gov.sandia.cognition.learning.algorithm.tree

Class RegressionTreeLearner<InputType>

java.lang.Object

gov.sandia.cognition.util.AbstractCloneableSerializable

gov.sandia.cognition.algorithm.AbstractIterativeAlgorithm

gov.sandia.cognition.learning.algorithm.tree.AbstractDecisionTreeLearner<InputType,java.lang.Double>

gov.sandia.cognition.learning.algorithm.tree.RegressionTreeLearner<InputType>

Type Parameters:

InputType - The type of the input to the tree.

All Implemented Interfaces:

IterativeAlgorithm, BatchLearner<java.util.Collection<? extends InputOutputPair<? extends InputType,java.lang.Double>>,RegressionTree<InputType>>, SupervisedBatchLearner<InputType,java.lang.Double,RegressionTree<InputType>>, CloneableSerializable, java.io.Serializable, java.lang.Cloneable

public class RegressionTreeLearner<InputType>
extends AbstractDecisionTreeLearner<InputType,java.lang.Double>
implements SupervisedBatchLearner<InputType,java.lang.Double,RegressionTree<InputType>>

The RegressionTreeLearner class implements a learning algorithm for a regression tree that makes use of a decider learner and a regression learner. The tree grows as a decision tree until it gets to a leaf node (determined by a minimum number of nodes), and then learns a regression function at the leaf node.

Since:

2.0

Author:

Justin Basilico

See Also:

Serialized Form

Field Summary

Fields

Modifier and Type	Field and Description
static int	DEFAULT_LEAF_COUNT_THRESHOLD The default threshold for making a leaf node based on count.
static int	DEFAULT_MAX_DEPTH The default maximum depth to grow the tree to.
protected int	leafCountThreshold The threshold for making a node a leaf, determined by how many instances fall in the threshold.
protected int	maxDepth The maximum depth for the tree.
protected BatchLearner<? super java.util.Collection<? extends InputOutputPair<? extends InputType,java.lang.Double>>,? extends Evaluator<? super InputType,java.lang.Double>>	regressionLearner The learning algorithm for the regression function.

Fields inherited from class gov.sandia.cognition.learning.algorithm.tree.AbstractDecisionTreeLearner

deciderLearner

Fields inherited from class gov.sandia.cognition.algorithm.AbstractIterativeAlgorithm

DEFAULT_ITERATION, iteration

Constructor Summary

Constructors

Constructor and Description
RegressionTreeLearner() Creates a new instance of RegressionTreeLearner
RegressionTreeLearner(DeciderLearner<? super InputType,java.lang.Double,?,?> deciderLearner) Creates a new instance of CategorizationTreeLearner with a mean node learner
RegressionTreeLearner(DeciderLearner<? super InputType,java.lang.Double,?,?> deciderLearner, BatchLearner<? super java.util.Collection<? extends InputOutputPair<? extends InputType,java.lang.Double>>,? extends Evaluator<? super InputType,java.lang.Double>> regressionLearner) Creates a new instance of CategorizationTreeLearner.
RegressionTreeLearner(DeciderLearner<? super InputType,java.lang.Double,?,?> deciderLearner, BatchLearner<? super java.util.Collection<? extends InputOutputPair<? extends InputType,java.lang.Double>>,? extends Evaluator<? super InputType,java.lang.Double>> regressionLearner, int leafCountThreshold, int maxDepth) Creates a new instance of CategorizationTreeLearner.

