org.apache.ignite.ml.composition.bagging

Class BaggedTrainer<L>

java.lang.Object

org.apache.ignite.ml.trainers.DatasetTrainer<BaggedModel,L>

org.apache.ignite.ml.composition.bagging.BaggedTrainer<L>

Type Parameters:

L - Type of labels.

public class BaggedTrainer<L>
extends DatasetTrainer<BaggedModel,L>

Trainer encapsulating logic of bootstrap aggregating (bagging). This trainer accepts some other trainer and returns bagged version of it. Resulting model consists of submodels results of which are aggregated by a specified aggregator.

Bagging is done on both samples and features (Samples bagging, Features bagging).

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
BaggedTrainer(DatasetTrainer<? extends IgniteModel,L> tr, PredictionsAggregator aggregator, int ensembleSize, double subsampleRatio, int featuresVectorSize, int featureSubspaceDim) Construct instance of this class with given parameters.

Method Summary

Modifier and Type	Method and Description
<K,V> BaggedModel	fitWithInitializedDeployingContext(DatasetBuilder<K,V> datasetBuilder, Preprocessor<K,V> preprocessor) Trains model based on the specified data.
static int[]	getMapping(int featuresVectorSize, int maximumFeaturesCntPerMdl, long seed) Get mapping R^featuresVectorSize -> R^maximumFeaturesCntPerMdl.
boolean	isUpdateable(BaggedModel mdl) This method is never called, instead of constructing logic of update from DatasetTrainer.isUpdateable(M) and DatasetTrainer.updateModel(M, org.apache.ignite.ml.dataset.DatasetBuilder<K, V>, org.apache.ignite.ml.preprocessing.Preprocessor<K, V>) in this class we explicitly override update method.
<K,V> BaggedModel	update(BaggedModel mdl, DatasetBuilder<K,V> datasetBuilder, Preprocessor<K,V> preprocessor) Gets state of model in arguments, compare it with training parameters of trainer and if they are fit then trainer updates model in according to new data and return new model.
protected <K,V> BaggedModel	updateModel(BaggedModel mdl, DatasetBuilder<K,V> datasetBuilder, Preprocessor<K,V> preprocessor) This method is never called, instead of constructing logic of update from DatasetTrainer.isUpdateable(M) and DatasetTrainer.updateModel(M, org.apache.ignite.ml.dataset.DatasetBuilder<K, V>, org.apache.ignite.ml.preprocessing.Preprocessor<K, V>) in this class we explicitly override update method.
BaggedTrainer<L>	withEnvironmentBuilder(LearningEnvironmentBuilder envBuilder) Changes learning Environment.

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

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

Methods inherited from class java.lang.Object

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

Constructor Detail

BaggedTrainer

public BaggedTrainer(DatasetTrainer<? extends IgniteModel,L> tr,
                     PredictionsAggregator aggregator,
                     int ensembleSize,
                     double subsampleRatio,
                     int featuresVectorSize,
                     int featureSubspaceDim)

Construct instance of this class with given parameters.

Parameters:

tr - Trainer for making bagged.

aggregator - Aggregator of models.

ensembleSize - Size of ensemble.

subsampleRatio - Ratio (subsample size) / (initial dataset size).

featuresVectorSize - Dimensionality of feature vector.

featureSubspaceDim - Dimensionality of feature subspace.

Method Detail

getMapping

public static int[] getMapping(int featuresVectorSize,
                               int maximumFeaturesCntPerMdl,
                               long seed)

Get mapping R^featuresVectorSize -> R^maximumFeaturesCntPerMdl.

Parameters:

featuresVectorSize - Features vector size (Dimension of initial space).

maximumFeaturesCntPerMdl - Dimension of target space.

seed - Seed.

Returns:

Mapping R^featuresVectorSize -> R^maximumFeaturesCntPerMdl.

fitWithInitializedDeployingContext

public <K,V> BaggedModel fitWithInitializedDeployingContext(DatasetBuilder<K,V> datasetBuilder,
                                                            Preprocessor<K,V> preprocessor)

Trains model based on the specified data.

Specified by:

fitWithInitializedDeployingContext in class DatasetTrainer<BaggedModel,L>

Type Parameters:

K - Type of a key in upstream data.

V - Type of a value in upstream data.

Parameters:

datasetBuilder - Dataset builder.

preprocessor - Extractor of UpstreamEntry into LabeledVector.

Returns:

Model.

update

public <K,V> BaggedModel update(BaggedModel mdl,
                                DatasetBuilder<K,V> datasetBuilder,
                                Preprocessor<K,V> preprocessor)

Gets state of model in arguments, compare it with training parameters of trainer and if they are fit then trainer updates model in according to new data and return new model. In other case trains new model.

Overrides:

update in class DatasetTrainer<BaggedModel,L>

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.

preprocessor - Extractor of UpstreamEntry into LabeledVector.

Returns:

Updated model.

withEnvironmentBuilder

public BaggedTrainer<L> withEnvironmentBuilder(LearningEnvironmentBuilder envBuilder)

Changes learning Environment.

Overrides:

withEnvironmentBuilder in class DatasetTrainer<BaggedModel,L>

Parameters:

envBuilder - Learning environment builder.

isUpdateable

public boolean isUpdateable(BaggedModel mdl)

This method is never called, instead of constructing logic of update from DatasetTrainer.isUpdateable(M) and DatasetTrainer.updateModel(M, org.apache.ignite.ml.dataset.DatasetBuilder<K, V>, org.apache.ignite.ml.preprocessing.Preprocessor<K, V>) in this class we explicitly override update method.

Specified by:

isUpdateable in class DatasetTrainer<BaggedModel,L>

Parameters:

mdl - Model.

Returns:

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

updateModel

protected <K,V> BaggedModel updateModel(BaggedModel mdl,
                                        DatasetBuilder<K,V> datasetBuilder,
                                        Preprocessor<K,V> preprocessor)

This method is never called, instead of constructing logic of update from DatasetTrainer.isUpdateable(M) and DatasetTrainer.updateModel(M, org.apache.ignite.ml.dataset.DatasetBuilder<K, V>, org.apache.ignite.ml.preprocessing.Preprocessor<K, V>) in this class we explicitly override update method.

Specified by:

updateModel in class DatasetTrainer<BaggedModel,L>

Type Parameters:

K - Type of a key in upstream data.

V - Type of a value in upstream data.

Parameters:

mdl - Model.

datasetBuilder - Dataset builder.

preprocessor - Extractor of UpstreamEntry into LabeledVector.

Returns:

Updated model.