On Fri, May 8, 2015 at 9:40 AM, Skipper Seabold wrote:
> Hi,
>
> Is there anyway to specify the backend used by joblib? It's not
> obvious to me and this doesn't seem to work.
>
> from multiprocessing import set_start_method
> set_start_method("forks
Hi,
Is there anyway to specify the backend used by joblib? It's not
obvious to me and this doesn't seem to work.
from multiprocessing import set_start_method
set_start_method("forkserver")
I get a ton of "RuntimeError: context has already been set" exceptions.
On Python 3.4.2, sklearn 0
On Wed, May 21, 2014 at 8:00 AM, Lars Buitinck wrote:
> 2014-05-21 13:47 GMT+02:00 Olivier Grisel :
>> This is a great trick. We might want to get rid of our own partial
>> copy of CBLAS at some point.
>
> I remember Radim (gensim maint) describing some trouble with BLAS ABIs
> on a mailing list s
On Thu, Apr 17, 2014 at 9:56 AM, Mathieu Blondel wrote:
> It seems to me that you assume the order in which a dataset is laid out is
> meaningful. I think there are cases when this order might be completely
> artificial and not reflect the true distribution of the data. For me, the
> order is an i
On Fri, Mar 7, 2014 at 2:01 PM, Vijay Desai wrote:
> It is actually commodities futures data.
>
> Another way to handle missing data could be to estimate covariance
> matrix by ignoring the missing values and then determine eigenvectors
> of the covariance matrix to obtain principal components.
>
Hi,
There have been some implementations of Theil-Sen floating around for inclusion
in statsmodels, but no PRs yet. IMO it might fit in a little better in
statsmodels.robust than sklearn unless their are some aspects of Theil-Sen I'm
not familiar with.
Skipper
Sent from my mobile
> On Jan 10
On Mon, Dec 2, 2013 at 3:11 PM, Gael Varoquaux <
gael.varoqu...@normalesup.org> wrote:
> On Mon, Dec 02, 2013 at 02:50:45PM +, Robert Kern wrote:
> > > +1. "Import *" is a really really bad habit. And hacked up interactive
> > > environments (with crazy start up scripts) make it really hard to
On Mon, Dec 2, 2013 at 11:29 AM, Joel Nothman wrote:
> I think it's great that scikit-learn keeps its objects to modular
> namespaces, and doesn't litter one space as does numpy, pyplot, etc. Yet,
> when writing quick scripts it can be frustrating to have to import from
> pipeline, grid_search, li
On Fri, Nov 8, 2013 at 6:30 PM, Karol Pysniak wrote:
> Hi All,
>
> Has there any been discussion on adding some automated benchmarks for both
> speed and accuracy of the algorithms we have? I think it would very
> interesting if such a script could be automatically executed after every
> commit so
On Wed, Sep 11, 2013 at 10:35 PM, David Reed wrote:
> I apologize if this is a tad of topic, but I figured this was the best place
> to ask if any.
>
> When presented with a new research idea or project, I consistently deal with
> the issue of saving figures and results out to my current working d
On Tue, Sep 3, 2013 at 6:31 PM, Olivier Grisel wrote:
> 2013/9/3 Gael Varoquaux :
>> On Tue, Sep 03, 2013 at 02:39:04PM -0400, Skipper Seabold wrote:
>>> from numpy.distutils.system_info import get_info
>>> blas_info = get_info('blas_opt', 0)
>>> bla
Hi,
I'm looking at the build of scikit-learn for guidance on some
statsmodels improvements, and I'm wondering about this in the
setup.py. What could this do?
from numpy.distutils.system_info import get_info
blas_info = get_info('blas_opt', 0)
blas_info.get('define_macros')
My numpy is linked aga
On Mon, Jul 8, 2013 at 1:56 PM, Olivier Grisel wrote:
> I am not sure what `sf` stands for, so it's probably a poor choice as
> we should not make the assumption that the library users will be well
> versed in stats acronyms.
