[mlpack] Automative benchmarking

[Ryan Curtin]

On Wed, Mar 05, 2014 at 09:23:43AM +0530, Ranjani wrote:
> Hi,
> Thanks for the link on Shogun's integration.I looked at their detailed
> expectations.
> I would like to discuss more on
> 1.Adding new algorithms
> 2.Improving accuracy and fixing Bottlenecks

This is a very vague description.  Can you describe more what you mean?
If you mean that you want to implement entirely new algorithms for
mlpack, then that would be outside the scope of the automatic
benchmarking project, but could still be a project.  What I mean to
point out here is that the benchmarking project itself is now a separate
codebase than mlpack.

If you meant adding new algorithms to the benchmarking project results,
be aware that adding an algorithm is usually done in a few days, with
not very many hours of work.  So just adding a few algorithms would not
be a large enough chunk of work to be a good Summer of Code project.

I don't know what you mean about "improving accuracy and fixing
bottlenecks"; can you elaborate?

Thanks,

Ryan

-- 
Ryan Curtin    | "Maybe the next time."
ryan at ratml.org |   - J.G. Ballard