Method Summary

Modifier and Type	Method and Description
RegressionTreeLearner<InputType>	clone() This makes public the clone method on the Object class and removes the exception that it throws.
int	getLeafCountThreshold() Gets the leaf count threshold, which determines the number of elements at which to learn a regression function.
int	getMaxDepth() Gets the maximum depth to grow the tree.
BatchLearner<? super java.util.Collection<? extends InputOutputPair<? extends InputType,java.lang.Double>>,? extends Evaluator<? super InputType,java.lang.Double>>	getRegressionLearner() Gets the regression learner that is to be used to fit a function approximator to the values in the tree.
RegressionTree<InputType>	learn(java.util.Collection<? extends InputOutputPair<? extends InputType,java.lang.Double>> data) The learn method creates an object of ResultType using data of type DataType, using some form of "learning" algorithm.
protected RegressionTreeNode<InputType,?>	learnNode(java.util.Collection<? extends InputOutputPair<? extends InputType,java.lang.Double>> data, AbstractDecisionTreeNode<InputType,java.lang.Double,?> parent) Recursively learns the regression tree using the given collection of data, returning the created node.
void	setLeafCountThreshold(int leafCountThreshold) Sets the leaf count threshold, which determines the number of elements at which to learn a regression function.
void	setMaxDepth(int maxDepth) Sets the maximum depth to grow the tree.
void	setRegressionLearner(BatchLearner<? super java.util.Collection<? extends InputOutputPair<? extends InputType,java.lang.Double>>,? extends Evaluator<? super InputType,java.lang.Double>> regressionLearner) Sets the regression learner that is to be used to fit a function approximator to the values in the tree.

Methods inherited from class gov.sandia.cognition.learning.algorithm.tree.AbstractDecisionTreeLearner

areAllOutputsEqual, getDeciderLearner, learnChildNodes, setDeciderLearner, splitData

Methods inherited from class gov.sandia.cognition.algorithm.AbstractIterativeAlgorithm

addIterativeAlgorithmListener, fireAlgorithmEnded, fireAlgorithmStarted, fireStepEnded, fireStepStarted, getIteration, getListeners, removeIterativeAlgorithmListener, setIteration, setListeners

Methods inherited from class java.lang.Object

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

Field Detail

DEFAULT_LEAF_COUNT_THRESHOLD

public static final int DEFAULT_LEAF_COUNT_THRESHOLD

The default threshold for making a leaf node based on count.

See Also:

Constant Field Values

DEFAULT_MAX_DEPTH

public static final int DEFAULT_MAX_DEPTH

The default maximum depth to grow the tree to.

See Also:

Constant Field Values

regressionLearner

protected BatchLearner<? super java.util.Collection<? extends InputOutputPair<? extends InputType,java.lang.Double>>,? extends Evaluator<? super InputType,java.lang.Double>> regressionLearner

The learning algorithm for the regression function.

leafCountThreshold

protected int leafCountThreshold

The threshold for making a node a leaf, determined by how many instances fall in the threshold.

maxDepth

protected int maxDepth

The maximum depth for the tree. Ignored if less than 1.

Constructor Detail

RegressionTreeLearner

public RegressionTreeLearner()

Creates a new instance of RegressionTreeLearner

RegressionTreeLearner

public RegressionTreeLearner(DeciderLearner<? super InputType,java.lang.Double,?,?> deciderLearner)

Creates a new instance of CategorizationTreeLearner with a mean node learner

Parameters:

deciderLearner - The learner for the decision function.

RegressionTreeLearner

public RegressionTreeLearner(DeciderLearner<? super InputType,java.lang.Double,?,?> deciderLearner,
                             BatchLearner<? super java.util.Collection<? extends InputOutputPair<? extends InputType,java.lang.Double>>,? extends Evaluator<? super InputType,java.lang.Double>> regressionLearner)

Creates a new instance of CategorizationTreeLearner.

Parameters:

deciderLearner - The learner for the decision function.

regressionLearner - The learner for the regression function.

RegressionTreeLearner

public RegressionTreeLearner(DeciderLearner<? super InputType,java.lang.Double,?,?> deciderLearner,
                             BatchLearner<? super java.util.Collection<? extends InputOutputPair<? extends InputType,java.lang.Double>>,? extends Evaluator<? super InputType,java.lang.Double>> regressionLearner,
                             int leafCountThreshold,
                             int maxDepth)

Creates a new instance of CategorizationTreeLearner.

