[mlpack] [GSOC 2016] Contribute for MLPack

[Parijat Dewangan]

Hello Marcus and Ryan,

I am working on the first idea  - Implement tree types. I went through the
archive mails, followed the instructions provided there.

Following are the things I worked on till now -
1. Building of MLPack.
2. Going through the library of MLPack and using the already implemented
algorithms through command line.
3. Having a sound knowledge of the paper "Tree-Independent Dual-Tree
Algorithms". Since I have a good command over data structures and
algorithms, I understood the concept of space trees and the way it is used
in implementing algorithms like k-nearest neighbors, range search etc as
explained in the paper. I have qualified for various algorithmic
competitions so understanding various trees and its implementation won't be
a problem as I have used many trees algorithms before.

Currently, I am going through the codes of the already implemented tree
types in src/mlpack/core/tree/ as the new algorithms would be based on the
same template and style. Could you specify any tree type I should start
reading about?

While I am very interested in the previously mentioned project, I am also
eager to work on Decision trees' and 'Fast k-centers algorithms and its
implementation'.

Is there any work/reading that you think I should do in order to be better
prepared to write a good proposal for GSOC 2016.


I appreciate the documentation of MLPack. It is well written and
systematic. :)

Regards,
Parijat Dewangan
+91 - 9966756771


> On Fri, Mar 4, 2016 at 6:45 PM, Parijat Dewangan <parijat10 at gmail.com>
> wrote:
>
>> Thanks for the reply. Please do tell me if you want to know about my
>> major project more.
>>
>> Working on MLpack. I ll update you soon.
>>
>> Parijat
>>
>> On Fri, Mar 4, 2016 at 6:42 PM, Marcus Edel <marcus.edel at fu-berlin.de>
>> wrote:
>>
>>> Hello Parijat,
>>>
>>> the project face verification using unbiased learning sounds
>>> interesting, I
>>> worked on a similar project some time ago.
>>>
>>> Anyway, there's been some talk about this project in other threads on the
>>> mailing list. Here is a link (be sure to take a look at the replies):
>>>
>>> https://mailman.cc.gatech.edu/pipermail/mlpack/2016-March/000790.html
>>>
>>> If you have other questions, feel free to ask. Hopefully the link I
>>> gave will answer most of them.
>>>
>>> Thanks,
>>> Marcus
>>>
>>> > On 04 Mar 2016, at 12:12, Parijat Dewangan <parijat10 at gmail.com>
>>> wrote:
>>> >
>>> > Hello,
>>> >
>>> > I am Parijat Dewangan, a 3rd year undergraduate of IIIT Hyderabad.
>>> >
>>> > I have always found ML very interesting. I have studied and coded many
>>> approaches of classification like KNN, PCA, LDA and many more. One of many
>>> major project include face verification using unbiased learning.
>>> >
>>> > I have installed mlpack and going through the tutorials. I'll update
>>> you soon about any progress.
>>> >
>>> > I have a very strong understanding of the fundamental concepts of data
>>> structures and algorithms. As I take part in competitive programming
>>> regularly, I have knowledge about various trees and its implementation.
>>> >
>>> > My skills are C, C++, Python.
>>> >
>>> > I am interested in the project 'Implement tree types'. Can someone
>>> please guide me through the project ?
>>> >
>>> > Thank you
>>> >
>>> > Parijat Dewangan
>>> > BTech & MS in ECE
>>> > IIIT Hyderabad
>>> > _______________________________________________
>>> > mlpack mailing list
>>> > mlpack at cc.gatech.edu
>>> > https://mailman.cc.gatech.edu/mailman/listinfo/mlpack
>>>
>>>
>>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://mailman.cc.gatech.edu/pipermail/mlpack/attachments/20160307/07835cee/attachment.html>