01:21 < travis-ci> mlpack/ensmallen#738 (2.11.5 - a174206 : Ryan Curtin): The build passed.

01:21 < travis-ci> Change view : https://github.com/mlpack/ensmallen/compare/ef02b18a6f8a^...a174206f9985

01:21 < travis-ci> Build details : https://travis-ci.org/mlpack/ensmallen/builds/661333952

03:15 < LakshyaOjhaGitte> (edited) ... says >‘struct LossFunctionsTest::HingeEmbeddingLoss’ has no member named ‘Forward’

03:15 < LakshyaOjhaGitte> Same goes for backward. This comes in the >loss_function_test.cpp file. ... => ... says ‘struct LossFunctionsTest::HingeEmbeddingLoss’ has no member named ‘Forward’

03:15 < LakshyaOjhaGitte> Same goes for backward. This comes in the loss_function_test.cpp file. ...

03:16 < LakshyaOjhaGitte> (edited) ... says ‘struct LossFunctionsTest::HingeEmbeddingLoss’ has no member named ‘Forward’

03:16 < LakshyaOjhaGitte> Same ... => ... says _‘struct LossFunctionsTest::HingeEmbeddingLoss’ has no member named ‘Forward’_

03:16 < LakshyaOjhaGitte> Same ...

03:16 < LakshyaOjhaGitte> (edited) ... says _‘struct LossFunctionsTest::HingeEmbeddingLoss’ has no member named ‘Forward’_

03:16 < LakshyaOjhaGitte> Same ... => ... says **‘struct LossFunctionsTest::HingeEmbeddingLoss’ has no member named ‘Forward’**

03:16 < LakshyaOjhaGitte> Same ...

03:17 < LakshyaOjhaGitte> (edited) ... the loss_function_test.cpp file.

03:17 < LakshyaOjhaGitte> Thanks => ... the _loss_function_test.cpp file_.

03:17 < LakshyaOjhaGitte> Thanks

03:17 < LakshyaOjhaGitte> (edited) ... the _loss_function_test.cpp file_.

03:17 < LakshyaOjhaGitte> Thanks => ... the **loss_function_test.cpp file**.

03:17 < LakshyaOjhaGitte> Thanks

03:44 < PrinceGuptaGitte> Hi @ojhalakshya I think it's because of `HingeEmbeddingLoss module;` I think it should be `HingeEmbeddingLoss<> module;`

04:28 < favre49> rcurtin The permissions seem fine

04:28 < favre49> drwxr-xr-x for everything

04:29 < favre49> oh wait my bad i misread it

04:29 < favre49> You're right, oddly enough the include folder is owned by root. I missed that

04:29 < favre49> I wonder why

04:29 < favre49> Thanks for the help

04:47 < LakshyaOjhaGitte> Hi @prince776 I tried that already it gives another error then **/home/cyrus_dux/Desktop/mlpack/src/mlpack/tests/loss_functions_test.cpp:522:3: error: ‘LossFunctionsTest::HingeEmbeddingLoss’ is not a template

04:47 < LakshyaOjhaGitte> 522 | HingeEmbeddingLoss<> module;

04:47 < LakshyaOjhaGitte> **

04:49 < LakshyaOjhaGitte> I think it might be related to line 23 in hinge_embedding_loss_impl.hpp

04:49 < LakshyaOjhaGitte> Its just an intuition but I don't seem to find anything related to it.

05:13 < jeffin143[m]> @rcurtin:matrix.org @zoq may be someday you should share your screen through zoom when you code something which is very important feature

05:13 < jeffin143[m]> I want to see how many times you stackoverflow :-p

05:45 < pickle-rick[m]> Lol

07:11 < jenkins-mlpack2> Project docker mlpack nightly build build #638: STILL FAILING in 2 hr 57 min: http://ci.mlpack.org/job/docker%20mlpack%20nightly%20build/638/

07:19 < himanshu_pathak[> jeffin143 Yeah it will be a great thing to see :)

09:30 < jenkins-mlpack2> Project mlpack - git commit test build #344: FAILURE in 22 min: http://ci.mlpack.org/job/mlpack%20-%20git%20commit%20test/344/

11:25 < rcurtin> himanshu_pathak[: agreed, I don't have any great way around that :( a lot of the boost libraries can be somewhat overwhelming during debugging

11:25 < rcurtin> favre49: no worries, glad that we got it figured out :)

11:27 < rcurtin> ahhh, it looks like master got broken. I will open a PR in just a few minutes... I bet the ANN rvalue reference refactoring did not handle changes from some PRs that got merged between when the rvalue reference PR was opened and when it was merged

11:27 < rcurtin> not fun to wake up to "build failed" messages... oops

11:27 < kartikdutt18Gitt> Hi @rcurtin, I have already opened a PR to fix the same

11:28 < zoq> kartikdutt18Gitt: Great!

11:28 < kartikdutt18Gitt> Could you please take a look at #2281

11:29 < rcurtin> kartikdutt18Gitt: awesome, thank you! I had not gotten to the end of my new emails yet

11:29 < rcurtin> when I was writing that message, I thought, "I wonder if someone's already fixed this, I should probably check before I write this message..."

