public class Evalb extends AbstractEval
This implementation assumes that the guess/gold input files are of equal length, and have one tree per line.
This implementation was last validated against EVALB20080701 (http://nlp.cs.nyu.edu/evalb/) by Spence Green on 22 Jan. 2010. Notwithstanding this, Sekine and collins' EVALB script has been the common standard for constituency evaluation of parsers for the last decade. We always validate any numbers we report with it, and we suggest that you do the same.
| Modifier and Type | Class and Description |
|---|---|
static class |
Evalb.CBEval |
AbstractEval.CatErrorEval, AbstractEval.RuleErrorEval, AbstractEval.ScoreEvalcurF1, num, runningAverages, str| Constructor and Description |
|---|
Evalb(java.lang.String str,
boolean runningAverages) |
| Modifier and Type | Method and Description |
|---|---|
void |
evaluate(Tree guess,
Tree gold,
java.io.PrintWriter pw) |
static void |
main(java.lang.String[] args)
Run the Evalb scoring metric on guess/gold input.
|
protected java.util.Set<Constituent> |
makeObjects(Tree tree)
evalb only evaluates phrasal categories, thus constituents() does not
return objects for terminals and pre-terminals.
|
display, display, evaluate, evaluate, getEvalbF1, getEvalbF1Percent, getExact, getExactPercent, getLastF1, getNum, getSentAveF1, precisionprotected java.util.Set<Constituent> makeObjects(Tree tree)
makeObjects in class AbstractEvalpublic void evaluate(Tree guess, Tree gold, java.io.PrintWriter pw)
evaluate in interface Evalevaluate in class AbstractEvalpublic static void main(java.lang.String[] args)
args -