Although the manual includes references here and there, I feel that
oftentimes, it doesn't highlight the reference papers enough. Also, we
should make it easier for people to cite the papers corresponding to the
estimators they use. For example, it would be nice to be able to retrieve
the corresponding bibtex in just one click. There might be a sphinx plugin
that allows to centralize the citations in one bibtex file and cite papers
using latex syntax.
Mathieu
On Wed, Aug 29, 2012 at 3:10 PM, Andreas Mueller
<[email protected]>wrote:
> Hi Ronnie.
> I don't really understand the comment about the kernel approximation.
> Each of the algorithms has a reference to the paper it is from.
> As Robert said, the others are textbook algorithms. But even these
> have often References, mostly in the documentation, though, as for
> example naive Bayes:
> http://scikit-learn.org/dev/modules/naive_bayes.html
>
> In general "Elements of Statistical Learning" is a textbook we reference
> often.
> It is available online for free. Another standard reference is Chris
> Bishop's
> "Pattern Recognition and Machine Learning".
>
> If you want to start contributing, I can talk you through some
> of the issues. I try to hang out in IRC this week. I can also just sent
> you a list and we can discuss on the issue tracker.
>
> Cheers,
> Andy
>
>
>
> On 08/29/2012 12:59 AM, Robert Layton wrote:
>
> On 29 August 2012 08:47, Ronnie Ghose <[email protected]> wrote:
>
>> Honestly, I'm not sure if i'm wrong but for example:
>>
>>
>> https://github.com/scikit-learn/scikit-learn/blob/master/sklearn/kernel_approximation.py
>> I'm reasonably sure kernel approximation is from a paper?
>>
>>
>> https://github.com/scikit-learn/scikit-learn/blob/master/sklearn/cluster/_k_means.pyx
>> K means clustering
>>
>>
>> https://github.com/scikit-learn/scikit-learn/blob/master/sklearn/neighbors/classification.py
>> This
>> doesn't seem to ...?
>>
>> I'm looking at ones that do not look like
>> https://github.com/scikit-learn/scikit-learn/blob/master/sklearn/naive_bayes.py
>> ,
>> which has a nice list at the bottom.
>>
>> Please do tell me if i'm completely wrong as that is likely.
>>
>> As a sidenote, where would you suggest I begin working on the project -
>> I was thinking of fixing issues listed in the Github Issues tab.
>>
>> Thanks,
>> Ronnie
>>
>>
>> On 28 August 2012 18:21, Robert Layton <[email protected]> wrote:
>>
>>> On 29 August 2012 08:17, Ronnie Ghose <[email protected]> wrote:
>>>
>>>> Where do you guys get your algorithms from? Some algorithms have
>>>> different variations and I personally don't really see any sources.
>>>>
>>>> Thanks,
>>>> Ronnie
>>>>
>>>>
>>>> ------------------------------------------------------------------------------
>>>> Live Security Virtual Conference
>>>> Exclusive live event will cover all the ways today's security and
>>>> threat landscape has changed and how IT managers can respond.
>>>> Discussions
>>>> will include endpoint security, mobile security and the latest in
>>>> malware
>>>> threats. http://www.accelacomm.com/jaw/sfrnl04242012/114/50122263/
>>>> _______________________________________________
>>>> Scikit-learn-general mailing list
>>>> [email protected]
>>>> https://lists.sourceforge.net/lists/listinfo/scikit-learn-general
>>>>
>>>>
>>> Hi Ronnie,
>>>
>>> Could you give some examples? The algorithms I've coded have
>>> references to the original paper, and most algorithms should have them too.
