Class LogisticOp

Provides outgoing messages for Logistic(Double), given random arguments to the function.

Inheritance

Object

LogisticOp

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), "Logistic", new Type[]{typeof(double)})]
[Quality(QualityBand.Stable)]
[Buffers(new string[]{"falseMsg"})]
public class LogisticOp

Methods

AverageLogFactor(Gaussian, Beta, Beta)

Evidence message for VMP.

Declaration

public static double AverageLogFactor(Gaussian x, Beta logistic, Beta to_logistic)

Parameters

Type	Name	Description
Gaussian	x	Incoming message from x. Must be a proper distribution. If uniform, the result will be uniform.
Beta	logistic	Incoming message from logistic.
Beta	to_logistic	Previous outgoing message to logistic.

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.

Exceptions

Type	Condition
ImproperMessageException	x is not a proper distribution.

AverageLogFactor(Double, Double)

Evidence message for VMP.

Declaration

public static double AverageLogFactor(double logistic, double x)

Parameters

Type	Name	Description
Double	logistic	Constant value for logistic.
Double	x	Constant value for x.

Returns

Type	Description
Double	Zero.

Remarks

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

FalseMsg(Beta, Gaussian, Gaussian)

Update the buffer falseMsg.

Declaration

public static Gaussian FalseMsg(Beta logistic, Gaussian x, Gaussian falseMsg)

Parameters

Type	Name	Description
Beta	logistic	Incoming message from logistic. Must be a proper distribution. If uniform, the result will be uniform.
Gaussian	x	Incoming message from x. Must be a proper distribution. If uniform, the result will be uniform.
Gaussian	falseMsg	Buffer falseMsg.

Returns

Type	Description
Gaussian	New value of buffer falseMsg.

Remarks

Exceptions

Type	Condition
ImproperMessageException	logistic is not a proper distribution.
ImproperMessageException	x is not a proper distribution.

FalseMsgInit()

Initialize the buffer falseMsg.

Declaration

public static Gaussian FalseMsgInit()

Returns

Type	Description
Gaussian	Initial value of buffer falseMsg.

Remarks

LogAverageFactor(Beta, Gaussian, Gaussian)

Evidence message for EP.

Declaration

public static double LogAverageFactor(Beta logistic, Gaussian x, Gaussian falseMsg)

Parameters

Type	Name	Description
Beta	logistic	Incoming message from logistic.
Gaussian	x	Incoming message from x.
Gaussian	falseMsg	Buffer falseMsg.

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_(logistic,x) p(logistic,x) factor(logistic,x)).

LogAverageFactor(Beta, Double)

Evidence message for EP.

Declaration

public static double LogAverageFactor(Beta logistic, double x)

Parameters

Type	Name	Description
Beta	logistic	Incoming message from logistic.
Double	x	Constant value for x.

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_(logistic) p(logistic) factor(logistic,x)).

LogAverageFactor(Double, Gaussian)

Evidence message for EP.

Declaration

public static double LogAverageFactor(double logistic, Gaussian x)

Parameters

Type	Name	Description
Double	logistic	Constant value for logistic.
Gaussian	x	Incoming message from x.

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_(x) p(x) factor(logistic,x)).

LogAverageFactor(Double, Double)

Evidence message for EP.

Declaration

public static double LogAverageFactor(double logistic, double x)

Parameters

Type	Name	Description
Double	logistic	Constant value for logistic.
Double	x	Constant value for x.

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(logistic,x)).

LogEvidenceRatio(Beta, Gaussian, Gaussian, Beta)

Evidence message for EP.

Declaration

public static double LogEvidenceRatio(Beta logistic, Gaussian x, Gaussian falseMsg, Beta to_logistic)

Parameters

Type	Name	Description
Beta	logistic	Incoming message from logistic.
Gaussian	x	Incoming message from x.
Gaussian	falseMsg	Buffer falseMsg.
Beta	to_logistic	Outgoing message to logistic.

