01:55 < robotcatorGitter> Congratulations to those who were selected on GSoC.

03:01 < nishantkr18[m]> Congratulations everyone! Looking forward to an awesome summer ahead!

04:20 < favre49> Congrats everyone, looking forward to a good summer

04:47 < chopper_inbound[> Wow! Congratulations everyone and good luck for the journey ahead :)

13:44 < abernauer[m]> rcurtin: So the R bindings are a no go on your side of things going forward?

13:49 < rcurtin> abernauer[m]: I'm not sure what you mean?

13:52 < abernauer[m]> Well I am still interested in completing the project. Just saw you accepted another candidate this year for the project.

13:53 < rcurtin> abernauer[m]: I wasn't aware of that---it's basically been a year since I've heard much about the project from your end

13:53 < rcurtin> in any case, Yashwant already has a working prototype and is not far from having something that's basically ready

13:54 < rcurtin> sorry if toes got stepped on; maybe if you like, you could help with the review process / contribute other fixes / etc.?

14:29 < abernauer[m]> <rcurtin "abernauer: I'm not sure what you"> You Dirk and James are going forward on the project with someone else?

15:27 < zoq> hemal[m]: Hey, will send you a mail.

16:47 < hemal[m]> That would be great, thanks

18:37 < PranavReddyP16Gi> Quick question, Can I still contribute and expect feedback over the course of the summer or will everyone be too busy with GSoC?

18:40 < zoq> Param-29Gitter[m: Absolutely, more than happy to provide feedback.

18:40 < zoq> ahh, wrong name

18:40 < zoq> PranavReddyP16Gi: ee message above :)

18:59 < PranavReddyP16Gi> Great thanks @zoq :)

19:02 < jeffin143[m]> zoq (@freenode_zoq:matrix.org): Is there any way to make print loss print validation loss too , ummm

19:03 < jeffin143[m]> I guess shrit was working on it and got stuck some where

19:03 < jeffin143[m]> Print loss as in ens callbacks

19:04 < zoq> jeffin143[m]: Currently no, but you are right you can prety much use #165 to get that part working.

19:06 < zoq> jeffin143[m]: Maybe you can open an issue in the mlpack repo?

19:11 < shrit[m]> jeffin143 zoq The callback is pretty in a good shape, only some review required, and adding some tests. However, I have no idea how to do the tests

19:13 < zoq> shrit[m]: let me comment on the PR.

19:27 < jeffin143[m]> I will also follow up shrit , I need it for gsoc

19:27 < jeffin143[m]> zoq (@freenode_zoq:matrix.org): should I open the issue in mlpack or ens ?

19:27 < rcurtin> I thought there was already an issue open for it? seems like no need for a new one

19:28 < jeffin143[m]> Acurracy would be ? 1- objective ?

19:28 < jeffin143[m]> Correct

19:29 < zoq> Maybe there is already an issue, in which case no need to open another one.

19:48 < jeffin143[m]> rcurtin (@freenode_rcurtin:matrix.org): vector<size_t> throws an error when markdown_binding = on

19:48 < jeffin143[m]> Did we map that vector type ??

19:49 < jeffin143[m]> This is in reference to a build failure in 2325 , for linux markdown , for other it passes

19:49 < rcurtin> I don't think that we did...

19:49 < rcurtin> I haven't looked at #2325 yet

19:49 < rcurtin> I dunno, take a look in src/mlpack/bindings/cli/ to see if you see references to vector<size_t>, which would mean that it's at least mapped for CLI bindings

19:50 < rcurtin> then if it doesn't exist in src/mlpack/bindings/markdown/, it should be ...somewhat easy to add... I hope? :)

20:42 < himanshu_pathak[> Hey zoq: I was testing NTM Using ReberGrammer. It is not learning upto two trails and during the third it shows some progress and in the fourth trails same thing happened.

20:44 < rcurtin> himanshu_pathak[: I don't think we confirmed that it was passing all the tests when it was originally written (that's part of why it wasn't yet merged); so there might be a bug...?

20:44 < himanshu_pathak[> I tried to print prediction only in the third trial every slice has different prediction in other cases every slice containing different prediction

20:45 < himanshu_pathak[> * I tried to print prediction only in the third trial every slice has different prediction in other cases every slice containing same prediction

20:45 < himanshu_pathak[> rcurtin: I think I have to go through all the maths :)

20:45 < rcurtin> himanshu_pathak[: yeah, that is the fun part :)

20:46 < himanshu_pathak[> rcurtin: Ok I will try and may come up with solution

20:47 < rcurtin> good luck, let us know if we can help out :)

20:47 < rcurtin> I've been wanting to review the work you were doing on that PR but haven't gotten to it yet

20:47 < rcurtin> I think I am still trying to find the time to figure out the Conda windows PR...

20:50 < himanshu_pathak[> rcurtin: Yes Conda Windows PR is also left right now I don't have any idea I am finding what can be the reason but not yet successful

20:52 < himanshu_pathak[> May be we will get it soon working fingers crossed 🙂

20:52 < rcurtin> :)