I agree with Andy that so far it is not that much of a bother, and the
pain created by blocking such mail might exceed the gain.
Of course, if we start having too many of such mails...
G
--
Free Next-Gen Firewall Hardwar
Someone registered for the ml and probably posted this by hand.
So we can only go by the content. The ml is hosted by sourceforge, I
would not expect
them do do any custom / advanced spam filtering.
It is not that big a problem at the moment, though, is it?
--
On Mon, Feb 11, 2013 at 11:21:14PM +0100, Lars Buitinck wrote:
> What's the backporting procedure? Just copy the files?
And adjust test, and chmod.
> We could put it in the py3k branch...
No, I want this in the master branch, because we need to benefit from an
important bugfix. I'll do it tomorr
2013/2/11 Gael Varoquaux :
> I need to backport it to scikit-learn. I have been waiting a been for
> feedback on the latest beta release that I did recently.
What's the backporting procedure? Just copy the files?
We could put it in the py3k branch...
--
Lars Buitinck
Scientific programmer, ILPS
> Btw, I forget to report one important thing: I did not touch the
> joblib code at all. It didn't seem wise to me to merge Olivier's 11
> months old commits with an updated version of an external dependency.
> Should we tackle joblib py3 compatibility outside the scikit-learn
> project first?
We
2013/2/11 Kenneth C. Arnold :
> Unfortunately, Scipy 0.11.0 is broken on Python 3.3.
> http://projects.scipy.org/scipy/ticket/1739
There goes the "skip to 3.3" advice :(
Btw, I forget to report one important thing: I did not touch the
joblib code at all. It didn't seem wise to me to merge Olivier
Interesting to see this thread revived! FYI I've made hyperopt a lot
friendlier since that original posting.
http://jaberg.github.com/hyperopt/
pip install hyperopt
1. It has docs.
2. The minimization interface is based on an fmin() function, that
should be pretty accessible.
3. It can be instal
ML4H is interesting, but I don't know if everything inside can be done as
is in Python. Wii It Python only handles the first chapter or so :/ Thus
there is a public for a ML4P book :(
Or perhaps if we make each of us one chapter on a specific topic and
aggregate the total in a book?
2013/2/11 Fre
Thanks to all for your speedy replies! I definitely appreciate that there
are non-trivial issues in committing to a project like that. I've got a
copy of ML4H, and will thus follow @ogrisel's suggestion (thanks!)
Best,
Fred.
On 11 February 2013 09:40, Jacob Vanderplas wrote:
> Last week an O'R
2013/2/11 Kenneth C. Arnold :
> tl;dr: Try Python 3.2 with MacPorts.
>
> Unfortunately, Scipy 0.11.0 is broken on Python 3.3.
> http://projects.scipy.org/scipy/ticket/1739
Indeed, that might be the issue I am facing.
> This is fixed in their master branch. I just made a successful build of that
>
tl;dr: Try Python 3.2 with MacPorts.
Unfortunately, Scipy 0.11.0 is broken on Python 3.3.
http://projects.scipy.org/scipy/ticket/1739
This is fixed in their master branch. I just made a successful build of
that branch on OS X
10.8:https://trac.macports.org/ticket/37400#comment:15There may be
some
I'll be in northern Europe (Denmark & Germany) for vacation during the
first part of April. It's unlikely I'll be able to change my itinerary to
get down to Paris afterward, but if the sprint is the week of April 14th I
might be able to work something out. Don't plan it around me, though -
it's p
Last week an O'Reilly editor I've worked with got in touch with me asking
if I'd be willing to write such a book. I can't consider it because of my
current commitments, but if anyone from the dev team (perhaps a few folks
working together) is interested, I'd be happy to forward them your
informati
I offered a table of contents, but then no answer from them afterwards :/
As Andreas said, there is not much financial interest to write such a book,
you have to work on it for 3 months full time or 6 months all evenings, and
there are not a lot of people that can afford such a commitment! In a wa
2013/2/11 Matthieu Brucher :
> Hi Fred,
>
> Packt asked some people here to write something like that. But I don't know
> where this project is, they stopped communicating (at least) with me.
I don't know what is the status of the Packt book (I don't even know
if anybody is working on such a book)
Hi Fred.
Afaik no-one is working on a book currently, though some people have
showed interest.
Most devs are researchers or have full-time jobs. Writing a book would
probably mean quitting jobs
for a couple of month, stalling research and basically not making any
money (From what I read, writi
Hi Fred,
Packt asked some people here to write something like that. But I don't know
where this project is, they stopped communicating (at least) with me.
Cheers,
Matthieu
2013/2/11 Fred Mailhot
> Hi list,
>
> Is anyone working on a book showcasing scikit-learn? I'm thinking
> something alon
Hi list,
Is anyone working on a book showcasing scikit-learn? I'm thinking something
along the lines of "Mahout In Action", that would showcase each of the
parts of scikit-learn and provide a dead-tree reference with a lot of
worked-out examples. I suppose it would make sense to wait for a 1.0
rel
2013/2/11 Lars Buitinck :
> 2013/2/11 Olivier Grisel :
>> But I am also having a hard time compiling a version of scipy that
>> builds against a matching version of numpy under python 3 for some
>> reason. I did not have that problem last time I tried so I might have
>> an issue with my python 3 in
2013/2/11 Olivier Grisel :
> But I am also having a hard time compiling a version of scipy that
> builds against a matching version of numpy under python 3 for some
> reason. I did not have that problem last time I tried so I might have
> an issue with my python 3 install...
You're a Mac user, rig
2013/2/11 Lars Buitinck :
> 2013/2/10 Olivier Grisel :
>> I see that setup.py still does 2to3 translation in the build folder in
>> that branch. Have you tried to get rid of it?
>
> I didn't even try running it under Py3 yet. The problem is that
> mock_urllib2 is replaced by something called Urlope
2013/2/11 Gael Varoquaux :
> On Sat, Feb 09, 2013 at 11:04:19AM +, Vlad Niculae wrote:
>> Is (the end of) March completely out of the question?
>
> As far as I am concerned, yes.
Same here: I will already spend most of March in Santa Clara for PyCon.
--
Olivier
http://twitter.com/ogrisel - h
On Sat, Feb 09, 2013 at 11:04:19AM +, Vlad Niculae wrote:
> Is (the end of) March completely out of the question?
As far as I am concerned, yes.
G
--
Free Next-Gen Firewall Hardware Offer
Buy your Sophos next-gen fir
On Mon, Feb 11, 2013 at 4:39 PM, Wei LI wrote:
> In my point of view, to optimize the hyperparameters can not use standard
> optimization techniques(or else it will become a parameters and cannot be
> set empirically?) So some heuristic in brute force searching maybe a good
> idea. I am thinking a
indeed SVM (libsvm / liblinear) could benefit also from a path strategy.
Alex
On Mon, Feb 11, 2013 at 8:39 AM, Wei LI wrote:
> In my point of view, to optimize the hyperparameters can not use standard
> optimization techniques(or else it will become a parameters and cannot be
> set empirically?)
25 matches
Mail list logo
