io.improbable.keanu.vertices.intgr.probabilistic

Class PoissonVertex

java.lang.Object

io.improbable.keanu.vertices.Vertex<IntegerTensor>

io.improbable.keanu.vertices.intgr.IntegerVertex

io.improbable.keanu.vertices.intgr.probabilistic.PoissonVertex

All Implemented Interfaces:

Variable<IntegerTensor,VertexState<IntegerTensor>>, HasShape, io.improbable.keanu.kotlin.IntegerOperators<IntegerVertex>, io.improbable.keanu.kotlin.NumberOperators<IntegerVertex>, ProbabilisticInteger, LogProbGraphSupplier, Observable<IntegerTensor>, Probabilistic<IntegerTensor>, Samplable<IntegerTensor>, SamplableWithManyScalars<IntegerTensor>, SamplableWithShape<IntegerTensor>

public class PoissonVertex
extends IntegerVertex
implements ProbabilisticInteger, SamplableWithManyScalars<IntegerTensor>, LogProbGraphSupplier

Constructor Summary

Constructors

Constructor and Description
PoissonVertex(double mu)
PoissonVertex(DoubleVertex mu) One to one constructor for mapping some shape of mu to a matching shaped Poisson.
PoissonVertex(long[] shape, double mu)
PoissonVertex(long[] shape, DoubleVertex mu) One mu that must match a proposed tensor shape of Poisson.
PoissonVertex(Vertex<? extends NumberTensor> mu)

Method Summary

Modifier and Type	Method and Description
java.util.Map<Vertex,DoubleTensor>	dLogProb(IntegerTensor value, java.util.Set<? extends Vertex> withRespectTo) The partial derivatives of the natural log prob.
DoubleVertex	getMu()
double	logProb(IntegerTensor value) This is the natural log of the probability at the supplied value.
LogProbGraph	logProbGraph()
IntegerTensor	sampleWithShape(long[] shape, KeanuRandom random)

Methods inherited from class io.improbable.keanu.vertices.intgr.IntegerVertex

abs, concat, div, div, divideBy, divideBy, divideBy, equalTo, getValue, greaterThan, greaterThanOrEqualTo, lambda, lambda, lessThan, lessThanOrEqualTo, loadValue, max, min, minus, minus, minus, multiply, multiply, multiply, notEqualTo, observe, observe, plus, plus, plus, pow, pow, reshape, reverseDiv, reverseMinus, saveValue, setAndCascade, setAndCascade, setValue, setValue, slice, sum, sum, take, times, times, toDouble, unaryMinus

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

addChild, addParent, addParents, equals, eval, getChildren, getConnectedGraph, getDegree, getId, getIndentation, getLabel, getObservedValue, getParents, getRank, getReference, getShape, getState, getValue, hashCode, hasValue, isDifferentiable, isObserved, isProbabilistic, lazyEval, observe, observeOwnValue, print, print, removeLabel, save, setAndCascade, setLabel, setLabel, setParents, setParents, setState, setValue, toString, unobserve

Methods inherited from class java.lang.Object

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

Methods inherited from interface io.improbable.keanu.vertices.intgr.probabilistic.ProbabilisticInteger

dLogPmf, dLogPmf, dLogPmf, logPmf, logPmf, logPmf

Methods inherited from interface io.improbable.keanu.vertices.Probabilistic

dLogProb, dLogProbAtValue, dLogProbAtValue, getValue, keepOnlyProbabilisticVertices, logProbAtValue

Methods inherited from interface io.improbable.keanu.vertices.Observable

getObservedValue, isObserved, observe, unobserve

Methods inherited from interface io.improbable.keanu.vertices.Samplable

sample, sample

Methods inherited from interface io.improbable.keanu.vertices.SamplableWithManyScalars

sample, sampleManyScalars, sampleManyScalars

Methods inherited from interface io.improbable.keanu.vertices.SamplableWithShape

sampleWithShape

Methods inherited from interface io.improbable.keanu.algorithms.variational.optimizer.HasShape

getShape

Constructor Detail

PoissonVertex

public PoissonVertex(long[] shape,
                     DoubleVertex mu)

One mu that must match a proposed tensor shape of Poisson.

If all provided parameters are scalar then the proposed shape determines the shape

Parameters:

shape - the desired shape of the vertex

mu - the mu of the Poisson with either the same shape as specified for this vertex or a scalar

PoissonVertex

public PoissonVertex(long[] shape,
                     double mu)

PoissonVertex

public PoissonVertex(DoubleVertex mu)

One to one constructor for mapping some shape of mu to a matching shaped Poisson.

Parameters:

mu - mu with same shape as desired Poisson tensor or scalar

PoissonVertex

public PoissonVertex(Vertex<? extends NumberTensor> mu)

PoissonVertex

public PoissonVertex(double mu)

Method Detail

getMu

public DoubleVertex getMu()

logProb

public double logProb(IntegerTensor value)

Description copied from interface: Probabilistic

This is the natural log of the probability at the supplied value. In the case of continuous vertices, this is actually the log of the density, which will differ from the probability by a constant.

Specified by:

logProb in interface Probabilistic<IntegerTensor>

Parameters:

value - The supplied value.

Returns:

The natural log of the probability function at the supplied value. For continuous variables this is called the PDF (probability density function). For discrete variables this is called the PMF (probability mass function).

logProbGraph

public LogProbGraph logProbGraph()

Specified by:

logProbGraph in interface LogProbGraphSupplier

dLogProb

public java.util.Map<Vertex,DoubleTensor> dLogProb(IntegerTensor value,
                                                   java.util.Set<? extends Vertex> withRespectTo)

Description copied from interface: Probabilistic

The partial derivatives of the natural log prob.

Specified by:

dLogProb in interface Probabilistic<IntegerTensor>

Parameters:

value - at a given value

withRespectTo - list of parents to differentiate with respect to

Returns:

the partial derivatives of the log of the probability function at the supplied value. For continuous variables this is called the PDF (probability density function). For discrete variables this is called the PMF (probability mass function).

sampleWithShape

public IntegerTensor sampleWithShape(long[] shape,
                                     KeanuRandom random)

Specified by:

sampleWithShape in interface SamplableWithShape<IntegerTensor>