Le 13 mars 2012 07:53, Alejandro Weinstein
a écrit :
> On Tue, Mar 13, 2012 at 6:37 AM, Shankar Satish wrote:
>> Do you think my proposal about implementing reinforcement-learning
>> algorithms (subject line: "GSOC project idea: online learning algorithms")
>> is something that is well suited for
Hi scikit-learn community,
I'm interested in reducing a large adjacency matrix (1M x 1M) to 2
dimensions (1M x 2)
This is very similar to force-based graph embedding (drawing):
http://en.wikipedia.org/wiki/Force-based_algorithms_(graph_drawing)
But instead of using force for optimization, i wan
I tried with .99 and the plots show correctly.
When trying to get .98 to build, there were gcc errors,
which I didn't want to bother with.
Therefore it would be good to know which version the server
actually runs or if we could have access to the error logs.
Any ideas?
> I know that several of th
On 14 March 2012 08:54, Lars Buitinck wrote:
> 2012/3/13 Robert Layton :
> > Lars, you are right, it should have metric='precomputed' in it.
> > However by passing the distance matrix without a metric, the features
> become
> > "distance to point i", which act as sort of meta-features anyway,
> a
2012/3/13 Robert Layton :
> Lars, you are right, it should have metric='precomputed' in it.
> However by passing the distance matrix without a metric, the features become
> "distance to point i", which act as sort of meta-features anyway, allowing
> training to happen. This means that it works with
On 14 March 2012 08:05, Andreas wrote:
> Hi Lars.
> All I can say is that it worked for me by passing X directly:
>
> http://scikit-learn.org/dev/auto_examples/cluster/plot_cluster_comparison.html
>
> I'm deadlining right now, hopefully I have time to work on Olivier's
> "quadratic_fit" (or whate
Hi Lars.
All I can say is that it worked for me by passing X directly:
http://scikit-learn.org/dev/auto_examples/cluster/plot_cluster_comparison.html
I'm deadlining right now, hopefully I have time to work on Olivier's
"quadratic_fit" (or whatever) proposal afterward.
Cheers,
Andy
On 03/13/2012
Hi all,
A colleague approached me today asking how the scikit-learn DBSCAN
algorithm should be applied and I must admit that the documentation
and example was confusing even to me. The fit docstring says
X: array [n_samples, n_samples] or [n_samples, n_features]
Array of distances bet
Le 13 mars 2012 01:44, Emanuele Olivetti a écrit :
> Hi,
>
> I guess the correct link of the video is now:
> http://pyvideo.org/video/622/introduction-to-interactive-predictive-analytics
> or
> https://www.youtube.com/watch?v=Zd5dfooZWG4
Thanks, indeed the URL has changed.
--
Olivier
http://tw
On Tue, Mar 13, 2012 at 1:37 PM, Shankar Satish wrote:
> Hi Paolo and others,
>
> Do you think my proposal about implementing reinforcement-learning
> algorithms (subject line: "GSOC project idea: online learning algorithms")
> is something that is well suited for integration into scikit-learn? Do
Thanks Andy, have added to the issue.
On 13 March 2012 15:16, Andreas wrote:
> Hi Martin.
>
> This might be solved by issue 565
> https://github.com/scikit-learn/scikit-learn/issues/565.
> Maybe you should add there that keeping the estimators in memory also
> prevents pickling.
>
> Cheers,
> And
Hi Martin.
This might be solved by issue 565
https://github.com/scikit-learn/scikit-learn/issues/565.
Maybe you should add there that keeping the estimators in memory also
prevents pickling.
Cheers,
Andy
On 03/13/2012 03:11 PM, Martin Fergie wrote:
> Good Afternoon,
>
> I'm trying to pickle a
On Tue, Mar 13, 2012 at 6:37 AM, Shankar Satish wrote:
> Do you think my proposal about implementing reinforcement-learning
> algorithms (subject line: "GSOC project idea: online learning algorithms")
> is something that is well suited for integration into scikit-learn? Do you
> think it makes mor
2012/3/13 Frédéric Bastien :
> I would also add that probably ensemble are slower to train then
> prunned tree. In academic, this is not a too big problem, but in
> industries it can be important in some case.
And slower to make predictions too, I guess.
Mathieu
-
Good Afternoon,
I'm trying to pickle a class that contains a reference to a
GridSearchCV object and it is failing due to an instancemethod type. I
was under the impression that scikit-learn objects should be
picklable, is this correct? If so, would it be appropriate for me to
raise an issue?
