ChanServ changed the topic of #mlpack to: "mlpack: a fast, flexible machine learning library :: We don't always respond instantly, but we will respond; please be patient :: Logs at http://www.mlpack.org/irc/
travis-ci has joined #mlpack
< travis-ci> mlpack/mlpack#8043 (3.2.0 - 105185f : Ryan Curtin): The build passed.
< travis-ci> Change view : https://github.com/mlpack/mlpack/compare/fb748d6bf5b8^...105185f5aec0
travis-ci has left #mlpack []
travis-ci has joined #mlpack
< travis-ci> mlpack/mlpack#8047 (master - f24c075 : Ryan Curtin): The build has errored.
travis-ci has left #mlpack []
< jenkins-mlpack2> Project docker ensmallen nightly build build #30: ABORTED in 2 hr 30 min: http://ci.mlpack.org/job/docker%20ensmallen%20nightly%20build/30/
< zoq> rcurtin: Great work with the release.
khizirsiddiqui has joined #mlpack
< rcurtin> zoq: thanks, it took me a little while to fix my release scripts but they are getting better
khizirsiddiqui has quit [Ping timeout: 260 seconds]
khizirsiddiqui has joined #mlpack
jeffin143 has joined #mlpack
< jeffin143> zoq , rcurtin : is Mlpack going to participate in Google code in ?
jeffin143 has quit [Read error: No route to host]
jeffin143 has joined #mlpack
< jeffin143> 4th point
< zoq> jeffin143: No plans, the main issue I see is to come up with enough simple tasks. The learning curve to contribute something is quite high.
< zoq> jeffin143: Nice :)
< rcurtin> zoq: when the matrix build succeeds, do you want to update/commit the date of the release to HISTORY.md and run `rel/ensmallen-release.sh 2 10 3`?
< zoq> rcurtin: Yes, will do.
ImQ009 has joined #mlpack
< jenkins-mlpack2> Project docker ensmallen nightly build build #31: FAILURE in 31 min: http://ci.mlpack.org/job/docker%20ensmallen%20nightly%20build/31/
< zoq> Not sure why this one failed, looks like one system just stopped.
< zoq> Let's test again with OMP_NUM_THREADS=1.
< jenkins-mlpack2> Yippee, build fixed!
< jenkins-mlpack2> Project docker ensmallen nightly build build #32: FIXED in 31 min: http://ci.mlpack.org/job/docker%20ensmallen%20nightly%20build/32/
< rcurtin> :)
< jeffin143> Did anyone Register for ACM-ICPC ?
< rcurtin> I don't think I am allowed to anymore :-O
jeffin143 has quit [Read error: Connection reset by peer]
jeffin143 has joined #mlpack
< jeffin143> rcurtin : did you participate when you were in college ?
jeffin has joined #mlpack
jeffin143 has quit [Ping timeout: 276 seconds]
< rcurtin> I didn't, actually... I was electrical engineering as an undergraduate :)
< jeffin> Ohhh my god
< jeffin> How are you so good at computers
< jeffin> ?
< jeffin> Wooohh
< rcurtin> well keep in mind that was over ten years ago when I got the undergraduate degree :)
< jeffin> Ahhaa
< jeffin> rcurtin , have you anytime came across gold linker ?
< jeffin> Linux
< jeffin> rcurtin , I tried build mlpack from master branch , if we don't have ensmallen , will run cmake download it for us
< jeffin> ?
< jeffin> Running*
< khizirsiddiqui> @jeffin: yes, it downloads automatically
khizirsiddiqui79 has joined #mlpack
khizirsiddiqui79 has quit [Remote host closed the connection]
abernauer has joined #mlpack
< abernauer> rcurtin: Are there any C reference books you would recommend? I would use cpp reference, but it's C coverage is a bit sparse.
< rcurtin> jeffin: not familiar with the gold linker
< rcurtin> abernauer: the K&R C book maybe? (kernighan and ritchie) "The C Programming Language"
< rcurtin> that's the usual standard textbook, although it's quite old now (don't know how recently it's been updated)
< abernauer> Yeah I have heard good things about that book. I get the feeling I will need to pick up a used copy for the R bindings.
