Class GetItem2DOp<T>

Provides outgoing messages for GetItem2D<T>(T[,], Int32, Int32), given random arguments to the function.

Inheritance

Object

GetItem2DOp<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(Collection), "GetItem2D<>", new Type[]{})]
[Quality(QualityBand.Stable)]
public static class GetItem2DOp<T>

Type Parameters

Name	Description
T	The type of an item.

Methods

ArrayAverageConditional<Distribution, DistributionArray>(Distribution, Int32, Int32, DistributionArray)

EP message to array.

Declaration

public static DistributionArray ArrayAverageConditional<Distribution, DistributionArray>(Distribution item, int index1, int index2, DistributionArray result)
    where Distribution : SettableTo<Distribution> where DistributionArray : IArray2D<Distribution>

Parameters

Type	Name	Description
Distribution	item	Incoming message from item. Must be a proper distribution. If uniform, the result will be uniform.
Int32	index1	Constant value for index1.
Int32	index2	Constant value for index2.
DistributionArray	result	Modified to contain the outgoing message.

Returns

Type	Description
DistributionArray	result

Type Parameters

Name	Description
Distribution	The type of the distribution over an item.
DistributionArray	The type of the outgoing message.

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_(item) p(item) factor(item,array,index1,index2)]/p(array).

Exceptions

Type	Condition
ImproperMessageException	item is not a proper distribution.

ArrayAverageConditional<Distribution, DistributionArray>(T, Int32, Int32, DistributionArray)

EP message to array.

Declaration

public static DistributionArray ArrayAverageConditional<Distribution, DistributionArray>(T item, int index1, int index2, DistributionArray result)
    where Distribution : HasPoint<T> where DistributionArray : IArray2D<Distribution>

Parameters

Type	Name	Description
T	item	Incoming message from item.
Int32	index1	Constant value for index1.
Int32	index2	Constant value for index2.
DistributionArray	result	Modified to contain the outgoing message.

Returns

Type	Description
DistributionArray	result

Type Parameters

Name	Description
Distribution	The type of the distribution over an item.
DistributionArray	The type of the outgoing message.

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_(item) p(item) factor(item,array,index1,index2)]/p(array).

ArrayAverageLogarithm<Distribution, DistributionArray>(Distribution, Int32, Int32, DistributionArray)

VMP message to array.

Declaration

public static DistributionArray ArrayAverageLogarithm<Distribution, DistributionArray>(Distribution item, int index1, int index2, DistributionArray result)
    where Distribution : SettableTo<Distribution> where DistributionArray : IArray2D<Distribution>

Parameters

Type	Name	Description
Distribution	item	Incoming message from item. Must be a proper distribution. If uniform, the result will be uniform.
Int32	index1	Constant value for index1.
Int32	index2	Constant value for index2.
DistributionArray	result	Modified to contain the outgoing message.

Returns

Type	Description
DistributionArray	result

Type Parameters

Name	Description
Distribution	The type of the distribution over an item.
DistributionArray	The type of the outgoing message.

Remarks

The outgoing message is the factor viewed as a function of array with item integrated out. The formula is sum_item p(item) factor(item,array,index1,index2).

Exceptions

Type	Condition
ImproperMessageException	item is not a proper distribution.

ArrayAverageLogarithm<Distribution, DistributionArray>(T, Int32, Int32, DistributionArray)

VMP message to array.

Declaration

public static DistributionArray ArrayAverageLogarithm<Distribution, DistributionArray>(T item, int index1, int index2, DistributionArray result)
    where Distribution : HasPoint<T> where DistributionArray : IArray2D<Distribution>

Parameters

Type	Name	Description
T	item	Incoming message from item.
Int32	index1	Constant value for index1.
Int32	index2	Constant value for index2.
DistributionArray	result	Modified to contain the outgoing message.

Returns

Type	Description
DistributionArray	result

Type Parameters

Name	Description
Distribution	The type of the distribution over an item.
DistributionArray	The type of the outgoing message.

Remarks

The outgoing message is the factor viewed as a function of array with item integrated out. The formula is sum_item p(item) factor(item,array,index1,index2).

AverageLogFactor()

Evidence message for VMP.

Declaration

public static double AverageLogFactor()

Returns

Type	Description
Double	Zero.

Remarks

The formula for the result is log(factor(item,array,index1,index2)). Adding up these values across all factors and variables gives the log-evidence estimate for VMP.

ItemAverageConditional<Distribution>(IArray2D<Distribution>, Int32, Int32, Distribution)

EP message to item.

