verne.freenode.net changed the topic of #mlpack to: http://www.mlpack.org/ -- We don't respond instantly... but we will respond. Give it a few minutes. Or hours. -- Channel logs: http://www.mlpack.org/irc/
sukrut has joined #mlpack
sukrut has quit [Client Quit]
tsathoggua has joined #mlpack
tsathoggua has quit [Quit: Konversation terminated!]
zoq has quit [Ping timeout: 240 seconds]
zoq has joined #mlpack
travis-ci has joined #mlpack
< travis-ci> mlpack/mlpack#549 (master - c25a2b6 : Ryan Curtin): The build passed.
travis-ci has left #mlpack []
< rcurtin> okay... Jenkins is back up with four slaves online (cabbie, dambala, shoeshine, samedi)
< rcurtin> now I guess we need to hook up shoeshine to the gsoc blog job again
< rcurtin> oh wow that was actually really easy, I just needed to modify the hostname in mlpack.org's apache configuration, so the GSoC blog is finally back up :)
jerone has joined #mlpack
< jerone> Hi, just to let you know - I'm still having an issue with: https://github.com/mlpack/mlpack/issues/515#issuecomment-185870952
< rcurtin> Hi Jerone, it's on my list for today
< rcurtin> I tested with the dataset you sent originally, but didn't have any issue
< jerone> Ok
< rcurtin> I don't have 5vs2Test.txt, which might make a difference
< rcurtin> but I was just running mlpack_det -t 5vs2Train.txt -v
< jerone> But when you train - you get more than 1 leaf node prior to pruning?
< rcurtin> yeah, hang on, let me give some output:
< rcurtin> (one second, I have to rebuild mlpack_det on this machine)
< jerone> ok, no worries
< rcurtin> ◈ ryan@humungus ☃ build-nodebug ◈ $ bin/mlpack_det -t 5vs2Train1.txt -v -f 0
< rcurtin> [INFO ] Loading '5vs2Train1.txt' as CSV data. Size is 784 x 896.
< rcurtin> [INFO ] Performing leave-one-out cross validation.
< rcurtin> [INFO ] 135 leaf nodes in the tree using full dataset; minimum alpha: 64.0377.
< rcurtin> [INFO ] 45 trees in the sequence; maximum alpha: 313.757.
< rcurtin> [INFO ] Optimal alpha: 298.193.
< rcurtin> [INFO ] 44 leaf nodes in the optimally pruned tree; optimal alpha: 298.193.
< rcurtin> I have to get lunch now, but can you make sure you've updated to the latest git master revision and rebuilt the mlpack_det program?
< jerone> Loading 'new/5vs2Train1.txt' as CSV data. Size is 784 x 896. [INFO ] Performing 10-fold cross validation. [INFO ] 1 leaf nodes in the tree using full dataset; minimum alpha: 1.79769e+308. [INFO ] 1 trees in the sequence; maximum alpha: 0. [INFO ] Optimal alpha: -1. [INFO ] 1 leaf nodes in the optimally pruned tree; optimal alpha: -1.79769e+308. [INFO ]
< jerone> I edited the file that you made a code edit too
< jerone> But, I don't think my mlpack has changed despite the change
< jerone> How can I correct that? Is there a way to uninstall mlpack completely and reinstall?
< jerone> to*
< jerone> I'll attempt to rebuilt my mlpack_det program
< zoq> If you run 'make install', you overwrite the existing or 'old' library and binaries.
kirizaki has joined #mlpack
< jerone> Cheers
< kirizaki> Hello
< rcurtin> jerone: back now, sorry for the delay
< rcurtin> so you rebuilt mlpack and you still have the issue?
< jerone> ok i just rebuilt it and tested it
< jerone> I get the same result as you now
< rcurtin> ok, great, so there is no problem now?
< jerone> 135 leaf nodes in the full tree
< jerone> thank you very much for your help
< rcurtin> awesome, great to hear it is worked out
< rcurtin> please don't hesitate to get in touch if you find more bugs :)
< jerone> Thanks, will do. Have a good evening.