Returns

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

Remarks

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

LogEvidenceRatio(Double, Gaussian)

Evidence message for EP.

Declaration

public static double LogEvidenceRatio(double logistic, Gaussian x)

Parameters

Type	Name	Description
Double	logistic	Constant value for logistic.
Gaussian	x	Incoming message from x.

Returns

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

Remarks

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

LogEvidenceRatio(Double, Double)

Evidence message for EP.

Declaration

public static double LogEvidenceRatio(double logistic, double x)

Parameters

Type	Name	Description
Double	logistic	Constant value for logistic.
Double	x	Constant value for x.

Returns

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

Remarks

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

LogisticAverageConditional(Beta, Gaussian, Gaussian, Gaussian)

EP message to logistic.

Declaration

public static Beta LogisticAverageConditional(Beta logistic, Gaussian x, Gaussian falseMsg, Gaussian to_x)

Parameters

Type	Name	Description
Beta	logistic	Incoming message from logistic.
Gaussian	x	Incoming message from x. Must be a proper distribution. If uniform, the result will be uniform.
Gaussian	falseMsg	Buffer falseMsg.
Gaussian	to_x	Previous outgoing message to x.

Returns

Type	Description
Beta	The outgoing EP message to the logistic argument.

Remarks

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

Exceptions

Type	Condition
ImproperMessageException	x is not a proper distribution.

LogisticAverageConditionalInit()

Declaration

public static Beta LogisticAverageConditionalInit()

Returns

Type	Description
Beta

Remarks

LogisticAverageLogarithm(Gaussian)

VMP message to logistic.

Declaration

public static Beta LogisticAverageLogarithm(Gaussian x)

Parameters

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

Returns

Type	Description
Beta	The outgoing VMP message to the logistic argument.

Remarks

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

Exceptions

Type	Condition
ImproperMessageException	x is not a proper distribution.

XAverageConditional(Beta, Gaussian)

EP message to x.

Declaration

public static Gaussian XAverageConditional(Beta logistic, Gaussian falseMsg)

Parameters

Type	Name	Description
Beta	logistic	Incoming message from logistic. Must be a proper distribution. If uniform, the result will be uniform.
Gaussian	falseMsg	Buffer falseMsg.

Returns

Type	Description
Gaussian	The outgoing EP message to the x argument.

Remarks

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

Exceptions

Type	Condition
ImproperMessageException	logistic is not a proper distribution.

XAverageConditional(Double)

EP message to x.

Declaration

public static Gaussian XAverageConditional(double logistic)

Parameters

Type	Name	Description
Double	logistic	Constant value for logistic.

Returns

Type	Description
Gaussian	The outgoing EP message to the x argument.

Remarks

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

XAverageLogarithm(Beta, Gaussian, Gaussian)

VMP message to x.

Declaration

public static Gaussian XAverageLogarithm(Beta logistic, Gaussian x, Gaussian to_X)

Parameters

Type	Name	Description
Beta	logistic	Incoming message from logistic. Must be a proper distribution. If uniform, the result will be uniform.
Gaussian	x	Incoming message from x. Must be a proper distribution. If uniform, the result will be uniform.
Gaussian	to_X	Previous outgoing message to X.

Returns

Type	Description
Gaussian	The outgoing VMP message to the x argument.

Remarks

The outgoing message is the factor viewed as a function of x with logistic integrated out. The formula is sum_logistic p(logistic) factor(logistic,x).

Exceptions

Type	Condition
ImproperMessageException	logistic is not a proper distribution.
ImproperMessageException	x is not a proper distribution.

XAverageLogarithm(Double)

VMP message to x.

Declaration

public static Gaussian XAverageLogarithm(double logistic)

Parameters

Type	Name	Description
Double	logistic	Constant value for logistic.

Returns

Type	Description
Gaussian	The outgoing VMP message to the x argument.

Remarks

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