Declaration

public static Distribution ItemAverageConditional<Distribution>([SkipIfAllUniform] IArray2D<Distribution> array, int index1, int index2, Distribution result)
    where Distribution : SettableTo<Distribution>

Parameters

Type	Name	Description
IArray2D<Distribution>	array	Incoming message from array. Must be a proper distribution. If all elements are uniform, the result will be uniform.
Int32	index1	Constant value for index1.
Int32	index2	Constant value for index2.
Distribution	result	Modified to contain the outgoing message.

Returns

Type	Description
Distribution	result

Type Parameters

Name	Description
Distribution	The type of the distribution over an item.

Remarks

The outgoing message is a distribution matching the moments of item as the random arguments are varied. The formula is proj[p(item) sum_(array) p(array) factor(item,array,index1,index2)]/p(item).

Exceptions

Type	Condition
ImproperMessageException	array is not a proper distribution.

ItemAverageLogarithm<Distribution>(IArray2D<Distribution>, Int32, Int32, Distribution)

VMP message to item.

Declaration

public static Distribution ItemAverageLogarithm<Distribution>([SkipIfAllUniform] IArray2D<Distribution> array, int index1, int index2, Distribution result)
    where Distribution : SettableTo<Distribution>

Parameters

Type	Name	Description
IArray2D<Distribution>	array	Incoming message from array. Must be a proper distribution. If all elements are uniform, the result will be uniform.
Int32	index1	Constant value for index1.
Int32	index2	Constant value for index2.
Distribution	result	Modified to contain the outgoing message.

Returns

Type	Description
Distribution	result

Type Parameters

Name	Description
Distribution	The type of the distribution over an item.

Remarks

The outgoing message is a distribution matching the moments of item as the random arguments are varied. The formula is proj[sum_(array) p(array) factor(item,array,index1,index2)].

Exceptions

Type	Condition
ImproperMessageException	array is not a proper distribution.

LogAverageFactor<Distribution>(Distribution, Distribution)

Evidence message for EP.

Declaration

public static double LogAverageFactor<Distribution>(Distribution item, Distribution to_item)
    where Distribution : CanGetLogAverageOf<Distribution>

Parameters

Type	Name	Description
Distribution	item	Incoming message from item.
Distribution	to_item	Outgoing message to item.

Returns

Type	Description
Double	Logarithm of the factor's average value across the given argument distributions.

Type Parameters

Name	Description
Distribution	The type of the distribution over an item.

Remarks

The formula for the result is log(sum_(item) p(item) factor(item,array,index1,index2)).

LogAverageFactor<Distribution>(T, IArray2D<Distribution>, Int32, Int32)

Evidence message for EP.

Declaration

public static double LogAverageFactor<Distribution>(T item, IArray2D<Distribution> array, int index1, int index2)
    where Distribution : CanGetLogProb<T>

Parameters

Type	Name	Description
T	item	Incoming message from item.
IArray2D<Distribution>	array	Incoming message from array.
Int32	index1	Constant value for index1.
Int32	index2	Constant value for index2.

Returns

Type	Description
Double	Logarithm of the factor's average value across the given argument distributions.

Type Parameters

Name	Description
Distribution	The type of the distribution over an item.

Remarks

The formula for the result is log(sum_(item,array) p(item,array) factor(item,array,index1,index2)).

LogEvidenceRatio<Distribution>(Distribution)

Evidence message for EP.

Declaration

public static double LogEvidenceRatio<Distribution>(Distribution item)

Parameters

Type	Name	Description
Distribution	item	Incoming message from item.

Returns

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

Type Parameters

Name	Description
Distribution	The type of the distribution over an item.

Remarks

The formula for the result is log(sum_(item) p(item) factor(item,array,index1,index2) / sum_item p(item) messageTo(item)). Adding up these values across all factors and variables gives the log-evidence estimate for EP.

LogEvidenceRatio<Distribution>(T, IArray2D<Distribution>, Int32, Int32)

Evidence message for EP.

Declaration

public static double LogEvidenceRatio<Distribution>(T item, IArray2D<Distribution> array, int index1, int index2)
    where Distribution : CanGetLogProb<T>

Parameters

Type	Name	Description
T	item	Incoming message from item.
IArray2D<Distribution>	array	Incoming message from array.
Int32	index1	Constant value for index1.
Int32	index2	Constant value for index2.

Returns

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

Type Parameters

Name	Description
Distribution	The type of the distribution over an item.

Remarks

The formula for the result is log(sum_(item,array) p(item,array) factor(item,array,index1,index2) / sum_item p(item) messageTo(item)). Adding up these values across all factors and variables gives the log-evidence estimate for EP.