smile.classification

Class RandomForest.Trainer

java.lang.Object

smile.classification.ClassifierTrainer<double[]>

smile.classification.RandomForest.Trainer

Enclosing class:

RandomForest

public static class RandomForest.Trainer
extends ClassifierTrainer<double[]>

Trainer for random forest classifiers.

Constructor Summary

Constructors

Constructor and Description
Trainer(smile.data.Attribute[] attributes, int ntrees) Constructor.
Trainer(int mtry, int ntrees) Constructor.

Method Summary

Modifier and Type	Method and Description
RandomForest.Trainer	setMaxNodes(int maxNodes) Sets the maximum number of leaf nodes.
RandomForest.Trainer	setNodeSize(int nodeSize) Sets the minimum size of leaf nodes.
RandomForest.Trainer	setNumRandomFeatures(int mtry) Sets the number of random selected features for splitting.
RandomForest.Trainer	setNumTrees(int ntrees) Sets the number of trees in the random forest.
RandomForest.Trainer	setSamplingRates(double subsample) Sets the sampling rate.
RandomForest.Trainer	setSplitRule(DecisionTree.SplitRule rule) Sets the splitting rule.
RandomForest	train(double[][] x, int[] y) Learns a classifier with given training data.

Methods inherited from class smile.classification.ClassifierTrainer

setAttributes

Methods inherited from class java.lang.Object

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

Constructor Detail

Trainer

public Trainer(int mtry,
               int ntrees)

Constructor.

Parameters:

mtry - the number of random selected features to be used to determine

ntrees - the number of trees.

Trainer

public Trainer(smile.data.Attribute[] attributes,
               int ntrees)

Constructor.

Parameters:

attributes - the attributes of independent variable.

ntrees - the number of trees.

Method Detail

setSplitRule

public RandomForest.Trainer setSplitRule(DecisionTree.SplitRule rule)

Sets the splitting rule.

Parameters:

rule - the splitting rule.

setNumTrees

public RandomForest.Trainer setNumTrees(int ntrees)

Sets the number of trees in the random forest.

Parameters:

ntrees - the number of trees.

setNumRandomFeatures

public RandomForest.Trainer setNumRandomFeatures(int mtry)

Sets the number of random selected features for splitting.

Parameters:

mtry - the number of random selected features to be used to determine the decision at a node of the tree. floor(sqrt(p)) seems to give generally good performance, where p is the number of variables.

setMaxNodes

public RandomForest.Trainer setMaxNodes(int maxNodes)

Sets the maximum number of leaf nodes.

Parameters:

maxNodes - the maximum number of leaf nodes.

setNodeSize

public RandomForest.Trainer setNodeSize(int nodeSize)

Sets the minimum size of leaf nodes.

Parameters:

nodeSize - the number of instances in a node below which the tree will not split.

setSamplingRates

public RandomForest.Trainer setSamplingRates(double subsample)

Sets the sampling rate.

Parameters:

subsample - the sampling rate.

train

public RandomForest train(double[][] x,
                          int[] y)

Description copied from class: ClassifierTrainer

Learns a classifier with given training data.

Specified by:

train in class ClassifierTrainer<double[]>

Parameters:

x - the training instances.

y - the training labels.

Returns:

a trained classifier.