edu.rutgers.md.test

Class Test1

java.lang.Object

edu.rutgers.md.test.Test1

public class Test1
extends Object

This is a simple command-line application that demonstrates some ways of using the edu.rutgers.md.engine API. It has two modes. In one mode, it takes a CSV file with experts' predictions and carries out necessary computations on them. In the other mode, it creates a number of such predictions itself (using a random number generator), and caries out the same computations on them.

Constructor Summary

Constructors

Constructor and Description
Test1()

Method Summary

Modifier and Type	Method and Description
static void	main(String[] args) Runs the application.

Methods inherited from class java.lang.Object

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

Constructor Detail

Test1

public Test1()

Method Detail

main

public static void main(String[] args)
                 throws IOException,
                        IllegalInputException

Runs the application.

Usage:
To process an input file:

java [-DR=3] edu.rutgers.md.engine.Main K input_file [ground_truth]

To generate and process random predictions:

java [-DR=3] edu.rutgers.md.engine.Main K random [ground_truth]

The input file should be in the same format as the one used by the web application: a CSV file with one line per expert. Each line of this file should have the following format:

"expert_ID",p[1],...,p[K],p[never],alpha

This is the same format that is used by ImportCSV.processFile(int,LineNumberReader).

The "ground truth" data is supplied on the command line in the same format as it is entered in the web application; it is the same format as the one taken by Observables.Observables(String).

The -DR=value option, if supplied, activates weight scaling (AKA "grading on a curve") for the experts; the value of the option is the value of R=weightMax/weightMin. The value must be R≥1.0.

Running with a shell script.

Two shell script, test.sh and test-random.sh, are included into the distribution (md.zip). They illustrate using this application in both modes.

Running with Apache Ant.

The Apache Ant build script build.xml included into the distribution (md.zip) has two targets, run-test and run-test-random, which can be used to run this application with Ant:

      ant run-test
      ant run-test-random
      

Throws:

IOException

IllegalInputException