Package edu.cmu.tetrad.search.work_in_progress

Class Demixer

java.lang.Object

edu.cmu.tetrad.search.work_in_progress.Demixer

public class Demixer
extends Object

Uses expectation-maximization to sort a a data set with data sampled from two or more multivariate Gaussian distributions into its component data sets.

Version:

$Id: $Id

Author:

Madelyn Glymour

Constructor Summary

Constructors

Constructor	Description
Demixer(DataSet data, int k)	Constructor.

Method Summary

Modifier and Type	Method	Description
MixtureModel	demix()	Runs the E-M algorithm iteratively until the weights array converges.
boolean	isDemixed()	isDemixed.

Methods inherited from class java.lang.Object

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

Constructor Details

Demixer

public Demixer(DataSet data,
 int k)

Constructor. Initializes the means, weights, and covariance matrices for each model.

Parameters:

data - the data set

k - the number of models

Method Details

demix

public MixtureModel demix()

Runs the E-M algorithm iteratively until the weights array converges. Returns a MixtureModel object containing the final values of the means, covariance matrices, weights, and gammas arrays.

Returns:

a MixtureModel object

isDemixed

public boolean isDemixed()

isDemixed.

Returns:

a boolean