On Tue, Jan 31, 2012 at 05:09:54PM +0100, Gael Varoquaux wrote:
> On Tue, Jan 31, 2012 at 05:05:53PM +0100, Lars Buitinck wrote:
> > I don't have a NumPy 2 installation and I haven't followed its
> > development closely. Could you open an issue for this?
> https://github.com/scikit-learn/scikit-le
On Tue, Jan 31, 2012 at 05:05:53PM +0100, Lars Buitinck wrote:
> I don't have a NumPy 2 installation and I haven't followed its
> development closely. Could you open an issue for this?
https://github.com/scikit-learn/scikit-learn/issues/600
I'll do better: I'll try to fix the problem when I get t
2012/1/23 Gael Varoquaux :
> On Sat, Jan 21, 2012 at 03:49:24PM +0100, Lars Buitinck wrote:
>> This is very strange; I get no such error.
>
> I have numpy 2 (dev). The rules for strides in array creation have
> changed.
I don't have a NumPy 2 installation and I haven't followed its
development clo
On Sat, Jan 21, 2012 at 03:49:24PM +0100, Lars Buitinck wrote:
> This is very strange; I get no such error.
I have numpy 2 (dev). The rules for strides in array creation have
changed.
Gael
--
Try before you buy = See our
On Sat, Jan 21, 2012 at 11:15 PM, Gael Varoquaux
wrote:
> sklearn.svm.liblinear.decision_function_wrap
> (sklearn/svm/liblinear.c:2775)
> ValueError: ndarray is not Fortran contiguous
I think this error message sometimes also appears if the array is 2d
instead of 1d.
Mathieu
--
2012/1/21 Gael Varoquaux :
> ==
> ERROR: sklearn.svm.tests.test_svm.test_liblinear_set_coef
> --
> Traceback (most recent call last):
> File "/usr/lib/pymodules/p
I haven't followed the recent changes to the SVM array ordering, but I
must say that I have the following test failing on my box:
==
ERROR: sklearn.svm.tests.test_svm.test_liblinear_set_coef
---
