Class ArrayFromVectorOp

Provides outgoing messages for ArrayFromVector(Vector), given random arguments to the function.

Inheritance

Object

ArrayFromVectorOp

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), "ArrayFromVector", new Type[]{})]
[Quality(QualityBand.Stable)]
public static class ArrayFromVectorOp

Methods

ArrayAverageConditional<GaussianList>(IList<Gaussian>, VectorGaussian, GaussianList)

EP message to array.

Declaration

public static GaussianList ArrayAverageConditional<GaussianList>(IList<Gaussian> array, VectorGaussian vector, GaussianList result)
    where GaussianList : IList<Gaussian>

Parameters

Type	Name	Description
IList<Gaussian>	array	Incoming message from array.
VectorGaussian	vector	Incoming message from vector. Must be a proper distribution. If any element is uniform, the result will be uniform.
GaussianList	result	Modified to contain the outgoing message.

Returns

Type	Description
GaussianList	result

Type Parameters

Name	Description
GaussianList	The type of the resulting 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_(vector) p(vector) factor(array,vector)]/p(array).

Exceptions

Type	Condition
ImproperMessageException	vector is not a proper distribution.

ArrayAverageConditional<GaussianList>(IList<Gaussian>, VectorGaussianMoments, GaussianList)

EP message to array.

Declaration

public static GaussianList ArrayAverageConditional<GaussianList>(IList<Gaussian> array, VectorGaussianMoments vector, GaussianList result)
    where GaussianList : IList<Gaussian>

Parameters

Type	Name	Description
IList<Gaussian>	array	Incoming message from array.
VectorGaussianMoments	vector	Incoming message from vector. Must be a proper distribution. If any element is uniform, the result will be uniform.
GaussianList	result	Modified to contain the outgoing message.

Returns

Type	Description
GaussianList	result

Type Parameters

Name	Description
GaussianList	The type of the resulting 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_(vector) p(vector) factor(array,vector)]/p(array).

Exceptions

Type	Condition
ImproperMessageException	vector is not a proper distribution.

ArrayAverageConditionalInit(VectorGaussian)

Declaration

public static DistributionStructArray<Gaussian, double> ArrayAverageConditionalInit(VectorGaussian vector)

Parameters

Type	Name	Description
VectorGaussian	vector	Incoming message from vector.

Returns

Type	Description
DistributionStructArray<Gaussian, Double>

Remarks

ArrayAverageLogarithm<GaussianList>(VectorGaussian, GaussianList)

VMP message to array.

Declaration

public static GaussianList ArrayAverageLogarithm<GaussianList>(VectorGaussian vector, GaussianList result)
    where GaussianList : IList<Gaussian>

Parameters

Type	Name	Description
VectorGaussian	vector	Incoming message from vector. Must be a proper distribution. If any element is uniform, the result will be uniform.
GaussianList	result	Modified to contain the outgoing message.

Returns

Type	Description
GaussianList	result

Type Parameters

Name	Description
GaussianList	The type of the resulting array.

Remarks

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

Exceptions

Type	Condition
ImproperMessageException	vector is not a proper distribution.

AverageLogFactor(IList<Gaussian>, VectorGaussian)

Evidence message for VMP.

Declaration

public static double AverageLogFactor(IList<Gaussian> array, VectorGaussian vector)

Parameters

Type	Name	Description
IList<Gaussian>	array	Incoming message from array.
VectorGaussian	vector	Incoming message from vector.

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.

AverageLogFactor(IList<Gaussian>, VectorGaussianMoments)

Evidence message for VMP.

Declaration

public static double AverageLogFactor(IList<Gaussian> array, VectorGaussianMoments vector)

Parameters

Type	Name	Description
IList<Gaussian>	array	Incoming message from array.
VectorGaussianMoments	vector	Incoming message from vector.

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.

AverageLogFactor(IList<Gaussian>, Vector)

Evidence message for VMP.

Declaration

public static double AverageLogFactor(IList<Gaussian> array, Vector vector)

Parameters

Type	Name	Description
IList<Gaussian>	array	Incoming message from array.
Vector	vector	Constant value for vector.

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.

AverageLogFactor(Double[], VectorGaussian)

Evidence message for VMP.

Declaration

public static double AverageLogFactor(double[] array, VectorGaussian vector)

Parameters

Type	Name	Description
Double[]	array	Constant value for array.
VectorGaussian	vector	Incoming message from vector.

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.

AverageLogFactor(Double[], VectorGaussianMoments)

Evidence message for VMP.

