[mlpack] [GSOC 2016] Contribute for MLPack
Parijat Dewangan
parijat10 at gmail.com
Fri Mar 11 07:30:28 EST 2016
Hi,
I went through various research papers to have a better understanding of
the dual tree algorithms and various trees. Currently, I am focusing on
vantage point trees, by referring the following papers.
- "IMPROVING DUAL-TREE ALGORITHMS"
<http://www.ratml.org/pub/pdf/2015improving.pdf> Thesis by Ryan Curtin .
- "Data Structures and Algorithms for Nearest Neighbor Search in
General Metric Spaces by Peter N. Yianilos*"
I am completely comfortable with the MLPack API after going through the
above thesis of Ryan Curtin. So, I was thinking of coding vantage point
trees. What do you suggest ?
Should I provide you with the pseudo code of Vantage Point Trees? Or should
I try fixing some issues? I was thinking of taking up issue #275.
https://github.com/mlpack/mlpack/issues/275.
Please guide me further.
Parijat Dewangan
+91-9966756771
On Tue, Mar 8, 2016 at 8:41 PM, Ryan Curtin <ryan at ratml.org> wrote:
> 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
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://mailman.cc.gatech.edu/pipermail/mlpack/attachments/20160311/b766aba5/attachment.html>
More information about the mlpack
mailing list