[jira] [Commented] (SPARK-13600) Use approxQuantile from DataFrame stats in QuantileDiscretizer

["Oliver Pierson (JIRA)" <ji...@apache.org>]

    [ https://issues.apache.org/jira/browse/SPARK-13600?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=15184954#comment-15184954 ] 

Oliver Pierson commented on SPARK-13600:
----------------------------------------

Sure thing.  I think the best way to fix the bug is to use approxQuantile so I've updated the ticket rather than open a new one.  If you'd prefer a new JIRA instead, let me know and I'll open it.  Thanks.

> Use approxQuantile from DataFrame stats in QuantileDiscretizer
> --------------------------------------------------------------
>
>                 Key: SPARK-13600
>                 URL: https://issues.apache.org/jira/browse/SPARK-13600
>             Project: Spark
>          Issue Type: Bug
>          Components: MLlib
>    Affects Versions: 1.6.0, 2.0.0
>            Reporter: Oliver Pierson
>            Assignee: Oliver Pierson
>
> For consistency and code reuse, QuantileDiscretizer should use approxQuantile to find splits in the data rather than implement it's own method.
> Additionally, making this change should remedy a bug where QuantileDiscretizer fails to calculate the correct splits in certain circumstances, resulting in an incorrect number of buckets/bins.
> E.g.
> val df = sc.parallelize(1.0 to 10.0 by 1.0).map(Tuple1.apply).toDF("x")
> val discretizer = new QuantileDiscretizer().setInputCol("x").setOutputCol("y").setNumBuckets(5)
> discretizer.fit(df).getSplits
> gives:
> Array(-Infinity, 2.0, 4.0, 6.0, 8.0, 10.0, Infinity)
> which corresponds to 6 buckets (not 5).



--
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