06:11 < jenkins-mlpack2> Project mlpack - git commit test build #337: UNSTABLE in 1 hr 20 min: http://ci.mlpack.org/job/mlpack%20-%20git%20commit%20test/337/

08:17 < jenkins-mlpack2> Project docker mlpack nightly build build #634: STILL FAILING in 3 hr 3 min: http://ci.mlpack.org/job/docker%20mlpack%20nightly%20build/634/

10:17 < tejasvi[m]> I finished up https://github.com/mlpack/mlpack/pull/2127. Review appreciated.

12:28 < metahost> chopper_inbound[: https://www.mlpack.org/community.html and https://www.mlpack.org/gsoc.html should be helpful.

13:42 < hemal[m]> Reference paper link: https://arxiv.org/abs/1609.02907

14:46 < rcurtin> hemal[m]: thanks, I think I've resolved the 64/32-bit issues now, so hopefully we can do a 3.3.0 release soon

14:47 < rcurtin> today I'm cleaning up a car and taking pictures of it to sell it, but once I get the ad posted I should be able to start coordinating the 3.3.0 release...

14:48 < rcurtin> I think the use of sparse matrices to represent graphs could be nice, hopefully the underlying format could be effective and efficient

14:48 < rcurtin> (it's not just CSC, it's actually a "hybrid" of CSC for linear algebra operations, and std::map<> for access/modification)

14:52 < jeffin143[m]> Yahoo release

14:52 < jeffin143[m]> There is a lot of work/updates recently so it would be a great release :)

15:46 < himanshu_pathak[> Woah new release :)

15:53 < hemal[m]> mlpack_3.3.0 let's have it ! 😊

18:32 < jeffin143[m]> Hi Naman Jain (Gitter) , we should dependences as much as possible

18:32 < jeffin143[m]> There is no hard and fast rule , but avoiding dependancy is our primary target since Mlpack is already a heavy project

18:32 < jeffin143[m]> And adding dependancy will make it more heavy

18:32 < jeffin143[m]> As such if integrating it with existing is really giving promising results , we won't mind doing it

18:33 < jeffin143[m]> I was also working on the same project , and trying to comp up with something from scratch , I do agree maintaining it will be cumber some but yes we could definitely make it our own :)

18:35 < jeffin143[m]> > Hi Naman Jain (Gitter) , we should avoid dependencies as much as possible

18:35 < jeffin143[m]> Edited -> avoid*

18:46 < NamanJainGitter[> Hi, @jeffin143. Thanks for the quick response. Are you currently working on this project? What kind of support are you planning to provide. Since the project would require realtime updates for logging various metrics, I believe we would require some optimized c++ resource which could handle the realtime elements properly. Do you know any good resource for plotting (realtime) with c++? I found QT Charts library which

18:46 < NamanJainGitter[> to an extent does solve this but again comes with heavy dependency issue i guess...

18:47 < NamanJainGitter[> (edited) Hi, @jeffin143. Thanks ... => Hi, @jeffin143:matrix.org. Thanks ...

19:13 < SaraanshTandonGi> Hey! Is someone here familiar with grid sampling. If yes can you please look at #2250 .

19:25 < jeffin143[m]> Naman Jain (Gitter): we can't add it our core project , but can always setup a different repository , such as models repo and also benchmark repo , so similarly planning to come up with mlpack-board repo

19:26 < jeffin143[m]> The frontend could be react or similar library , but has a wider support among peers more than qt

19:44 < metahost>

19:53 < PranavReddyP16Gi> @zoq can you help me figure out how one would implement a CPPN in mlpack? Here is a relevant link : https://towardsdatascience.com/understanding-compositional-pattern-producing-networks-810f6bef1b88

21:10 < SaraanshTandonGi> I want to implement a network module which can have any FFN with specific input and ouput dimensions as a part of it.

21:10 < SaraanshTandonGi> Can i use the FFN class as a part of the module.