summaryrefslogtreecommitdiff
path: root/gi/pyp-topics/src/train-contexts.cc
diff options
context:
space:
mode:
authorphilblunsom@gmail.com <philblunsom@gmail.com@ec762483-ff6d-05da-a07a-a48fb63a330f>2010-06-29 22:38:22 +0000
committerphilblunsom@gmail.com <philblunsom@gmail.com@ec762483-ff6d-05da-a07a-a48fb63a330f>2010-06-29 22:38:22 +0000
commit4e1c6e53e121aeb54df2b3ac825548fa4878fed2 (patch)
tree707beb9afc0294a39ad9ed7c573ba8b7c0d87ee9 /gi/pyp-topics/src/train-contexts.cc
parent9f211fce859163cd26b1d8a0b7632cb300b0f67f (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.cc3
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)