[ https://issues.apache.org/jira/browse/SPARK-6617?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=15233025#comment-15233025 ]
Joseph K. Bradley edited comment on SPARK-6617 at 4/8/16 10:32 PM: ------------------------------------------------------------------- [~mengxr] Does this still need to stay open? was (Author: josephkb): [~mengxr] Does this still need to stay open? Is this issue with numerical precision only, or does partitioning play a bigger role than that? (I haven't looked into the implementation deeply.) > Word2Vec is nondeterministic > ---------------------------- > > Key: SPARK-6617 > URL: https://issues.apache.org/jira/browse/SPARK-6617 > Project: Spark > Issue Type: Bug > Components: MLlib > Affects Versions: 1.3.0 > Reporter: Xiangrui Meng > Priority: Minor > > Word2Vec uses repartition: > https://github.com/apache/spark/blob/v1.3.0/mllib/src/main/scala/org/apache/spark/mllib/feature/Word2Vec.scala#L291, > which doesn't provide deterministic ordering. This makes QA a little harder. -- This message was sent by Atlassian JIRA (v6.3.4#6332) --------------------------------------------------------------------- To unsubscribe, e-mail: issues-unsubscr...@spark.apache.org For additional commands, e-mail: issues-h...@spark.apache.org