eus.ixa.ixa.pipe.ml.eval

Class SequenceLabelerEvaluate

java.lang.Object

eus.ixa.ixa.pipe.ml.eval.SequenceLabelerEvaluate

public class SequenceLabelerEvaluate
extends Object

Evaluation class mostly using SequenceLabelerEvaluator.

Version:

2016-07-04

Author:

ragerri

Constructor Summary

Constructors

Constructor and Description
SequenceLabelerEvaluate(Properties props) Construct an evaluator.

Method Summary

Modifier and Type	Method and Description
void	detailEvaluate() Evaluate and print the precision, recall and F measure per sequence class.
void	evalError() Evaluate and print every error.
void	evaluate() Evaluate and print precision, recall and F measure.
void	evaluateAccuracy()

Methods inherited from class java.lang.Object

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

Constructor Detail

SequenceLabelerEvaluate

public SequenceLabelerEvaluate(Properties props)
                        throws IOException

Construct an evaluator. It takes from the properties a model, a testset and the format of the testset. Every other parameter set in the training, e.g., beamsize, decoding, etc., is serialized in the model.

Parameters:

props - the properties parameter

Throws:

IOException - the io exception

Method Detail

evaluate

public final void evaluate()
                    throws IOException

Evaluate and print precision, recall and F measure.

Throws:

IOException - if test corpus not loaded

evaluateAccuracy

public final void evaluateAccuracy()
                            throws IOException

Throws:

IOException

detailEvaluate

public final void detailEvaluate()
                          throws IOException

Evaluate and print the precision, recall and F measure per sequence class.

Throws:

IOException - if test corpus not loaded

evalError

public final void evalError()
                     throws IOException

Evaluate and print every error.

Throws:

IOException - if test corpus not loaded