Github user oliverpierson commented on the pull request:
https://github.com/apache/spark/pull/11402#issuecomment-190740482
@mengxr You're correct that backporting `setSeed` will make the new test
pass, however we'd still be left with issue above, namely that
`QuantileDiscretizer` can return the wrong number of buckets. I think that the
wrong bucket number issue is minor since (I believe) it only affects the
min/max values in your data. In that case, it may be find to leave the
`findSplitCandidates` as is and just backport `setSeed`.
---
If your project is set up for it, you can reply to this email and have your
reply appear on GitHub as well. If your project does not have this feature
enabled and wishes so, or if the feature is enabled but not working, please
contact infrastructure at infrastruct...@apache.org or file a JIRA ticket
with INFRA.
---
---------------------------------------------------------------------
To unsubscribe, e-mail: reviews-unsubscr...@spark.apache.org
For additional commands, e-mail: reviews-h...@spark.apache.org