Declaration

public static double AverageLogFactor(double[] array, VectorGaussianMoments vector)

Parameters

Type	Name	Description
Double[]	array	Constant value for array.
VectorGaussianMoments	vector	Incoming message from vector.

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.

AverageLogFactor(Double[], Vector)

Evidence message for VMP.

Declaration

public static double AverageLogFactor(double[] array, Vector vector)

Parameters

Type	Name	Description
Double[]	array	Constant value for array.
Vector	vector	Constant value for vector.

Returns

Type	Description
Double	Zero.

Remarks

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

LogAverageFactor(IList<Gaussian>, VectorGaussian, VectorGaussian)

Evidence message for EP.

Declaration

public static double LogAverageFactor(IList<Gaussian> array, VectorGaussian vector, VectorGaussian to_vector)

Parameters

Type	Name	Description
IList<Gaussian>	array	Incoming message from array.
VectorGaussian	vector	Incoming message from vector.
VectorGaussian	to_vector	Outgoing message to vector.

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_(array,vector) p(array,vector) factor(array,vector)).

LogAverageFactor(IList<Gaussian>, VectorGaussianMoments, VectorGaussianMoments)

Evidence message for EP.

Declaration

public static double LogAverageFactor(IList<Gaussian> array, VectorGaussianMoments vector, VectorGaussianMoments to_vector)

Parameters

Type	Name	Description
IList<Gaussian>	array	Incoming message from array.
VectorGaussianMoments	vector	Incoming message from vector.
VectorGaussianMoments	to_vector	Outgoing message to vector.

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_(array,vector) p(array,vector) factor(array,vector)).

LogAverageFactor(Double[], VectorGaussian)

Evidence message for EP.

Declaration

public static double LogAverageFactor(double[] array, VectorGaussian vector)

Parameters

Type	Name	Description
Double[]	array	Constant value for array.
VectorGaussian	vector	Incoming message from vector.

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_(vector) p(vector) factor(array,vector)).

LogAverageFactor(Double[], VectorGaussianMoments)

Evidence message for EP.

Declaration

public static double LogAverageFactor(double[] array, VectorGaussianMoments vector)

Parameters

Type	Name	Description
Double[]	array	Constant value for array.
VectorGaussianMoments	vector	Incoming message from vector.

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_(vector) p(vector) factor(array,vector)).

LogAverageFactor(Double[], Vector)

Evidence message for EP.

Declaration

public static double LogAverageFactor(double[] array, Vector vector)

Parameters

Type	Name	Description
Double[]	array	Constant value for array.
Vector	vector	Constant value for vector.

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(array,vector)).

LogEvidenceRatio(IList<Gaussian>, VectorGaussian, VectorGaussian, IList<Gaussian>)

Evidence message for EP.

Declaration

public static double LogEvidenceRatio(IList<Gaussian> array, VectorGaussian vector, VectorGaussian to_vector, IList<Gaussian> to_array)

Parameters

Type	Name	Description
IList<Gaussian>	array	Incoming message from array.
VectorGaussian	vector	Incoming message from vector.
VectorGaussian	to_vector	Outgoing message to vector.
IList<Gaussian>	to_array	Outgoing message to array.

Returns

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

Remarks

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

LogEvidenceRatio(IList<Gaussian>, VectorGaussianMoments, VectorGaussianMoments, IList<Gaussian>)

Evidence message for EP.

Declaration

public static double LogEvidenceRatio(IList<Gaussian> array, VectorGaussianMoments vector, VectorGaussianMoments to_vector, IList<Gaussian> to_array)

Parameters

Type	Name	Description
IList<Gaussian>	array	Incoming message from array.
VectorGaussianMoments	vector	Incoming message from vector.
VectorGaussianMoments	to_vector	Outgoing message to vector.
IList<Gaussian>	to_array	Outgoing message to array.

Returns

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

Remarks

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

LogEvidenceRatio(Double[], VectorGaussian)

Evidence message for EP.

Declaration

public static double LogEvidenceRatio(double[] array, VectorGaussian vector)

Parameters

Type	Name	Description
Double[]	array	Constant value for array.
VectorGaussian	vector	Incoming message from vector.

Returns

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

Remarks

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

LogEvidenceRatio(Double[], VectorGaussianMoments)

Evidence message for EP.

Declaration

public static double LogEvidenceRatio(double[] array, VectorGaussianMoments vector)

Parameters

Type	Name	Description
Double[]	array	Constant value for array.
VectorGaussianMoments	vector	Incoming message from vector.

Returns

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

Remarks

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

LogEvidenceRatio(Double[], Vector)

Evidence message for EP.

