04:35 < jenkins-mlpack2> Project mlpack - git commit test build #588: STILL FAILING in 1 hr 35 min: http://ci.mlpack.org/job/mlpack%20-%20git%20commit%20test/588/

07:39 < iamarchit123Gitt> hey how to add a tag like documentation in pull request

08:56 < iamarchit123Gitt> hey how can I add label to my pull request? I dont see any option for that. Or is it only editable by maintainers?

12:01 < zoq> iamarchit123Gitt: When you open a PR you can set tags, there is a panel in the right side.

14:32 < ayush_nlp> Hi zoq, I was able to build successfully. You mentioned https://github.com/mlpack/mlpack/issues/2296 the previous time. I am a little confused, I was going through Mrityunjay Tripathi's blog and didn't he already implement this?

16:04 < MattGomes> Has anyone had any luck using Visual Studio (Cmake integration) to build mlpack? I keep getting "compiler out of heap space" errors when building the tests :p

16:10 < rcurtin> MattGomes: yeah, mlpack is pretty hard on compiler RAM... what version of Visual Studio are you using? we seem to have better success in our build farm with VS2019, whereas VS2017 always runs out of heap space

16:10 < rcurtin> you can probably also disable parallel builds so only one core is used; it will be slower, but it will use less RAM

16:11 < MattGomes> rcurtin VS2019 as well. I managed to resolve some errors in the "FindArmadillo" script to use my system to go through with linking. However, 10 minutes into compilation I get the heap space issue Lol

16:12 < MattGomes> I'll have a look to see if I can increate the heap space on VS, although I think I probably can't as it seems like a very unsafe thing for MS to allow :D

16:12 < rcurtin> here's our CI configuration, maybe that is helpful:

16:12 < rcurtin> https://github.com/mlpack/mlpack/blob/master/.ci/windows-steps.yaml#L70

16:13 < MattGomes> rcurtin thanks man, will report back if I succeed

16:13 < rcurtin> so basically we make sure it runs only on one core, and at least on the build servers that was sufficient

16:13 < rcurtin> how much RAM do you have in your computer?

16:13 < rcurtin> you could also try freeing RAM from other processes by doing things like closing browser tabs and other applications

16:13 < rcurtin> but the biggest, first thing to check would be whether it's trying to build in parallel, and if it is, disabling that

16:13 < MattGomes> rcurtin I have 16gb ram, 6 cores so I would assume it would be enough for this project

16:13 < rcurtin> yeah, that should be sufficient

16:14 < rcurtin> we do distribute an .msi on our website, so if you want precompiled mlpack.dll/mlpack.lib to link against in your own projects, you might not need to build mlpack by hand

16:14 < rcurtin> https://www.mlpack.org/getstarted.html

16:15 < MattGomes> rcurtin thanks dude, I saw the precompiled options. I thought I would compile it locally to start messing around with it as I plan to contribute somehow. I usually start with compiling the whole thing to begin to understand how it all comes together.

16:19 < alex40> Hi, I have a small question, in binary of mlpack_test, how do you run a single testcase given the test name. I saw that there is areguyment -c which run the section but even when I do so, it still run the whole set of test cases.

16:20 < rcurtin> alex40: you could just specify the name of the test suite or even the test case, like this:

16:20 < rcurtin> mlpack_test "OneClusterTest"

16:20 < rcurtin> mlpack_test "[DBSCANTest]"

16:21 < alex40> Oh that's great! So the one in the bracket is the test suite right?

16:21 < rcurtin> yeah :)

16:21 < alex40> thanks!

16:22 < rcurtin> sure! I think there are lots of other interesting options to the test executable, but that's the only one I know about :)

18:45 < iamarchit123Gitt> Hey folks small query here how to enable printing of Log::Info and Log::Warning tried passing -DDEBUG -g -rdynamic while linking with mlpack but still those logs not coming

18:55 < zoq> iamarchit123Gitt: Set mlpack::Log::Warn.ignoreInput = false; and mlpack::Log::Info.ignoreInput = false;

22:02 < rcurtin> ok, got the slack inviter back online:

22:02 < rcurtin> http://slack-inviter.mlpack.org:4000/