diff options
| author | philblunsom@gmail.com <philblunsom@gmail.com@ec762483-ff6d-05da-a07a-a48fb63a330f> | 2010-06-29 22:38:22 +0000 |
|---|---|---|
| committer | philblunsom@gmail.com <philblunsom@gmail.com@ec762483-ff6d-05da-a07a-a48fb63a330f> | 2010-06-29 22:38:22 +0000 |
| commit | 32c356e5dd29b6d0ade9423043a44fb52b5e1653 (patch) | |
| tree | 6da850c2a5722b470aa7224e58e33d7b7ce26512 /gi/pyp-topics/src/train-contexts.cc | |
| parent | fb515d6fe01c65c924d0806619fa938688335579 (diff) | |
Added singleton filter.
git-svn-id: https://ws10smt.googlecode.com/svn/trunk@70 ec762483-ff6d-05da-a07a-a48fb63a330f
Diffstat (limited to 'gi/pyp-topics/src/train-contexts.cc')
| -rw-r--r-- | gi/pyp-topics/src/train-contexts.cc | 3 |
1 files changed, 2 insertions, 1 deletions
diff --git a/gi/pyp-topics/src/train-contexts.cc b/gi/pyp-topics/src/train-contexts.cc index c5ab8734..833565cd 100644 --- a/gi/pyp-topics/src/train-contexts.cc +++ b/gi/pyp-topics/src/train-contexts.cc @@ -43,6 +43,7 @@ int main(int argc, char **argv) ("topic-words-out,w", value<string>(), "file to write the topic word distribution to") ("samples,s", value<int>()->default_value(10), "number of sampling passes through the data") ("backoff-type", value<string>(), "backoff type: none|simple") + ("filter-singleton-contexts", "filter singleton contexts") ; store(parse_command_line(argc, argv, cmdline_options), vm); notify(vm); @@ -80,7 +81,7 @@ int main(int argc, char **argv) } ContextsCorpus contexts_corpus; - contexts_corpus.read_contexts(vm["data"].as<string>(), backoff_gen); + contexts_corpus.read_contexts(vm["data"].as<string>(), backoff_gen, vm.count("filter-singleton-contexts")); model.set_backoff(contexts_corpus.backoff_index()); if (backoff_gen) |