BUG: Output all columns even if subsetting columns after groupby

[nagatanik]

• I have checked that this issue has not already been reported.

• I have confirmed this bug exists on the latest version of pandas.

• (optional) I have confirmed this bug exists on the master branch of pandas.

---

Problem

I would like create a dictionary from pandas dataframe using groupby, which the key is the group which is used in pandas groupby and the value is list of the groupby output.

# create a dataframe 
import pandas as pd
df = pd.DataFrame({'A': [1, 2, 3], 'B': [2, 3, 4], 'C': [3, 4, 5]})

When subsetting a single column after groupby, it works well,

# get a single column after groupby
dict(df.groupby('A')['B'].apply(list))

# output
{1: [2], 2: [3], 3: [4]}

but when subsetting multiple columns, it outputs the list of dataframe.columns.

# get multiple columns after groupby
dict(df.groupby('A')[['B', 'C']].apply(list))

# output
{1: ['B', 'C'], 2: ['B', 'C'], 3: ['B', 'C']}

I checked raw output when subsetting columns after groupby and found that it outputs all columns even if subsetting columns after groupby.

# get raw output when subsetting columns after groupby
list(df.groupby('A')[['B', 'C']])

# output
[(1,
     A  B  C
  0  1  2  3),
 (2,
     A  B  C
  1  2  3  4),
 (3,
     A  B  C
  2  3  4  5)]

Expected Output

For 1st example:

# get multiple columns after groupby
dict(df.groupby('A')[['B', 'C']].apply(list))

# output
{1: [2, 3], 2: [3, 4], 3: [4, 5]}

And for 2nd example:

# get raw output when subsetting columns after groupby
list(df.groupby('A')[['B', 'C']])

# output
[(1,
     B  C
  0  2  3),
 (2,
     B  C
  1  3  4),
 (3,
     B  C
  2  4  5)]

Output of pd.show_versions()

INSTALLED VERSIONS

commit : 2a7d332
python : 3.8.5.final.0
python-bits : 64
OS : Linux
OS-release : 4.15.0-29-generic
Version : #31-Ubuntu SMP Tue Jul 17 15:39:52 UTC 2018
machine : x86_64
processor : x86_64
byteorder : little
LC_ALL : None
LANG : None
LOCALE : en_US.UTF-8

pandas : 1.1.2
numpy : 1.19.1
pytz : 2020.1
dateutil : 2.8.1
pip : 20.1.1
setuptools : 47.1.0
Cython : None
pytest : None
hypothesis : None
sphinx : None
blosc : None
feather : None
xlsxwriter : None
lxml.etree : None
html5lib : None
pymysql : None
psycopg2 : None
jinja2 : 2.11.2
IPython : 7.18.1
pandas_datareader: None
bs4 : 4.9.3
bottleneck : None
fsspec : 0.8.4
fastparquet : 0.4.1
gcsfs : None
matplotlib : 3.3.1
numexpr : None
odfpy : None
openpyxl : None
pandas_gbq : None
pyarrow : 0.16.0
pytables : None
pyxlsb : None
s3fs : None
scipy : 1.5.3
sqlalchemy : None
tables : None
tabulate : None
xarray : None
xlrd : None
xlwt : None
numba : 0.51.2

[rhshadrach]

apply consistently acts on the entire DataFrame. The output of list(df) is the columns, and the output of list(ser) is the values. From a groupby perspective, you're getting what I think are the expected results. There is already an issue whether or not applying list to a DataFrame vs a Series is something that should be changed. Closing.