| Age | Commit message (Collapse) | Author |
|---|
|
|
|
bit less than half of the data
|
|
before
|
|
|
|
|
|
|
|
later is now retrained. Once the grammar is compiled with the .probs file now it doesn't need anything more to do robust parsing. The robustness itself is controlled by the flags 'heuristic_search_factor', 'meta_prob' and 'meta_token_prob' in ParseEngAbs.gf
|
|
|
|
|
|
abstract syntax are not tagged with their senses. There is a new statistical model too
|
|
the statistical parser for robust chunking
|
|
addition there are a number of small other changes in the dictionary. The statistical model is updated and is now moved to treebanks/PennTreebank/ParseEngAbs.probs
|
|
in ParseEng
|
|
|
|
|
