You are viewing a plain text version of this content. The canonical link for it is here.

Posted to issues@spark.apache.org by "Xiangrui Meng (JIRA)" <ji...@apache.org> on 2014/09/03 08:03:51 UTC

[jira] [Updated] (SPARK-3195) Can you add some statistics to do logistic regression better in mllib?

     [ https://issues.apache.org/jira/browse/SPARK-3195?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ]

Xiangrui Meng updated SPARK-3195:
---------------------------------
    Affects Version/s:     (was: 1.3.0)

> Can you add some statistics to do logistic regression better in mllib?
> ----------------------------------------------------------------------
>
>                 Key: SPARK-3195
>                 URL: https://issues.apache.org/jira/browse/SPARK-3195
>             Project: Spark
>          Issue Type: New Feature
>          Components: MLlib
>            Reporter: miumiu
>              Labels: test
>             Fix For: 1.3.0
>
>   Original Estimate: 1m
>  Remaining Estimate: 1m
>
> HI，
>     In logistic regression model practice，Test of regression coefficient and whole model fitting are very important.Can you add some effective support on these Aspects?
>     Such as,The likelihood ratio test or the wald test is offer used for test of coefficient,and the Hosmer-Lemeshow test is used for evaluate the model fitting.
>     Learning that we have ROC and Precision-Recall already,but can you also provide KS statistic,which is mostly used in Model evaluation aspect?



--
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