ChanServ changed the topic of #mlpack to: Due to ongoing spam on freenode, we've muted unregistered users. See http://www.mlpack.org/ircspam.txt for more information, or also you could join #mlpack-temp and chat there.
vivekp has quit [Ping timeout: 272 seconds]
vivekp has joined #mlpack
travis-ci has joined #mlpack
< travis-ci>
mlpack/mlpack#5585 (master - 0287b6f : Ryan Curtin): The build passed.
jenkins-mlpack2 has quit [Ping timeout: 252 seconds]
jenkins-mlpack2 has joined #mlpack
caiocarvalho has joined #mlpack
caiocarvalho has quit [Ping timeout: 240 seconds]
< zoq>
davida: I think for the RES net model we would have to add a new layer that skips connections.
< davida>
zoq: So ResNets not currently possible... Will need to skip that Coursera example for now. Do you plan to add that functionality in? Seems a necessary part of such a library. The ability to combine/string partial models together to build a larger model is key to ResNets. I also noticed that some ResNets models are designed to view output from earlier parts of the network. Is that feasible or also needs to be added?
< zoq>
I think the implementation is straightforward, just need some time to do it.
< ShikharJ>
rcurtin: I'll push a PR shortly today.
< ShikharJ>
rcurtin: Did the ensmallen paper happen to get through for MLSYS?
davida has quit [Quit: Leaving]
< rcurtin>
ShikharJ: haven't heard back from MLsys yet
< rcurtin>
I've been working on automating other parts of the release pipeline, so when the PR is ready I'll take a look and when it's ready for release it should hopefully take a little less time :)
< ShikharJ>
rcurtin: Ah, I thought since the accpeptance notification day is behind us, you might have received the notification.
< rcurtin>
actually let's see if I can access the reviews even if I got no notification email
< ShikharJ>
rcurtin: Also do you think we can talk privately in a sub-channel? I can't seem to send you private messages.
< rcurtin>
that's strange
< rcurtin>
I just tried sending you a message, let me know if you got it
< rcurtin>
also no review yet for the ensmallen paper. I guess probably they will send that later today :)
< ShikharJ>
Yeah I did, but everytime I do "/msg" I can't seem to send you the messages.
vivekp has quit [Read error: Connection reset by peer]
vivekp has joined #mlpack
cjlcarvalho_ has quit [Remote host closed the connection]
manish7294 has joined #mlpack
< manish7294>
rcurtin: zoq: Hi, I have sent a mail over mailing list regarding an open-source hackathon on behalf of coding club of my college. I request you guys and all mlpack members to consider their proposal.
manish7294 has quit [Quit: Page closed]
< rcurtin>
I saw that, hopefully someone is interested; it sounds like a cool opportunity :)
davida has joined #mlpack
< davida>
zoq: I have been studying the YOLO algorithm for object detection using CNNs. This algorithm requires a set of labels that contain 1 or 0 for an objects presence then if 1 labels will specify the bounding box, whereas if 0 the bounding box labels are DO NOT CARE. I am wondering if there is a way to specify such a DO NOT CARE state with MLPACK?
vivekp has quit [Ping timeout: 244 seconds]
vivekp has joined #mlpack
< zoq>
davida: So the labels matrix has the same width/height as the input image?
< rcurtin>
ok, let's try allowing anyone to chat again... I haven't seem spam in a while now
ChanServ changed the topic of #mlpack to: "mlpack: a scalable C++ machine learning library :: We don't always respond instantly, but we will respond; please be patient :: Logs at http://www.mlpack.org/irc/
< rcurtin>
I think that was the old topic or close to it :)
gendron5000 has joined #mlpack
< zoq>
I think we removed the "C++"
< zoq>
maybe not, not sure :)
< rcurtin>
oh right, technically it is not all C++ :)
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/