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
Op 18 maart 2012 21:10 heeft Alexandre Gramfort
het volgende geschreven:
>> Another minor variation: make a second libsvm wrapper constructor that
>> only uses alpha, never C.
>> e.g.: svm = SVCa(1e-3)
>
> I am -1 on that.
>
> that would probably confuse users and won't prevent them from using an
> Another minor variation: make a second libsvm wrapper constructor that
> only uses alpha, never C.
> e.g.: svm = SVCa(1e-3)
I am -1 on that.
that would probably confuse users and won't prevent them from using an
SVC with GridSearchCV with scale_C=False.
Alex
--
On Sun, Mar 18, 2012 at 12:22 PM, Andreas wrote:
> On 03/18/2012 05:07 PM, James Bergstra wrote:
>> On Sat, Mar 17, 2012 at 11:55 PM, Mathieu Blondel
>> wrote:
>>
The alpha specified this way could (should?) have the same name and
interpretation as the l2_regularization coefficient in
On 03/18/2012 05:07 PM, James Bergstra wrote:
> On Sat, Mar 17, 2012 at 11:55 PM, Mathieu Blondel
> wrote:
>
>>> The alpha specified this way could (should?) have the same name and
>>> interpretation as the l2_regularization coefficient in the
>>> SGDClassifier.
>>>
>> Would you conve
On Sun, Mar 18, 2012 at 9:42 AM, Alexandre Gramfort
wrote:
>> I agree that it's a good idea to correct C for sample size when moving
>> from a sub-problem to the full thing. I just wouldn't use the word
>> "optimal" to describe the new value of C that you get this way - it's
>> an extrapolation,
On Sat, Mar 17, 2012 at 11:55 PM, Mathieu Blondel wrote:
>> The alpha specified this way could (should?) have the same name and
>> interpretation as the l2_regularization coefficient in the
>> SGDClassifier.
>
> Would you convert alpha into a C internal value or would you patch
> libsvm / liblinea
> I agree that it's a good idea to correct C for sample size when moving
> from a sub-problem to the full thing. I just wouldn't use the word
> "optimal" to describe the new value of C that you get this way - it's
> an extrapolation, a good guess... possibly provably better than the
> un-corrected
> 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.
--
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'],
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
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
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
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
14 matches
Mail list logo
