org.apache.ignite.ml.composition.combinators.parallel

Class TrainersParallelComposition<I,O,L>

java.lang.Object

org.apache.ignite.ml.trainers.DatasetTrainer<IgniteModel<I,List<O>>,L>

org.apache.ignite.ml.composition.combinators.parallel.TrainersParallelComposition<I,O,L>

Type Parameters:

I - Type of trainers inputs.

O - Type of trainers outputs.

L - Type of dataset labels.

public class TrainersParallelComposition<I,O,L>
extends DatasetTrainer<IgniteModel<I,List<O>>,L>

This class represents a parallel composition of trainers. Parallel composition of trainers is a trainer itself which trains a list of trainers with same input and output. Training is done in following manner:

     1. Independently train all trainers on the same dataset and get a list of models.
     2. Combine models produced in step (1) into a ModelsParallelComposition.
 

Updating is made in a similar fashion. Like in other trainers combinators we avoid to include type of contained trainers in type parameters because otherwise compositions of compositions would have a relatively complex generic type which will reduce readability.

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
TrainersParallelComposition(List<T> trainers) Construct an instance of this class from a list of trainers.

Method Summary

Modifier and Type	Method and Description
<K,V> IgniteModel<I,List<O>>	fitWithInitializedDeployingContext(DatasetBuilder<K,V> datasetBuilder, Preprocessor<K,V> preprocessor) Trains model based on the specified data.
boolean	isUpdateable(IgniteModel<I,List<O>> 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.
static <I,O,M extends IgniteModel<I,O>,T extends DatasetTrainer<M,L>,L> TrainersParallelComposition<I,O,L>	of(List<T> trainers) Create parallel composition of trainers contained in a given list.
<K,V> IgniteModel<I,List<O>>	update(IgniteModel<I,List<O>> 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> IgniteModel<I,List<O>>	updateModel(IgniteModel<I,List<O>> mdl, DatasetBuilder<K,V> datasetBuilder, Preprocessor<K,V> preprocessor) This method is never called, instead of constructing logic of update from DatasetTrainer.isUpdateable(IgniteModel) and DatasetTrainer.updateModel(IgniteModel, DatasetBuilder, Preprocessor) in this class we explicitly override update method.

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

fit, fit, fit, fit, fit, fit, getLastTrainedModelOrThrowEmptyDatasetException, identityTrainer, learningEnvironment, 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

TrainersParallelComposition

public TrainersParallelComposition(List<T> trainers)

Construct an instance of this class from a list of trainers.

Type Parameters:

T - Type of trainer.

Parameters:

trainers - Trainers.

Method Detail

of

public static <I,O,M extends IgniteModel<I,O>,T extends DatasetTrainer<M,L>,L> TrainersParallelComposition<I,O,L> of(List<T> trainers)

Create parallel composition of trainers contained in a given list.

Type Parameters:

I - Type of input of model priduced by trainers.

O - Type of output of model priduced by trainers.

M - Type of model priduced by trainers.

T - Type of trainers.

L - Type of input of labels.

Parameters:

trainers - List of trainers.

Returns:

Parallel composition of trainers contained in a given list.

fitWithInitializedDeployingContext

public <K,V> IgniteModel<I,List<O>> fitWithInitializedDeployingContext(DatasetBuilder<K,V> datasetBuilder,
                                                                       Preprocessor<K,V> preprocessor)

Trains model based on the specified data.

Specified by:

fitWithInitializedDeployingContext in class DatasetTrainer<IgniteModel<I,List<O>>,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> IgniteModel<I,List<O>> update(IgniteModel<I,List<O>> 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<IgniteModel<I,List<O>>,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.

isUpdateable

public boolean isUpdateable(IgniteModel<I,List<O>> 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<IgniteModel<I,List<O>>,L>

Parameters:

mdl - Model.

Returns:

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

updateModel

protected <K,V> IgniteModel<I,List<O>> updateModel(IgniteModel<I,List<O>> mdl,
                                                   DatasetBuilder<K,V> datasetBuilder,
                                                   Preprocessor<K,V> preprocessor)

This method is never called, instead of constructing logic of update from DatasetTrainer.isUpdateable(IgniteModel) and DatasetTrainer.updateModel(IgniteModel, DatasetBuilder, Preprocessor) in this class we explicitly override update method.

Specified by:

updateModel in class DatasetTrainer<IgniteModel<I,List<O>>,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.