Info
I would also add that probably ensemble are slower to train then
prunned tree. In academic, this is not a too big problem, but in
industries it can be important in some case.
Fred
On Tue, Mar 13, 2012 at 7:34 AM, Paolo Losi wrote:
> On Tue, Mar 13, 2012 at 12:09 PM, Andreas wrote:
>> On 03/13/2
Hi Paolo and others,
Do you think my proposal about implementing reinforcement-learning
algorithms (subject line: "GSOC project idea: online learning algorithms")
is something that is well suited for integration into scikit-learn? Do you
think it makes more sense to start a new scikit focussed on
I've added my personal item in the GoS 2012 wish list [1].
That list is a very good source of ideas. I would urge all GoS applicants
to look at it before formulating a proposal.
Thanks
Paolo
[1]
https://github.com/scikit-learn/scikit-learn/wiki/A-list-of-topics-for-a-google-summer-of-code-(gso
On Tue, Mar 13, 2012 at 12:09 PM, Andreas wrote:
> On 03/13/2012 12:11 PM, Paolo Losi wrote:
>> Since ensemble methods consistently outperform "traditional" tree building
>> (where variance is controlled by pruning), what are the advantages of
>> implementing
>> pruning in sklearn?
>>
>>
> I think
2012/3/13 Andreas
> On 03/13/2012 12:11 PM, Paolo Losi wrote:
> > Since ensemble methods consistently outperform "traditional" tree
> building
> > (where variance is controlled by pruning), what are the advantages of
> > implementing
> > pruning in sklearn?
> >
> >
> I think the idea would be to
On 03/13/2012 12:11 PM, Paolo Losi wrote:
> Since ensemble methods consistently outperform "traditional" tree building
> (where variance is controlled by pruning), what are the advantages of
> implementing
> pruning in sklearn?
>
>
I think the idea would be to have an easy to interpret model.
T
Since ensemble methods consistently outperform "traditional" tree building
(where variance is controlled by pruning), what are the advantages of
implementing
pruning in sklearn?
Paolo
N.B. The question is not directed specifically to Brain
but to GoS applicants and sklearn contributors.
On Tue,
Hi Andy,
Thanks for the information. I read the thread by Vikram, and would
gladly share my work with him. My particular interest is in model
selection for decision trees and at this stage I would like to test how
different prunings can improve generalisation.
Best,
Charanpal
Le 13/03/2012 11:18
On 13 March 2012 18:16, Andreas Mueller wrote:
> On 03/13/2012 07:49 AM, Olivier Grisel wrote:
> > Le 12 mars 2012 17:49, Robert Layton a écrit :
> >> I'll work off that template, and when I work out the details of the
> >> shrinking parameters (specifically which one is more in use), I'll
> bra
Decision trees tend to overfit, so they are most often used (unpruned) in a
forest. That said, I think it would be a useful contribution to our offering.
Brian
-Original Message-
From: Charanpal Dhanjal
Date: Tue, 13 Mar 2012 11:20:45
To:
Reply-To: scikit-learn-general@lists.sourcefo
Hi Charanpal.
In the recent GSoC-Thread, Vikram Kamath has proposed this and other
improvements
as a GSoC project. This idea is still in a pretty early stage, though.
In general, this is definitely a useful feature.
Cheers,
Andy
--
I noticed that decision trees are currently unpruned, and wondered if
anyone was working on this (or has been)? If not, I might implement
pruning myself.
Charanpal
--
Keep Your Developer Skills Current with LearnDevNow!
T
Hi,
I guess the correct link of the video is now:
http://pyvideo.org/video/622/introduction-to-interactive-predictive-analytics
or
https://www.youtube.com/watch?v=Zd5dfooZWG4
Best,
Emanuele
On 03/11/2012 09:52 AM, Olivier Grisel wrote:
> Hi all,
>
> The video of the tutorial I gave on Thursday
On 03/13/2012 07:49 AM, Olivier Grisel wrote:
> Le 12 mars 2012 17:49, Robert Layton a écrit :
>> I'll work off that template, and when I work out the details of the
>> shrinking parameters (specifically which one is more in use), I'll branch
>> and submit a PR.
> Great. I think the nearest centro
> Great. I think the nearest centroid is a very nice baseline classifier
> for sanity check: fast to fit, fast to predict, zero hyper-paramete
> and yet make reasonable assumption for many classification dataset (a
> good example of high bias, low variance, the opposite of deep decision
> trees or
30 matches
Mail list logo
