[
https://issues.apache.org/jira/browse/SPARK-18686?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
Yanbo Liang updated SPARK-18686:
--------------------------------
Description:
Several cleanup and improvements for {{spark.logit}}:
* {{summary}} should return coefficients matrix, and should output labels for
each class if the model is multinomial logistic regression model.
* {{summary}} should not return {{areaUnderROC, roc, pr, ...}}, since most of
them are DataFrame which are less important for R users. Meanwhile, these
metrics ignore instance weights (setting all to 1.0) which will be changed in
later Spark version. In case it will introduce breaking changes, we do not
expose them currently.
* SparkR test improvement: comparing the training result with native R glmnet.
was:
Several cleanup and improvements for {{spark.logit}}:
* {{summary}} should return coefficients matrix, and should output labels for
each class if the model is multinomial logistic regression model.
* {{summary}} should not return {{areaUnderROC, roc, pr, ...}}, since most of
them are DataFrame which are less important for R users. Meanwhile, these
metrics ignore instance weights (setting all to 1.0) which will be changed in
later Spark version. In case it will introduce breaking changes, we do not
expose them currently.
> Several cleanup and improvements for spark.logit
> ------------------------------------------------
>
> Key: SPARK-18686
> URL: https://issues.apache.org/jira/browse/SPARK-18686
> Project: Spark
> Issue Type: Improvement
> Components: ML, SparkR
> Reporter: Yanbo Liang
>
> Several cleanup and improvements for {{spark.logit}}:
> * {{summary}} should return coefficients matrix, and should output labels for
> each class if the model is multinomial logistic regression model.
> * {{summary}} should not return {{areaUnderROC, roc, pr, ...}}, since most of
> them are DataFrame which are less important for R users. Meanwhile, these
> metrics ignore instance weights (setting all to 1.0) which will be changed in
> later Spark version. In case it will introduce breaking changes, we do not
> expose them currently.
> * SparkR test improvement: comparing the training result with native R glmnet.
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