Survival function. 1 - cdf but often with better precision in the
tails
On Mon, Jul 8, 2013 at 1:20 PM, Bertrand Thirion
wrote:
>
> De: "Jacob Vanderplas"
> À: scikit-learn-general@lists.sourceforge.net
> Envoyé: Dimanche 7 Juillet 2013 19:10:38
> Objet: [Scikit-learn-general] Defining a Density Estimation Interface
>
>
> Hi,
> I've been working on a big rewrite of t
Hi,
Does anyone have any code for computing rotations of components after
PCA or FactorAnalysis, etc. E.g., varimax?
Thanks,
Skipper
--
Precog is a next-generation analytics platform capable of advanced
analytics on sem
On Tue, Jan 22, 2013 at 4:14 PM, Lars Buitinck wrote:
> 2013/1/22 Andreas Mueller :
> > I think a system like that will be a good way to go forward.
>
> Me too -- I've been thinking about something like this for a
> scikit-sequence-learn project. Not that I have the time to start one,
> though :)
On Thu, Aug 2, 2012 at 11:05 AM, Olivier Grisel
wrote:
> 2012/8/2 Olivier Grisel :
>> 2012/8/2 Jim Vickroy :
>>> On 8/2/2012 8:27 AM, Brian Holt wrote:
Thanks Jim,
Could you try it again with
X = np.array([[0]])
Note the double "[" bracket - this is what causes t
On Sat, Jul 28, 2012 at 3:13 PM, Zach Bastick wrote:
> The docs do not indicate whether there is anyway to do a stepwise
> regression in scikit-learn or in Python.
> All there seems to be is linear_model.LinearRegression().
>
> This function outputs resulting x-values/beta-values/coefficents that
On Sat, Jun 23, 2012 at 7:42 PM, Daniel Duckworth wrote:
> Hello everyone,
>
> I would like to ask for those willing to begin reviewing my new
> sklearn.kalman module (found
> here: https://github.com/scikit-learn/scikit-learn/pull/862 ). It is a
> module that implements the Kalman Filter, Kalman
On Mon, May 14, 2012 at 4:55 PM, Daniel Duckworth wrote:
> Hello everyone,
>
> I noticed that scikit-learn (and Python in general) seems to be missing a
> decent module for State Space Models. State Space Models are a type of
> generative model wherein one attempts to estimate the hidden state of
On Thu, Dec 8, 2011 at 10:34 AM, Yaroslav Halchenko wrote:
> it might be handy to know which particular features of 2.6 are you
> aiming to use which aren't in 2.5... ?
>
I'm not a contributor to scikit-learn, but I would use the
getter/setter/deleter decorators for properties, if I didn't have t
On Fri, Oct 28, 2011 at 11:31 AM, Olivier Grisel
wrote:
> 2011/10/28 Skipper Seabold :
>> On Fri, Oct 28, 2011 at 11:05 AM, Alexandre Passos
>> wrote:
>>> On Fri, Oct 28, 2011 at 11:02, Olivier Grisel
>>> wrote:
>>>> 2011/10/28 Kenneth C. Arnol
On Fri, Oct 28, 2011 at 11:05 AM, Alexandre Passos
wrote:
> On Fri, Oct 28, 2011 at 11:02, Olivier Grisel
> wrote:
>> 2011/10/28 Kenneth C. Arnold :
>>> I just implemented Latent Dirichlet Allocation with collapsed Gibbs
>>> sampling and made a demo on 20 Newsgroups. If there's interest in
>>> h
On Thu, Oct 13, 2011 at 11:10 PM, Robert Layton wrote:
> I'm working on adding Adjusted Mutual Information, and need to calculate the
> Mutual Information.
> I think I have the algorithm itself correct, except for the fact that
> whenever the contingency matrix is 0, a nan happens and propogates t
On Fri, Sep 23, 2011 at 2:03 PM, Mathieu Blondel wrote:
>
> On Sat, Sep 24, 2011 at 2:57 AM, Alexandre Gramfort
> wrote:
> > I feel there is 2 different things: on one side the package name in
> > pypi, the url and the doc
> > and the other side the import name. scikit-learn will continue to be
>
On Fri, Sep 23, 2011 at 9:04 AM, Nelle Varoquaux
wrote:
> Out of curiosity, did any of the other scikit move to a new name prefixed
> with sk? Is this something the scikits community is willing to do long term?
> Cheers,
I can't say for sure, but I'm fairly certain that statsmodels is
planning to
26 matches
Mail list logo
