[mlpack] GSoc 2019

[Marcus Edel]

Hello Bishwa,

thanks for getting in touch; mlpack already supports some GAN types, see:

https://github.com/mlpack/mlpack/tree/master/src/mlpack/methods/ann/gan

for more information. Also, the blog posts from Shikhar should be interesting:

http://mlpack.org/gsocblog/ShikharJaiswalPage.html

Anyway, the goal of the project is to implement a fast and well tests GAN, so my
recommendation is to focus on one or two models. Writing meaningful tests takes
a lot of time.

I hope anything I said was helpful, let me know if I should clarify anything.

Thanks,
Marcus

> On 28. Feb 2019, at 07:05, Bishwa Karkee <karkeebishwa1 at gmail.com> wrote:
> 
> This message is not encrypted but sent from a verified user on the dmail blockchain <https://dmail.io/>Dear sir,
> 
> I am Bishwa Karki, a 4th year Computer Engineering student from one of the prestigious engineering college, Paschimanchal Campus, of Nepal. I have practical experience in C/C++, Java and Python. 
> As of my major project I am implementing "Text to Image Synthesis" using Generative Adversarial Network(GAN) and found similar project in the section " Essential Deep Learning Module"  of Mlpack. So I am keenly interested in contributing mlpack in doing GAN project in this summer code 2019. 
> 
> For this I have already gone through paper listed in the GAN section under "Essential Deep Learning Module" as it was also the same paper for my project. But coming upto the description section in the ideas list it became vague for me to understand, should we have to implement all the deep learning modules listed in that section or can choose any of those, like GAN only? 
> 
> At the last, from mlpack.org <http://www.mlpack.org/docs/mlpack-git/doxygen/cli_quickstart.html> I saw mlpack library can be implemented with Python and as per the requirement of GSoC should we have to implement it in C++ only or can choose Python ? 
> 
> Regards,
> Bishwa Karki
>  
> _______________________________________________
> mlpack mailing list
> mlpack at lists.mlpack.org
> http://knife.lugatgt.org/cgi-bin/mailman/listinfo/mlpack

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://knife.lugatgt.org/pipermail/mlpack/attachments/20190228/978567e2/attachment.html>