io.improbable.keanu.vertices.dbl.probabilistic

Class BetaVertex

java.lang.Object

io.improbable.keanu.vertices.Vertex<java.lang.Double>

io.improbable.keanu.vertices.dbl.DoubleVertex

io.improbable.keanu.vertices.dbl.probabilistic.ProbabilisticDouble

io.improbable.keanu.vertices.dbl.probabilistic.BetaVertex

All Implemented Interfaces:

Identifiable, io.improbable.keanu.kotlin.DoubleOperators<DoubleVertex>, io.improbable.keanu.kotlin.Operators<DoubleVertex>

public class BetaVertex
extends ProbabilisticDouble

Field Summary

Fields inherited from class io.improbable.keanu.vertices.Vertex

idGenerator

Constructor Summary

Constructors

Constructor and Description
BetaVertex(DoubleVertex alpha, DoubleVertex beta)
BetaVertex(DoubleVertex alpha, DoubleVertex beta, java.util.Random random)

Method Summary

Modifier and Type	Method and Description
java.util.Map<java.lang.String,java.lang.Double>	convertDualNumbersToDiff(double dPdAlpha, double dPdBeta, double dPdx)
java.util.Map<java.lang.String,java.lang.Double>	dDensityAtValue() This returns the derivative of the density function with respect to any dependent vertices.
double	density(java.lang.Double value) This is the value of the probability density at the supplied value.
java.util.Map<java.lang.String,java.lang.Double>	dlnDensityAtValue() This is the same as dDensityAtValue except for the log of the density.
double	logDensity(java.lang.Double value) This is the value of the natural log of the probability density at the supplied value.
java.lang.Double	sample()

Methods inherited from class io.improbable.keanu.vertices.dbl.probabilistic.ProbabilisticDouble

getDualNumber, isProbabilistic, lazyEval, updateValue

Methods inherited from class io.improbable.keanu.vertices.dbl.DoubleVertex

abs, div, div, divideBy, divideBy, divideBy, lambda, minus, minus, minus, multiply, multiply, multiply, plus, plus, plus, pow, times, times, unaryMinus

Methods inherited from class io.improbable.keanu.vertices.Vertex

addChild, addParent, addParents, densityAtValue, equals, getChildren, getConnectedGraph, getId, getParents, getValue, hashCode, isObserved, logDensityAtValue, observe, observeOwnValue, setAndCascade, setParents, setParents, setValue, unobserve, updateChildren

Methods inherited from class java.lang.Object

clone, finalize, getClass, notify, notifyAll, toString, wait, wait, wait

Constructor Detail

BetaVertex

public BetaVertex(DoubleVertex alpha,
                  DoubleVertex beta,
                  java.util.Random random)

BetaVertex

public BetaVertex(DoubleVertex alpha,
                  DoubleVertex beta)

Method Detail

density

public double density(java.lang.Double value)

Description copied from class: Vertex

This is the value of the probability density at the supplied value.

Specified by:

density in class Vertex<java.lang.Double>

Parameters:

value - The supplied value.

Returns:

The probability.

logDensity

public double logDensity(java.lang.Double value)

Description copied from class: Vertex

This is the value of the natural log of the probability density at the supplied value.

Overrides:

logDensity in class Vertex<java.lang.Double>

Parameters:

value - The supplied value.

Returns:

The probability.

dDensityAtValue

public java.util.Map<java.lang.String,java.lang.Double> dDensityAtValue()

Description copied from class: Vertex

This returns the derivative of the density function with respect to any dependent vertices.

Specified by:

dDensityAtValue in class Vertex<java.lang.Double>

Returns:

a Map containing { dependent vertex Id -> density slope w.r.t. dependent vertex}

dlnDensityAtValue

public java.util.Map<java.lang.String,java.lang.Double> dlnDensityAtValue()

Description copied from class: Vertex

This is the same as dDensityAtValue except for the log of the density. For numerical stability a vertex may chose to override this method but if not overridden, the chain rule is used to calculate the derivative of the log of the density.

dlog(P)/dx = (dP/dx)*(1/P(x))

Overrides:

dlnDensityAtValue in class Vertex<java.lang.Double>

convertDualNumbersToDiff

public java.util.Map<java.lang.String,java.lang.Double> convertDualNumbersToDiff(double dPdAlpha,
                                                                                 double dPdBeta,
                                                                                 double dPdx)

sample

public java.lang.Double sample()

Specified by:

sample in class Vertex<java.lang.Double>

Returns:

a sample from the vertex's distribution. For non-probabilistic vertices, this will always be the same value.