< jerone> Or afternoon
< rcurtin> you too :)
< rcurtin> (early afternoon here. I wish it was evening, then I would be relaxing instead of working!)
< jerone> Soon enough!
< rcurtin> :)
jerone has quit [Ping timeout: 252 seconds]
< zoq> awsome :)
< zoq> .... awesome :)
tsathoggua has joined #mlpack
< rcurtin> now the mailing list will explode...
< zoq> really nice page this year
gaut has joined #mlpack
< rcurtin> I didn't get an email about it, I only saw it because the Shogun guys were celebrating because the list was posted
< rcurtin> so I went and looked and saw that mlpack was on the list too
gaut has quit [Client Quit]
< zoq> If you like we can also celebrate :)
< rcurtin> I would celebrate but I am in a meeting, no celebration allowed, only boring project updates :(
< kirizaki> congratulations! ;)
< rcurtin> apparently 180 accepted organizations this year, vs. 137 last year
< rcurtin> I have no idea how many slots we have yet though
< rcurtin> wow, #gsoc is crazy right now
< rcurtin> I guess that shouldn't be surprising...
< zoq> crazy in a positive sense?
< rcurtin> crazy as in lots of messages and activity :)
< aditya123> zoq: Are the project list the same which were described on GSoC2016 page of mlpack ?
< rcurtin> aditya123: yeah, that is the ideas list, but you are always welcome to propose your own project idea too, so don't feel constrained by the ideas listed there
< kirizaki> how many slots mlpack had last year?
< rcurtin> none in 2015 (we weren't accepted), 5 in 2014, 3 in 2013
< rcurtin> I'd probably expect somewhere between 2 and 5 this year
< kirizaki> nice
hercky has joined #mlpack
mudit has joined #mlpack
< mudit> Hi everyone, I am new to this community and willing to contribute to it. Can anyone guide me through it?
jagsxi has joined #mlpack
< rcurtin> mudit: this webpage might be helpful: http://mlpack.org/involved.html
< kirizaki> it's just begun :)
jagsxi has quit [Quit: Page closed]
< mudit> @rcurtin thanks
< rcurtin> sure, I hope the page is helpful :)
< mudit> well I am interested in working for some long term project. I have previous experience on Deep Learning and C++.
< mudit> Can you tell me simple issue that can be solved ?
< mudit> So that I can get to know project structure.
< rcurtin> from the webpage I linked to:
< rcurtin> "You can find a bug to solve on the issues list. Issues (generally) have a difficulty label, which may help in selecting an issue that is interesting to you."
< rcurtin> then you could search for issues labeled "easy": https://github.com/mlpack/mlpack/issues?q=is%3Aopen+is%3Aissue+label%3A%22D%3A+easy%22
< mudit> okay
travis-ci has joined #mlpack
< travis-ci> mlpack/mlpack#550 (master - a2e57d6 : Marcus Edel): The build was broken.
travis-ci has left #mlpack []
decltypeme has joined #mlpack
< zoq> Thanks travis great introduction!
< decltypeme> Hi all, I was asking about the status of the deep learning modules, it is listed from last year, so has any progress been done with it?
< zoq> decltypeme: Actually I've updated the project idea, to reflect the changes made since then (removed the recurrent neural networks and convolution neural networks).
< decltypeme> great, thanks
kirizaki has quit [Quit: Konversation terminated!]
Rishabh has joined #mlpack
sanket has joined #mlpack
__anonymous__ has joined #mlpack
__anonymous__ has quit [Remote host closed the connection]
__anonymous__ has joined #mlpack
__anonymous__ is now known as oyqzp
oyqzp has quit [Remote host closed the connection]
__anonymous__ has joined #mlpack
__anonymous__ is now known as anveshi
anveshi has quit [Remote host closed the connection]
anveshi has joined #mlpack
anveshi has quit [Quit: Page closed]
anveshi has joined #mlpack
anveshi has quit [Client Quit]
Rishabh has quit [Read error: Connection reset by peer]
Rishabh has joined #mlpack