[
https://issues.apache.org/jira/browse/SPARK-10207?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=14712677#comment-14712677
]
Sean Owen commented on SPARK-10207:
-----------------------------------
[~fliang] see my comments on https://issues.apache.org/jira/browse/SPARK-7751
You have a lot of JIRAs for logically similar changes and I think it creates
some problems. I'm not objecting so much as asking for some discussion. But at
least, there is no parent JIRA here tying the "Specify schema" tickets -- would
that be better to keep them grouped?
> Specify schema during Word2Vec.save to avoid reflection
> -------------------------------------------------------
>
> Key: SPARK-10207
> URL: https://issues.apache.org/jira/browse/SPARK-10207
> Project: Spark
> Issue Type: Improvement
> Components: MLlib
> Reporter: Feynman Liang
> Priority: Minor
>
> [Word2Vec.save|https://github.com/apache/spark/blob/7cfc0750e14f2c1b3847e4720cc02150253525a9/mllib/src/main/scala/org/apache/spark/mllib/feature/Word2Vec.scala#L615]
> currently infers a schema from a case class when the schema is known and
> should be manually provided.
> See parent JIRA for rationale.
--
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
