An interface which can be implemented by anything that evaluates one tree at a time and then prints out a summary when done.
A framework for Set-based precision/recall/F1 evaluation.
This class counts which categories are over and underproposed in trees.
This isn't really a kind of AbstractEval: we're sort of cheating here.
Applies an AbstractEval to a list of trees to pick the best tree using F1 measure.
A Java re-implementation of the evalb bracket scoring metric (Collins, 1997) that accepts Unicode input.
Computes labeled precision and recall (evalb) at the constituent category level.
An AbstractEval which doesn't just evaluate all constituents, but lets you provide a filter to only pay attention to constituents formed from certain subtrees.
Implementation of the Leaf Ancestor metric first described by Sampson and Babarczy (2003) and later analyzed more completely by Clegg and Shepherd (2005).
Computes POS tagging P/R/F1 from guess/gold trees.
Provides a method for deciding how similar two trees are.
Dependency unlabeled attachment score.
Stanford NLP Group