[mlpack] Regarding the scope for a beginner in GSoC’17

[Ryan Curtin]

On Tue, Feb 21, 2017 at 09:20:01AM +0530, Sankalp Sanand wrote:
> Hello,
>  I am a second year undergraduate student in Computer Science branch.
> I’m currently doing a course on Kadenze Academy , Creative
> Applications of Deep Learning. I wanted to ask that will it be
> feasible for me to work or prepare a project proposal for the project
> “Essential Deep Learning Modules”, since it was rated as 5/10 in
> difficulty level so I thought it is comparatively easier for a
> beginner like me. I have a good knowledge of C++, and am learning
> Python currently, which covers the necessary knowledge section for the
> given project idea. I will begin by fixing bugs and entrance level
> tasks first on GitHub for mlpack so that I get an idea of how the
> codebase is like. I had one more question, will we have to implement
> all of the modules or only some of it, because I think it depends on
> the deliverability of the particular module. Please guide me for how
> should I proceed further.

Hi Sankalp,

We haven't yet been accepted into GSoC 2017, but, if we are, everyone is
welcome to submit an application, so you should also feel free to submit
one.

I think that to be successful at this project, you should have a good
knowledge of deep learning; i.e., you should be familiar with the way
deep neural networks are typically built and trained, and certainly you
should be familiar with the individual components that you plan to
implement.

You are welcome to propose to implement any or all of the four modules
that are currently listed (RBM, RBFN, BRN, GAN), or you can even propose
other modules that mlpack does not currently have implemented.

Another useful thing to do will be to search the mailing list archives
to see other previous discussions about this project.

I hope this is helpful; let me know if I can clarify anything.  I'm not
the mentor for that project, so it's possible I have misstated something
(Marcus, please correct me if I wrote something wrong!).

Thanks,

Ryan

-- 
Ryan Curtin    | "So I just, uh... I just cut them up like regular
ryan at ratml.org | chickens?" - Henry Spencer