Repository: spark Updated Branches: refs/heads/master 4a01bfc2a -> b79bf1df6
[SPARK-9337] [MLLIB] Add an ut for Word2Vec to verify the empty vocabulary check jira: https://issues.apache.org/jira/browse/SPARK-9337 Word2Vec should throw exception when vocabulary is empty Author: Yuhao Yang <[email protected]> Closes #7660 from hhbyyh/ut4Word2vec and squashes the following commits: 17a18cb [Yuhao Yang] add ut for word2vec Project: http://git-wip-us.apache.org/repos/asf/spark/repo Commit: http://git-wip-us.apache.org/repos/asf/spark/commit/b79bf1df Tree: http://git-wip-us.apache.org/repos/asf/spark/tree/b79bf1df Diff: http://git-wip-us.apache.org/repos/asf/spark/diff/b79bf1df Branch: refs/heads/master Commit: b79bf1df6238c087c3ec524344f1fc179719c5de Parents: 4a01bfc Author: Yuhao Yang <[email protected]> Authored: Sun Jul 26 14:02:20 2015 +0100 Committer: Sean Owen <[email protected]> Committed: Sun Jul 26 14:02:20 2015 +0100 ---------------------------------------------------------------------- .../org/apache/spark/mllib/feature/Word2VecSuite.scala | 10 ++++++++++ 1 file changed, 10 insertions(+) ---------------------------------------------------------------------- http://git-wip-us.apache.org/repos/asf/spark/blob/b79bf1df/mllib/src/test/scala/org/apache/spark/mllib/feature/Word2VecSuite.scala ---------------------------------------------------------------------- diff --git a/mllib/src/test/scala/org/apache/spark/mllib/feature/Word2VecSuite.scala b/mllib/src/test/scala/org/apache/spark/mllib/feature/Word2VecSuite.scala index 4cc8d11..a864eec 100644 --- a/mllib/src/test/scala/org/apache/spark/mllib/feature/Word2VecSuite.scala +++ b/mllib/src/test/scala/org/apache/spark/mllib/feature/Word2VecSuite.scala @@ -45,6 +45,16 @@ class Word2VecSuite extends SparkFunSuite with MLlibTestSparkContext { assert(newModel.getVectors.mapValues(_.toSeq) === word2VecMap.mapValues(_.toSeq)) } + test("Word2Vec throws exception when vocabulary is empty") { + intercept[IllegalArgumentException] { + val sentence = "a b c" + val localDoc = Seq(sentence, sentence) + val doc = sc.parallelize(localDoc) + .map(line => line.split(" ").toSeq) + new Word2Vec().setMinCount(10).fit(doc) + } + } + test("Word2VecModel") { val num = 2 val word2VecMap = Map( --------------------------------------------------------------------- To unsubscribe, e-mail: [email protected] For additional commands, e-mail: [email protected]