< rcurtin> it's not a bad book to have a copy of anyway :) I think there is some really nice general programming wisdom in there (not just for C)
< jeffin> rcurtin : if we ensmallen and we do fresh build , it doesn't throw an error to upgrade
< jeffin> May be it's good to inform user to update
< rcurtin> I'm not sure what you mean, I think maybe you forgot a verb? :) "if we ensmallen"?
< rcurtin> cmake should fail to configure if the version of ensmallen is older than 2.10.0
< rcurtin> do you mean if you have ensmallen 1.x on the system and an already-configured cmake build directory, that the build will fail once you upgrade with git to the latest master branch?
< rcurtin> I'm not sure if I described that well, but anyway if you can walk me through what you mean I can understand and answer better :)
< jeffin> Wait
< jeffin> I will let you know
< rcurtin> sure, sounds good
< jeffin> So , I removed my build directory and then created a fresh build directory and then ran cmake ../ , and got this https://ibb.co/26xxpL9
< jeffin> It was successful although I have ens 1.*.*
< rcurtin> and that's the git master branch?
< jeffin> Yes that's git master branch
< rcurtin> can you just double check that that master branch is up to date?
< jeffin> I have checked CMakeList manually too
< rcurtin> https://github.com/mlpack/mlpack/blob/master/CMakeLists.txt#L366 is the same in your CMakeLists.txt?
< jeffin> Yes sure I will do that , and let you know once more
< jeffin> Yes I am dead sure it's that and before reporting it here , to be more sure I added a required flag just like arma has found ( **** required)
< rcurtin> hmm, interesting
< rcurtin> ok, I managed to reproduce it
< jeffin> Umm , I did it again and it happens
< jeffin> So I was correct :)
< rcurtin> right, I just am not sure why this is happening
< rcurtin> according to the CMake documentation, it should be checking the extracted version value against 2.10.0
< jeffin> Yup , I also thought about it because it always threw an error if my boost or arma was not updated
< rcurtin> right, so I am looking at FindEnsmallen.cmake now but I am not seeing what the error is here
< rcurtin> oh wow
< rcurtin> I can't believe the fix for this
< rcurtin> let me open a PR
< rcurtin> zoq: awesome, thanks for releasing 2.10.3, hopefully you didn't have a problem with any of the scripts :)
< rcurtin> I'd love to use the Github API to simply POST the release instead of having to do it manually, but I haven't figured out how yet
< zoq> rcurtin: I had to adapt the sed command since I did the release on mac OS; sed on mac OS requires a backup file ...
< zoq> rcurtin: The next time I just us linux :)
< jeffin> rcurtin : e vs E ! :-p
< rcurtin> zoq: ah, ok, if you want, you can make it more portable too :)
< rcurtin> jeffin: yeah! I was quite surprised...
< zoq> rcurtin: Do you mind if I use ssh instead of https? I use 2FA, which disables push over https.
< zoq> For ensmallen.org that is.
< rcurtin> oh yeah that's just fine, whichever you like works
< rcurtin> I don't think I have ssh set up, but maybe it is finally time for me to do that :)
< zoq> okay, perfect
< rcurtin> great to see that the ensmallen.org rebuild worked
< zoq> yep
abernauer has quit [Remote host closed the connection]
ImQ009 has quit [Quit: Leaving]
< jeffin> rcurtin : you there
< jeffin> Now there is a new issues
< jeffin> Suppose consider I have ens 1.*.* Something , and I pull your branch ( pr2032) , and run cmake ../ and build , it successfully downloads the new ens and successfully builds it . But if I deleted the build directory and again make a directory and run cmake ../ it again downloads ens 2.*.*
< jeffin> The problem is once downloaded it is not replacing the old ens* which is there in use/local/include , it does search and reports unsuitable version and downloads the new one , but again if you delete the build folder and again create the same it will again, Is there any ways we can make the ens download persistent that means once downloaded the newer version will over write the older one
< rcurtin> the download is directly to the build folder, so, not really
< rcurtin> the intention is to remove the automatic downloader once ensmallen is widely available in package managers, etc.
< jeffin> Ohh , ok I didn't know that was the plan
jeffin has quit [Quit: AndroIRC - Android IRC Client ( http://www.androirc.com )]
< rcurtin> :)
khizirsiddiqui has quit [Quit: Connection closed for inactivity]