Repository: spark Updated Branches: refs/heads/branch-1.4 e12fbd80c -> eda1ff4ee
[SPARK-11813][MLLIB] Avoid serialization of vocab in Word2Vec jira: https://issues.apache.org/jira/browse/SPARK-11813 I found the problem during training a large corpus. Avoid serialization of vocab in Word2Vec has 2 benefits. 1. Performance improvement for less serialization. 2. Increase the capacity of Word2Vec a lot. Currently in the fit of word2vec, the closure mainly includes serialization of Word2Vec and 2 global table. the main part of Word2vec is the vocab of size: vocab * 40 * 2 * 4 = 320 vocab 2 global table: vocab * vectorSize * 8. If vectorSize = 20, that's 160 vocab. Their sum cannot exceed Int.max due to the restriction of ByteArrayOutputStream. In any case, avoiding serialization of vocab helps decrease the size of the closure serialization, especially when vectorSize is small, thus to allow larger vocabulary. Actually there's another possible fix, make local copy of fields to avoid including Word2Vec in the closure. Let me know if that's preferred. Author: Yuhao Yang <hhb...@gmail.com> Closes #9803 from hhbyyh/w2vVocab. (cherry picked from commit e391abdf2cb6098a35347bd123b815ee9ac5b689) Signed-off-by: Xiangrui Meng <m...@databricks.com> Project: http://git-wip-us.apache.org/repos/asf/spark/repo Commit: http://git-wip-us.apache.org/repos/asf/spark/commit/eda1ff4e Tree: http://git-wip-us.apache.org/repos/asf/spark/tree/eda1ff4e Diff: http://git-wip-us.apache.org/repos/asf/spark/diff/eda1ff4e Branch: refs/heads/branch-1.4 Commit: eda1ff4eede3968c24a0d1338432eae5682e8432 Parents: e12fbd8 Author: Yuhao Yang <hhb...@gmail.com> Authored: Wed Nov 18 13:25:15 2015 -0800 Committer: Xiangrui Meng <m...@databricks.com> Committed: Wed Nov 18 13:25:54 2015 -0800 ---------------------------------------------------------------------- .../src/main/scala/org/apache/spark/mllib/feature/Word2Vec.scala | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) ---------------------------------------------------------------------- http://git-wip-us.apache.org/repos/asf/spark/blob/eda1ff4e/mllib/src/main/scala/org/apache/spark/mllib/feature/Word2Vec.scala ---------------------------------------------------------------------- diff --git a/mllib/src/main/scala/org/apache/spark/mllib/feature/Word2Vec.scala b/mllib/src/main/scala/org/apache/spark/mllib/feature/Word2Vec.scala index 74e7dbf..3493186 100644 --- a/mllib/src/main/scala/org/apache/spark/mllib/feature/Word2Vec.scala +++ b/mllib/src/main/scala/org/apache/spark/mllib/feature/Word2Vec.scala @@ -141,8 +141,8 @@ class Word2Vec extends Serializable with Logging { private var trainWordsCount = 0 private var vocabSize = 0 - private var vocab: Array[VocabWord] = null - private var vocabHash = mutable.HashMap.empty[String, Int] + @transient private var vocab: Array[VocabWord] = null + @transient private var vocabHash = mutable.HashMap.empty[String, Int] private def learnVocab(words: RDD[String]): Unit = { vocab = words.map(w => (w, 1)) --------------------------------------------------------------------- To unsubscribe, e-mail: commits-unsubscr...@spark.apache.org For additional commands, e-mail: commits-h...@spark.apache.org