11:29 < rcurtin> then I wrote the message anyway :)

11:29 < zoq> kartikdutt18Gitt: approved from my side

11:30 < rcurtin> should we make sure the builds pass before merging it?

11:31 < rcurtin> ah, actually I can see that some have passed already; that's probably sufficient

11:31 < kartikdutt18Gitt> Great, Thanks. Sorry I missed it earlier. I think we should always rebase our branches locally before any merge that way if there is going to be an error we can know about before hand.

11:31 < zoq> rcurtin: yes, I think this is fine.

11:31 < rcurtin> kartikdutt18Gitt: yeah, agreed; actually I got bitten by this recently at my job, so I should have known better!

11:33 < kartikdutt18Gitt> Great, I will keep that in mind in my future PRs as well as for all the PRs that will get merged. Thanks a lot.

11:33 < zoq> We could reconfigure the jenkins job, to automate the process.

11:34 < rcurtin> hmm, how would we do that? the build would need to fire off for every open PR every time someone merges a PR to master

11:35 < zoq> true

11:36 < rcurtin> luckily, I think that this type of situation has never happened before with mlpack (I think?), so it's at least quite rare :)

12:08 < nishantkr18[m]> hello everyone! had a small question. There are no python bindings for reinforcement learning module right?

12:10 < zoq> nishantkr18[m]: That is correct.

12:11 < zoq> nishantkr18[m]: There is no easy way to pass the task to solve to the method.

12:15 < nishantkr18[m]> ok.. so all RL Algos are tested using https://github.com/zoq/gym_tcp_api ??

12:16 < zoq> nishantkr18[m]: No, we have some env implemented here: https://github.com/mlpack/mlpack/tree/master/src/mlpack/methods/reinforcement_learning/environment that we use for testing.

12:21 < nishantkr18[m]> <zoq "nishantkr18: No, we have some en"> Ohk.. So is there any documentation available explaining how to use them?

12:22 < zoq> nishantkr18[m]: Each rl related test should be helpful here; https://github.com/mlpack/mlpack/blob/master/src/mlpack/tests/q_learning_test.cpp is one example.

12:24 < nishantkr18[m]> <zoq "nishantkr18: Each rl related tes"> Alright, thanks for the help 🙂

12:30 < jenkins-mlpack2> Yippee, build fixed!

12:30 < jenkins-mlpack2> Project mlpack - git commit test build #345: FIXED in 59 min: http://ci.mlpack.org/job/mlpack%20-%20git%20commit%20test/345/

12:34 < SriramSKGitter[m> @nishantkr18 , @zoq : I'm working on a reinforcement learning tutorial similar to the ones for other methods and modules, hopefully I can open a PR within the week :)

12:36 < zoq> SriramSKGitter[m: awesome :)

14:09 < kritika12298Gitt> Please guide me on how to get started

14:09 < kritika12298Gitt> Hello, I am Kritika Gupta and am looking forward to contribute to the project Improving Tree Traversers.

14:10 < zoq> kritika12298Gitt: Hello, https://www.mlpack.org/community.html and https://www.mlpack.org/gsoc.html should you help get started.

14:48 < rcurtin> kartikdutt18Gitt: thanks for catching the issue with #2265---I'm going to go through and make a post on all the ANN-related PRs that master should be merged and tests re-run

14:50 < PrinceGuptaGitte> Hi @kartikdutt18 , I wanted to add VGG and inception models to the models repository, but since you are working on restructuring it I should add those to your PR instead of main repository right? Since you've already made LeNet and tested it, I think it should be fine to add models in this stage.

14:50 < PrinceGuptaGitte> But I wanted to confirm if it's ok to start working on them right now because you might be working on something that can potentially change how models are structured right now. So I thought I should ask first.

14:50 < PrinceGuptaGitte> Also about VGG, since there is already a PR for it, how should I proceed for adding it.

15:02 < kartikdutt18Gitt> Hi @prince776, I think for VGG, I would also have to implement an imagenet DataLoader. For VGGNet, The model is already implemented and I could just absorb it in my PR or you could open another (after restructuring). For inception net I think you could start working on it, It would be great to have that model. Once the restructuring is done, incase it has already been merged I'll make changes in my PR or you could

15:02 < kartikdutt18Gitt> rebase your PR then. Anything works for me.

15:02 < kartikdutt18Gitt> Also thanks @rcurtin, for the message you just shared on ANN related PRs.

15:04 < rcurtin> sure, I just blew up my inbox :-D

15:05 < PrinceGuptaGitte> So how should I start working on inception, since the restructured code in in your PR. Do I make PR on your forked repo or someother way

15:05 < shrit[m]> rcurtin 👍️

15:05 < PrinceGuptaGitte> (edited) ... code in in ... => ... code is in ...

