You are all right. Thank you! The problem was that I had sklearn installed
in both a path under PYTHONPATH and in my virtualenv. Once I removed the
path in PYTHONPATH, my virtualenv's pip reported the right version.

Thanks again,

Jacob


On Wed, Jul 3, 2013 at 6:00 AM, Eustache DIEMERT <[email protected]>wrote:

> Perhaps a stupid suggestion, but did you use "the" pip that belongs to
> your virtualenv ?
>
> Usually I activate my virtualenv (sthg like /path/to/venv/bin/activate),
> then I type "sudo `which pip` upgrade scikit-learn". Mind the backtips
> around "wich pip" who enforce the fact that it is "the pip" that belongs to
> the virtualenv that is used by the priviledged user to do the upgrade.
>
> If this doesn't work, maybe try to sudo `which pip` uninstall scikit-learn
> first...
>
> E/
>
>
> 2013/7/3 Josh Wasserstein <[email protected]>
>
>>  Hi,
>>
>> I am having problems upgrading to 0.13.1 from 0.13.  I have a local
>> (rootless) installation of Python, and have installed most of my packages
>> on a virtualenv with pip.
>>
>> Attached is the log of the upgrading process with:
>>
>> > pip install scikit-learn upgrade
>>
>> (the log was 60K and the mailing list limit is 40K so I had to compress
>> it).
>>
>> I then checked which version I have, and pip reports that I still have
>> 0.13.
>>
>> $ pip freeze
>> Cython==0.19.1
>> PIL==1.1.7
>> PyXB==1.2.2
>> PyYAML==3.10
>> argparse==1.2.1
>> distribute==0.6.34
>> epc==0.0.3
>> ipython==0.13.2
>> jedi==0.5b5
>> matplotlib==1.3.x
>> nltk==2.0.4
>> nose==1.3.0
>> numexpr==2.1
>> numpy==1.7.1
>> pandas==0.11.0
>> pyparsing==1.5.7
>> python-dateutil==2.1
>> pytz==2013b
>> rpy2==2.3.1
>> scikit-learn==0.13
>> scipy==0.12.0
>> sexpdata==0.0.2
>> six==1.3.0
>> stemming==1.0.1
>> -e git+
>> https://github.com/PyTables/PyTables.git@df7b20444b0737cf34686b5d88b4e674ec85575b#egg=tables-dev
>> tornado==3.0.1
>> wsgiref==0.1.2
>>
>> Any thoughts why? Note that this email follows up the effort to fixing
>> the problems I reported in a previous email in 
>> this<http://stackoverflow.com/questions/17435772/tracing-errors-in-sklearn-tests>
>>  StackOverflow
>> thread.
>>
>> Thank you,
>>
>> Jacob
>>
>>
>>
>> ------------------------------------------------------------------------------
>> This SF.net email is sponsored by Windows:
>>
>> Build for Windows Store.
>>
>> http://p.sf.net/sfu/windows-dev2dev
>> _______________________________________________
>> Scikit-learn-general mailing list
>> [email protected]
>> https://lists.sourceforge.net/lists/listinfo/scikit-learn-general
>>
>>
>
>
> ------------------------------------------------------------------------------
> This SF.net email is sponsored by Windows:
>
> Build for Windows Store.
>
> http://p.sf.net/sfu/windows-dev2dev
> _______________________________________________
> Scikit-learn-general mailing list
> [email protected]
> https://lists.sourceforge.net/lists/listinfo/scikit-learn-general
>
>
------------------------------------------------------------------------------
This SF.net email is sponsored by Windows:

Build for Windows Store.

http://p.sf.net/sfu/windows-dev2dev
_______________________________________________
Scikit-learn-general mailing list
[email protected]
https://lists.sourceforge.net/lists/listinfo/scikit-learn-general

Reply via email to