-
Notifications
You must be signed in to change notification settings - Fork 101
Discourse Parsers Details
Mihai Surdeanu edited this page Feb 20, 2015
·
7 revisions
Note: this is of limited utility to regular users; but it might be useful for developers interested in modifying (i.e., retraining) the discourse parsers.
To regenerate the constituent-syntax model (slower but better), run this command:
sbt 'run-main edu.arizona.sista.discourse.rstparser.RSTParser -train /data/nlp/corpora/RST_cached_preprocessing/rst_train -test /data/nlp/corpora/RST_cached_preprocessing/rst_test -model model.const.rst.gz'
This will generate the model file, model.const.rst.gz
, in the current directory.
To regenerate the dependency-syntax model (faster but slightly worse), run this command:
sbt 'run-main edu.arizona.sista.discourse.rstparser.RSTParser -train /data/nlp/corpora/RST_cached_preprocessing/rst_train -test /data/nlp/corpora/RST_cached_preprocessing/rst_test -model model.dep.rst.gz -dep'
Similarly, this command generates the model that uses only dependency information: model.dep.rst.gz
- Users (r--)
- Developers (-w-)
- Maintainers (--x)