Hi Andy,
I made the change and wrote a test for it but I'm having some problems
with testing now. To be specific, BLAS seems to fail on my machine in
the standard test suite (details below).
I wonder if this is something to do with my numpy configuration. I'm
using a standard installation of numpy 1.7.1, without any changes to
site.cfg, and it seems that it isn't using the local ATLAS installation.
The sklearn test script compained about this but didn't raise an error:
Atlas (http://math-atlas.sourceforge.net/) libraries not found.
Directories to search for the libraries can be specified in the
numpy/distutils/site.cfg file (section [atlas]) or by setting
the ATLAS environment variable.
warnings.warn(AtlasNotFoundError.__doc__)
Michal
Failing test:
======================================================================
FAIL: Check fast dot blas wrapper function
----------------------------------------------------------------------
Traceback (most recent call last):
File
"/vol/vipdata/users/mpr06/VIRTUALENV/biomedia07x/lib/python2.6/site-packages/nose/case.py",
line 197, in runTest
self.test(*self.arg)
File
"/vol/vipdata/users/mpr06/sklearn-dev/0.15dev/scikit-learn/sklearn/utils/tests/test_extmath.py",
line 360, in test_fast_dot
assert_almost_equal(C, C_)
File
"/vol/vipdata/users/mpr06/VIRTUALENV/biomedia07x/lib/python2.6/site-packages/numpy/testing/utils.py",
line 452, in assert_almost_equal
return assert_array_almost_equal(actual, desired, decimal, err_msg)
File
"/vol/vipdata/users/mpr06/VIRTUALENV/biomedia07x/lib/python2.6/site-packages/numpy/testing/utils.py",
line 812, in assert_array_almost_equal
header=('Arrays are not almost equal to %d decimals' % decimal))
File
"/vol/vipdata/users/mpr06/VIRTUALENV/biomedia07x/lib/python2.6/site-packages/numpy/testing/utils.py",
line 645, in assert_array_compare
raise AssertionError(msg)
AssertionError:
Arrays are not almost equal to 7 decimals
(mismatch 25.0%)
x: array([[ 1.83460963, 2.10736465],
[ 1.29582286, 1.32504284]], dtype=float32)
y: array([[ 1.83460951, 2.10736442],
[ 1.29582274, 1.32504272]], dtype=float32)
>> raise AssertionError('\nArrays are not almost equal to 7
decimals\n\n(mismatch 25.0%)\n x: array([[ 1.83460963, 2.10736465],\n
[ 1.29582286, 1.32504284]], dtype=float32)\n y: array([[
1.83460951, 2.10736442],\n [ 1.29582274, 1.32504272]],
dtype=float32)')
----------------------------------------------------------------------
On 11/11/13 15:56, scikit-learn-general-requ...@lists.sourceforge.net wrote:
> Subject:
> Re: [Scikit-learn-general] Random forest with zero features
> From:
> Andy <t3k...@gmail.com>
> Date:
> 11/11/13 04:37
>
> To:
> <scikit-learn-general@lists.sourceforge.net>
>
>
> Hi Michal.
> Thanks for wanting to work on this.
> Could you please open an issue? That makes it easier to track the progress.
> Could you also post the traceback / error from your example script there?
>
> Thanks,
> Andy
>
> On 11/08/2013 09:56 AM, Michal Romaniuk wrote:
>> Did anyone work on this problem (exceptions raised by classifiers in
>> grid search) since? I would be happy to do some work to fix this
>> problem, but would need some advice.
>>
>> It seems to me like the easiest way around the issue is to wrap the call
>> to clf.fit() in a try statement and catch the exception if one is
>> raised. In such case, there are (at least) 2 questions:
>>
>> 1. Should it catch any exception or just a specific type?
>> 2. What should go into the results table for the failing grid point?
>> NaN? Zero?
>>
>> Thanks,
>> Michal
>>
>
>
------------------------------------------------------------------------------
November Webinars for C, C++, Fortran Developers
Accelerate application performance with scalable programming models. Explore
techniques for threading, error checking, porting, and tuning. Get the most
from the latest Intel processors and coprocessors. See abstracts and register
http://pubads.g.doubleclick.net/gampad/clk?id=60136231&iu=/4140/ostg.clktrk
_______________________________________________
Scikit-learn-general mailing list
Scikit-learn-general@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/scikit-learn-general
