flink-issues mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From gaborhermann <...@git.apache.org>
Subject [GitHub] flink issue #2542: [FLINK-4613] [ml] Extend ALS to handle implicit feedback ...
Date Thu, 10 Nov 2016 13:11:09 GMT
Github user gaborhermann commented on the issue:

    https://github.com/apache/flink/pull/2542
  
    I agree. I already fixed the seed at the `ImplicitALSITSuite`. At the `ALSITSuite` the
seed is unset, but by default it's 0, that's why `ALSITSuite` is deterministic, so I don't
think it's a problem. Although, I would reconsider setting the default seed, because that
way two training for the same parameters yields the same result, and that might not be what
do user expects. The user might expect truly randomized result.
    
    Do you think we should modify the default optional seed `Some(0L)` to be `None`, and use
the system random generator by default? Again, this is in the original algorithm. Should we
do it in this PR or create another issue for refactoring ALS?


---
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 infrastructure@apache.org or file a JIRA ticket
with INFRA.
---

Mime
View raw message