01:43 < mee> where did GANs go in the new docs? I don't see them under mlpack::ann

01:51 < rcurtin> mee: should still be here: http://mlpack.org/doc/mlpack-git/doxygen/classmlpack_1_1ann_1_1GAN.html

01:52 < mee> That's odd. In the link you sent me it exists, but when I go by myself through the left pane in a new tab, GAN is not listed

01:52 < rcurtin> weird, I used 'namespace list' to narrow down to mlpack -> ann -> gan

01:52 < rcurtin> what route did you use?

01:52 < rcurtin> or, sorry, I used 'class list'

01:53 < rcurtin> the search box also works

01:53 < mee> I started here: http://mlpack.org/doc/mlpack-3.0.4/doxygen/index.html

01:53 < sreenik> Thanks zoq and rcurtin. I had thought of using combinations as the last resort but rcurtin’s comment should shorten up the Thanks :D

01:53 < rcurtin> uh, hang on, having trouble with the search box

01:53 < rcurtin> mee: that's the 3.0.4 documentation, change to the 'git' version and you can find it

01:53 < mee> ah

01:53 < rcurtin> GAN is not released as part of 3.0.4 :)

01:54 < mee> Yeah wasn't the concept founded in like 2014 or something?

01:54 < mee> Quite new

01:54 < mee> Compared to lots of other algorithms

01:54 < rcurtin> yep :)

01:55 < mee> Thanks

01:55 < rcurtin> sure :)

01:55 < rcurtin> looks like I have to figure out what's up with the search in the doxygen documentation...

01:56 < mee> The search doesn't work for me either,

01:56 < rcurtin> yeah, I suspect that it's some javascript issue

01:56 < rcurtin> not my favorite thing to debug :(

01:56 < mee> that's why I usually google search whatever + mlpack. Unfortunately the old docs are dead lol

01:56 < rcurtin> right, hopefully google will update its indexes soon

01:57 < rcurtin> if not, I'll have to add a symlink or something

01:57 < rcurtin> anyway, yeah, doxygen search is not great even when it works... google might be better

02:17 < mee> also, what makes a Sparse AutoEncoder "Sparse"?

05:02 < anduser> Is there any function to to do scaling on input data*..?

05:07 < anduser> Transofrm feature by scaling up or down*

05:51 < shine_> zoq: i went through the reinforcement learning code on mlpack. It already implements DQN and DDQN in its algorithm of deep reinforcement learning for atari games. I need a little help with proposal. Do i need to implement only the improvements made in other papers over the classic DQN or is there something i am missing?

05:52 < shine_> I have just started going through reinforcement learning tutorials for the project and don't know much about them. Sorry, if my questions seem a bit wierd

06:22 < shine_> By DDQN i mean Double DQN and not Dueling DQN

07:53 < jenkins-mlpack2> Yippee, build fixed!

07:53 < jenkins-mlpack2> Project docker mlpack nightly build build #275: FIXED in 3 hr 39 min: http://ci.mlpack.org/job/docker%20mlpack%20nightly%20build/275/

08:47 < riaash04> zoq: rcurtin: I have submitted a draft proposal using the dashboard, on the NEAT project. If you get time, it would be good to get some review and comments to improve it. Thanks.

11:24 < mulx10> zoq: should I create a PR for the Concat layer. Also we may need Reshape layer. It would be helpful for 'Application of ANN algorithms '

12:51 < zoq> mulx10: yeah, good idea

15:35 < ShikharJ> rcurtin: Did we end up finalizing the time and date for the meetup?

15:41 < rcurtin> ShikharJ: that was on my todo for today, did you get the email I sent?

15:50 < mulx10> In the documentation of python (mlpack-3.0.4) the import statements are wrong :( .

15:59 < rcurtin> mulx10: nice catch, just gotta drop the () from the generated documentation

16:32 < ShikharJ> rcurtin: Regarding the meeting? Yes. I forgot to fill in my timings though.

16:32 < rcurtin> sounds good

16:32 < rcurtin> once you do that, tonight I'll find the best time from there

16:33 < rcurtin> might need to email one or two people to see if they can move their availability a little bit

16:33 < ShikharJ> Great.

17:35 < Toshal> rcurtin: Hi sir, sorry this is quite unprofessional, are we allowed to join the meetup or it's only for members?

17:38 < rcurtin> Toshal: not unprofessional at all---the meeting will be open to everyone, I just wanted to find a time where all the primary maintainers can make it

17:38 < rcurtin> I'll send an email to the list once everything is finalized (next day or so)

17:42 < Toshal> rcurtin: Okay thanks, Should I reply to the mail you sent on mailing list for appending my name to the list?

17:43 < rcurtin> it'll be a public video call, you can just join, but if you'd like to RSVP you can feel free to

17:43 < rcurtin> (might as well wait until I send the actual meeting info)

17:51 < Toshal> Okay, I will reply to your mail. :) Thanks for the reply.

18:21 < kushagra> can you tell me something about the particle swarm optimisation idea

18:45 < Shubhangi> When some change is made in source code file do I need to run 'make' command again?

18:45 < Shubhangi> I am trying to change .pyx file but changes are not visible. Can anyone tell me why this is happening?

18:54 < rcurtin> Shubhangi: read the tutorial about the automatic bindings and how they work; the pyx files are autogenerated during the build

19:35 < Subham_> Hi Ryan,I(subhamrdn405@gmail.com) had sent email in mlpack mailing list.Although i got reply for my first-two queries(on QGMM and general purpose-EM),I had a l'il hard time deciding on few issues.I would be highly grateful if you could respond to my mail as it would help prepare my proposal

23:11 < jenkins-mlpack2> Project mlpack - git commit test build #136: UNSTABLE in 41 min: http://ci.mlpack.org/job/mlpack%20-%20git%20commit%20test/136/

23:11 < jenkins-mlpack2> Ryan Curtin: Merge branch 'whoopityDoop-master'