Re: [Scikit-learn-general] K-SVD implementation

2015-08-27 Thread Alexey Umnov
Hi,Some time ago I made a PR on the K-SVD algorithm, but it didn't made to a commit.Now I have time to work on this once again, so I am asking for some reviews on the code.The link to the PR: https://github.com/scikit-learn/scikit-learn/pull/3739--Alexey Umnov -

Re: [Scikit-learn-general] K-SVD implementation

2014-11-19 Thread Alexey Umnov
Hi, The work on the K-SVD implementation is mostly finished, so I wanted to ask you for comments and review.Here is the PR: https://github.com/scikit-learn/scikit-learn/pull/3739--Alexey Umnov  -- Download BIRT iHub F-Type

Re: [Scikit-learn-general] K-SVD implementation

2014-10-05 Thread Gael Varoquaux
On Sun, Oct 05, 2014 at 09:48:56PM +0400, Умнов Алексей (Alexey Umnov) wrote: > 2) Create a pull request with the current changes. Create a 'WIP' pull request: a pull request that you name with '[WIP] K-SVD'. WIP stands for 'work in progress'. G --

Re: [Scikit-learn-general] K-SVD implementation

2014-10-05 Thread Alexey Umnov
Hi, I implemented the K-SVD as a free function and created a small example that uses it.I wanted to ask for some feedback on it before starting polishing the code (also I need your advice on several points). What would be the most convenient way to do it?1) Post it to gist.github.com and send link

Re: [Scikit-learn-general] K-SVD implementation

2014-09-28 Thread Alexey Umnov
Yes, that is reasonable. Thank you! 28.09.2014, 19:45, "Alexandre Gramfort" :hi, I am trying to make the K-SVD implementation consistent with the existing sparse coding algorigthms (dict_learning and orthogonal_mp) and I am a bit confused by the notations. I'll give a summary of the notations and e

Re: [Scikit-learn-general] K-SVD implementation

2014-09-28 Thread Alexandre Gramfort
hi, > I am trying to make the K-SVD implementation consistent with the existing > sparse coding algorigthms (dict_learning and orthogonal_mp) and I am a bit > confused by the notations. I'll give a summary of the notations and explain > what I am not understanding. > > 1) sklearn.decomposition.dic

Re: [Scikit-learn-general] K-SVD implementation

2014-09-28 Thread Alexey Umnov
Hi, I am trying to make the K-SVD implementation consistent with the existing sparse coding algorigthms (dict_learning and orthogonal_mp) and I am a bit confused by the notations. I'll give a summary of the notations and explain what I am not understanding. 1) sklearn.decomposition.dict_learning Pr

Re: [Scikit-learn-general] K-SVD implementation

2014-09-23 Thread Alexey Umnov
Thanks, the code and the link will be helpful indeed! Alright, I am starting to work on the subject and will make a PR in some time. 23.09.2014, 16:17, "Kyle Kastner" : > I started some code here long ago > (https://gist.github.com/kastnerkyle/8143030) that isn't really > finished or cleaned up -

Re: [Scikit-learn-general] K-SVD implementation

2014-09-23 Thread Kyle Kastner
I started some code here long ago (https://gist.github.com/kastnerkyle/8143030) that isn't really finished or cleaned up - maybe it can give you some ideas/advice for implementing? I never got a chance to clean this up for PR, and it doesn't look like I will have time in the near future so your PR

Re: [Scikit-learn-general] K-SVD implementation

2014-09-23 Thread Gael Varoquaux
Hi Alexei, This is a well-known algorithm. I personnally think that it would be interesting to see it in scikit-learn. Therefor I would definitely welcome a pull request implementing it. Cheers, Gaël On Tue, Sep 23, 2014 at 02:44:14PM +0400, Умнов Алексей wrote: > Hello, > Is there anyone curr

[Scikit-learn-general] K-SVD implementation

2014-09-23 Thread Умнов Алексей
Hello, Is there anyone currently working on the K-SVD algorithm? If not, I can implement it, as I need it for my research anyway and there is already a good Orthogonal Matching Pursuit implementation in the library. -- Alexey Umnov --