Skip to content

ImportError (import pandas) #23798

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Jump to bottom
Closed
Alexander-Shukaev opened this issue Nov 19, 2018 · 12 comments
Closed

ImportError (import pandas) #23798

Alexander-Shukaev opened this issue Nov 19, 2018 · 12 comments
Labels
Needs Info Clarification about behavior needed to assess issue

Comments

@Alexander-Shukaev
Copy link

Problem description

First of all, for some reason pandas==0.23.4 does not even build when installed via pip (pip install pandas==0.23.4 --force --no-cache), it simply reports success but importing pandas fails in the same way as posted below.

Not building upon installation looked weird to me (no idea if it's a bug or intentional change) but I decided to give downgrade a go:

(.venv) pip install pandas==0.23.0 --force --no-cache
Collecting pandas==0.23.0
  Downloading https://files.pythonhosted.org/packages/45/12/1e1ba99fb65df9f7f3724d3232feef35cc044d18604d57492d561e90219f/pandas-0.23.0.tar.gz (13.1MB)
    100% |████████████████████████████████| 13.1MB 1.7MB/s 
Collecting python-dateutil>=2.5.0 (from pandas==0.23.0)
  Downloading https://files.pythonhosted.org/packages/74/68/d87d9b36af36f44254a8d512cbfc48369103a3b9e474be9bdfe536abfc45/python_dateutil-2.7.5-py2.py3-none-any.whl (225kB)
    100% |████████████████████████████████| 235kB 1.6MB/s 
Collecting pytz>=2011k (from pandas==0.23.0)
  Downloading https://files.pythonhosted.org/packages/f8/0e/2365ddc010afb3d79147f1dd544e5ee24bf4ece58ab99b16fbb465ce6dc0/pytz-2018.7-py2.py3-none-any.whl (506kB)
    100% |████████████████████████████████| 512kB 1.3MB/s 
Collecting numpy>=1.9.0 (from pandas==0.23.0)
  Downloading https://files.pythonhosted.org/packages/38/39/f73e104d44f19a6203e786d5204532e214443ea2954917b27f3229e7639b/numpy-1.15.4-cp37-cp37m-manylinux1_x86_64.whl (13.8MB)
    100% |████████████████████████████████| 13.9MB 1.5MB/s 
Collecting six>=1.5 (from python-dateutil>=2.5.0->pandas==0.23.0)
  Downloading https://files.pythonhosted.org/packages/67/4b/141a581104b1f6397bfa78ac9d43d8ad29a7ca43ea90a2d863fe3056e86a/six-1.11.0-py2.py3-none-any.whl
Installing collected packages: six, python-dateutil, pytz, numpy, pandas
  Found existing installation: six 1.11.0
    Uninstalling six-1.11.0:
      Successfully uninstalled six-1.11.0
  Found existing installation: python-dateutil 2.7.5
    Uninstalling python-dateutil-2.7.5:
      Successfully uninstalled python-dateutil-2.7.5
  Found existing installation: pytz 2018.7
    Uninstalling pytz-2018.7:
      Successfully uninstalled pytz-2018.7
  Found existing installation: numpy 1.15.4
    Uninstalling numpy-1.15.4:
      Successfully uninstalled numpy-1.15.4
  Found existing installation: pandas 0.23.0
    Uninstalling pandas-0.23.0:
      Successfully uninstalled pandas-0.23.0
  Running setup.py install for pandas ... done
Successfully installed numpy-1.15.4 pandas-0.23.0 python-dateutil-2.7.5 pytz-2018.7 six-1.11.0

This indeed built as it should have but importing still reports errors (same as for unbuilt 0.23.4):

.venv/lib/python3.7/site-packages/pandas/__init__.py:26: in <module>
    from pandas._libs import (hashtable as _hashtable,
.venv/lib/python3.7/site-packages/pandas/_libs/__init__.py:4: in <module>
    from .tslib import iNaT, NaT, Timestamp, Timedelta, OutOfBoundsDatetime
pandas/_libs/tslibs/conversion.pxd:11: in init pandas._libs.tslib
    ???
pandas/_libs/tslibs/conversion.pyx:1: in init pandas._libs.tslibs.conversion
    ???
pandas/_libs/tslibs/nattype.pyx:14: in init pandas._libs.tslibs.nattype
    ???
E   ImportWarning: can't resolve package from __spec__ or __package__, falling back on __name__ and __path__
ImportError while importing test module '<...>'.
Hint: make sure your test modules/packages have valid Python names.
Traceback:
.venv/lib/python3.7/site-packages/pandas/__init__.py:26: in <module>
    from pandas._libs import (hashtable as _hashtable,
.venv/lib/python3.7/site-packages/pandas/_libs/__init__.py:4: in <module>
    from .tslib import iNaT, NaT, Timestamp, Timedelta, OutOfBoundsDatetime
E   ImportError: cannot import name 'iNaT' from 'pandas._libs.tslib' (unknown location)

