Cheers,
Adrien
Le 29/10/2012 19:54, Stéfan van der Walt a écrit :
> On Mon, Oct 29, 2012 at 11:10 AM, Alexandre Gramfort
> wrote:
>> DTW in O(n^2) is super easy to implement. A good exercise with cython !
>> Some implementations exist in O(n) although they give only good
n some trivial 2D cases. See here:
https://gist.github.com/1989253
Remember, it's very fresh code, so I put it here just to give some
concrete matter to the discussion.
Cheers,
Adrien
>
> Gael
>
> --
>
es. Therefore, users can enjoy batch MLR until the
>> stochastic version is available, at which point everyone will switch to
>> SGD of course ;-).
> In general I agree that it would be a nice to have an "exact"
> solver instead of only a SGD on
Le 06/03/2012 20:23, Andreas a écrit :
> On 03/06/2012 08:17 PM, Adrien wrote:
>> Le 06/03/2012 19:19, Andreas Mueller a écrit :
>>
>>> Hi Adrien.
>>> Thanks for the offer and thanks for converting people from the dark side ;)
>>>
>>> I
Le 06/03/2012 19:19, Andreas Mueller a écrit :
> Hi Adrien.
> Thanks for the offer and thanks for converting people from the dark side ;)
>
> I'm not sure this is the way to go, though.
> There is already quite efficient SGD code in sklearn and this should
> probably
>
tell my advisor ;-)
As you can imagine, the result in such a short time is a bit
unsightly... For now, it's very rough and I barely tested in a simple 2D
case. I hope I can clean it up a bit, add tests and then make a pull
request. I'll let you know about my progress.
Cheers,
Adrien
>
ssfully in my own code for some time.
I cleaned it up a little bit and added a test. The pull request is here:
https://github.com/scikit-learn/scikit-learn/pull/649 and would have been
there a bit earlier if not for the weird DDOS att
lso some
voodoo engineering involved (line search...). It's doable, but it takes
some effort (for a reward I didn't get...).
My 2 cents,
Adrien
On 02/20/2012 03:36 PM, Nicholas Pilkington wrote:
I was wondering if there were any immediate plans to implement
Multiple Kernel SVMs in
quot; type contribute in a
relatively painless way for them? I believe the number of such users is
proportional to how easy it is to use the library, so with sklearn I'm
sure there are more than a few!
Sorry for being a
Le 23/01/2012 11:34, Olivier Grisel a écrit :
> 2012/1/23 Andreas:
>> On 01/23/2012 11:28 AM, Adrien wrote:
>>> Hello everyone,
>>>
>>> A quick question: why not use Nystrom instead?
>>>
>> That was on my GSoC wish list ;)
>> The application
ra Malik's PAMI
paper [1]).
Cheers,
Adrien
[1] Spectral grouping using the Nystrom method,
Fowlkes, C. and Belongie, S. and Chung, F. and Malik, J.
PAMI 2004
Le 23/01/2012 10:20, Mathieu Blondel a écrit :
> On Mon, Jan 23, 2012 at 6:06 PM, Andreas wrote:
>
>> It might be as easy
e, you probably need to
recompute kernel evaluations unless you cache them. Furthermore, the
best C values on my problems are high ones and, therefore, I have almost
all points as SV. That was also true in my experience on the Pascal VOC
challenge with RBF chi-square kernels on Bag-of-Features.
se vectors if necessary, but my guess
is that you would need really huge and very sparse vectors for it to be
worth your while.
My 1.99 cents (not really worth 2).
Adrien
> D. Sculley proposed a simpler algorithm based on a binary search in
> his web scale k-means paper. The linear-time
13 matches
Mail list logo
