[gsoc18] Request for Guidance

Previous Topic Next Topic
 
classic Classic list List threaded Threaded
3 messages Options
Reply | Threaded
Open this post in threaded view
|

[gsoc18] Request for Guidance

Boost - Dev mailing list
Hello, my name is Mohd Faisal. I want to work on Boost.uBLAS: linear
algebra and matrix computations. Is there a specific reason that only the
listed algorithms have to be implemented. Because there are many good ML
algorithms that can also be implemented like SVM.

_______________________________________________
Unsubscribe & other changes: http://lists.boost.org/mailman/listinfo.cgi/boost
Reply | Threaded
Open this post in threaded view
|

Re: [gsoc18] Request for Guidance

Boost - Dev mailing list
Hi Faisal,
David and I are mentoring this project. You may start with the competency
test, and when its
done, you can send it directly to us. Once the test looks good, we'll help
you with the proposal.

Is there a specific reason that only the
> listed algorithms have to be implemented. Because there are many good ML
> algorithms that can also be implemented like SVM.
>
True, the list is long, and the mentioned algorithms are not the only ones
we'd like to have. Infact,
the more, the better. We can discuss more about it, after you are done with
the competency test.

Should you have any questions, please feel free to ask.

Regards,
Sharique

_______________________________________________
Unsubscribe & other changes: http://lists.boost.org/mailman/listinfo.cgi/boost
Reply | Threaded
Open this post in threaded view
|

Re: [gsoc18] Request for Guidance

Boost - Dev mailing list
Yes indeed, you can implement many more. You're free to add or remove
algorithms, as long as they can fit in a generic library like uBLAS.
SVM are quite specific when compared to a linear algebra library like
ublas. That's why I proposed to focus on generic methods first.
Cheers,
David

On Sat, Mar 10, 2018 at 4:57 AM, mohammad sharique via Boost <
[hidden email]> wrote:

> Hi Faisal,
> David and I are mentoring this project. You may start with the competency
> test, and when its
> done, you can send it directly to us. Once the test looks good, we'll help
> you with the proposal.
>
> Is there a specific reason that only the
> > listed algorithms have to be implemented. Because there are many good ML
> > algorithms that can also be implemented like SVM.
> >
> True, the list is long, and the mentioned algorithms are not the only ones
> we'd like to have. Infact,
> the more, the better. We can discuss more about it, after you are done with
> the competency test.
>
> Should you have any questions, please feel free to ask.
>
> Regards,
> Sharique
>
> _______________________________________________
> Unsubscribe & other changes: http://lists.boost.org/
> mailman/listinfo.cgi/boost
>

_______________________________________________
Unsubscribe & other changes: http://lists.boost.org/mailman/listinfo.cgi/boost