io.improbable.keanu.algorithms.mcmc

Class MetropolisHastings.Sampler

java.lang.Object

io.improbable.keanu.algorithms.mcmc.MetropolisHastings.Sampler

All Implemented Interfaces:

SamplingAlgorithm

Enclosing class:

MetropolisHastings

public static class MetropolisHastings.Sampler
extends java.lang.Object
implements SamplingAlgorithm

Constructor Summary

Constructors

Constructor and Description
Sampler(java.util.List<Vertex> latentVertices, java.util.List<? extends Vertex> verticesToSampleFrom, io.improbable.keanu.algorithms.mcmc.MetropolisHastingsStep mhStep, MHStepVariableSelector variableSelector, double logProbabilityBeforeStep)

Method Summary

Modifier and Type	Method and Description
NetworkState	sample()
void	sample(java.util.Map<java.lang.Long,java.util.List<?>> samplesByVertex) Takes a sample with the algorithm and saves it in the supplied map
void	step() Same effect as a sample but the result isn't saved or returned.

Methods inherited from class java.lang.Object

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

Constructor Detail

Sampler

public Sampler(java.util.List<Vertex> latentVertices,
               java.util.List<? extends Vertex> verticesToSampleFrom,
               io.improbable.keanu.algorithms.mcmc.MetropolisHastingsStep mhStep,
               MHStepVariableSelector variableSelector,
               double logProbabilityBeforeStep)

Method Detail

step

public void step()

Description copied from interface: SamplingAlgorithm

Same effect as a sample but the result isn't saved or returned.

Specified by:

step in interface SamplingAlgorithm

sample

public void sample(java.util.Map<java.lang.Long,java.util.List<?>> samplesByVertex)

Description copied from interface: SamplingAlgorithm

Takes a sample with the algorithm and saves it in the supplied map

Specified by:

sample in interface SamplingAlgorithm

Parameters:

samplesByVertex - map to store sampled vertex values

sample

public NetworkState sample()

Specified by:

sample in interface SamplingAlgorithm

Returns:

a network state that represents the value of vertices at the end of the algorithm step