org.encog.ml.importance

Class PerturbationFeatureImportanceCalc

java.lang.Object

org.encog.ml.importance.AbstractFeatureImportance

org.encog.ml.importance.PerturbationFeatureImportanceCalc

All Implemented Interfaces:

FeatureImportance

public class PerturbationFeatureImportanceCalc
extends AbstractFeatureImportance

perturbation feature encoding can be used to determine the importance of features for any type of regression or classification model, with any compatible dataset. This method works by evaluating the performance of the model when each of the input's corrisponding data is scrambled. Features that are more important will result in worse errors when their data are scrambled. Source: Breiman, L. (2001). Random forests. Machine learning, 45(1), 5-32.

Constructor Summary

Constructors

Constructor and Description
PerturbationFeatureImportanceCalc()

Method Summary

Modifier and Type	Method and Description
GenerateRandom	getRnd()
void	performRanking() Perform the ranking, without using a specific training set.
void	performRanking(MLDataSet theDataset) Perform the ranking, using a specific training set.
void	setRnd(GenerateRandom rnd) Set the random number generator.

Methods inherited from class org.encog.ml.importance.AbstractFeatureImportance

getFeatures, getFeaturesSorted, getModel, init, toString

Methods inherited from class java.lang.Object

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

Constructor Detail

PerturbationFeatureImportanceCalc

public PerturbationFeatureImportanceCalc()

Method Detail

performRanking

public void performRanking()

Perform the ranking, without using a specific training set. Not all ranking algorithms support this.

performRanking

public void performRanking(MLDataSet theDataset)

Perform the ranking, using a specific training set. Not all ranking algorithms can make use of a dataset.

Parameters:

theDataset - The dataset.

getRnd

public GenerateRandom getRnd()

Returns:

The random number generator.

setRnd

public void setRnd(GenerateRandom rnd)

Set the random number generator.

Parameters:

rnd - The random number generator.