Class IndexOfMaximumOp

Provides outgoing messages for IndexOfMaximumDouble(IList<Double>), given random arguments to the function.

Inheritance

Object

IndexOfMaximumOp

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(MMath), "IndexOfMaximumDouble", new Type[]{})]
[Quality(QualityBand.Experimental)]
[Buffers(new string[]{"Buffer"})]
public static class IndexOfMaximumOp

Methods

Buffer<GaussianList>(IndexOfMaximumBuffer, GaussianList, Int32)

Update the buffer Buffer.

Declaration

public static IndexOfMaximumBuffer Buffer<GaussianList>(IndexOfMaximumBuffer Buffer, GaussianList list, int IndexOfMaximumDouble)
    where GaussianList : IList<Gaussian>

Parameters

Type	Name	Description
IndexOfMaximumBuffer	Buffer	Buffer Buffer.
GaussianList	list	Incoming message from list.
Int32	IndexOfMaximumDouble	Constant value for indexOfMaximumDouble.

Returns

Type	Description
IndexOfMaximumBuffer	New value of buffer Buffer.

Type Parameters

Name	Description
GaussianList	The type of an incoming message from list.

Remarks

BufferInit<GaussianList>(GaussianList)

Initialize the buffer Buffer.

Declaration

public static IndexOfMaximumBuffer BufferInit<GaussianList>(GaussianList list)
    where GaussianList : IList<Gaussian>

Parameters

Type	Name	Description
GaussianList	list	Incoming message from list.

Returns

Type	Description
IndexOfMaximumBuffer	Initial value of buffer Buffer.

Type Parameters

Name	Description
GaussianList	The type of an incoming message from list.

Remarks

listAverageConditional<GaussianList>(IndexOfMaximumBuffer, GaussianList, Int32)

EP message to list.

Declaration

public static GaussianList listAverageConditional<GaussianList>(IndexOfMaximumBuffer Buffer, GaussianList to_list, int IndexOfMaximumDouble)
    where GaussianList : IList<Gaussian>

Parameters

Type	Name	Description
IndexOfMaximumBuffer	Buffer	Buffer Buffer.
GaussianList	to_list	Previous outgoing message to list.
Int32	IndexOfMaximumDouble	Constant value for indexOfMaximumDouble.

Returns

Type	Description
GaussianList	The outgoing EP message to the list argument.

Type Parameters

Name	Description
GaussianList	The type of an incoming message from list.

Remarks

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

LogAverageFactor<GaussianList>(IndexOfMaximumBuffer, GaussianList, Int32)

Evidence message for EP.

Declaration

public static double LogAverageFactor<GaussianList>(IndexOfMaximumBuffer Buffer, GaussianList list, int IndexOfMaximumDouble)
    where GaussianList : IList<Gaussian>

Parameters

Type	Name	Description
IndexOfMaximumBuffer	Buffer	Buffer Buffer.
GaussianList	list	Incoming message from list.
Int32	IndexOfMaximumDouble	Constant value for indexOfMaximumDouble.

Returns

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

Type Parameters

Name	Description
GaussianList	The type of an incoming message from list.

Remarks

The formula for the result is log(sum_(list) p(list) factor(indexOfMaximumDouble,list)).

LogEvidenceRatio<GaussianList>(IndexOfMaximumBuffer, GaussianList, Int32)

Evidence message for EP.

Declaration

public static double LogEvidenceRatio<GaussianList>(IndexOfMaximumBuffer Buffer, GaussianList list, int IndexOfMaximumDouble)
    where GaussianList : IList<Gaussian>

Parameters

Type	Name	Description
IndexOfMaximumBuffer	Buffer	Buffer Buffer.
GaussianList	list	Incoming message from list.
Int32	IndexOfMaximumDouble	Constant value for indexOfMaximumDouble.

Returns

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

Type Parameters

Name	Description
GaussianList	The type of an incoming message from list.

Remarks

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