[
https://issues.apache.org/jira/browse/ARROW-3899?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=16702664#comment-16702664
]
Wes McKinney commented on ARROW-3899:
-------------------------------------
We can minimize memory use by passing the date32[day] values through a hash
table. This should be easier with the new hashing machinery
> [Python] Table.to_pandas converts Arrow date32[day] to pandas datetime64[ns]
> ----------------------------------------------------------------------------
>
> Key: ARROW-3899
> URL: https://issues.apache.org/jira/browse/ARROW-3899
> Project: Apache Arrow
> Issue Type: Improvement
> Components: Python
> Reporter: Wes McKinney
> Priority: Major
> Fix For: 0.13.0
>
>
> This issue was raised here:
> https://github.com/wesm/feather/issues/359
> I explored this minimally against Arrow master:
> https://gist.github.com/wesm/2ebe0ca2461d1ecfba6185777238ad1f
> While it's pretty memory-wasteful, it might be better to preserve the intent
> of the data type when converting to pandas data structures. It also allows
> the data to round trip successfully
--
This message was sent by Atlassian JIRA
(v7.6.3#76005)
