org.apache.ignite.ml.clustering.gmm

Class GmmTrainer

java.lang.Object

org.apache.ignite.ml.trainers.DatasetTrainer<M,Double>

org.apache.ignite.ml.trainers.SingleLabelDatasetTrainer<GmmModel>

org.apache.ignite.ml.clustering.gmm.GmmTrainer

public class GmmTrainer
extends SingleLabelDatasetTrainer<GmmModel>

Traner for GMM model.

Nested Class Summary

Nested classes/interfaces inherited from class org.apache.ignite.ml.trainers.DatasetTrainer

DatasetTrainer.EmptyDatasetException

Field Summary

Fields inherited from class org.apache.ignite.ml.trainers.DatasetTrainer

envBuilder, environment

Constructor Summary

Constructors

Constructor and Description
GmmTrainer() Creates an instance of GmmTrainer.
GmmTrainer(int countOfComponents) Creates an instance of GmmTrainer.
GmmTrainer(int countOfComponents, int maxCountOfIterations) Creates an instance of GmmTrainer.

Method Summary

Modifier and Type	Method and Description
<K,V> GmmModel	fitWithInitializedDeployingContext(DatasetBuilder<K,V> datasetBuilder, Preprocessor<K,V> extractor) Trains model based on the specified data.
boolean	isUpdateable(GmmModel mdl)
protected <K,V> GmmModel	updateModel(GmmModel mdl, DatasetBuilder<K,V> datasetBuilder, Preprocessor<K,V> extractor) Trains new model taken previous one as a first approximation.
GmmTrainer	withEps(double eps) Sets min divergence beween iterations.
GmmTrainer	withInitialCountOfComponents(int numberOfComponents) Sets numberOfComponents.
GmmTrainer	withInitialMeans(List<Vector> means) Sets initial means.
GmmTrainer	withMaxCountIterations(int maxCountOfIterations) Sets max count of iterations
GmmTrainer	withMaxCountOfClusters(int maxCountOfClusters) Sets maximum number of clusters in GMM.
GmmTrainer	withMaxCountOfInitTries(int maxCountOfInitTries) Sets MaxCountOfInitTries parameter.
GmmTrainer	withMaxLikelihoodDivergence(double maxLikelihoodDivergence) Sets maximum divergence between maximum of likelihood of vector in dataset and other for anomalies identification.
GmmTrainer	withMinClusterProbability(double minClusterProbability) Sets minimum requred probability for cluster.
GmmTrainer	withMinElementsForNewCluster(int minElementsForNewCluster) Sets minimum required anomalies in terms of maxLikelihoodDivergence for creating new cluster.

Methods inherited from class org.apache.ignite.ml.trainers.DatasetTrainer

fit, fit, fit, fit, fit, fit, getLastTrainedModelOrThrowEmptyDatasetException, identityTrainer, learningEnvironment, update, update, update, update, update, withConvertedLabels, withEnvironmentBuilder

Methods inherited from class java.lang.Object

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

Constructor Detail

GmmTrainer

public GmmTrainer()

Creates an instance of GmmTrainer.

GmmTrainer

public GmmTrainer(int countOfComponents,
                  int maxCountOfIterations)

Creates an instance of GmmTrainer.

Parameters:

countOfComponents - Count of components.

maxCountOfIterations - Max count of iterations.

GmmTrainer

public GmmTrainer(int countOfComponents)

Creates an instance of GmmTrainer.

Parameters:

countOfComponents - Count of components.

Method Detail

fitWithInitializedDeployingContext

public <K,V> GmmModel fitWithInitializedDeployingContext(DatasetBuilder<K,V> datasetBuilder,
                                                         Preprocessor<K,V> extractor)

Trains model based on the specified data.

Specified by:

fitWithInitializedDeployingContext in class DatasetTrainer<GmmModel,Double>

Type Parameters:

K - Type of a key in upstream data.

V - Type of a value in upstream data.

Parameters:

datasetBuilder - Dataset builder.

extractor - Extractor of UpstreamEntry into LabeledVector.

Returns:

Model.

withInitialCountOfComponents

public GmmTrainer withInitialCountOfComponents(int numberOfComponents)

Sets numberOfComponents.

Parameters:

numberOfComponents - Number of components.

Returns:

Trainer.

withInitialMeans

public GmmTrainer withInitialMeans(List<Vector> means)

Sets initial means.

Parameters:

means - Initial means for clusters.

Returns:

Trainer.

withMaxCountIterations

public GmmTrainer withMaxCountIterations(int maxCountOfIterations)

Sets max count of iterations

Parameters:

maxCountOfIterations - Max count of iterations.

Returns:

Trainer.

withEps

public GmmTrainer withEps(double eps)

Sets min divergence beween iterations.

Parameters:

eps - Eps.

Returns:

Trainer.

withMaxCountOfInitTries

public GmmTrainer withMaxCountOfInitTries(int maxCountOfInitTries)

Sets MaxCountOfInitTries parameter. If means initialization were unsuccessfull then algorithm try to reinitialize means randomly MaxCountOfInitTries times.

Parameters:

maxCountOfInitTries - Max count of init tries.

Returns:

Trainer.

withMaxCountOfClusters

public GmmTrainer withMaxCountOfClusters(int maxCountOfClusters)

Sets maximum number of clusters in GMM.

Parameters:

maxCountOfClusters - Max count of clusters.

Returns:

Trainer.

withMaxLikelihoodDivergence

public GmmTrainer withMaxLikelihoodDivergence(double maxLikelihoodDivergence)

Sets maximum divergence between maximum of likelihood of vector in dataset and other for anomalies identification.

Parameters:

maxLikelihoodDivergence - Max likelihood divergence.

Returns:

Trainer.

withMinElementsForNewCluster

public GmmTrainer withMinElementsForNewCluster(int minElementsForNewCluster)

Sets minimum required anomalies in terms of maxLikelihoodDivergence for creating new cluster.

Parameters:

minElementsForNewCluster - Min elements for new cluster.

Returns:

Trainer.

withMinClusterProbability

public GmmTrainer withMinClusterProbability(double minClusterProbability)

Sets minimum requred probability for cluster. If cluster has probability value less than this value then this cluster will be eliminated.

Parameters:

minClusterProbability - Min cluster probability.

Returns:

Trainer.

isUpdateable

public boolean isUpdateable(GmmModel mdl)

Specified by:

isUpdateable in class DatasetTrainer<GmmModel,Double>

Parameters:

mdl - Model.

Returns:

True if current critical for training parameters correspond to parameters from last training.

updateModel

protected <K,V> GmmModel updateModel(GmmModel mdl,
                                     DatasetBuilder<K,V> datasetBuilder,
                                     Preprocessor<K,V> extractor)

Trains new model taken previous one as a first approximation.

Specified by:

updateModel in class DatasetTrainer<GmmModel,Double>

Type Parameters:

K - Type of a key in upstream data.

V - Type of a value in upstream data.

Parameters:

mdl - Learned model.

datasetBuilder - Dataset builder.

extractor - Extractor of UpstreamEntry into LabeledVector.

Returns:

Updated model.