>>>
>>> - Robert
>>>
>>> --
>>>
>>> Public key at: http://pgp.mit.edu/ Search for this email address and
>>> select the key from "2011-08-19" (key id: 54BA8735)
>>>
>>>
>>>
>>> ------------------------------------------------------------------------------
>>> Live Security Virtual Conference
>>> Exclusive live event will cover all the ways today's security and
>>> threat landscape has changed and how IT managers can respond. Discussions
>>> will include endpoint security, mobile security and the latest in malware
>>> threats. http://www.accelacomm.com/jaw/sfrnl04242012/114/50122263/
>>> _______________________________________________
>>> Scikit-learn-general mailing list
>>> [email protected]
>>> https://lists.sourceforge.net/lists/listinfo/scikit-learn-general
>>>
>>>
>>
>>
>> ------------------------------------------------------------------------------
>> Live Security Virtual Conference
>> Exclusive live event will cover all the ways today's security and
>> threat landscape has changed and how IT managers can respond. Discussions
>> will include endpoint security, mobile security and the latest in malware
>> threats. http://www.accelacomm.com/jaw/sfrnl04242012/114/50122263/
>> _______________________________________________
>> Scikit-learn-general mailing list
>> [email protected]
>> https://lists.sourceforge.net/lists/listinfo/scikit-learn-general
>>
>>
> Hi Ronnie,
>
> Thanks for that.
> The k-nearest neighbours is a fairly obvious algorithm I don't believe it
> is generally *cited* in academic papers anymore. k-means is a bit similar
> (that does have a reference in the kmeans.py file), and Naive Bayes less so.
> Most of the files do have (sometimes poor) references in the "notes"
> section of the specific classes, but some of the ones you pointed to do
> not. Thanks for pointing that out!
>
> It's hard to say exactly where to start. We generally point new
> contributors to the easyfix bugs in the bug tracker:
> https://github.com/scikit-learn/scikit-learn/issues?direction=desc&labels=EasyFix&page=1&sort=created&state=open
> If you aren't a very strong coder, you could learn a lot by helping us
> review pull requests (this is true even if you are a strong coder).
> If you really aren't a strong coder, start with the documentation and
> examples - do all the examples work? Do the tutorials make sense?
> If you *are* a strong coder, this issue (
> https://github.com/scikit-learn/scikit-learn/issues/339) may be a good
> spot to start -- some of the tests are slow, which makes building annoying.
> We have improved significantly and have a test suite which is very useful:
> http://jenkins-scikit-learn.github.com/scikit-learn-speed/
>
> Thanks!
>
> - Robert
>
>
> --
>
> Public key at: http://pgp.mit.edu/ Search for this email address and
> select the key from "2011-08-19" (key id: 54BA8735)
>
>
>
> ------------------------------------------------------------------------------
> Live Security Virtual Conference
> Exclusive live event will cover all the ways today's security and
> threat landscape has changed and how IT managers can respond. Discussions
> will include endpoint security, mobile security and the latest in malware
> threats. http://www.accelacomm.com/jaw/sfrnl04242012/114/50122263/
>
>
>
> _______________________________________________
> Scikit-learn-general mailing
> [email protected]https://lists.sourceforge.net/lists/listinfo/scikit-learn-general
>
>
>
>
> ------------------------------------------------------------------------------
> Live Security Virtual Conference
> Exclusive live event will cover all the ways today's security and
> threat landscape has changed and how IT managers can respond. Discussions
> will include endpoint security, mobile security and the latest in malware
> threats. http://www.accelacomm.com/jaw/sfrnl04242012/114/50122263/
> _______________________________________________
> Scikit-learn-general mailing list
> [email protected]
> https://lists.sourceforge.net/lists/listinfo/scikit-learn-general
>
>
------------------------------------------------------------------------------
Live Security Virtual Conference
Exclusive live event will cover all the ways today's security and
threat landscape has changed and how IT managers can respond. Discussions
will include endpoint security, mobile security and the latest in malware
threats. http://www.accelacomm.com/jaw/sfrnl04242012/114/50122263/
_______________________________________________
Scikit-learn-general mailing list
[email protected]
https://lists.sourceforge.net/lists/listinfo/scikit-learn-general
