[
https://issues.apache.org/jira/browse/SPARK-11696?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=15002311#comment-15002311
]
Narine Kokhlikyan commented on SPARK-11696:
-------------------------------------------
I've done some investigations about existing solutions and this is how the
optimization output looks like for Scipy:
http://docs.scipy.org/doc/scipy/reference/generated/scipy.optimize.OptimizeResult.html#scipy.optimize.OptimizeResult
> MLLIB:Optimization - Extend optimizer output for GradientDescent and LBFGS
> --------------------------------------------------------------------------
>
> Key: SPARK-11696
> URL: https://issues.apache.org/jira/browse/SPARK-11696
> Project: Spark
> Issue Type: Improvement
> Components: ML, MLlib
> Affects Versions: 1.6.0
> Reporter: Narine Kokhlikyan
>
> Hi there,
> in current implementation the Optimization:optimize() method returns only the
> weights for the features.
> However, we could make it more transparent and provide more parameters about
> the optimization, e.g. number of iteration, error, etc.
> As discussed in bellow jira, this will be useful:
> https://issues.apache.org/jira/browse/SPARK-5575
> What do you think ?
> Thanks,
> Narine
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
