smile.sampling

Class Bagging

java.lang.Object

smile.sampling.Bagging

public class Bagging
extends java.lang.Object

Bagging (Bootstrap aggregating) is a way to improve the classification by combining classifications of randomly generated training sets.

Field Summary

Fields

Modifier and Type	Field and Description
int[][]	samples Samples.
int	size The number of samples in this bag.

Constructor Summary

Constructors

Constructor and Description
Bagging(int k, int[] y, int[] classWeight, double subsample) Stratified sampling.

Method Summary

Methods inherited from class java.lang.Object

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

Field Detail

size

public int size

The number of samples in this bag.

samples

public int[][] samples

Samples. The first column is the sample index while the second column is the number of samples.

Constructor Detail

Bagging

public Bagging(int k,
               int[] y,
               int[] classWeight,
               double subsample)

Stratified sampling.

Parameters:

k - the number of classes.

y - class labels.

classWeight - Priors of the classes. The weight of each class is roughly the ratio of samples in each class. For example, if there are 400 positive samples and 100 negative samples, the classWeight should be [1, 4] (assuming label 0 is of negative, label 1 is of positive).

subsample - sampling rate. Draw samples with replacement if it is 1.0.