edu.stanford.nlp.parser.lexparser

Class EvaluateTreebank

java.lang.Object

edu.stanford.nlp.parser.lexparser.EvaluateTreebank

public class EvaluateTreebank
extends java.lang.Object

Field Summary

Fields

Modifier and Type	Field and Description
protected Function<java.util.List<? extends HasWord>,java.util.ArrayList<TaggedWord>>	tagger The tagger optionally used before parsing.

Constructor Summary

Constructors

Constructor and Description
EvaluateTreebank(LexicalizedParser parser)
EvaluateTreebank(Options op, Lexicon lex, LexicalizedParser pqFactory)

Method Summary

Methods

Modifier and Type	Method and Description
double	getLBScore()
double	getTagScore()
void	processResults(ParserQuery pq, Tree goldTree, java.io.PrintWriter pwErr, java.io.PrintWriter pwOut, java.io.PrintWriter pwFileOut, java.io.PrintWriter pwStats, TreePrint treePrint)
double	testOnTreebank(Treebank testTreebank) Test the parser on a treebank.

Methods inherited from class java.lang.Object

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

Field Detail

tagger

protected Function<java.util.List<? extends HasWord>,java.util.ArrayList<TaggedWord>> tagger

The tagger optionally used before parsing.
We keep it here as a function rather than a MaxentTagger so that we can distribute a version of the parser that doesn't include the entire tagger.
TODO: pass this in rather than create it here if we wind up using this in more place. Right now it's only used in testOnTreebank.

Constructor Detail

EvaluateTreebank

public EvaluateTreebank(LexicalizedParser parser)

EvaluateTreebank

public EvaluateTreebank(Options op,
                Lexicon lex,
                LexicalizedParser pqFactory)

Method Detail

getLBScore

public double getLBScore()

getTagScore

public double getTagScore()

processResults

public void processResults(ParserQuery pq,
                  Tree goldTree,
                  java.io.PrintWriter pwErr,
                  java.io.PrintWriter pwOut,
                  java.io.PrintWriter pwFileOut,
                  java.io.PrintWriter pwStats,
                  TreePrint treePrint)

testOnTreebank

public double testOnTreebank(Treebank testTreebank)

Test the parser on a treebank. Parses will be written to stdout, and various other information will be written to stderr and stdout, particularly if op.testOptions.verbose is true.

Parameters:

testTreebank - The treebank to parse

Returns:

The labeled precision/recall F₁ (EVALB measure) of the parser on the treebank.