04:10 < Param-29Gitter[m> following are the commands on terminal and the outputs which I think will answer all the queries @matrixbot

04:22 < rcurtin> why are you using mlpack 3.0.4? that's quite old

04:23 < rcurtin> and your LD_LIBRARY_PATH is /mlpackEverything/mlpack-3.0.4/build/lib/, not ~/mlpackEverything/mlpack-3.0.4/build/lib/

04:24 < rcurtin> for what it's worth, I think it may be worth spending some time reading some resources on C++ development from a command-line environment

05:07 < Param-29Gitter[m> @matrixbot adding a ~ solved my problem. Where can i find latest version of mlpack? also can you suggest me some useful links that would help me write programs in python and c++ using mlpack. I am trying to solve an issue #2119 to understand mlpack so if there are any useful links for understanding of coding style of mlpack it would really be helpful.

05:54 < Hunting-Party10[> Hi i was looking at issue #1660 and #2031 it seems that the changes have been merged but is the issue completely resolved since i would like to start working on it .

05:58 < Hunting-Party10[> Cause its still open on github

06:21 < tejasvi[m]> I can see theres a lot on the table. If someone has time please review #2127

07:09 < HimanshuPathakGi> tejasvi can you please force push your branch to re-run the builds. So, that we can see builds of your PR for some reasons your build got deleted.

07:11 < kartikdutt18Gitt> yes, I think they failed in CTest , I have left a comment for the same.

07:12 < HimanshuPathakGi> Thanks for the help @kartikdutt18

08:11 < jenkins-mlpack2> Project docker mlpack nightly build build #588: UNSTABLE in 2 hr 57 min: http://ci.mlpack.org/job/docker%20mlpack%20nightly%20build/588/

09:32 < Hunting-Party10[> Those packages aren't for bionic , download and build them from source or let cmake download them for you

09:36 < Param-29Gitter[m> cmake did download some files but even then (make) command shows error.

09:39 < Hunting-Party10[> What does it say

09:42 < Param-29Gitter[m> :point_up: [January 20, 2020 3:00 PM](https://gitter.im/mlpack/mlpack?at=5e25731fa1dfaa3ba14e5453) this

09:48 < Hunting-Party10[> Does cmake ../ give all okay ?( Except the man pages will not be generated warning )

10:41 < Param-29Gitter[m> Yes it is completely fine.

12:58 < Param-29Gitter[m> :point_up: [January 20, 2020 4:11 PM](https://gitter.im/mlpack/mlpack?at=5e2583e8ad195a0f673de575) i meant cmake../ is completely fine. I still cannot install v3.2.1

15:15 < Hunting-Party10[> Why are you using an older version tho? Still anyways it should compile just fine. Can you write the exact steps to reproduce this? Since mine is building just fine.

15:17 < Hunting-Party10[> Check your dependencies as well instead of pasting `sudo apt-get install libboost-math-dev libarmadillo-dev binutils-dev python-pandas python-numpy cython python-setuptools libensmallen-dev libstb-dev` do it individually it may have missed out some due to that package not found error

16:35 < kumaran[m]> A nice article about modern cmake.

16:35 < kumaran[m]> https://kubasejdak.com/modern-cmake-is-like-inheritance

16:37 < Nakul[m]> @kumaran:matrix.org: thanks for sharing

17:30 < Param-29Gitter[m> on 1st october 2019 someone had a talk the same issue :point_up: [January 20, 2020 3:12 PM](https://gitter.im/mlpack/mlpack?at=5e2575eb7148837898a81343) and this is because i use bionic ( 16:28 < rcurtin> I see, 'libstb-dev' is only available on disco/eoan on ubuntu) so did anyone solve this problem yet?

17:31 < Param-29Gitter[m> If yes please let me know how do i solve this problem.

17:31 < rcurtin> Param-29Gitter[m: you aren't actually giving any error messages with your output. You said that it fails to link mlpack_test, but we would need the actual failure message in order to help

17:32 < rcurtin> and I'm still unclear on why you are trying to use 3.2.1, when that isn't the newest version

17:32 < rcurtin> it may be worth trying the git master branch instead

17:34 < Param-29Gitter[m> ok i'll try and let you know.

17:34 < rcurtin> sounds good, and if it has an error, be sure to post the full error message :)

17:34 < rcurtin> I'd also suggest trying on a "clean" system and following the build tutorial instructions exactly

17:41 < Param-29Gitter[m> this is what it shows:

17:42 < Param-29Gitter[m> and after this execution of make command stops.

17:45 < rcurtin> have you installed an older version of mlpack on your system?

17:45 < rcurtin> if you have an old version of mlpack installed, this can cause issues when developing with a new version

17:46 < rcurtin> i.e., if /usr/local/lib/libmlpack.so exists (or it may be somewhere else depending on your system), and that libmlpack.so corresponds to an older version of mlpack (for instance, 3.0.4...), this can cause problems

17:47 < rcurtin> I think you had also set LD_LIBRARY_PATH earlier... if it is set to somewhere that has an old libmlpack.so in it, then you should also unset LD_LIBRARY_PATH

18:57 < Nakul[m]> for softmax_regression

18:58 < Nakul[m]> it's throw this error ```critical check stream.str().length() > 0 failed [0 <= 0]

18:58 < Nakul[m]> ```

18:58 < Nakul[m]> * it's throw this error ```

18:58 < Nakul[m]> ```

19:00 < rcurtin> I think you're going to have to dig in and figure out why nothing is being printed to `stream`. you may need to use a debugger like gdb and other techniques

19:01 < abernauer[m]> Yeah I was going to say something along those lines.

