weka.classifiers.evaluation
Class EvaluationUtils

java.lang.Object
  |
  +--weka.classifiers.evaluation.EvaluationUtils

public class EvaluationUtils

extends java.lang.Object

Contains utility functions for generating lists of predictions in various manners.

Author:

Len Trigg (len@intelligenesis.net)

Constructor Summary

EvaluationUtils()

Method Summary

FastVector	getCVPredictions(DistributionClassifier classifier, Instances data, int numFolds) Generate a bunch of predictions ready for processing, by performing a cross-validation on the supplied dataset.
Prediction	getPrediction(DistributionClassifier classifier, Instance test) Generate a single prediction for a test instance given the pre-trained classifier.
int	getSeed() Gets the seed for randomization during cross-validation
FastVector	getTestPredictions(DistributionClassifier classifier, Instances test) Generate a bunch of predictions ready for processing, by performing a evaluation on a test set assuming the classifier is already trained.
FastVector	getTrainTestPredictions(DistributionClassifier classifier, Instances train, Instances test) Generate a bunch of predictions ready for processing, by performing a evaluation on a test set after training on the given training set.
void	setSeed(int seed) Sets the seed for randomization during cross-validation

Methods inherited from class java.lang.Object

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

Constructor Detail

EvaluationUtils

public EvaluationUtils()

Method Detail

setSeed

public void setSeed(int seed)

Sets the seed for randomization during cross-validation

getSeed

public int getSeed()

Gets the seed for randomization during cross-validation

getCVPredictions

public FastVector getCVPredictions(DistributionClassifier classifier,
                                   Instances data,
                                   int numFolds)
                            throws java.lang.Exception

Generate a bunch of predictions ready for processing, by performing a cross-validation on the supplied dataset.

Parameters:

classifier - the DistributionClassifier to evaluate

data - the dataset

numFolds - the number of folds in the cross-validation.

Throws:

java.lang.Exception - if an error occurs

getTrainTestPredictions

public FastVector getTrainTestPredictions(DistributionClassifier classifier,
                                          Instances train,
                                          Instances test)
                                   throws java.lang.Exception

Generate a bunch of predictions ready for processing, by performing a evaluation on a test set after training on the given training set.

Parameters:

classifier - the DistributionClassifier to evaluate

train - the training dataset

test - the test dataset

Throws:

java.lang.Exception - if an error occurs

getTestPredictions

public FastVector getTestPredictions(DistributionClassifier classifier,
                                     Instances test)
                              throws java.lang.Exception

Generate a bunch of predictions ready for processing, by performing a evaluation on a test set assuming the classifier is already trained.

Parameters:

classifier - the pre-trained DistributionClassifier to evaluate

test - the test dataset

Throws:

java.lang.Exception - if an error occurs

getPrediction

public Prediction getPrediction(DistributionClassifier classifier,
                                Instance test)
                         throws java.lang.Exception

Generate a single prediction for a test instance given the pre-trained classifier.

Parameters:

classifier - the pre-trained DistributionClassifier to evaluate

test - the test instance

Throws:

java.lang.Exception - if an error occurs