Re: [Scikit-learn-general] GoS wish list updated

Hi Andy and others, I was just going through the Python Software Foundation's GSoC page, and it says that accepted projects need to be listed on the PSF page: http://wiki.python.org/moin/SummerOfCode/2012 Also, the project mentors apparently need to contact ArcRiley

Re: [Scikit-learn-general] GoS wish list updated

Hello everyone, Based on this discussion, i shall go ahead and craft a more detailed GSoC proposal for Bayesian Networks in Scikits-learn over the next few days. In the meanwhile, do keep your suggestions / concerns coming in :) Also, i feel including structure learning as part of the GSoC propos

Re: [Scikit-learn-general] GoS wish list updated

> The idea was to give each node potential manually. So there will be > a dict of marginal distributions that is as big as the graph. > Given that, I don't think that the DAG representation will play such a > big role. > I meant to say dict of conditional distributions. My bad. --

Re: [Scikit-learn-general] GoS wish list updated

On 03/18/2012 01:07 PM, Lars Buitinck wrote: > Op 18 maart 2012 08:00 heeft Shankar Satish > het volgende geschreven: > >> The first thing to decide would be how to represent the DAG. For that, i >> could either use something like py_graph, or roll my own, like so: >> >> dag = {'A': ['B', 'C'],

Re: [Scikit-learn-general] GoS wish list updated

Op 18 maart 2012 08:00 heeft Shankar Satish het volgende geschreven: > The first thing to decide would be how to represent the DAG. For that, i > could either use something like py_graph, or roll my own, like so: > > dag = {'A': ['B', 'C'], >            'B': ['C', 'D']} I don't think such a repre

Re: [Scikit-learn-general] GoS wish list updated

Ohh wow, so scikit-learn IS part of GSoC!?! I am so happy to know that! :) regards, shankar. On Sun, Mar 18, 2012 at 6:20 PM, Andreas Mueller wrote: > Hi Shankar. > > Thank you very much for offering to mentor my project! But sadly, it > > looks like sklearn is not in the GSoC organization lis

Re: [Scikit-learn-general] GoS wish list updated

Hi Shankar. > Thank you very much for offering to mentor my project! But sadly, it > looks like sklearn is not in the GSoC organization list :(... > Scikit-learn is not on the organzation list but it is part of GSoC through the Python Software Foundation.. > To answer your question about how i'll

Re: [Scikit-learn-general] GoS wish list updated

Hi Andy, Thank you very much for offering to mentor my project! But sadly, it looks like sklearn is not in the GSoC organization list :(... I still want to go ahead and implement bayes-nets in sklearn. However, i will also be applying to other places at GSoC, so i'll have to go slow on the bayes

Re: [Scikit-learn-general] GoS wish list updated

Hi Shankar. I am not really saying that graphical models are not useful tools. What I am questioning is whether it is reasonable to formulate these problems as general as you do and how they fit into the scikit-learn. By the way, there are several "Bayes nets" already implemented in the scikit-le

Re: [Scikit-learn-general] GoS wish list updated

Hi all, A bayesian network (BN) is a directed acyclic graph whose nodes represent random variables X1,...Xn. For each node Xi, we can encode a conditional probability distribution P(Xi | parents(Xi)), conditioned on the node's parents. The entire BN represents a joint probability distribution over

Re: [Scikit-learn-general] GoS wish list updated

FYR, here is a pretty good comparison of the existing graphical software http://www.cs.ubc.ca/~murphyk/Software/bnsoft.html . However, I am really concerned about the efficiency issue. On Thu, Mar 15, 2012 at 6:03 AM, Andreas Mueller wrote: > On 03/14/2012 10:52 PM, Alexandre Gramfort wrote: > >

Re: [Scikit-learn-general] GoS wish list updated

On 03/14/2012 10:52 PM, Alexandre Gramfort wrote: hi shankar, that sounds interesting to me. Can you come up with a few references and are you aware of existing implementations? I think the best reference is "Bishop: Machine Learning and Pattern Recognition". Though it is not used for classifi

Re: [Scikit-learn-general] GoS wish list updated

hi shankar, that sounds interesting to me. Can you come up with a few references and are you aware of existing implementations? Alex On Wed, Mar 14, 2012 at 6:18 PM, Shankar Satish wrote: > Thank you so much for your valuable inputs, Alejandro, Paolo, and Olivier. > Based on your feedback, i ha

Re: [Scikit-learn-general] GoS wish list updated

Thank you so much for your valuable inputs, Alejandro, Paolo, and Olivier. Based on your feedback, i have decided to drop the idea of RL-learning for sklearn. Instead, here is an alternate proposal: Supervised learning with Bayesian Networks. Bayesian networks provide highly interpretable models,

Re: [Scikit-learn-general] GoS wish list updated

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

Re: [Scikit-learn-general] GoS wish list updated

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

Re: [Scikit-learn-general] GoS wish list updated

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

Re: [Scikit-learn-general] GoS wish list updated

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

[Scikit-learn-general] GoS wish list updated

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