public class NUTS
extends java.lang.Object
Modifier and Type | Method and Description |
---|---|
static NetworkSamples |
getPosteriorSamples(BayesianNetwork bayesNet,
java.util.List<? extends Vertex> sampleFromVertices,
int sampleCount,
int adaptCount,
double targetAcceptanceProb,
KeanuRandom random)
Sample from the posterior of a Bayesian Network using the No-U-Turn-Sampling algorithm
|
static NetworkSamples |
getPosteriorSamples(BayesianNetwork bayesNet,
java.util.List<DoubleVertex> sampleFromVertices,
int sampleCount,
int adaptCount,
double targetAcceptanceProb)
Sample from the posterior of a Bayesian Network using the No-U-Turn-Sampling algorithm
|
public static NetworkSamples getPosteriorSamples(BayesianNetwork bayesNet, java.util.List<DoubleVertex> sampleFromVertices, int sampleCount, int adaptCount, double targetAcceptanceProb)
bayesNet
- the bayesian network to sample fromsampleFromVertices
- the vertices to sample fromsampleCount
- the number of samples to takeadaptCount
- the number of samples for which the stepsize will be tuned. For the remaining samples
in which it is not tuned, the stepsize will be frozen to its last calculated valuetargetAcceptanceProb
- the target acceptance probability, a suggested value of this is 0.65,
Beskos et al., 2010; Neal, 2011public static NetworkSamples getPosteriorSamples(BayesianNetwork bayesNet, java.util.List<? extends Vertex> sampleFromVertices, int sampleCount, int adaptCount, double targetAcceptanceProb, KeanuRandom random)
bayesNet
- the bayesian network to sample fromsampleFromVertices
- the vertices inside the bayesNet to sample fromsampleCount
- the number of samples to takeadaptCount
- the number of samples for which the stepsize will be tuned. For the remaining samples
in which it is not tuned, the stepsize will be frozen to its last calculated valuetargetAcceptanceProb
- the target acceptance probability, a suggested value of this is 0.65,
Beskos et al., 2010; Neal, 2011random
- the source of randomness