During handling of the above exception, another exception occurred:
.venv/lib/python3.7/site-packages/_pytest/python.py:481: in _importtestmodule
    mod = self.fspath.pyimport(ensuresyspath=importmode)
.venv/lib/python3.7/site-packages/py/_path/local.py:668: in pyimport
    __import__(modname)
.venv/lib/python3.7/site-packages/_pytest/assertion/rewrite.py:296: in load_module
    six.exec_(co, mod.__dict__)
.venv/lib/python3.7/site-packages/pandas/__init__.py:35: in <module>
    "the C extensions first.".format(module))
E   ImportError: C extension: 'iNaT' from 'pandas._libs.tslib' (unknown location) not built. If you want to import pandas from the source directory, you may need to run 'python setup.py build_ext --inplace --force' to build the C extensions first.

Any ideas how to go about it?

P.S. Please, no conda solutions.
@TomAugspurger
Copy link
Contributor

Your blog shows pandas==0.23.0. 0.23.0 doesn't support python 3.7, so there's no wheel.

Can you post the full log from

  1. creating a new venv
  2. pip install pandas 0.23.4 with verbose
  3. pip, setuptools and OS and version info.

@TomAugspurger TomAugspurger added the Needs Info Clarification about behavior needed to assess issue label Nov 20, 2018
@Alexander-Shukaev
Copy link
Author

Attaching venv.log (a virtual environment crearted from scratch). All requested versions are inside and much more. The OS is obviously Linux.

@TomAugspurger
Copy link
Contributor

TomAugspurger commented Nov 21, 2018 via email

@Alexander-Shukaev
Copy link
Author

Not really, I mean this is the first package that I'm having such a problem with. I don't know the internals of pandas so I cannot speak about why these multiple exceptions are thrown in the end when I actually import it. I can only judge from their messages where they claim that they perhaps cannot locate the C shared libraries carrying the extensions? Where those are supposed to be after installation?

@TomAugspurger
Copy link
Contributor

TomAugspurger commented Nov 21, 2018 via email

@Alexander-Shukaev
Copy link
Author

Outside, it's source code. It says that because this module imports pandas and that actually throws, see that stack.

@Alexander-Shukaev
Copy link
Author

As soon as I comment out import pandas, everything is good.

@Alexander-Shukaev
Copy link
Author

Digging through <cython/cython#1720> and e.g. a corresponding fix <IntelPython/mkl_random/pull/1>, given my traceback:

test/xxx/test_utils.py:41: in <module>
    import pandas     as pd
.venv/lib/python3.7/site-packages/pandas/__init__.py:28: in <module>
    from pandas._libs import (hashtable as _hashtable,
.venv/lib/python3.7/site-packages/pandas/_libs/__init__.py:6: in <module>
    from .tslib import iNaT, NaT, Timestamp, Timedelta, OutOfBoundsDatetime
pandas/_libs/tslibs/conversion.pxd:11: in init pandas._libs.tslib
    ???
pandas/_libs/tslibs/conversion.pyx:1: in init pandas._libs.tslibs.conversion
    ???
pandas/_libs/tslibs/nattype.pyx:14: in init pandas._libs.tslibs.nattype
    ???
E   ImportWarning: can't resolve package from __spec__ or __package__, falling back on __name__ and __path__

I believe we should add

from __future__ import absolute_import

at the top of <pandas/_libs/tslibs/nattype.pyx> and most likely <pandas/_libs/tslibs/conversion.pyx> and <pandas/_libs/tslibs/conversion.pxd> as well. Could somebody do the changes, please?

@Alexander-Shukaev
Copy link
Author

By the way, a workaround from #22788 also works of course, but why not simply fix it?

@TomAugspurger
Copy link
Contributor

TomAugspurger commented Nov 24, 2018 via email

@TomAugspurger
Copy link
Contributor

@Alexander-Shukaev @WillAyd and you check if #23940 will fix this?

@WillAyd
Copy link
Member

WillAyd commented Apr 24, 2019

I don't think this issue is relevant any longer so closing but if you are still having the issue @Alexander-Shukaev feel free to reopen

@WillAyd WillAyd closed this as completed Apr 24, 2019
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Assignees
No one assigned
Labels
Needs Info Clarification about behavior needed to assess issue
Projects
None yet
Milestone
No milestone
Development

No branches or pull requests
3 participants
@WillAyd @Alexander-Shukaev @TomAugspurger