2012/7/4 Buddy Rich <[email protected]>
>
> Hi,
> I have a simple black image with three kinds of white objects in it (circle, 
> vertical rectangle and horizontal rectangle).
>
> I learn a dictionary with k atoms, then I would like to recombine the 
> "correct" atoms for rebuilding my three objects.
>
> Is there a method for this task?

If you put a sparsity constraint on the number of "active" atoms to
use for each reconstruction then this is typically named "sparse
coding". For instance :

http://scikit-learn.org/dev/auto_examples/decomposition/plot_sparse_coding.html

Otherwise this is just a (multi-output) regression problem. It can be
solved with RidgeRegression, (soon) random forests or multiple single
output regression model such as SVR trained independently.

--
Olivier
http://twitter.com/ogrisel - http://github.com/ogrisel

------------------------------------------------------------------------------
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

Reply via email to