You are viewing a plain text version of this content. The canonical link for it is here.
Posted to issues@spark.apache.org by "Sean Owen (JIRA)" <ji...@apache.org> on 2016/07/25 19:01:20 UTC
[jira] [Resolved] (SPARK-16653) Make convergence tolerance param in
ANN default value consistent with other algorithm using LBFGS
[ https://issues.apache.org/jira/browse/SPARK-16653?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ]
Sean Owen resolved SPARK-16653.
-------------------------------
Resolution: Fixed
Fix Version/s: 2.1.0
Issue resolved by pull request 14286
[https://github.com/apache/spark/pull/14286]
> Make convergence tolerance param in ANN default value consistent with other algorithm using LBFGS
> -------------------------------------------------------------------------------------------------
>
> Key: SPARK-16653
> URL: https://issues.apache.org/jira/browse/SPARK-16653
> Project: Spark
> Issue Type: Improvement
> Components: ML, Optimizer
> Reporter: Weichen Xu
> Fix For: 2.1.0
>
> Original Estimate: 24h
> Remaining Estimate: 24h
>
> The default value of convergence tolerance param in ANN is 1e-4,
> but other algorithm (such as LinearRegression, LogisticRegression, and so on) using LBFGS optimizers this param's default values are all 1e-6.
> I think make them the same will be better.
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)
---------------------------------------------------------------------
To unsubscribe, e-mail: issues-unsubscribe@spark.apache.org
For additional commands, e-mail: issues-help@spark.apache.org