[mlpack] GSOC 2014: collaborative filtering package improvements

[Pulkit Yadav]

Hello,

I am Pulkit Yadav, a senior year Computer Science and Engineering
undergraduate at IIT Delhi. I am passionate for Machine Learning and have
also taken relevant courses like Machine Learning, Artificial Neural
Networks, Artificial Intelligence, etc. at my university. Having done a
large number of projects in C++, I am quite proficient in the language. I
have also been actively involved in research projects and have two
publications to my name. So, reading latest literature on efficient
algorithms for implementation would not be an issue.

I am very excited in contributing to mlpack library. I am particularly
interested in the project idea "collaborative filtering package
improvements" and would like to work on it.

The project ideas page (http://www.mlpack.org/trac/wiki/SummerOfCodeIdeas)
says that QUIC-SVD needs to be implemented. However the project wiki (
http://www.mlpack.org/trac/wiki/cf) says that QUIC-SVD using cosine trees
has already been implemented. Could you please clarify which algorithms
have been implemented and what are some suggestions for implementation?

I have installed mlpack and tried the cf module. Currently, I am studying
the source code of cf module and the tree abstractions
(src/mlpack/core/tree/).

How can I start contributing to the project now in some way? Kindly give
suggestions how should I proceed from here.

Thanks,
Pulkit
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mailman.cc.gatech.edu/pipermail/mlpack/attachments/20140312/09d6ba72/attachment-0002.html>