GroupBy and Agg Breaking with Multiple Parameters #3262
Description
System information
- OS Platform and Distribution (e.g., Linux Ubuntu 16.04):
- Modin version (
modin.__version__): 0.9.1 - Python version: 3.7.9 Intel Corporation
- Code we can use to reproduce: please contact alyssa.unell@intel.com for more information on the code and the dataset
te = gf.iloc[px].groupby(col)[tar].agg(['mean','count'])
Describe the problem
When using the agg command on a SeriesGroupBy object, an error occurs regarding a key error for the columns, even though the column name matches this key error. When only one command is passed in as a parameter to agg instead of a list of parameters, this discrepancy doesn't occur.
Source code / logs
Please contact alyssa.unell@intel.com for more information on the code and dataset.
Traceback (most recent call last):
File "deeplearning/RecSys2020/02_ModelsCompetition/XGBoost1/1-reply-build-TE.py", line 369, in
train[fname] = MultiTE_gpu( train, c, t, 20, 0 )
File "deeplearning/RecSys2020/02_ModelsCompetition/XGBoost1/1-reply-build-TE.py", line 265, in MultiTE_gpu
te = gf.iloc[px].groupby(col)[tar].agg(['mean','count'])
File "/localdisk/alyssa/anaconda3/envs/aikit-modin/lib/python3.7/site-packages/modin/pandas/groupby.py", line 394, in aggregate
**kwargs,
File "/localdisk/alyssa/anaconda3/envs/aikit-modin/lib/python3.7/site-packages/modin/pandas/groupby.py", line 980, in _default_to_pandas
return self._df._default_to_pandas(groupby_on_multiple_columns, *args, **kwargs)
File "/localdisk/alyssa/anaconda3/envs/aikit-modin/lib/python3.7/site-packages/modin/pandas/base.py", line 400, in _default_to_pandas
result = op(pandas_obj, *args, **kwargs)
File "/localdisk/alyssa/anaconda3/envs/aikit-modin/lib/python3.7/site-packages/modin/pandas/groupby.py", line 974, in groupby_on_multiple_columns
by=by, axis=self._axis, squeeze=self._squeeze, **self._kwargs
File "/localdisk/alyssa/anaconda3/envs/aikit-modin/lib/python3.7/site-packages/pandas/core/frame.py", line 6727, in groupby
dropna=dropna,
File "/localdisk/alyssa/anaconda3/envs/aikit-modin/lib/python3.7/site-packages/pandas/core/groupby/groupby.py", line 568, in init
dropna=self.dropna,
File "/localdisk/alyssa/anaconda3/envs/aikit-modin/lib/python3.7/site-packages/pandas/core/groupby/grouper.py", line 811, in get_grouper
raise KeyError(gpr)
KeyError: 'b_user_id'