DOC: How to install nightly pandas wheels

[AlenkaF]

Pandas version checks

• I have checked that the issue still exists on the latest versions of the docs on main here

Location of the documentation

https://pandas.pydata.org/pandas-docs/stable/getting_started/install.html

Documentation problem

It would be great to have the information about the installation location of nightly pandas builds available in the official documentation.

Suggested fix for documentation

Add to the install docs the possibility of installing nightly versions of pandas from scipy-wheels-nightly/packages/pandas

[phofl]

Good idea! We can basically use https://scikit-learn.org/stable/developers/advanced_installation.html as a template

[MarcoGorelli]

thanks @AlenkaF

btw, I think the instructions in https://anaconda.org/scipy-wheels-nightly/pandas are wrong, as they don't give the latest nightly

If I try that, it definitely doesn't install the latest nightly, as this is old behaviour:

(.venv) marcogorelli@DESKTOP-U8OKFP3:~/tmp$ pip uninstall pandas -y
Found existing installation: pandas 1.4.4
Uninstalling pandas-1.4.4:
  Successfully uninstalled pandas-1.4.4
(.venv) marcogorelli@DESKTOP-U8OKFP3:~/tmp$ pip install -i https://pypi.anaconda.org/scipy-wheels-nightly/simple pandas
Looking in indexes: https://pypi.anaconda.org/scipy-wheels-nightly/simple
Collecting pandas
  Downloading https://pypi.anaconda.org/scipy-wheels-nightly/simple/pandas/0%2Bunknown/pandas-0%2Bunknown-cp38-cp38-manylinux_2_17_x86_64.manylinux2014_x86_64.whl (12.1 MB)
     ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 12.1/12.1 MB 9.1 MB/s eta 0:00:00
Requirement already satisfied: numpy>=1.18.5 in ./.venv/lib/python3.8/site-packages (from pandas) (1.23.5)
Requirement already satisfied: pytz>=2020.1 in ./.venv/lib/python3.8/site-packages (from pandas) (2022.6)
Requirement already satisfied: python-dateutil>=2.8.1 in ./.venv/lib/python3.8/site-packages (from pandas) (2.8.2)
Requirement already satisfied: six>=1.5 in ./.venv/lib/python3.8/site-packages (from python-dateutil>=2.8.1->pandas) (1.16.0)
Installing collected packages: pandas
ERROR: pip's dependency resolver does not currently take into account all the packages that are installed. This behaviour is the source of the following dependency conflicts.
streamlit 1.16.0 requires pandas>=0.21.0, but you have pandas 0+unknown which is incompatible.
altair 4.2.0 requires pandas>=0.18, but you have pandas 0+unknown which is incompatible.
Successfully installed pandas-0+unknown
(.venv) marcogorelli@DESKTOP-U8OKFP3:~/tmp$ python
Python 3.8.16 (default, Dec  7 2022, 01:12:06)
[GCC 11.3.0] on linux
Type "help", "copyright", "credits" or "license" for more information.
>>> import pandas
>>> pandas.__version__
'0+unknown'
>>> pandas.to_datetime(['13/01/2000 00:00', '12/01/2000 00:00'])
DatetimeIndex(['2000-01-13', '2000-12-01'], dtype='datetime64[ns]', freq=None)

Using the command as in the sklearn docs, it works:

(.venv) marcogorelli@DESKTOP-U8OKFP3:~/tmp$ pip uninstall pandas -y
Found existing installation: pandas 0+unknown
Uninstalling pandas-0+unknown:
  Successfully uninstalled pandas-0+unknown
(.venv) marcogorelli@DESKTOP-U8OKFP3:~/tmp$ pip install --pre --extra-index https://pypi.anaconda.org/scipy-wheels-night
ly/simple pandas
Looking in indexes: https://pypi.org/simple, https://pypi.anaconda.org/scipy-wheels-nightly/simple
Collecting pandas
  Downloading https://pypi.anaconda.org/scipy-wheels-nightly/simple/pandas/2.0.0.dev0%2B1147.g7cb7592523/pandas-2.0.0.dev0%2B1147.g7cb7592523-cp38-cp38-manylinux_2_17_x86_64.manylinux2014_x86_64.whl (47.1 MB)
     ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 47.1/47.1 MB 8.7 MB/s eta 0:00:00
Requirement already satisfied: python-dateutil>=2.8.2 in ./.venv/lib/python3.8/site-packages (from pandas) (2.8.2)
Requirement already satisfied: pytz>=2020.1 in ./.venv/lib/python3.8/site-packages (from pandas) (2022.6)
Requirement already satisfied: numpy>=1.20.3 in ./.venv/lib/python3.8/site-packages (from pandas) (1.23.5)
Requirement already satisfied: six>=1.5 in ./.venv/lib/python3.8/site-packages (from python-dateutil>=2.8.2->pandas) (1.16.0)
Installing collected packages: pandas
Successfully installed pandas-2.0.0.dev0+1147.g7cb7592523
(.venv) marcogorelli@DESKTOP-U8OKFP3:~/tmp$ python
Python 3.8.16 (default, Dec  7 2022, 01:12:06)
[GCC 11.3.0] on linux
Type "help", "copyright", "credits" or "license" for more information.
>>> import pandas
>>> pandas.to_datetime(['13/01/2000 00:00', '12/01/2000 00:00'])
<stdin>:1: UserWarning: Parsing dates in %d/%m/%Y %H:%M format when dayfirst=False (the default) was specified. Pass `dayfirst=True` or specify a format to silence this warning.
DatetimeIndex(['2000-01-13', '2000-01-12'], dtype='datetime64[ns]', freq=None)

So yeah, this really needs updating - are you interested in making a pull request?