[
https://issues.apache.org/jira/browse/SPARK-7780?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
Xiangrui Meng updated SPARK-7780:
---------------------------------
Target Version/s: 1.7.0 (was: 1.6.0)
> The intercept in LogisticRegressionWithLBFGS should not be regularized
> ----------------------------------------------------------------------
>
> Key: SPARK-7780
> URL: https://issues.apache.org/jira/browse/SPARK-7780
> Project: Spark
> Issue Type: Bug
> Components: MLlib
> Reporter: DB Tsai
>
> The intercept in Logistic Regression represents a prior on categories which
> should not be regularized. In MLlib, the regularization is handled through
> `Updater`, and the `Updater` penalizes all the components without excluding
> the intercept which resulting poor training accuracy with regularization.
> The new implementation in ML framework handles this properly, and we should
> call the implementation in ML from MLlib since majority of users are still
> using MLlib api.
> Note that both of them are doing feature scalings to improve the convergence,
> and the only difference is ML version doesn't regularize the intercept. As a
> result, when lambda is zero, they will converge to the same solution.
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
