Skip to content

Commit

Permalink
Register cudf.core.groupby.Grouper objects to dask `grouper_dispatc…
Browse files Browse the repository at this point in the history
…h` (rapidsai#10838)

This PR registers uses the (presumably shortly merged) dask `Grouper` dispatch to register `cudf.core.groupby.Grouper` objects to `cudf.DataFrame` objects. This should allow our own Grouper objects to be used in critical places in dask rather than pandas objects. 

This solution is favorable IMO rather than changing cuDF to handle pandas grouper objects directly. 

Xref dask/dask#9074

Authors:
  - https://github.com/brandon-b-miller

Approvers:
  - GALI PREM SAGAR (https://github.com/galipremsagar)

URL: rapidsai#10838
  • Loading branch information
brandon-b-miller committed May 26, 2022
1 parent b4674a1 commit aee42fe
Showing 1 changed file with 10 additions and 0 deletions.
10 changes: 10 additions & 0 deletions python/dask_cudf/dask_cudf/backends.py
Original file line number Diff line number Diff line change
Expand Up @@ -296,6 +296,16 @@ def is_categorical_dtype_cudf(obj):
return cudf.api.types.is_categorical_dtype(obj)


try:
from dask.dataframe.dispatch import grouper_dispatch

@grouper_dispatch.register((cudf.Series, cudf.DataFrame))
def get_grouper_cudf(obj):
return cudf.core.groupby.Grouper

except ImportError:
pass

try:
try:
from dask.array.dispatch import percentile_lookup
Expand Down

0 comments on commit aee42fe

Please sign in to comment.