org.apache.ignite.ml.regressions.linear

Class LinearRegressionSGDTrainer<P extends Serializable>

java.lang.Object

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

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

org.apache.ignite.ml.regressions.linear.LinearRegressionSGDTrainer<P>

public class LinearRegressionSGDTrainer<P extends Serializable>
extends SingleLabelDatasetTrainer<LinearRegressionModel>

Trainer of the linear regression model based on stochastic gradient descent algorithm.

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
LinearRegressionSGDTrainer(UpdatesStrategy<? super MultilayerPerceptron,P> updatesStgy) Constructs a new instance of linear regression SGD trainer.
LinearRegressionSGDTrainer(UpdatesStrategy<? super MultilayerPerceptron,P> updatesStgy, int maxIterations, int batchSize, int locIterations, long seed) Constructs a new instance of linear regression SGD trainer.

Method Summary

Modifier and Type	Method and Description
<K,V> LinearRegressionModel	fitWithInitializedDeployingContext(DatasetBuilder<K,V> datasetBuilder, Preprocessor<K,V> extractor) Trains model based on the specified data.
int	getBatchSize() Get the batch size.
int	getLocIterations() Get the amount of local iterations.
int	getMaxIterations() Get the max amount of iterations.
long	getSeed() Get the seed for random generator.
UpdatesStrategy<? super MultilayerPerceptron,P>	getUpdatesStgy() Get the update strategy.
boolean	isUpdateable(LinearRegressionModel mdl)
protected <K,V> LinearRegressionModel	updateModel(LinearRegressionModel mdl, DatasetBuilder<K,V> datasetBuilder, Preprocessor<K,V> extractor) Trains new model taken previous one as a first approximation.
LinearRegressionSGDTrainer<P>	withBatchSize(int batchSize) Set up the batchSize parameter.
LinearRegressionSGDTrainer<P>	withLocIterations(int amountOfLocIterations) Set up the amount of local iterations of SGD algorithm.
LinearRegressionSGDTrainer<P>	withMaxIterations(int maxIterations) Set up the max amount of iterations before convergence.
LinearRegressionSGDTrainer<P>	withSeed(long seed) Set up the random seed parameter.

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

LinearRegressionSGDTrainer

public LinearRegressionSGDTrainer(UpdatesStrategy<? super MultilayerPerceptron,P> updatesStgy,
                                  int maxIterations,
                                  int batchSize,
                                  int locIterations,
                                  long seed)

Constructs a new instance of linear regression SGD trainer.

Parameters:

updatesStgy - Update strategy.

maxIterations - Max number of iteration.

batchSize - Batch size.

locIterations - Number of local iterations.

seed - Seed for random generator.

LinearRegressionSGDTrainer

public LinearRegressionSGDTrainer(UpdatesStrategy<? super MultilayerPerceptron,P> updatesStgy)

Constructs a new instance of linear regression SGD trainer.

Method Detail

fitWithInitializedDeployingContext

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

Trains model based on the specified data.

Specified by:

fitWithInitializedDeployingContext in class DatasetTrainer<LinearRegressionModel,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.

updateModel

protected <K,V> LinearRegressionModel updateModel(LinearRegressionModel 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<LinearRegressionModel,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.

isUpdateable

public boolean isUpdateable(LinearRegressionModel mdl)

Specified by:

isUpdateable in class DatasetTrainer<LinearRegressionModel,Double>

Parameters:

mdl - Model.

Returns:

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

withMaxIterations

public LinearRegressionSGDTrainer<P> withMaxIterations(int maxIterations)

Set up the max amount of iterations before convergence.

Parameters:

maxIterations - The parameter value.

Returns:

Model with new max number of iterations before convergence parameter value.

withBatchSize

public LinearRegressionSGDTrainer<P> withBatchSize(int batchSize)

Set up the batchSize parameter.

Parameters:

batchSize - The size of learning batch.

Returns:

Trainer with new batch size parameter value.

withLocIterations

public LinearRegressionSGDTrainer<P> withLocIterations(int amountOfLocIterations)

Set up the amount of local iterations of SGD algorithm.

Parameters:

amountOfLocIterations - The parameter value.

Returns:

Trainer with new locIterations parameter value.

withSeed

public LinearRegressionSGDTrainer<P> withSeed(long seed)

Set up the random seed parameter.

Parameters:

seed - Seed for random generator.

Returns:

Trainer with new seed parameter value.

getUpdatesStgy

public UpdatesStrategy<? super MultilayerPerceptron,P> getUpdatesStgy()

Get the update strategy.

Returns:

The property value.

getMaxIterations

public int getMaxIterations()

Get the max amount of iterations.

Returns:

The property value.

getBatchSize

public int getBatchSize()

Get the batch size.

Returns:

The property value.

getLocIterations

public int getLocIterations()

Get the amount of local iterations.

Returns:

The property value.

getSeed

public long getSeed()

Get the seed for random generator.

Returns:

The property value.