Mailing-List: contact issues-help@spark.apache.org; run by ezmlm
Precedence: bulk
Date: Wed, 4 May 2016 17:20:13 +0000 (UTC)
From: "Miao Wang (JIRA)" <jira@apache.org>
To: issues@spark.apache.org
Message-ID: <JIRA.12962372.1461614451000.110532.1462382413034@Atlassian.JIRA>
In-Reply-To: <JIRA.12962372.1461614451000@Atlassian.JIRA>
References: <JIRA.12962372.1461614451000@Atlassian.JIRA> <JIRA.12962372.1461614451519@arcas>
Subject: [jira] [Commented] (SPARK-14900) spark.ml classification metrics
 should include accuracy
MIME-Version: 1.0
Content-Type: text/plain; charset=utf-8
Content-Transfer-Encoding: 7bit
archived-at: Wed, 04 May 2016 17:20:14 -0000


    [ https://issues.apache.org/jira/browse/SPARK-14900?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=15271023#comment-15271023 ] 

Miao Wang commented on SPARK-14900:
-----------------------------------

Yes, "accuracy" makes more sense for binary classification problems. I will create a JIRA for it and work on it.

For the MulticalssMetrics, I will make changes to deprecate "precision" by accuracy.

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