groupby value_counts strips categorical information

[mwaskom]

Code Sample, a copy-pastable example if possible

import seaborn as sns
tips = sns.load_dataset("tips")
print(tips["day"].dtype)
counts = tips.groupby("size").day.value_counts().rename("count").reset_index()
print(counts["day"].dtype)

category
object

Problem description

When the categorical variable (day) becomes an index in the groupby and then gets reset, it loses its categorical metadata (and dtype). In the case of the tips dataset, this includes the proper ordering of days.

Interestingly, this isn't the case with the Series.value_counts method, which properly creates a CategoricalIndex with the order metadata.

I don't see an existing issue about this but maybe #9748 is related?

Expected Output

category
category

Output of pd.show_versions()

INSTALLED VERSIONS

commit: None
python: 3.6.2.final.0
python-bits: 64
OS: Darwin
OS-release: 16.7.0
machine: x86_64
processor: i386
byteorder: little
LC_ALL: None
LANG: en_US.UTF-8
LOCALE: en_US.UTF-8

pandas: 0.21.1
pytest: 3.2.1
pip: 9.0.1
setuptools: 36.5.0.post20170921
Cython: 0.26.1
numpy: 1.13.1
scipy: 0.19.1
pyarrow: None
xarray: None
IPython: 6.2.1
sphinx: 1.6.3
patsy: 0.4.1
dateutil: 2.6.1
pytz: 2017.2
blosc: None
bottleneck: 1.2.1
tables: 3.4.2
numexpr: 2.6.2
feather: None
matplotlib: 2.1.1
openpyxl: 2.4.8
xlrd: 1.1.0
xlwt: 1.2.0
xlsxwriter: 0.9.8
lxml: 3.8.0
bs4: 4.6.0
html5lib: 0.999999999
sqlalchemy: 1.1.13
pymysql: None
psycopg2: None
jinja2: 2.9.6
s3fs: None
fastparquet: None
pandas_gbq: None
pandas_datareader: None

[jreback]

thanksj duplicate of #18502