00:21 < rcurtin> abernauer: sounds good, hopefully you can get it worked out with those tips

02:12 < jenkins-mlpack2> Project docker mlpack weekly build build #62: STILL FAILING in 4 hr 25 min: http://ci.mlpack.org/job/docker%20mlpack%20weekly%20build/62/

07:04 < jenkins-mlpack2> Yippee, build fixed!

07:04 < jenkins-mlpack2> Project mlpack - git commit test build #223: FIXED in 59 min: http://ci.mlpack.org/job/mlpack%20-%20git%20commit%20test/223/

07:48 < jenkins-mlpack2> Project docker mlpack nightly build build #428: STILL FAILING in 3 hr 34 min: http://ci.mlpack.org/job/docker%20mlpack%20nightly%20build/428/

07:48 < jenkins-mlpack2> * Ryan Curtin: Deploy all of gcc including stdlib on new versions.

07:48 < jenkins-mlpack2> * Ryan Curtin: Fix syntax?

13:31 < sakshamB> ShikharJ: I am here

13:33 < ShikharJ> sakshamB: Great, let's begin

13:33 < ShikharJ> Toshal: Are you here?

13:34 < ShikharJ> sakshamB: As far as I can tell, only spectral norm is pending some work? Or is that complete as well?

13:35 < sakshamB> ShikharJ yes it does require some more work. I am having some trouble with deriving the gradient.

13:40 < ShikharJ> sakshamB: Okay, no worries there. Just mention it in the final report, we can later work out the gradient.

13:40 < sakshamB> ShikharJ: alright will do thanks :)

13:44 < sakshamB> ShikharJ is there anything else we need to discuss?

13:44 < ShikharJ> sakshamB: Okay, it'll probably be our last meeting here. It was wonderful to have you as a part of mlpack. It was also a really enriching experience for me, guiding you. I hope you had a good time, and will continue our collaboration in the future.

13:46 < ShikharJ> sakshamB: I think we only had three PRs pending right now in your case, and two of them are complete, so I'm really happy with the work product you've provided.

13:47 < sakshamB> ShikharJ: Thanks a lot for your help throughout the summer and it was great contributing to mlpack. I hope to be active with the community after gsoc as well. :)

13:49 < ShikharJ> sakshamB: Great, I think Toshal would not be able to join us, so I'll email him later I guess. Feel free to log off. Have a nice weekend :)

13:51 < sakshamB> ShikharJ: you too have a great weekend! :)

13:55 < AndroUser> Can anyone help me how to use conditional using

13:56 < jeffin143> Like

13:56 < jeffin143> If ( true ) using type = something; else using type = other ;

13:56 < jeffin143> I said later be able to use type, how to increase the scope

13:58 < jeffin143> Also i need to dynamic assign type to a variable something as

13:58 < jeffin143> If ( true ) int i ; else float i;

13:58 < jeffin143> And then later be able to use i

13:59 < lozhnikov> jeffin143: It doesn't work like that.

14:00 < lozhnikov> jeffin143: If you are going to use std::conditional then the condition should be defined at the compilation step, you can't use runtime dependent conditions.

14:00 < jeffin143> Umm, i wanted it to support charextract , and splitbyanyof , or else then i have to copy paste the code two times !

14:00 < lozhnikov> Probably you need boost::variant

14:00 < jeffin143> It looked too long so i thought this way i can reduce the lines but i will go with the brute force approach, i mean with increased length of code

14:01 < rcurtin> sakshamB: looking forward to reading the final report---I haven't had a chance to keep up very well all summer, but I think the work you've done has been pretty excellent :)

14:01 < jeffin143> Umm , i need to look at that , i haven't used it anywhere so i will check and will come back

14:02 < lozhnikov> jeffin143: Look at mlpack/methods/neighbor_search/ns_model.hpp

14:03 < lozhnikov> jeffin143: You need to implement something like the NSModel class.

19:53 < jenkins-mlpack2> Project docker mlpack nightly build build #429: STILL FAILING in 3 hr 17 min: http://ci.mlpack.org/job/docker%20mlpack%20nightly%20build/429/

20:37 < jeffin143> lozhnikov : i wrote summary at Mlpack/gsocblog , let me know if any correction is needed

20:38 < lozhnikov> jeffin143: ok, I'll read it in the morning.

20:38 < jeffin143> Also about boost::variant , i still need some sample codes to understand it, the nsmodel class had lot of apply_visitor and static_visitor which i need to understand and hence I will add support for charsplit

20:38 < jeffin143> Once I am comfortable with the all the boost::variant

20:38 < jeffin143> I will take a look at some of the examples, and then try to implement it.

20:40 < lozhnikov> jeffin143: Try to read cppreference first.

20:40 < lozhnikov> https://en.cppreference.com/w/cpp/utility/variant/visit

20:40 < lozhnikov> and https://en.cppreference.com/w/cpp/utility/variant

20:44 < jeffin143> Ok , thanks for the link :)

21:23 < rcurtin> jeffin143: awesome, I'll take a look at it too when I have a second. the work that you and Mikhail have been doing is great; I'm really looking forward to using it in the future

21:23 < rcurtin> I think you are also lucky to have such a good mentor, I see lozhnikov in here all the time working tirelessly to help out :)