io.improbable.keanu.algorithms.mcmc

Class MetropolisHastingsStep

java.lang.Object

io.improbable.keanu.algorithms.mcmc.MetropolisHastingsStep

public class MetropolisHastingsStep
extends java.lang.Object

Method Summary

Modifier and Type	Method and Description
io.improbable.keanu.algorithms.mcmc.MetropolisHastingsStep.StepResult	step(java.util.Set<Variable> chosenVariables, double logProbabilityBeforeStep)
io.improbable.keanu.algorithms.mcmc.MetropolisHastingsStep.StepResult	step(java.util.Set<Variable> chosenVariables, double logProbabilityBeforeStep, double temperature)

Methods inherited from class java.lang.Object

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

Method Detail

step

public io.improbable.keanu.algorithms.mcmc.MetropolisHastingsStep.StepResult step(java.util.Set<Variable> chosenVariables,
                                                                                  double logProbabilityBeforeStep)

step

public io.improbable.keanu.algorithms.mcmc.MetropolisHastingsStep.StepResult step(java.util.Set<Variable> chosenVariables,
                                                                                  double logProbabilityBeforeStep,
                                                                                  double temperature)

Parameters:

chosenVariables - variables to get a proposed change for

logProbabilityBeforeStep - The log of the previous state's probability

temperature - Temperature for simulated annealing. This should be constant if no annealing is wanted

Returns:

the log probability of the network after either accepting or rejecting the sample