Hi Mohit
I believe this proposal is very welcome. Just pointing out that it's
not about removing the dependency on liblinear: we will still depend
on it for the LinearSVC and its implementation of binary logistic
regression. It's just that we are completely missing a true
multinomial logistic
Hi everyone,
I'm wondering if scikit would wish to add reinforcement learning (rl)
modules (its basically machine learning). Since I'm implementing some of
its algorithms, it would be great if we can share a repository to test
different rl algorithms as a 'black box' to different reinforcement
Hi Issam.
I don't think adding RL is a good idea, as it requires a whole different
API.
Also, we try to focus on the current methods and provide an interface,
working up to a 1.0 release in the not-so-far-future.
Have you looked at PyBrain? I think the already implement some RL
algorithms.
On 05/01/2013 10:50 AM, Vlad Niculae wrote:
Hi Mohit
I believe this proposal is very welcome. Just pointing out that it's
not about removing the dependency on liblinear: we will still depend
on it for the LinearSVC and its implementation of binary logistic
regression. It's just that we are
Hi Andy,
That is true, RL requires a whole new API, just didn't find PyBrain
friendly enough :).
Thanks,
--Issam
On 5/1/2013 12:56 PM, Andreas Mueller wrote:
Hi Issam.
I don't think adding RL is a good idea, as it requires a whole different
API.
Also, we try to focus on the current methods
My library for large-scale linear classification / regression, lightning,
already supports multiclass logistic regression (by SGD and by coordinate
descent). It also includes many other goodies, including my upcoming paper
on sparse multiclass classification. For all my Cython code, I used a
My library for large-scale linear classification / regression, lightning,
already supports multiclass logistic regression (by SGD and by coordinate
descent). It also includes many other goodies, including my upcoming paper
on sparse multiclass classification. For all my Cython code, I used a
hi,
Is there a reason why this option could not be included in LassoCV?
lack of time :)
Apparently someone else had the same question:
http://stackoverflow.com/questions/14324976/what-non-negative-linear-models-are-supported-planned-in-scikit-learn
I hope it can be included in a future
I'm looking to do regularized regression with a non-negativity
constraint. Scikit-learn's Lasso method has a 'positive' option that
applies this constraint, so it seems like a good tool for the job. At
the same time, the automatic tuning of the regularization parameter that
is offered by
Hi all,
I spend a couple of hours fixing the build infrastructures:
- the travis CI mldata failures should be fixed by:
https://github.com/scikit-learn/scikit-learn/commit/0859653e096e7cc48fbdb1e56564482c92efc120
https://github.com/scikit-learn/scikit-learn/issues/1417
- the jenkins build was
2013/5/1 Olivier Grisel olivier.gri...@ensta.org:
I spend a couple of hours fixing the build infrastructures:
- the travis CI mldata failures should be fixed by:
https://github.com/scikit-learn/scikit-learn/commit/0859653e096e7cc48fbdb1e56564482c92efc120
2013/5/1 Lars Buitinck l.j.buiti...@uva.nl:
Here is the pep8 violations report:
https://jenkins.shiningpanda-ci.com/scikit-learn/job/python-2.7-numpy-1.6.2-scipy-0.10.1/violations/
A lot of this is joblib and six, though. Can we turn off pep8
reporting for sklearn/externals? I like to keep
On Wed, May 01, 2013 at 06:19:34PM +0200, Olivier Grisel wrote:
I spend a couple of hours fixing the build infrastructures:
Wow! Thank you so much. These are well-spent hours.
G
--
Introducing AppDynamics Lite, a free
Thanks for solving the Travis bug :)
On 1 May 2013 21:15, Gael Varoquaux gael.varoqu...@normalesup.org wrote:
On Wed, May 01, 2013 at 06:19:34PM +0200, Olivier Grisel wrote:
I spend a couple of hours fixing the build infrastructures:
Wow! Thank you so much. These are well-spent hours.
G
14 matches
Mail list logo
