[
https://issues.apache.org/jira/browse/SPARK-14900?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=15270563#comment-15270563
]
Sean Owen commented on SPARK-14900:
-----------------------------------
To me, we should also deprecate MulticlassMetrics.precision, since I don't
think there's such a thing as "precision" that isn't specific to a class in the
multi-class setting. What it implements already is really accuracy.
Accuracy is different from precision so adding it to the
BinaryClassificationMetrics class entails a little more, implementing a new
metric, but, that would also make some sense to me.
> spark.ml classification metrics should include accuracy
> -------------------------------------------------------
>
> Key: SPARK-14900
> URL: https://issues.apache.org/jira/browse/SPARK-14900
> Project: Spark
> Issue Type: New Feature
> Components: ML
> Reporter: Joseph K. Bradley
> Priority: Minor
>
> To compute "accuracy" (0/1 classification accuracy), users can use
> {{precision}} in MulticlassMetrics and
> MulticlassClassificationEvaluator.metricName. We should also support
> "accuracy" directly as an alias to help users familiar with that name.
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
