[jira] [Commented] (SPARK-38082) Update minimum numpy version

["Maciej Szymkiewicz (Jira)" <ji...@apache.org>]

    [ https://issues.apache.org/jira/browse/SPARK-38082?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=17485327#comment-17485327 ] 

Maciej Szymkiewicz commented on SPARK-38082:
--------------------------------------------

cc [~hyukjin.kwon] [~WeichenXu123] [~huaxingao]

> Update minimum numpy version
> ----------------------------
>
>                 Key: SPARK-38082
>                 URL: https://issues.apache.org/jira/browse/SPARK-38082
>             Project: Spark
>          Issue Type: Improvement
>          Components: ML, MLlib, PySpark
>    Affects Versions: 3.3.0
>            Reporter: Maciej Szymkiewicz
>            Priority: Major
>
> Currently, we use set numpy version in {{extras_require}} to be {{>=1.7}}.
> However, 1.7 has been released over almost 9 years old and since then, some methods that we use have been deprecated in favor of new additions, and new API ({{numpy.typing}}, that is of some interest to us, has been added.
> We should update minimum version requirement to:
> - {{>=1.9.0}} ‒ this is minimum reasonable bound, that will allow us to replace deprecated {{tostring}} calls with {{tobytes}}.
> - {{>=1.15.0}} (released 2018-07-23) ‒ this is reasonable bound to match our minimum supported pandas version.
> - {{>=1.20.0}} (released 2021-01-30) ‒ to fully utilize numpy typing.
> The last one might be somewhat controversial, but 1.15 shouldn't require much discussion.



--
This message was sent by Atlassian Jira
(v8.20.1#820001)

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