Github user jkbradley commented on the pull request:
https://github.com/apache/spark/pull/2313#issuecomment-56135962
Philosophically, I agree with @erikerlandson about it being OK for random
generators to be, well, random. If problems are caused by the output of a
randomized process not being reproducible, then then output probably isn't
being used/tested correctly.
Practically, I second @mengxr in saying we should encourage reproducibility
by requiring numpy in MLlib. But avoiding it where possible sounds good,
assuming the performance hit is not too bad.
---
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
