[
https://issues.apache.org/jira/browse/FLINK-1992?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=14553957#comment-14553957
]
ASF GitHub Bot commented on FLINK-1992:
---------------------------------------
Github user thvasilo commented on a diff in the pull request:
https://github.com/apache/flink/pull/692#discussion_r30785537
--- Diff:
flink-staging/flink-ml/src/main/scala/org/apache/flink/ml/optimization/Regularization.scala
---
@@ -95,6 +95,7 @@ abstract class DiffRegularization extends Regularization {
regParameter: Double)
}
+// TODO(tvas): I think NoRegularization should extend DiffRegularization
--- End diff --
Because the idea is to disallow setting non-differentiable regularization
in solver like L-BFGS which require differentiable loss, through type-matching.
By setting NoRegularization to extend DiffRegularization we would be able
to use it with L-BFGS
> Add convergence criterion to SGD optimizer
> ------------------------------------------
>
> Key: FLINK-1992
> URL: https://issues.apache.org/jira/browse/FLINK-1992
> Project: Flink
> Issue Type: Improvement
> Components: Machine Learning Library
> Reporter: Till Rohrmann
> Assignee: Theodore Vasiloudis
> Priority: Minor
> Labels: ML
> Fix For: 0.9
>
>
> Currently, Flink's SGD optimizer runs for a fixed number of iterations. It
> would be good to support a dynamic convergence criterion, too.
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)
