[
https://issues.apache.org/jira/browse/FLINK-1994?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=15021887#comment-15021887
]
ASF GitHub Bot commented on FLINK-1994:
---------------------------------------
Github user tillrohrmann commented on the pull request:
https://github.com/apache/flink/pull/1384#issuecomment-158896082
Hi @rawkintrevo, thanks for your contribution.
You have some scala style violations in `GradientDescent.scala` line 46 and
168. The lines mustn't be longer than 100 characters as the maven log output
indicates.
Furthermore, could you change the title to the format "[FLINK-1994] [ml]
Add different gain calculation schemes to SGD". That way, the JIRA issues will
be automatically updated with the what happens on Github.
In Flink, we always rebase our changes on top of the current master branch.
Therefore, it would be good if you could revert the last merge commit. If you
want to update on the latest master, then do a rebase and a force push to the
branch in your repository (here `master`).
> Add different gain calculation schemes to SGD
> ---------------------------------------------
>
> Key: FLINK-1994
> URL: https://issues.apache.org/jira/browse/FLINK-1994
> Project: Flink
> Issue Type: Improvement
> Components: Machine Learning Library
> Reporter: Till Rohrmann
> Assignee: Trevor Grant
> Priority: Minor
> Labels: ML, Starter
>
> The current SGD implementation uses as gain for the weight updates the
> formula {{stepsize/sqrt(iterationNumber)}}. It would be good to make the gain
> calculation configurable and to provide different strategies for that. For
> example:
> * stepsize/(1 + iterationNumber)
> * stepsize*(1 + regularization * stepsize * iterationNumber)^(-3/4)
> See also how to properly select the gains [1].
> Resources:
> [1] http://arxiv.org/pdf/1107.2490.pdf
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)