Parameters:

deciderLearner - The learner for the decision function.

regressionLearner - The learner for the regression function.

leafCountThreshold - The leaf count threshold, which determines the number of elements at which to learn a regression function.

maxDepth - The maximum depth to learn the tree. Must be positive.

Method Detail

clone

public RegressionTreeLearner<InputType> clone()

Description copied from class: AbstractCloneableSerializable

This makes public the clone method on the Object class and removes the exception that it throws. Its default behavior is to automatically create a clone of the exact type of object that the clone is called on and to copy all primitives but to keep all references, which means it is a shallow copy. Extensions of this class may want to override this method (but call super.clone() to implement a "smart copy". That is, to target the most common use case for creating a copy of the object. Because of the default behavior being a shallow copy, extending classes only need to handle fields that need to have a deeper copy (or those that need to be reset). Some of the methods in ObjectUtil may be helpful in implementing a custom clone method. Note: The contract of this method is that you must use super.clone() as the basis for your implementation.

Specified by:

clone in interface CloneableSerializable

Overrides:

clone in class AbstractDecisionTreeLearner<InputType,java.lang.Double>

Returns:

A clone of this object.

learn

public RegressionTree<InputType> learn(java.util.Collection<? extends InputOutputPair<? extends InputType,java.lang.Double>> data)

Description copied from interface: BatchLearner

The learn method creates an object of ResultType using data of type DataType, using some form of "learning" algorithm.

Specified by:

learn in interface BatchLearner<java.util.Collection<? extends InputOutputPair<? extends InputType,java.lang.Double>>,RegressionTree<InputType>>

Parameters:

data - The data that the learning algorithm will use to create an object of ResultType.

Returns:

The object that is created based on the given data using the learning algorithm.

learnNode

protected RegressionTreeNode<InputType,?> learnNode(java.util.Collection<? extends InputOutputPair<? extends InputType,java.lang.Double>> data,
                                                    AbstractDecisionTreeNode<InputType,java.lang.Double,?> parent)

Recursively learns the regression tree using the given collection of data, returning the created node.

Specified by:

learnNode in class AbstractDecisionTreeLearner<InputType,java.lang.Double>

Parameters:

data - The set of data to learn a node from.

parent - The parent node.

Returns:

The regression tree node learned from the given data.

getRegressionLearner

public BatchLearner<? super java.util.Collection<? extends InputOutputPair<? extends InputType,java.lang.Double>>,? extends Evaluator<? super InputType,java.lang.Double>> getRegressionLearner()

Gets the regression learner that is to be used to fit a function approximator to the values in the tree.

Returns:

The regression learner.

setRegressionLearner

public void setRegressionLearner(BatchLearner<? super java.util.Collection<? extends InputOutputPair<? extends InputType,java.lang.Double>>,? extends Evaluator<? super InputType,java.lang.Double>> regressionLearner)

Sets the regression learner that is to be used to fit a function approximator to the values in the tree.

Parameters:

regressionLearner - The regression learner.

getLeafCountThreshold

public int getLeafCountThreshold()

Gets the leaf count threshold, which determines the number of elements at which to learn a regression function.

Returns:

The leaf count threshold.

setLeafCountThreshold

public void setLeafCountThreshold(int leafCountThreshold)

Sets the leaf count threshold, which determines the number of elements at which to learn a regression function.

Parameters:

leafCountThreshold - The leaf count threshold. Must be non-negative.

getMaxDepth

public int getMaxDepth()

Gets the maximum depth to grow the tree.

Returns:

The maximum depth to grow the tree. Zero or less means no maximum depth.

setMaxDepth

public void setMaxDepth(int maxDepth)

Sets the maximum depth to grow the tree.

Parameters:

maxDepth - The maximum depth to grow the tree. Zero or less means no maximum depth.