Package jdistlib.inference

Class SparseSubsetRjSampler

java.lang.Object

jdistlib.inference.SparseSubsetRjSampler

public final class SparseSubsetRjSampler
extends Object

Allocation-conscious add/drop/swap RJMCMC for very large sparse candidate universes.

Constructor Summary

Constructors

Constructor	Description
SparseSubsetRjSampler(SparseCandidateProposal candidates, SparseCoefficientProposal coefficients, double initialWithinScale, double targetWithinAcceptance)
SparseSubsetRjSampler(SparseCandidateProposal candidates, SparseCoefficientProposal coefficients, double initialWithinScale, double targetWithinAcceptance, double[] moveWeights)

Method Summary

Modifier and Type	Method	Description
SparseSubsetResult	resume(SparseSubsetTarget target, SparseSubsetCheckpoint checkpoint, SparseSubsetSamplingOptions options)
SparseSubsetResult	sample(SparseSubsetTarget target, SparseSubsetState initialState, SparseSubsetSamplingOptions options, RandomEngine random)

Methods inherited from class java.lang.Object

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

Constructor Details

SparseSubsetRjSampler

public SparseSubsetRjSampler(SparseCandidateProposal candidates,
 SparseCoefficientProposal coefficients,
 double initialWithinScale,
 double targetWithinAcceptance)

SparseSubsetRjSampler

public SparseSubsetRjSampler(SparseCandidateProposal candidates,
 SparseCoefficientProposal coefficients,
 double initialWithinScale,
 double targetWithinAcceptance,
 double[] moveWeights)

Method Details

sample

public SparseSubsetResult sample(SparseSubsetTarget target,
 SparseSubsetState initialState,
 SparseSubsetSamplingOptions options,
 RandomEngine random)

resume

public SparseSubsetResult resume(SparseSubsetTarget target,
 SparseSubsetCheckpoint checkpoint,
 SparseSubsetSamplingOptions options)