05:04 < akhandait[m]> zoq: Yeah, a clean build on master :/

06:45 < jenkins-mlpack2> Project docker ensmallen nightly build build #28: ABORTED in 2 hr 30 min: http://ci.mlpack.org/job/docker%20ensmallen%20nightly%20build/28/

07:04 < jenkins-mlpack2> Project docker mlpack nightly build build #471: STILL UNSTABLE in 2 hr 50 min: http://ci.mlpack.org/job/docker%20mlpack%20nightly%20build/471/

12:01 < zoq> akhandait[m]: Hm, so you can build with -DDOWNLOAD_STB_IMAGE=OFF but we should see if we can somehow solve the issue.

12:02 < zoq> akhandait[m]: The error message isn't really helpful here, so I guess we have to find a system to reproduce the issue, can you post some more information about your setup, boost, armadillo version, dist?

12:27 < make-me-smile> Hey! I was looking to start contributing. I am familiar with the basics of Machine Learning. I found this inactive PR #1836. The good-first-issue #1965 specifies few other stale PRs, does it mean that PR #1836 is no longer required?

12:27 < make-me-smile> https://github.com/mlpack/mlpack/issues/1965

12:28 < make-me-smile> https://github.com/mlpack/mlpack/pull/1875

13:06 < rcurtin> make-me-smile: I'm not sure if your nick is a reference to the band Chicago but I now have the associated song stuck in my head :)

13:06 < rcurtin> anyway, sure, I think it's fine if you want to resurrect #1836 and #1875

14:14 < make_me_smile> Thanks rcurtin. But now when I see, I have this problem building MLPack on Ubuntu 18.04.

14:15 < make_me_smile> I am following as instructed on the website, but keep getting this error:

14:15 < make_me_smile> [ 49%] Built target generate_pyx_kmeans[ 49%] Built target generate_pyx_kdeMakefile:140: recipe for target 'all' failedmake: *** [all] Error 2

14:19 < rcurtin> okay, that's not the actual error though, could you paste the part of the compilation that failed? :)

14:22 < make_me_smile> it produces too many warnings about mlpackMain() defined but not used, and parameter 'objective' set but not used.

14:22 < make_me_smile> And yes: error: ‘class ens::VanillaUpdate’ has no member named ‘Initialize’ updater.Initialize(learningNetwork.Parameters().n_rows,

14:24 < make_me_smile> https://pastebin.com/eby1Sgu1

14:24 < make_me_smile> for the complete log

14:25 < make_me_smile> btw, posting pastebin links is allowed right?

14:35 < rcurtin> khizirsiddiqui: sure, no problem, sorry about the slow response from my end

14:36 < rcurtin> what version of mlpack are you building? there is an issue with 3.1.1 that means it will not build properly with the newest ensmallen version (which it automatically downloads)

14:36 < rcurtin> so I'd suggest building the git master branch until we release 3.2.0 (which we will do in just a couple days)

14:37 < khizirsiddiqui> Oh, cool. I will use the git master branch then.

14:49 < rcurtin> awesome; sorry about the inconvenience

15:04 < khizirsiddiqui> I could build it now. Thanks.

15:04 < khizirsiddiqui> BTW, why are we getting that error?

15:05 < rcurtin> khizirsiddiqui: the underlying reason is a little stupid and it's my fault ...

15:06 < rcurtin> basically, if ensmallen (a dependency) is not available during the CMake configuration, then CMake will go download https://ensmallen.org/files/ensmallen-newest.tar.gz

15:06 < rcurtin> and use that as a dependency

15:06 < rcurtin> but then we released ensmallen 2.10.0, which is not reverse-compatible with the previous versions; thus, mlpack 3.1.1 requires ensmallen 1.x, but the build configuration now automatically downloads 2.x,

15:06 < rcurtin> because ensmallen-newest.tar.gz is just a symlink to the newest version...

15:07 < rcurtin> so I had simply overlooked the clear fact that ensmallen-newest.tar.gz might not always be compatible with the mlpack 3.1.1 release

15:08 < rcurtin> I guess I could release 3.1.2 with a really simple patch fix, but we are so close to a 3.2.0 release that it makes more sense to just wait a bit

15:10 < khizirsiddiqui> Ah, yes yes.

20:06 < ameyaosd> I'm trying to install mlpack on windows reading this tutorial. https://www.mlpack.org/doc/stable/doxygen/build_windows.html. If I have to install it on Visual Studio 2019 which kind of project should I make? A Console App or a CMake Project or something else?

20:08 < zoq> ameyaosd: After the cmake step, open "C:\mlpack\mlpack\build\mlpack.sln", this will open the correct project for you, so there is no need to create one.

20:13 < ameyaosd> Thanks zoq. When it says `Unzip mlpack to "C:\mlpack\mlpack"` from where does it mean that? Cause I downloaded the installer from https://www.mlpack.org/ and then ran it.

20:15 < zoq> ameyaosd: With installer you mean the tar.gz file?

20:15 < zoq> ameyaosd: Or I guess you mean the msi?

20:15 < ameyaosd> the msi

20:16 < zoq> ameyaosd: If you just wanted to use mlpack, all you need to do is to run the installer, so the steps mentioned in the wiki aren't necessary.

20:17 < ameyaosd> I wanted to run mlpack but I later plan to solve an issue on GitHub. That's why

20:18 < zoq> ameyaosd: In this case, I would use the tar.gz file instead of msi and follow the steps.

20:18 < zoq> ameyaosd: The tar.gz file should contain a mlpack folder which you can copy to the path mentioned in the tutorial.

20:18 < ameyaosd> Alright sure.