[ https://issues.apache.org/jira/browse/SPARK-15027?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=15269218#comment-15269218 ]
Xiangrui Meng commented on SPARK-15027: --------------------------------------- Ah, I see the problems now. We do need the hash partitioner to accelerate queries from the driver and probably joins. What if we convert the factors using `repartition(blocks, "id")` before we return the factors? It should come with a hash partitioner, but it might be different from the one we used in ALS. #2 seems like a bug. Could you provide a minimal example that can reproduce it? Given the pending issues, it seems that we should target this to 2.1. Sounds good? > ALS.train should use DataFrame instead of RDD > --------------------------------------------- > > Key: SPARK-15027 > URL: https://issues.apache.org/jira/browse/SPARK-15027 > Project: Spark > Issue Type: Improvement > Components: ML, PySpark > Affects Versions: 2.0.0 > Reporter: Xiangrui Meng > > We should also update `ALS.train` to use `Dataset/DataFrame` instead of `RDD` > to be consistent with other APIs under spark.ml and it also leaves space for > Tungsten-based optimization. -- 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