org.apache.ignite.ml.tree

Class DecisionTree<T extends ImpurityMeasure<T>>

java.lang.Object

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

org.apache.ignite.ml.tree.DecisionTree<T>

Type Parameters:

T - Type of impurity measure.

Direct Known Subclasses:

DecisionTreeClassificationTrainer, DecisionTreeRegressionTrainer

public abstract class DecisionTree<T extends ImpurityMeasure<T>>
extends DatasetTrainer<DecisionTreeNode,Double>

Distributed decision tree trainer that allows to fit trees using row-partitioned dataset.

Nested Class Summary

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

DatasetTrainer.EmptyDatasetException

Field Summary

Fields

Modifier and Type	Field and Description
protected boolean	usingIdx Use index structure instead of using sorting while learning.

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

environment

Method Summary

Modifier and Type	Method and Description
protected boolean	checkState(DecisionTreeNode mdl)
<K,V> DecisionTreeNode	fit(Dataset<EmptyContext,DecisionTreeData> dataset)
<K,V> DecisionTreeNode	fit(DatasetBuilder<K,V> datasetBuilder, IgniteBiFunction<K,V,Vector> featureExtractor, IgniteBiFunction<K,V,Double> lbExtractor) Trains model based on the specified data.
protected abstract ImpurityMeasureCalculator<T>	getImpurityMeasureCalculator(Dataset<EmptyContext,DecisionTreeData> dataset) Returns impurity measure calculator.
static String	printTree(DecisionTreeNode node, boolean pretty) Represents DecisionTree as String.
<K,V> DecisionTreeNode	updateModel(DecisionTreeNode mdl, DatasetBuilder<K,V> datasetBuilder, IgniteBiFunction<K,V,Vector> featureExtractor, IgniteBiFunction<K,V,Double> lbExtractor) Trains new model based on dataset because there is no valid approach to update decision trees.

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

fit, fit, fit, fit, getLastTrainedModelOrThrowEmptyDatasetException, setEnvironment, update, update, update, update, update

Methods inherited from class java.lang.Object

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

Field Detail

usingIdx

protected boolean usingIdx

Use index structure instead of using sorting while learning.

Method Detail

fit

public <K,V> DecisionTreeNode fit(DatasetBuilder<K,V> datasetBuilder,
                                  IgniteBiFunction<K,V,Vector> featureExtractor,
                                  IgniteBiFunction<K,V,Double> lbExtractor)

Trains model based on the specified data.

Specified by:

fit in class DatasetTrainer<DecisionTreeNode,Double>

Type Parameters:

K - Type of a key in upstream data.

V - Type of a value in upstream data.

Parameters:

datasetBuilder - Dataset builder.

featureExtractor - Feature extractor.

lbExtractor - Label extractor.

Returns:

Model.

updateModel

public <K,V> DecisionTreeNode updateModel(DecisionTreeNode mdl,
                                          DatasetBuilder<K,V> datasetBuilder,
                                          IgniteBiFunction<K,V,Vector> featureExtractor,
                                          IgniteBiFunction<K,V,Double> lbExtractor)

Trains new model based on dataset because there is no valid approach to update decision trees.

Specified by:

updateModel in class DatasetTrainer<DecisionTreeNode,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.

featureExtractor - Feature extractor.

lbExtractor - Label extractor.

Returns:

New model based on new dataset.

checkState

protected boolean checkState(DecisionTreeNode mdl)

Specified by:

checkState in class DatasetTrainer<DecisionTreeNode,Double>

Parameters:

mdl - Model.

Returns:

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

fit

public <K,V> DecisionTreeNode fit(Dataset<EmptyContext,DecisionTreeData> dataset)

getImpurityMeasureCalculator

protected abstract ImpurityMeasureCalculator<T> getImpurityMeasureCalculator(Dataset<EmptyContext,DecisionTreeData> dataset)

Returns impurity measure calculator.

Parameters:

dataset - Dataset.

Returns:

Impurity measure calculator.

printTree

public static String printTree(DecisionTreeNode node,
                               boolean pretty)

Represents DecisionTree as String.

Parameters:

node - Decision tree.

pretty - Use pretty mode.