[mlpack] [GSOC 2016] Contribute for MLPack

Ryan Curtin ryan at ratml.org
Tue Mar 8 10:11:05 EST 2016


On Tue, Mar 08, 2016 at 12:48:08PM +0530, Parijat Dewangan wrote:
> Thanks Ryan for the quick reply.
> 
> Since I am done with the documentation part, I am now going to focus on the
> literature part of various trees.
> 
> Would it be beneficial for me to solve bugs related to project with respect
> to GSOC or should I concentrate on literature part?
> 
> Also, do you have any priorities for the projects?

Hi Parijat,

There are not any specific priorities for projects, and no decisions
will be made on which projects to allocate slots to until after the
application period has ended.

I don't think that there are any bugs open relating to the decision tree
code, but it would still be very useful to take a look through the code,
use it on some toy example datasets, and familiarize yourself with its
design.  But, if you can find any problems or make any improvements
(like, for instance, making the implementation faster), those would
definitely be useful contributions.

I would say that it is definitely important to read lots of papers
relating to trees, in order to understand the tradeoffs associated with
each tree type and to better design your code.  Again, don't feel
restricted to any particular tree type; if you find something
interesting that I didn't list, then that's just fine to include in your
proposal (as long as it satisfies the definition of "space tree" given
in the tree-independent dual-tree algorithms paper, otherwise it won't
work with the dual-tree algorithms we have implemented).

Thanks,

Ryan

-- 
Ryan Curtin    | "Get off my lawn!"
ryan at ratml.org |   - Kowalski


More information about the mlpack mailing list