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/
< rcurtin>
alright, after almost two years I now finally own the pypi package name 'mlpack'---so I will be able to update our documentation so our users can run 'pip install mlpack' not 'pip install mlpack3'
khizirsiddiqui has quit [Quit: Connection closed for inactivity]
< zoq>
rcurtin: Wow, through the package claim program?
xiaohong has joined #mlpack
xiaohong has quit [Remote host closed the connection]
xiaohong has joined #mlpack
xiaohong has quit [Remote host closed the connection]
xiaohong has joined #mlpack
xiaohong has quit [Remote host closed the connection]
xiaohong has joined #mlpack
xiaohong_ has joined #mlpack
xiaohong has quit [Read error: Connection reset by peer]
xiaohong has joined #mlpack
xiaohong_ has quit [Read error: Connection reset by peer]
khizirsiddiqui has joined #mlpack
xiaohong has quit [Remote host closed the connection]
khizirsiddiqui43 has joined #mlpack
khizirsiddiqui43 has quit [Remote host closed the connection]
khizirsiddiqui97 has joined #mlpack
< khizirsiddiqui97>
rcurtin: In PR 1836, you suggested the metrics to have same interface as other metrics (lmetric_imp.hpp) and not evaluate model. To maintain uniformity, I will have to change the implementation of other already implemented metrics like accuracy, f1, mse, recall, average, etc.
< khizirsiddiqui97>
Should I change them all or just implement the new metrics in lmetric_imp.hpp way.
jeffin143 has quit [Read error: Connection reset by peer]
coatless has joined #mlpack
abernauer has joined #mlpack
< rcurtin>
zoq: jeffin143: let's make the 3.2.1 patch release include #2032, #2033, #2034, and #2029; all of those are really close to merged, probably in the next day or so
< rcurtin>
(and if anything else from the 3.2.1 milestone gets in, great, if not, no need to wait)
< rcurtin>
zoq: and yeah, it was through that issue I opened on the pypa repo a few years ago; we actually managed to get in touch with the owner of the 'mlpack' package and he said he'd be willing to give it up
< rcurtin>
khizirsiddiqui97: sorry for the slow response; are you sure it was #1836? I'm not seeing my comment that you're referencing
< zoq>
rcurtin: nice, no more confusion about the package name
< rcurtin>
yeah, although now we have a new set of confusion which means we have to keep *both* the mlpack and mlpack3 pypi packages up to date, otherwise anyone who has installed mlpack3 will stop receiving updates :)
< zoq>
rcurtin: Yep, let's include #2032, #2033, #2034 as well.
< zoq>
True, haven't thought about that.
< rcurtin>
jeffin143: nice, do you think you might try some of the mlpack string processing utilities to play with the codesearchnet challenge?
coatless has quit [Remote host closed the connection]
< abernauer>
rcurtin: I' am trouble shooting the cli util code and getting weird syntax errors and macro errors from included files.
< rcurtin>
abernauer: I can try to take a look at the error trace; what have you tried to debug the issues?
< abernauer>
I did some google searching. This is a new issue. I can run the code through gdb, I guess.
< rcurtin>
if it doesn't compile, it doesn't seem like you can use gdb
< abernauer>
The code seg faults due to an invalid memory reference.
< rcurtin>
okay, but that's not "weird syntax errors and macro errors" then?
< abernauer>
Well yeah. Some armadillo code is aborting the process. Though I doubt that is the main problem.
< abernauer>
Oh I think it's an issue with the order the header files for R and the C++ headers.
< abernauer>
rcurtin: Problem resolved.
< rcurtin>
abernauer: great, sorry for the slow response from my end (I am in and out of meetings)
< abernauer>
It's fine code compiles now. Still core dumps oddly in gdb. Openmp support is bizarre with R.
< rcurtin>
if R doesn't support OpenMP, be sure to compile without -fopenmp
< abernauer>
It does on Linux the flags are a bit unforgiving though. I think I just need to set an environment variable.
abernauer has left #mlpack []
travis-ci has joined #mlpack
< travis-ci>
mlpack/mlpack#8061 (master - ff4a033 : Ryan Curtin): The build passed.