edu.stanford.nlp.sentiment

Class AbstractEvaluate

java.lang.Object

edu.stanford.nlp.sentiment.AbstractEvaluate

Direct Known Subclasses:

Evaluate

public abstract class AbstractEvaluate
extends Object

Author:

John Bauer, Michael Haas (extracted this abstract class from Evaluate)

Field Summary

Fields

Modifier and Type	Field and Description
protected static NumberFormat	NF

Constructor Summary

Constructors

Constructor and Description
AbstractEvaluate(RNNOptions options)

Method Summary

Modifier and Type	Method and Description
protected static double[]	approxAccuracy(int[][] confusion, int[][] classes)
protected static double	approxCombinedAccuracy(int[][] confusion, int[][] classes)
protected int	countLengthAccuracy(Tree tree)
protected void	countRoot(Tree tree)
protected void	countTree(Tree tree)
void	eval(List<Tree> trees)
void	eval(Tree tree)
double	exactNodeAccuracy()
double	exactRootAccuracy()
Counter<Integer>	lengthAccuracies()
abstract void	populatePredictedLabels(List<Tree> trees) Sets the predicted sentiment label for all trees given.
protected static void	printConfusionMatrix(String name, int[][] confusion)
void	printLengthAccuracies()
void	printSummary()
void	reset()

Methods inherited from class java.lang.Object

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

Field Detail

NF

protected static final NumberFormat NF

Constructor Detail

AbstractEvaluate

public AbstractEvaluate(RNNOptions options)

Method Detail

printConfusionMatrix

protected static void printConfusionMatrix(String name,
                                           int[][] confusion)

approxAccuracy

protected static double[] approxAccuracy(int[][] confusion,
                                         int[][] classes)

approxCombinedAccuracy

protected static double approxCombinedAccuracy(int[][] confusion,
                                               int[][] classes)

reset

public void reset()

eval

public void eval(List<Tree> trees)

eval

public void eval(Tree tree)

countLengthAccuracy

protected int countLengthAccuracy(Tree tree)

countTree

protected void countTree(Tree tree)

countRoot

protected void countRoot(Tree tree)

exactNodeAccuracy

public double exactNodeAccuracy()

exactRootAccuracy

public double exactRootAccuracy()

lengthAccuracies

public Counter<Integer> lengthAccuracies()

printLengthAccuracies

public void printLengthAccuracies()

printSummary

public void printSummary()

populatePredictedLabels

public abstract void populatePredictedLabels(List<Tree> trees)

Sets the predicted sentiment label for all trees given. This method sets the RNNCoreAnnotations.PredictedClass annotation for all nodes in all trees.

Parameters:

trees - List of Trees to be annotated