[FactorMethod(typeof(Factor), "GammaFromShapeAndRate", new Type[]{}, Default = false)]
[Buffers(new string[]{"Q"})]
[Quality(QualityBand.Experimental)]
public class GammaFromShapeAndRateOp_Laplace : GammaFromShapeAndRateOpBase

Methods

LogAverageFactor(Gamma, Double, Gamma, Gamma)

Evidence message for EP.

Declaration

public static double LogAverageFactor(Gamma sample, double shape, Gamma rate, Gamma q)

Parameters

Type	Name	Description
Gamma	sample	Incoming message from `sample`.
Double	shape	Constant value for `shape`.
Gamma	rate	Incoming message from `rate`.
Gamma	q	Buffer `q`.

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_(sample,rate) p(sample,rate) factor(sample,shape,rate)).

LogEvidenceRatio(Gamma, Double, Gamma, Gamma, Gamma)

Evidence message for EP.

Declaration

public static double LogEvidenceRatio(Gamma sample, double shape, Gamma rate, Gamma to_sample, Gamma q)

Parameters

Type	Name	Description
Gamma	sample	Incoming message from `sample`.
Double	shape	Constant value for `shape`.
Gamma	rate	Incoming message from `rate`.
Gamma	to_sample	Previous outgoing message to `sample`.
Gamma	q	Buffer `q`.

Returns

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

Remarks

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

Q(Gamma, Double, Gamma)

Update the buffer Q.

Declaration

public static Gamma Q(Gamma sample, double shape, Gamma rate)

Parameters

Type	Name	Description
Gamma	sample	Incoming message from `sample`.
Double	shape	Constant value for `shape`.
Gamma	rate	Incoming message from `rate`. Must be a proper distribution. If uniform, the result will be uniform.

Returns

Type	Description
Gamma	New value of buffer `Q`.

Remarks

Exceptions

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

QInit()

Initialize the buffer Q.

Declaration

public static Gamma QInit()

Returns

Type	Description
Gamma	Initial value of buffer `Q`.

Remarks

RateAverageConditional(Gamma, Double, Gamma, Gamma)

EP message to rate.

Declaration

public static Gamma RateAverageConditional(Gamma sample, double shape, Gamma rate, Gamma q)

Parameters

Type	Name	Description
Gamma	sample	Incoming message from `sample`. Must be a proper distribution. If uniform, the result will be uniform.
Double	shape	Constant value for `shape`.
Gamma	rate	Incoming message from `rate`. Must be a proper distribution. If uniform, the result will be uniform.
Gamma	q	Buffer `q`.

Returns

Type	Description
Gamma	The outgoing EP message to the `rate` argument.

Remarks

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

Exceptions

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

SampleAverageConditional(Gamma, Double, Gamma, Gamma)

EP message to sample.

Declaration

public static Gamma SampleAverageConditional(Gamma sample, double shape, Gamma rate, Gamma q)

Parameters

Type	Name	Description
Gamma	sample	Incoming message from `sample`.
Double	shape	Constant value for `shape`.
Gamma	rate	Incoming message from `rate`. Must be a proper distribution. If uniform, the result will be uniform.
Gamma	q	Buffer `q`.

Returns

Type	Description
Gamma	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_(rate) p(rate) factor(sample,shape,rate)]/p(sample).

Exceptions

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