diff options
| author | trevor.cohn <trevor.cohn@ec762483-ff6d-05da-a07a-a48fb63a330f> | 2010-07-15 00:34:58 +0000 |
|---|---|---|
| committer | trevor.cohn <trevor.cohn@ec762483-ff6d-05da-a07a-a48fb63a330f> | 2010-07-15 00:34:58 +0000 |
| commit | 40ee5446b84b5cdbc4e4a613e4c1aa19231c42d3 (patch) | |
| tree | deb10ba93df13ec5cce90aa59d5fb8fe5a678a55 /gi/posterior-regularisation | |
| parent | 2775fc13d1e8d3ad45c8ddf94226397403e0e373 (diff) | |
Massacred the pipeline to support source language phrases and contexts.
git-svn-id: https://ws10smt.googlecode.com/svn/trunk@255 ec762483-ff6d-05da-a07a-a48fb63a330f
Diffstat (limited to 'gi/posterior-regularisation')
| -rw-r--r-- | gi/posterior-regularisation/prjava/src/phrase/Corpus.java | 4 |
1 files changed, 2 insertions, 2 deletions
diff --git a/gi/posterior-regularisation/prjava/src/phrase/Corpus.java b/gi/posterior-regularisation/prjava/src/phrase/Corpus.java index 81264ab9..d57f3c04 100644 --- a/gi/posterior-regularisation/prjava/src/phrase/Corpus.java +++ b/gi/posterior-regularisation/prjava/src/phrase/Corpus.java @@ -151,7 +151,7 @@ public class Corpus for (int i = 0; i < c.size(); ++i) { if (i > 0) b.append(" "); - if (i == c.size() / 2) b.append("<PHRASE> "); + //if (i == c.size() / 2) b.append("<PHRASE> "); b.append(wordLexicon.lookup(c.get(i))); } return b.toString(); @@ -198,7 +198,7 @@ public class Corpus while (ctxStrtok.hasMoreTokens()) { String token = ctxStrtok.nextToken(); - if (!token.equals("<PHRASE>")) + //if (!token.equals("<PHRASE>")) ctx.add(c.wordLexicon.insert(token)); } int contextId = c.contextLexicon.insert(ctx); |