org.apache.ignite.ml.svm

Class SVMLinearClassificationTrainer

java.lang.Object

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

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

org.apache.ignite.ml.svm.SVMLinearClassificationTrainer

public class SVMLinearClassificationTrainer
extends SingleLabelDatasetTrainer<SVMLinearClassificationModel>

Base class for a soft-margin SVM linear classification trainer based on the communication-efficient distributed dual coordinate ascent algorithm (CoCoA) with hinge-loss function.

This trainer takes input as Labeled Dataset with 0 and 1 labels for two classes and makes binary classification.

The paper about this algorithm could be found here https://arxiv.org/abs/1409.1458.

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
SVMLinearClassificationTrainer()

Method Summary

Modifier and Type	Method and Description
<K,V> SVMLinearClassificationModel	fitWithInitializedDeployingContext(DatasetBuilder<K,V> datasetBuilder, Preprocessor<K,V> preprocessor) Trains model based on the specified data.
int	getAmountOfIterations() Get the amount of outer iterations of SCDA algorithm.
int	getAmountOfLocIterations() Get the amount of local iterations of SCDA algorithm.
double	getLambda() Get the regularization lambda.
long	getSeed() Get the seed number.
boolean	isUpdateable(SVMLinearClassificationModel mdl)
protected <K,V> SVMLinearClassificationModel	updateModel(SVMLinearClassificationModel mdl, DatasetBuilder<K,V> datasetBuilder, Preprocessor<K,V> preprocessor) Trains new model taken previous one as a first approximation.
SVMLinearClassificationTrainer	withAmountOfIterations(int amountOfIterations) Set up the amount of outer iterations of SCDA algorithm.
SVMLinearClassificationTrainer	withAmountOfLocIterations(int amountOfLocIterations) Set up the amount of local iterations of SCDA algorithm.
SVMLinearClassificationTrainer	withLambda(double lambda) Set up the regularization parameter.
SVMLinearClassificationTrainer	withSeed(long seed) Set up the seed.

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

SVMLinearClassificationTrainer

public SVMLinearClassificationTrainer()

Method Detail

fitWithInitializedDeployingContext

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

Trains model based on the specified data.

Specified by:

fitWithInitializedDeployingContext in class DatasetTrainer<SVMLinearClassificationModel,Double>

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.

updateModel

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

Trains new model taken previous one as a first approximation.

Specified by:

updateModel in class DatasetTrainer<SVMLinearClassificationModel,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.

preprocessor - Extractor of UpstreamEntry into LabeledVector.

Returns:

Updated model.

isUpdateable

public boolean isUpdateable(SVMLinearClassificationModel mdl)

Specified by:

isUpdateable in class DatasetTrainer<SVMLinearClassificationModel,Double>

Parameters:

mdl - Model.

Returns:

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

withLambda

public SVMLinearClassificationTrainer withLambda(double lambda)

Set up the regularization parameter.

Parameters:

lambda - The regularization parameter. Should be more than 0.0.

Returns:

Trainer with new lambda parameter value.

getLambda

public double getLambda()

Get the regularization lambda.

Returns:

The property value.

getAmountOfIterations

public int getAmountOfIterations()

Get the amount of outer iterations of SCDA algorithm.

Returns:

The property value.

withAmountOfIterations

public SVMLinearClassificationTrainer withAmountOfIterations(int amountOfIterations)

Set up the amount of outer iterations of SCDA algorithm.

Parameters:

amountOfIterations - The parameter value.

Returns:

Trainer with new amountOfIterations parameter value.

getAmountOfLocIterations

public int getAmountOfLocIterations()

Get the amount of local iterations of SCDA algorithm.

Returns:

The property value.

withAmountOfLocIterations

public SVMLinearClassificationTrainer withAmountOfLocIterations(int amountOfLocIterations)

Set up the amount of local iterations of SCDA algorithm.

Parameters:

amountOfLocIterations - The parameter value.

Returns:

Trainer with new amountOfLocIterations parameter value.

getSeed

public long getSeed()

Get the seed number.

Returns:

The property value.

withSeed

public SVMLinearClassificationTrainer withSeed(long seed)

Set up the seed.

Parameters:

seed - The parameter value.

Returns:

Model with new seed parameter value.