mml

Class Test

java.lang.Object

mml.Test

public class Test
extends java.lang.Object

main(java.lang.String[]) runs a few simple tests on mml. This is nowhere near an exhaustive set of tests.

Constructor Summary

Constructors

Constructor and Description
Test()

Method Summary

Modifier and Type	Method and Description
static void	compete(int trials, int N, double AoM, Model gen, Estimator e1, Estimator e2) Version of compete(int, int, mml.Model, mml.Estimator, mml.Estimator) with a specified 'AoM', thus allowing for competition between Continuous.M Models.
static void	compete(int trials, int N, Model gen, Estimator e1, Estimator e2) Do a number of 'trials', each time using Model 'gen' to generate a data-set of 'N' values, and then fitting Models by estimators 'e1' and 'e2' to that data-set to see which one wins in the minimum message length challenge.
static void	exercise(Estimator est, Vector ds) Use Estimator 'est' to fit a Model 'm' to data 'ds', then exercise(m,ds).
static void	exercise(Model m, Vector ds) Run Model 'm' through a few simple tests on data-set 'ds'.
static void	main(java.lang.String[] argv) Run a few, very(!) simple tests.

Methods inherited from class java.lang.Object

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

Constructor Detail

Test

public Test()

Method Detail

exercise

public static void exercise(Estimator est,
                            Vector ds)

Use Estimator 'est' to fit a Model 'm' to data 'ds', then exercise(m,ds).

exercise

public static void exercise(Model m,
                            Vector ds)

Run Model 'm' through a few simple tests on data-set 'ds'. Also tries m.random() (which not all Models implement). Also see exercise(Estimator,Vector).

compete

public static void compete(int trials,
                           int N,
                           Model gen,
                           Estimator e1,
                           Estimator e2)

Do a number of 'trials', each time using Model 'gen' to generate a data-set of 'N' values, and then fitting Models by estimators 'e1' and 'e2' to that data-set to see which one wins in the minimum message length challenge. Also see compete(int,int,double,Model,Estimator,Estimator).

compete

public static void compete(int trials,
                           int N,
                           double AoM,
                           Model gen,
                           Estimator e1,
                           Estimator e2)

Version of compete(int, int, mml.Model, mml.Estimator, mml.Estimator) with a specified 'AoM', thus allowing for competition between Continuous.M Models.

main

public static void main(java.lang.String[] argv)

Run a few, very(!) simple tests.