|
|||||||||||
PREV CLASS NEXT CLASS | FRAMES NO FRAMES | ||||||||||
SUMMARY: NESTED | FIELD | CONSTR | METHOD | DETAIL: FIELD | CONSTR | METHOD |
java.lang.Object edu.ksu.cis.bnj.bbn.inference.Inference edu.ksu.cis.bnj.bbn.inference.ApproximateInference edu.ksu.cis.bnj.bbn.inference.approximate.sampling.MCMC edu.ksu.cis.bnj.bbn.inference.approximate.sampling.SelfImportance
Self-importance sampling, much like adaptive importance sampling, gets samples by instantiating each node of the network (using the likelihood of each state given the state of the parent nodes) in topological order, using the probabilities in the importance function (which are initially set equal to the conditional probability tables). In this sampling process, evidence nodes are automatically instantiated to the value given in the evidence file. After 10% of the samples have been taken, the approximate probabilities for those samples are used to update the importance function. This allows the probabilities for each node to change to reflect the values of the evidence nodes.
Field Summary | |
static int |
defaultUpdateIteration
|
static double |
defaultWeight
|
protected int |
updateIteration
|
protected double |
weight
|
Fields inherited from class edu.ksu.cis.bnj.bbn.inference.approximate.sampling.MCMC |
abort, abortTimer, defaultMaxIteration, generateSamples, listeners, maxIteration, NO_TIME_LIMIT, OPT_GENERATE_DATA, OPT_MAX_ITERATION, OPT_RUNNING_TIME_LIMIT, OPT_USE_MARKOV_BLANKET, random, runningTimeLimit, tuples, useMarkovBlanketScore |
Fields inherited from class edu.ksu.cis.bnj.bbn.inference.ApproximateInference |
rmseWriter |
Fields inherited from class edu.ksu.cis.bnj.bbn.inference.Inference |
graph, inferenceClassName, MAP, MARGINALS, marginalsResult, MPE, OPT_RUN_TYPE, options |
Fields inherited from interface edu.ksu.cis.kdd.util.gui.Optionable |
OPT_OUTPUT_FILE |
Constructor Summary | |
SelfImportance()
|
|
SelfImportance(BBNGraph g)
The constructor for SIS, which initializes the graph on which to run inference. |
|
SelfImportance(BBNGraph g,
int i)
An alternate constructor, which intializes the graph on which to run inference and the number of samples to use |
Method Summary | |
InferenceResult |
getMarginals()
Computes the SIS algorithm and calculates the posterior probabilities of the network |
java.lang.String |
getName()
Gets the name of the inference algorithm |
int |
getUpdateIteration()
Gets the number of samples taken before updating the ICPTs |
double |
getWeight()
Returns the weight of samples |
static void |
main(java.lang.String[] args)
|
void |
setUpdateIteration(int updateIteration)
Sets the updateIteration. |
void |
setWeight(double weight)
Sets the weight. |
Methods inherited from class edu.ksu.cis.bnj.bbn.inference.approximate.sampling.MCMC |
abort, addListener, cancelAbortTimer, generateData, generateData, getData, getDefaultOptions, getMarkovBlanketScore, getMaxIteration, getOptionsDialog, getRunningTimeLimit, isUseMarkovBlanketScore, removeListener, sendEvent, setAbortTimer, setMaxIteration, setRunningTimeLimit, setUseMarkovBlanketScore |
Methods inherited from class edu.ksu.cis.bnj.bbn.inference.ApproximateInference |
computeRMSE, getRMSEWriter, setRMSEfile |
Methods inherited from class edu.ksu.cis.bnj.bbn.inference.Inference |
execute, getCurrentOptions, getGraph, getMAP, getMarginalsResult, getMPE, getOutputFile, getRunType, load, setGraph, setOption, setOptions, setOutputFile, setRunType |
Methods inherited from class java.lang.Object |
clone, equals, finalize, getClass, hashCode, notify, notifyAll, toString, wait, wait, wait |
Methods inherited from interface edu.ksu.cis.bnj.bbn.datagen.DataGenerator |
generateData |
Field Detail |
public static final int defaultUpdateIteration
public static final double defaultWeight
protected int updateIteration
protected double weight
Constructor Detail |
public SelfImportance()
public SelfImportance(BBNGraph g)
g
- - the BBNGraph on which to run inferencepublic SelfImportance(BBNGraph g, int i)
g
- - the BBNGraph on which to run inferencei
- - the number of samples to generate during inferenceMethod Detail |
public java.lang.String getName()
getName
in interface DataGenerator
getName
in class Inference
public InferenceResult getMarginals()
getMarginals
in class Inference
edu.ksu.cis.bnj.bbn.inference.approximate.sampling.MCMC#getMarginalsImpl(boolean)
public int getUpdateIteration()
public double getWeight()
public void setUpdateIteration(int updateIteration)
updateIteration
- - the updateIteration to setpublic void setWeight(double weight)
weight
- - the weight to setpublic static void main(java.lang.String[] args)
|
|||||||||||
PREV CLASS NEXT CLASS | FRAMES NO FRAMES | ||||||||||
SUMMARY: NESTED | FIELD | CONSTR | METHOD | DETAIL: FIELD | CONSTR | METHOD |