You are viewing a plain text version of this content. The canonical link for it is here.
Posted to issues@arrow.apache.org by "Benjamin (Jira)" <ji...@apache.org> on 2020/02/19 16:34:00 UTC
[jira] [Comment Edited] (ARROW-7885) [Python] pyarrow.serialize
does not support dask dataframe
[ https://issues.apache.org/jira/browse/ARROW-7885?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=17040215#comment-17040215 ]
Benjamin edited comment on ARROW-7885 at 2/19/20 4:33 PM:
----------------------------------------------------------
From [this|#serializing-pandas-objects]] my understanding is that arrow has some built-in optimization when serializing pandas DataFrames. By "efficient" I mean that it should be smaller and/or serialize faster.
When serializing dask DataFrames I now use pickle but cannot make use of these optimizations.
With df.to_parquet('arrow') it is possible to use arrow serialization but that requires data to be written to the file system which I would like to avoid (thus I am bringing up databases).
My use case is to serialize dask dataframes (preferably with pyarrow) and then store them in MongoDB (similar to [this library|[https://github.com/man-group/arctic]]).
was (Author: jonen):
From [this|[https://arrow.apache.org/docs/python/ipc.html#serializing-pandas-objects]] my understanding is that arrow has some built-in optimization when serializing pandas DataFrames. By "efficient" I mean that it should be smaller and/or serialize faster.
When serializing dask DataFrames I now use pickle but cannot make use of these optimizations.
With df.to_parquet('arrow') it is possible to use arrow serialization but that requires data to be written to the file system which I would like to avoid (thus I am bringing up databases).
My use case is to serialize dask dataframes (preferably with pyarrow) and then store them in MongoDB (similar to [this library|[https://github.com/man-group/arctic]].
> [Python] pyarrow.serialize does not support dask dataframe
> ----------------------------------------------------------
>
> Key: ARROW-7885
> URL: https://issues.apache.org/jira/browse/ARROW-7885
> Project: Apache Arrow
> Issue Type: Wish
> Components: Python
> Reporter: Benjamin
> Priority: Minor
>
> Currently pyarrow knows how to serialize pandas dataframes but not dask dataframes.
> {code:java}
> SerializationCallbackError: pyarrow does not know how to serialize objects of type <class 'dask.dataframe.core.DataFrame'>. {code}
> Pickling the dask dataframe foregoes the benefits of using pyarrow for the sub dataframes.
> Pyarrow support for serializing dask dataframes would allow storing dataframes efficiently in a database instead of a file system (e.g. parquet).
--
This message was sent by Atlassian Jira
(v8.3.4#803005)