LBJ2.learn
Class Accuracy

java.lang.Object
  LBJ2.learn.Accuracy

All Implemented Interfaces:

TestingMetric

public class Accuracy
extends java.lang.Object
implements TestingMetric

Returns the accuracy of a discrete classifier with respect to the oracle as the fraction of examples for which its prediction was correct.

Field Summary

private boolean

print Whether or not to print a table of results to STDOUT when test(Classifier,Classifier,Parser) is called.

Constructor Summary

Accuracy()
Creates an Accuracy testing metric that does not print a table of results.

Accuracy(boolean p)
Creates an Accuracy testing metric that prints a table of results if requested.

Method Summary

java.lang.String	getName() Returns the name of the testing metric.
double	test(Classifier classifier, Classifier oracle, Parser parser) Evaluates a classifier against an oracle on the data provided by a parser.

Methods inherited from class java.lang.Object

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

Field Detail

print

private boolean print

Whether or not to print a table of results to STDOUT when test(Classifier,Classifier,Parser) is called.

Constructor Detail

Accuracy

public Accuracy()

Creates an Accuracy testing metric that does not print a table of results.

Accuracy

public Accuracy(boolean p)

Creates an Accuracy testing metric that prints a table of results if requested.

Parameters:

p - Whether or not to print a table of results when test(Classifier,Classifier,Parser) is called.

Method Detail

getName

public java.lang.String getName()

Returns the name of the testing metric.

Specified by:

getName in interface TestingMetric

test

public double test(Classifier classifier,
                   Classifier oracle,
                   Parser parser)

Evaluates a classifier against an oracle on the data provided by a parser.

Specified by:

test in interface TestingMetric

Parameters:

classifier - The classifier whose accuracy is being measured.

oracle - A classifier that returns the label of each example.

parser - A parser to supply the example objects.

Returns:

The fraction of examples for which the classifier's prediction was correct.