Add baseform keyword argument to NombankCorpusReader.instances()
…acilitate searching the corpus by lemma. (Fixes bug with previous pull request.)
Fix bug in NombankInstance.__str__(): baseform, not basename
…acilitate searching the corpus by lemma.
make code compatible to NLTK book example.
Switched epydoc-style comments to Sphinx-compatible ones.
Score in LinThesaurusCorpusReader is now stored as float instead of string.
Reworked the demo to not depend upon the items attr.
Added missing import for the demo.
By referencing 'items' on the treebank object, an AttributeError was getting raised. Fixes issue #262.
Small test running improvements + better pypy support
…temmer is fixed.
… this on ETS time (and contributed it with their permission).