15:06 < rcurtin> I think the inbox fun is just getting started, I assume all those messages I posted are going to get responses and questions and then my inbox will really be on fire :-D

15:07 < PrinceGuptaGitte> Thanks @rcurtin for the message on all my PRs just now. (I save you 5 message s :)

15:07 < kartikdutt18Gitt> It's upto you. I am fine either way. You can also make a PR to the main repo and I can pick up the changes (seems a bit redundant) or what you suggested. I'll be pushing a chunk of changes today for LSTMs and hopefully VAE tomorrow.

15:07 < PrinceGuptaGitte> (edited) ... (I save you 5 ... => ... (I saved your inbox 5 ...

15:07 < rcurtin> PrinceGuptaGitte: :)

15:08 < PrinceGuptaGitte> Ok thanks. I will start working on it now.

15:08 < shrit[m]> rcurtin: I am surprised of how fast you were able to answer all of these issues and pull request

15:09 < kartikdutt18Gitt> @rcurtin, Since I have already learned about, I'll try to answer some of then to the best of my ability, I hope that's okay.

15:24 < rcurtin> shrit[m]: copy-paste :-D

15:25 < rcurtin> I opened all the PRs whose titles looked like they had to do with ANN, then just flipped tabs and pasted in the message

15:25 < rcurtin> kartikdutt18Gitt: thanks; don't feel obligated---I'll try and respond quickly when I can

15:59 < kartikdutt18Gitt> Great. Happy to help in anyway I can.

16:09 < SakshamRastogiGi> @kartikdutt18 Hi what all models are implemented like resnet, inception or someone is working on if you have any idea ?

16:09 < SakshamRastogiGi> Or something is there in your mind I could work on

16:23 < kartikdutt18Gitt> Hi @codeboy5, Currently there are PR open for VGG, AlexNet, LeNet and after addition of Grouped Convolution, MobileNet. InceptionNet is taken up @prince776. For Resnet , mlpack has residual layer. I think you could start by getting familiar with codebase models repo. I am currently in the middle of restructuring the repo so you could also take a look at that PR. I think there are tons of models that would be fun to

16:23 < kartikdutt18Gitt> implement, I think starting off small would be the best idea. After restructuring the repo I'll probably have to open some issues that I couldn't take complete in the currently open PR so if you would like you could also work on them. It might be a couple of days though before I end up opening the issues.

16:26 < SakshamRastogiGi> @kartikdutt18 I’ve already gone through some of the models codebase. Resnet sounds interesting to me

16:26 < SakshamRastogiGi> Thanks a lot for your help , looking forward to your PR !

16:33 < kartikdutt18Gitt> @codeboy5, It's already open, I would be pushing another commit today, If you would like you could review it then. I would like to get any style issue fixed or description related issue fixed as I work through it. It can be found at models/#60 . No obligation though. Thanks.

16:42 < AbishaiEbenezerG> I'm sorry if this is a little too dumb, but i need some guidance here

16:44 < AbishaiEbenezerG> i've been through the q_learning_test.cpp , mountain_car.hpp and q_learning.hpp code. I understand it to a certain extent in a very theoretical way and it might take me a bit more time to understand it inside out

16:44 < AbishaiEbenezerG> the thing is

16:45 < AbishaiEbenezerG> i do not know how to test and run these modules locally on my machine

16:45 < AbishaiEbenezerG> how do you guys make test your code locally?

16:45 < AbishaiEbenezerG> and especially when i wanted to test it with the gym envs

16:45 < AbishaiEbenezerG> i had absolutely no clue

16:47 < zoq> AbishaiEbenezerG: Have a look at: https://github.com/mlpack/mlpack/wiki/Testing-Guidelines

16:49 < AbishaiEbenezerG> sure @zoq. I should have known about this page earlier

16:49 < AbishaiEbenezerG> sorry if this was too trivial

16:49 < zoq> AbishaiEbenezerG: Nahh, no worries.

17:19 < PrinceGuptaGitte> Hi @zoq did you get a chance to see the mail I sent to mailing list about GSOC proposal.

17:19 < PrinceGuptaGitte> It would be great if you can give some opinions about it.

17:19 < PrinceGuptaGitte> Thanks.

17:39 < SriramSKGitter[m> Hello @nisha1729 , did you have a look at https://www.mlpack.org/community.html ?

17:42 < SakshamRastogiGi> @nisha1729 also you can take a look at https://www.mlpack.org/gsoc.html

17:43 < NishaGeorgeGitte> Yes, I did

17:44 < SakshamRastogiGi> @kartikdutt18 Thanks a lot for your help, i would take a look at the issues and wait for your PR to find something to work on ?

17:44 < NishaGeorgeGitte> Great, thanks. This should be useful. I found a list of issues but I'm a little confused on how to begin.

17:52 < kartikdutt18Gitt> Great. Hopefully I won't keep you waiting long. Thanks.

19:01 < SakshamRastogiGi> @kartikdutt18 I was going through some recent architectures and DenseNet sounds interesting to me, i would start working on this for now ( https://arxiv.org/abs/1608.06993 )

20:48 < zoq> PrinceGuptaGitte: Just responded.