EnricoMi commented on PR #37360:
URL: https://github.com/apache/spark/pull/37360#issuecomment-1272901268
@HyukjinKwon Two options here:
- provide an alternative for `applyInPandas` that takes the same user
function signature in batch mode
- Python (`wrap_grouped_batch_map_pandas_udf`) calls that function
multiple times for one invocation from Scala
- provide an alternative for `applyInPandas` that takes a different user
function signature in batch mode
- provides a `pandas.DataFrameGroupBy` to provide user code access to
multiple groups at once, and Pandas Group API
Both alternatives could be supported based on annotations of the user
function (`inspect.getfullargspec`).
I wouldn't touch `applyInPandas` as this might introduce some penalty user
code that has large groups.
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
To unsubscribe, e-mail: [email protected]
For queries about this service, please contact Infrastructure at:
[email protected]
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
