Hi Michael,
my answer is more a warning than an answer.
Beyond 2 classes the "best" would be to use something like L1/L2 mixed norm
with a real multinomial loss. Unfortunately we don't have it in the scikit.
Also looking at the weights of a sparse logistic or L2 model (logistic or SVM)
working w
Hi all,
I have a LogisticRegression model I'm training in a 3-class scenario. I'd
like to examine the coefficients for the models. As the default for
LogisitcRegression is to do one-vs-all classification, my clf.coef_ array
is shape 3 x nfeat.
My question is how to interpret the sign of the coe
Hi Olivier,
Regarding your concern about the efficiency of the algorithms, as far as I
can tell, as the field is developing rapidly, the current state-of-the-art
RPCA algorithm is actually very scalable. In a nutshell, the cost of the
RPCA is about the cost of 20s SVD. For a 1000x1000 matrix, it i
On Mon, Mar 5, 2012 at 12:22 AM, xinfan meng wrote:
> LaSVM seems like an on-line SVM training method. How does it compare to
> MIRA?
You could use LaSVM in a strict online setting as in MIRA if you
really wanted to but that's not the point of LaSVM in my opinion [*].
For me, the main innovation
LaSVM seems like an on-line SVM training method. How does it compare to
MIRA?
On Sun, Mar 4, 2012 at 11:11 PM, Mathieu Blondel wrote:
> On Sun, Mar 4, 2012 at 8:10 PM, Olivier Grisel
> wrote:
>
> > We should also add the scalable SVM / LaSVM proposal, if someone wants
> > to volunteer.
>
> I'm p
On Sun, Mar 4, 2012 at 8:10 PM, Olivier Grisel wrote:
> We should also add the scalable SVM / LaSVM proposal, if someone wants
> to volunteer.
I'm planning to work on LaSVM. Don't expect a PR too soon but it's
definitely on my mid-term todo list.
Mathieu
---
Hi Shishir.
It is actually 0.10, meaning the one after 0.9. Not really a decimal
number ;)
We hope to get to 1.0 soon but didn't want to do a major release just yet.
Cheers,
Andy
On 03/04/2012 01:47 PM, Shishir Pandey wrote:
Hi
I think the version of the present stable release is suppose to
Hi
I think the version of the present stable release is suppose to be 1.0
instead of 0.1 as mentioned on the website.
Thanks,
--sp
--
Virtualization & Cloud Management Using Capacity Planning
Cloud computing makes use of
2012/3/4 Andreas Mueller :
> On 03/04/2012 12:10 PM, Olivier Grisel wrote:
>> I started structuring this discussion on the wiki:
>>
>>
>> https://github.com/scikit-learn/scikit-learn/wiki/A-list-of-topics-for-a-google-summer-of-code-%28gsoc%29-2012
>>
>> We should also add the scalable SVM / La
On 03/04/2012 12:10 PM, Olivier Grisel wrote:
> I started structuring this discussion on the wiki:
>
>
> https://github.com/scikit-learn/scikit-learn/wiki/A-list-of-topics-for-a-google-summer-of-code-%28gsoc%29-2012
>
> We should also add the scalable SVM / LaSVM proposal, if someone wants
> to
I started structuring this discussion on the wiki:
https://github.com/scikit-learn/scikit-learn/wiki/A-list-of-topics-for-a-google-summer-of-code-%28gsoc%29-2012
We should also add the scalable SVM / LaSVM proposal, if someone wants
to volunteer.
--
Olivier
-
2012/3/4 Kerui Min :
> Hi all,
>
> I'm a graduate student at UIUC who is currently pursuing the research work
> related to low-rank matrices recovery & Robust PCA. This kind of techniques
> turned out to be very useful in applications in different areas (e.g.,
> matrix completion for the Netflix-li
Hi all,
I'm a graduate student at UIUC who is currently pursuing the research work
related to low-rank matrices recovery & Robust PCA. This kind of techniques
turned out to be very useful in applications in different areas (e.g.,
matrix completion for the Netflix-like recommendation systems, image
13 matches
Mail list logo
