

EP message to index.

Declaration

public static Discrete IndexAverageConditional(Bernoulli sample, double[] ProbTrue, Discrete result)

Parameters

Type	Name	Description
Bernoulli	sample	Incoming message from `sample`. Must be a proper distribution. If uniform, the result will be uniform.
Double[]	ProbTrue	Constant value for `probTrue`.
Discrete	result	Modified to contain the outgoing message.

Returns

Type	Description
Discrete	`result`

Remarks

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

Exceptions

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

IndexAverageConditional(Boolean, Double[], Discrete)

EP message to index.

Declaration

public static Discrete IndexAverageConditional(bool sample, double[] ProbTrue, Discrete result)

Parameters

Type	Name	Description
Boolean	sample	Constant value for `sample`.
Double[]	ProbTrue	Constant value for `probTrue`.
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 index conditioned on the given values.

IndexAverageLogarithm(Bernoulli, Double[], Discrete)

VMP message to index.

Declaration

public static Discrete IndexAverageLogarithm(Bernoulli sample, double[] ProbTrue, Discrete result)

Parameters

Type	Name	Description
Bernoulli	sample	Incoming message from `sample`.
Double[]	ProbTrue	Constant value for `probTrue`.
Discrete	result	Modified to contain the outgoing message.

Returns

Type	Description
Discrete	`result`

Remarks

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

IndexAverageLogarithm(Boolean, Double[], Discrete)

VMP message to index.

Declaration

public static Discrete IndexAverageLogarithm(bool sample, double[] ProbTrue, Discrete result)

Parameters

Type	Name	Description
Boolean	sample	Constant value for `sample`.
Double[]	ProbTrue	Constant value for `probTrue`.
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 index conditioned on the given values.

IndexConditional(Boolean, Double[], Discrete)

Gibbs message to index.

Declaration

public static Discrete IndexConditional(bool sample, double[] ProbTrue, Discrete result)

Parameters

Type	Name	Description
Boolean	sample	Constant value for `sample`.
Double[]	ProbTrue	Constant value for `probTrue`.
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 index conditioned on the given values.

LogAverageFactor(Boolean, Discrete, Double[])

Evidence message for EP.

Declaration

public static double LogAverageFactor(bool sample, Discrete index, double[] probTrue)

Parameters

Type	Name	Description
Boolean	sample	Constant value for `sample`.
Discrete	index	Incoming message from `index`.
Double[]	probTrue	Constant value for `probTrue`.

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_(index) p(index) factor(sample,index,probTrue)).

LogEvidenceRatio(Boolean, Discrete, Double[])

Evidence message for EP.

Declaration

public static double LogEvidenceRatio(bool sample, Discrete index, double[] probTrue)

Parameters

Type	Name	Description
Boolean	sample	Constant value for `sample`.
Discrete	index	Incoming message from `index`.
Double[]	probTrue	Constant value for `probTrue`.

Returns

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

Remarks

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

SampleAverageConditional(Discrete, Double[])

EP message to sample.

Declaration

public static Bernoulli SampleAverageConditional(Discrete index, double[] ProbTrue)

Parameters

Type	Name	Description
Discrete	index	Incoming message from `index`.
Double[]	ProbTrue	Constant value for `probTrue`.

Returns

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

Remarks

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

SampleAverageConditional(Int32, Double[])

EP message to sample.

Declaration

public static Bernoulli SampleAverageConditional(int index, double[] ProbTrue)

Parameters

Type	Name	Description
Int32	index	Constant value for `index`.
Double[]	ProbTrue	Constant value for `probTrue`.

Returns

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

Remarks

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

SampleAverageLogarithm(Discrete, Double[])

VMP message to sample.

Declaration

public static Bernoulli SampleAverageLogarithm(Discrete index, double[] ProbTrue)

Parameters

Type	Name	Description
Discrete	index	Incoming message from `index`.
Double[]	ProbTrue	Constant value for `probTrue`.

Returns

Type	Description
Bernoulli	The outgoing VMP message to the `sample` argument.

Remarks

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

SampleAverageLogarithm(Int32, Double[])

VMP message to sample.

Declaration

public static Bernoulli SampleAverageLogarithm(int index, double[] ProbTrue)

Parameters

Type	Name	Description
Int32	index	Constant value for `index`.
Double[]	ProbTrue	Constant value for `probTrue`.

Returns

Type	Description
Bernoulli	The outgoing VMP message to the `sample` argument.

Remarks

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

SampleConditional(Int32, Double[])

Gibbs message to sample.

Declaration

public static Bernoulli SampleConditional(int index, double[] ProbTrue)

Parameters

Type	Name	Description
Int32	index	Constant value for `index`.
Double[]	ProbTrue	Constant value for `probTrue`.

Returns

Type	Description
Bernoulli	The outgoing Gibbs message to the `sample` argument.

Remarks

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