I have Python 2.6.6, I installed scikit-learn-0.13.1.win32-py2.6.exe
from sourceforge.net/projects/scikit-learn/files/
whyen I run nosetests sklearn --exe I get:
Ran 591 tests in 19.176s
FAILED (SKIP=1, errors=35)
when I run python -c "import sklearn; sklearn.test()" I get
Ran 609 tests in 20.528
No, I installed numpy from the official website (with scipy).
2013/2/27 Andreas Mueller
> On 02/27/2013 12:08 PM, Vlad Niculae wrote:
> > The second run ShNaYkHs posted looks like a good install though,
> > despite the test failure.
> Are these your binaries on that website?
> It also has 64bit
No, my binaries are only on sourceforge and pypi.
Vlad
On Wed, Feb 27, 2013 at 5:46 PM, Andreas Mueller
wrote:
> On 02/27/2013 12:08 PM, Vlad Niculae wrote:
>> The second run ShNaYkHs posted looks like a good install though,
>> despite the test failure.
> Are these your binaries on that website?
On 02/27/2013 12:08 PM, Vlad Niculae wrote:
> The second run ShNaYkHs posted looks like a good install though,
> despite the test failure.
Are these your binaries on that website?
It also has 64bit versions.
They say they require the mkl numpy install.
Shnaykhs: did you install the mkl numpy from t
The second run ShNaYkHs posted looks like a good install though,
despite the test failure.
On Wed, Feb 27, 2013 at 11:06 AM, Vlad Niculae wrote:
> I built the binaries, is this because of the version of numpy I
> compiled against?
>
> On Tue, Feb 26, 2013 at 4:51 PM, ShNaYkHs ShNaYkHs wrote:
>>
I built the binaries, is this because of the version of numpy I
compiled against?
On Tue, Feb 26, 2013 at 4:51 PM, ShNaYkHs ShNaYkHs wrote:
> Now I re-installed numpy, scipy matplotlib and scikit-learn from
> http://www.lfd.uci.edu/~gohlke/pythonlibs/#scikit-learn , I choose the
> versions ending
Now I re-installed numpy, scipy matplotlib and scikit-learn from
http://www.lfd.uci.edu/~gohlke/pythonlibs/#scikit-learn , I choose the
versions ending with ".win32-py2.6.exe". Now I get the following new
failure log:
http://pastebin.com/MggL5Lyz
Ran 1601 tests in 102.300s
FAILED (SKIP=15, failures
Well, I see that I have numpy 1.6.0 and scipy 0.9.0 installed
2013/2/26 Olivier Grisel
> 2013/2/26 ShNaYkHs ShNaYkHs :
> > I used scikit-learn-0.13.1.win32-py2.6.exe that I downloaded from this
> URL
> > http://sourceforge.net/projects/scikit-learn/files/
> > P.S. I have windows seven 64 bits, o
How can I verify which versions of numpy and scipy are installed, someone
else already installed them on my computer the last year ..
2013/2/26 Olivier Grisel
> 2013/2/26 ShNaYkHs ShNaYkHs :
> > I used scikit-learn-0.13.1.win32-py2.6.exe that I downloaded from this
> URL
> > http://sourceforge.n
2013/2/26 ShNaYkHs ShNaYkHs :
> I used scikit-learn-0.13.1.win32-py2.6.exe that I downloaded from this URL
> http://sourceforge.net/projects/scikit-learn/files/
> P.S. I have windows seven 64 bits, on Intel i7 processor
Alright, how did you install numpy and scipy and which version / from
which si
I used scikit-learn-0.13.1.win32-py2.6.exe that I downloaded from this URL
http://sourceforge.net/projects/scikit-learn/files/
P.S. I have windows seven 64 bits, on Intel i7 processor
2013/2/26 Olivier Grisel
> How did you install sklearn? Did you use some precompiled windows
> binaries? If so w
How did you install sklearn? Did you use some precompiled windows
binaries? If so which version from which URL?
--
Everyone hates slow websites. So do we.
Make your web apps faster with AppDynamics
Download AppDynamics Lit
2013/2/26 ShNaYkHs ShNaYkHs :
> This is the full log of the test (nosetests sklearn --exe):
> http://pastebin.com/qJZwedTL
Looks like some integer type mismatches with MSVC -- I was afraid this
was going to happen at some point. Who built the Windows binaries?
--
Lars Buitinck
Scientific program
whyen I run nosetests sklearn --exe I get:
Ran 591 tests in 19.176s
FAILED (SKIP=1, errors=35)
when I run python -c "import sklearn; sklearn.test()" I get
Ran 609 tests in 20.528s
FAILED (SKIP=1, errors=51, failures=1)
Is that normal ? Can I use scikit-learn anyway ?
This is the full log of the
14 matches
Mail list logo