You are viewing a plain text version of this content. The canonical link for it is here.
Posted to issues@spark.apache.org by "zhengruifeng (JIRA)" <ji...@apache.org> on 2019/05/27 02:29:00 UTC

[jira] [Created] (SPARK-27847) One-Pass MultilabelMetrics & MulticlassMetrics

zhengruifeng created SPARK-27847:
------------------------------------

             Summary: One-Pass MultilabelMetrics & MulticlassMetrics
                 Key: SPARK-27847
                 URL: https://issues.apache.org/jira/browse/SPARK-27847
             Project: Spark
          Issue Type: Improvement
          Components: ML
    Affects Versions: 3.0.0
            Reporter: zhengruifeng


In MultilabelMetrics, one sperate pass is needed to compute:

{color:#93a6f5}numDocs/numLabels/{color}{color:#93a6f5}subsetAccuracy/accuracy/hammingLoss/precision/recall/f1Measure/{color}{color:#93a6f5}tpPerClass/fpPerClass/fpPerClass/fnPerClass{color}

 

And In MulticlassMetrics, one sperate pass is needed to compute:

{color:#93a6f5}labelCountByClass{color}{color:#93a6f5}/tpByClass/{color}{color:#93a6f5}fpByClass/confusions{color}

 

however, all above intermeidate metrics can be computed at once.



--
This message was sent by Atlassian JIRA
(v7.6.3#76005)

---------------------------------------------------------------------
To unsubscribe, e-mail: issues-unsubscribe@spark.apache.org
For additional commands, e-mail: issues-help@spark.apache.org