[
https://issues.apache.org/jira/browse/SPARK-4111?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=14186514#comment-14186514
]
Yanbo Liang commented on SPARK-4111:
------------------------------------
We had implement regression metrics such as explained variance score, MAE, MSE
and R2 score to evaluate the regression model.
If there is no evaluation metrics, users can not know stand or fall of this
model and tuning parameter for better result.
I will submit PR for this issue.
> [MLlib] Implement regression model evaluation metrics
> -----------------------------------------------------
>
> Key: SPARK-4111
> URL: https://issues.apache.org/jira/browse/SPARK-4111
> Project: Spark
> Issue Type: New Feature
> Components: MLlib
> Affects Versions: 1.2.0
> Reporter: Yanbo Liang
>
> Supervised machine learning include classification and regression. There is
> classification metrics (BinaryClassificationMetrics) in MLlib, we also need
> regression metrics to evaluate the regression model and tunning parameter.
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)
---------------------------------------------------------------------
To unsubscribe, e-mail: issues-unsubscr...@spark.apache.org
For additional commands, e-mail: issues-h...@spark.apache.org