19:03 < Nakul[m]> ok so it's time to learn some debugging technique apart from printf/cout : XD

19:04 < rcurtin> :) printf is still a useful tool too; in my experience the debugger doesn't completely replace that

19:04 < rcurtin> but yeah, gdb is a super powerful tool (the same is true with other debuggers)

19:04 < abernauer[m]> Yeah it might be some issue with memory of a temporary being freed.

19:04 < rcurtin> so if you put down a bit of time learning how to work with it, it can be used to get to the bottom of a complex issue pretty fast

19:08 < sreenik[m]> rcurtin: For the gsoc ideas list wiki, is there any system of pull requests? Remember the idea we discussed in the last video meet: I want to float that but I'm skeptical of doing it without you or zoq taking a look at it (in case there is no PR system and it is direct editing).

19:11 < rcurtin> oh, do you mean like PRs for the GSoC ideas page instead of people just randomly editing it?

19:11 < rcurtin> I could agree with that... but I don't think Github Wiki pages support something like that---correct me if I'm wrong

19:12 < rcurtin> mostly in the past, interested mentors have just written down ideas that they think would be nice to mentor

19:13 < sreenik[m]> Oh okay, I will go ahead with it then

19:13 < rcurtin> and then when the applications all come in, mentors see what students/applications are interesting (which may not even be the ideas they wrote on the ideas list in some cases), and then accept a student (or don't, if they don't find anyone they like)

19:13 < rcurtin> if you want feedback on the idea, I'd say just add it and then we can take a look :)

19:13 < rcurtin> an alternative idea would be just to send an email back and forth to workshop the idea

19:15 < sreenik[m]> Got it. Both sound good. Either I'll update it there and ping you once in here, i.e., this IRC, or I'll mail it to our mailing list and discuss it over there.

19:15 < rcurtin> sounds good

19:15 < sreenik[m]> :)

19:16 < rcurtin> I think I misunderstood your question at first; I thought you wanted to set up a whole process for reviewing ideas :)

19:16 < rcurtin> but then I realized you were actually just talking about one idea :)

19:16 < sreenik[m]> Oh :)

19:51 < HimanshuPathakGi> Hey rcurtin I was trying to enable windows build https://github.com/conda-forge/ensmallen-feedstock/pull/14 forensmallen

19:52 < HimanshuPathakGi> And I am getting this error in build Unsatisfiable dependencies for platform win-64: {"vc[version='14.*,9,>=14,<15.0a0']"}

19:54 < HimanshuPathakGi> and I added - vc ==14.* # [win]

19:54 < HimanshuPathakGi> - vc ==9 # [win] because both version are required

19:55 < HimanshuPathakGi> > And I am getting this error in build Unsatisfiable dependencies for platform win-64: {"vc[version='14.*,9,>=14,<15.0a0']"}

19:55 < HimanshuPathakGi> I don't know why it is behaving like this

20:16 < rcurtin> I don't have any idea at all about conda, and especially not conda on Windows

20:17 < rcurtin> have you tried asking the conda maintainers or digging in deeper to find out what the problem is?

20:18 < HimanshuPathakGi> Sure I will ask from him

22:17 < ibtihaj> I am new in open source contribution. I am having trouble with compiling mlpack repository cloned from github .

22:21 < zoq> ibtihaj: Hello, if you can provide some more informations, we might be able to help you to solve the issue.

22:21 < HimanshuPathakGi> Can you please elaborate what problem are you facing I you are struggling with build read this https://www.mlpack.org/doc/mlpack-git/doxygen/build.html

22:22 < HimanshuPathakGi> (edited) ... facing I you ... => ... facing If you ...

22:25 < HimanshuPathakGi> Hey zoq there is a problem with windows vs16 build it can't able to configure armadillo https://dev.azure.com/mlpack/mlpack/_build/results?buildId=315&view=logs&jobId=934eac7e-378e-5e05-ed87-f05e34fa9fdc&j=934eac7e-378e-5e05-ed87-f05e34fa9fdc&t=b61c7632-ff67-55d3-d287-5cb333ca7b45

22:25 < HimanshuPathakGi> Please check this

22:27 < zoq> Looks like the armadillo download failed, let me restart the build.

22:27 < HimanshuPathakGi> I tried it two times same thing happened

22:28 < ibtihaj> I am following this link to compile mlpack : https://www.mlpack.org/doc/mlpack-git/doxygen/build.html but in this tutorial we don't clone the repo from github instead we download a "tar.gz" extension file and then build it . Does the src directory holds all the files present in repo and we can edit and compile them by "make" . I had followed all

22:28 < ibtihaj> the steps mentioned in the tutorial and in the end there is a build directory and source directory etc so if i want to recompile the code or just want to edit the code and compile it which directory holds all the code in tar.gz file?

22:29 < HimanshuPathakGi> Yes you are right after cloning create a build directory

22:30 < HimanshuPathakGi> and follow use the following commands

22:31 < HimanshuPathakGi> $cmake ../

22:31 < HimanshuPathakGi> and then

22:31 < HimanshuPathakGi> make

22:32 < HimanshuPathakGi> inside build directory

22:32 < ibtihaj> Thank you HimanshuPathakGi it worked .

22:32 < zoq> ibtihaj: The archive (tar.gz) packaged a stable but specific version, it contains all source files, you can also checkout the version on github.

22:33 < HimanshuPathakGi> You'r welcome

22:33 < zoq> ibtihaj: If you want to contribute at some point, I recommend to checkout the git repository, instead of using the archive.

22:34 < zoq> HimanshuPathak: Okay, I see let me check the config.

22:35 < ibtihaj> zoq thank you for your response . I am going through the files on git repo

22:46 < zoq> HimanshuPathak: Commented on the PR, not sure that will solve the problem, but let's see.