[scikit-learn] How to test on PYTHON_ARCH=32 with mac?

[Nelson Liu]

Your assumption is correct, appveyor does run windows.

Nelson

On Wed, Jul 20, 2016 at 9:36 AM lin yenchen <yenchenlin1994 at gmail.com>
wrote:

> Well, but the tests failed on CI
> <https://ci.appveyor.com/project/agramfort/scikit-learn/build/1.0.505> of
> my PR <https://github.com/scikit-learn/scikit-learn/pull/6913> passed on
> my local 32 bit scikit-learn.
> Do this mean that the problem exists on the Windows? (I
> assumed appveyor is running Window because of C:\ in the console)
>
> Sorry if this is a nonsense assumption.
>
> Best,
> Yen-Chen
>
> 2016-07-21 0:25 GMT+08:00 lin yenchen <yenchenlin1994 at gmail.com>:
>
>> Thanks for you guys' precious inputs.
>>
>> I've successfully built a 32-bit python version scikit-learn and check it
>> by printing `sys.maxint`,
>> and all the tests passed on my mac. (I'm running the newest dev version
>> though)
>>
>> Best,
>> Yen-Chen
>>
>>
>> 2016-07-21 0:00 GMT+08:00 Matthew Brett <matthew.brett at gmail.com>:
>>
>>> On Wed, Jul 20, 2016 at 9:16 AM, Matthew Brett <matthew.brett at gmail.com>
>>> wrote:
>>> > On Wed, Jul 20, 2016 at 9:09 AM, Olivier Grisel
>>> > <olivier.grisel at ensta.org> wrote:
>>> >>> I believe this `arch -i386`  only works as a prefix for Python.org
>>> Python, but I'm happy to be corrected.
>>> >>
>>> >> Then the following should work:
>>> >>
>>> >> arch -i386 python -c "import nose; nose.main()" sklearn
>>> >
>>> > Sorry - I should have been clear - this does work in selecting 32-bit
>>> > for the tests, using a nosetests installed into a Python.org Python
>>> > environment:
>>> >
>>> > arch -i386 nosetests sklearn
>>>
>>> Actually, I took the liberty of adding the OSX 32-bit tests to the
>>> wheel build tests, and scikit-learn 0.17.1 has one failure on 32-bit
>>> OSX:
>>>
>>> https://travis-ci.org/MacPython/scikit-learn-wheels/jobs/146127267#L11761
>>>
>>> ```
>>>
>>> ======================================================================
>>> ERROR: sklearn.tree.tests.test_tree.test_huge_allocations
>>> ----------------------------------------------------------------------
>>> Traceback (most recent call last):
>>>   File
>>> "/Users/travis/build/MacPython/scikit-learn-wheels/venv/lib/python3.5/site-packages/nose/case.py",
>>> line 198, in runTest
>>>     self.test(*self.arg)
>>>   File
>>> "/Users/travis/build/MacPython/scikit-learn-wheels/venv/lib/python3.5/site-packages/sklearn/tree/tests/test_tree.py",
>>> line 1032, in test_huge_allocations
>>>     assert_raises(MemoryError, clf.fit, X, y)
>>>   File
>>> "/Library/Frameworks/Python.framework/Versions/3.5/lib/python3.5/unittest/case.py",
>>> line 727, in assertRaises
>>>     return context.handle('assertRaises', args, kwargs)
>>>   File
>>> "/Library/Frameworks/Python.framework/Versions/3.5/lib/python3.5/unittest/case.py",
>>> line 176, in handle
>>>     callable_obj(*args, **kwargs)
>>>   File
>>> "/Users/travis/build/MacPython/scikit-learn-wheels/venv/lib/python3.5/site-packages/sklearn/tree/tree.py",
>>> line 348, in fit
>>>     max_leaf_nodes)
>>>   File "sklearn/tree/_tree.pyx", line 291, in
>>> sklearn.tree._tree.BestFirstTreeBuilder.__cinit__
>>>
>>> (/private/var/folders/gw/_2jq29095y7b__wtby9dg_5h0000gn/T/pip-j_xtpbj9-build/sklearn/tree/_tree.c:4461)
>>> OverflowError: Python int too large to convert to C long
>>> ```
>>>
>>> No failures for 32-bit Linux though.
>>>
>>> Best,
>>>
>>> Matthew
>>> _______________________________________________
>>> scikit-learn mailing list
>>> scikit-learn at python.org
>>> https://mail.python.org/mailman/listinfo/scikit-learn
>>>
>>
>>
> _______________________________________________
> scikit-learn mailing list
> scikit-learn at python.org
> https://mail.python.org/mailman/listinfo/scikit-learn
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.python.org/pipermail/scikit-learn/attachments/20160720/7f6e3f26/attachment.html>