Declaration

public static double LogEvidenceRatio(double[] array, Vector vector)

Parameters

Type	Name	Description
Double[]	array	Constant value for array.
Vector	vector	Constant value for vector.

Returns

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

Remarks

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

VectorAverageConditional(IList<Gaussian>, VectorGaussian)

EP message to vector.

Declaration

public static VectorGaussian VectorAverageConditional([SkipIfAllUniform] IList<Gaussian> array, VectorGaussian result)

Parameters

Type	Name	Description
IList<Gaussian>	array	Incoming message from array. Must be a proper distribution. If all elements are uniform, the result will be uniform.
VectorGaussian	result	Modified to contain the outgoing message.

Returns

Type	Description
VectorGaussian	result

Remarks

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

Exceptions

Type	Condition
ImproperMessageException	array is not a proper distribution.

VectorAverageConditional(IList<Gaussian>, VectorGaussianMoments)

EP message to vector.

Declaration

public static VectorGaussianMoments VectorAverageConditional([SkipIfAllUniform] IList<Gaussian> array, VectorGaussianMoments result)

Parameters

Type	Name	Description
IList<Gaussian>	array	Incoming message from array. Must be a proper distribution. If all elements are uniform, the result will be uniform.
VectorGaussianMoments	result	Modified to contain the outgoing message.

Returns

Type	Description
VectorGaussianMoments	result

Remarks

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

Exceptions

Type	Condition
ImproperMessageException	array is not a proper distribution.

VectorAverageConditional(Double[], VectorGaussian)

EP message to vector.

Declaration

public static VectorGaussian VectorAverageConditional(double[] array, VectorGaussian result)

Parameters

Type	Name	Description
Double[]	array	Constant value for array.
VectorGaussian	result	Modified to contain the outgoing message.

Returns

Type	Description
VectorGaussian	result

Remarks

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

VectorAverageConditional(Double[], VectorGaussianMoments)

EP message to vector.

Declaration

public static VectorGaussianMoments VectorAverageConditional(double[] array, VectorGaussianMoments result)

Parameters

Type	Name	Description
Double[]	array	Constant value for array.
VectorGaussianMoments	result	Modified to contain the outgoing message.

Returns

Type	Description
VectorGaussianMoments	result

Remarks

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

VectorAverageLogarithm(IList<Gaussian>, VectorGaussian)

VMP message to vector.

Declaration

public static VectorGaussian VectorAverageLogarithm([SkipIfAllUniform] IList<Gaussian> array, VectorGaussian result)

Parameters

Type	Name	Description
IList<Gaussian>	array	Incoming message from array. Must be a proper distribution. If all elements are uniform, the result will be uniform.
VectorGaussian	result	Modified to contain the outgoing message.

Returns

Type	Description
VectorGaussian	result

Remarks

The outgoing message is the factor viewed as a function of vector with array integrated out. The formula is sum_array p(array) factor(array,vector).

Exceptions

Type	Condition
ImproperMessageException	array is not a proper distribution.

VectorAverageLogarithm(IList<Gaussian>, VectorGaussianMoments)

VMP message to vector.

Declaration

public static VectorGaussianMoments VectorAverageLogarithm([SkipIfAllUniform] IList<Gaussian> array, VectorGaussianMoments result)

Parameters

Type	Name	Description
IList<Gaussian>	array	Incoming message from array. Must be a proper distribution. If all elements are uniform, the result will be uniform.
VectorGaussianMoments	result	Modified to contain the outgoing message.

Returns

Type	Description
VectorGaussianMoments	result

Remarks

The outgoing message is the factor viewed as a function of vector with array integrated out. The formula is sum_array p(array) factor(array,vector).

Exceptions

Type	Condition
ImproperMessageException	array is not a proper distribution.

VectorAverageLogarithm(Double[], VectorGaussian)

VMP message to vector.

Declaration

public static VectorGaussian VectorAverageLogarithm(double[] array, VectorGaussian result)

Parameters

Type	Name	Description
Double[]	array	Constant value for array.
VectorGaussian	result	Modified to contain the outgoing message.

Returns

Type	Description
VectorGaussian	result

Remarks

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

VectorAverageLogarithm(Double[], VectorGaussianMoments)

VMP message to vector.

Declaration

public static VectorGaussianMoments VectorAverageLogarithm(double[] array, VectorGaussianMoments result)

Parameters

Type	Name	Description
Double[]	array	Constant value for array.
VectorGaussianMoments	result	Modified to contain the outgoing message.

Returns

Type	Description
VectorGaussianMoments	result

Remarks

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