[
https://issues.apache.org/jira/browse/FLINK-1994?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=15110476#comment-15110476
]
ASF GitHub Bot commented on FLINK-1994:
---------------------------------------
Github user tillrohrmann commented on a diff in the pull request:
https://github.com/apache/flink/pull/1397#discussion_r50391309
--- Diff:
flink-staging/flink-ml/src/main/scala/org/apache/flink/ml/regression/MultipleLinearRegression.scala
---
@@ -107,6 +107,11 @@ class MultipleLinearRegression extends
Predictor[MultipleLinearRegression] {
this
}
+ def setOptimizationMethod(optimizationMethod: String): this.type = {
--- End diff --
Furthermore, if you expose the optimization method, then you should also
expose the decay parameter. But to be honest I'm not so sure whether this is
the right way to do, because you would have to add to every algorithm every new
parameter of the underlying solver. I guess it would be better to expose a
parameter where you can set a non-default solver which allows you to set these
things directly. What do you think?
> 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)
