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/
rf_sust2018 has quit [Quit: Leaving.]
bhairavm has joined #mlpack
< bhairavm>
have there been any discussions about the simulator GSOC project?
manish7294 has quit [Ping timeout: 256 seconds]
manish7294 has joined #mlpack
bhairavm has quit [Quit: Page closed]
manish7294 has quit [Ping timeout: 245 seconds]
manish7294 has joined #mlpack
ShikharJ has joined #mlpack
ShikharJ has quit [Quit: Page closed]
travis-ci has joined #mlpack
< travis-ci>
ShikharJ/mlpack#102 (master - 0e14562 : Ryan Curtin): The build has errored.
csoni has quit [Read error: Connection reset by peer]
csoni has joined #mlpack
s1998_ has joined #mlpack
csoni2 has joined #mlpack
csoni3 has joined #mlpack
csoni2 has quit [Read error: Connection reset by peer]
csoni has quit [Ping timeout: 260 seconds]
csoni3 has quit []
csoni has joined #mlpack
thepro has joined #mlpack
< thepro>
can Adaboost be only used with the decision tree stumps or any other kind of classifiers?
< rcurtin>
thepro: you should be able to use it with any learner that supports instance weights
< rcurtin>
but AdaBoost is generally most successful when used with weaker learners
< rcurtin>
one example of weak learner you could use is the perceptron
csoni has quit [Read error: Connection reset by peer]
Atharva has joined #mlpack
csoni has joined #mlpack
< Atharva>
The readme file in tutorials folder has wrong link to tutorials webpage, in the first line.
yashsharan has joined #mlpack
< yashsharan>
hello.If I'm interested in two different ideas should I prepare two proposals or one? I'm interested in reinforcement learning as well as Bidirectional RNN idea for my GSoC project.
< yashsharan>
Also I was wondering for the reinforcement learning idea will the algorithm be implemented only on cartpole and mountain cart environment since these are the only two environment available in mlpack repo?
< rcurtin>
Atharva: can you open an issue or PR for that? I'm not at my desk right now so it's likely I'll forget
< rcurtin>
yashsharan: you can do either, but usually I would suggest that it's better to focus on one proposal instead of two
< Atharva>
rcurtin: I will open a PR
daniel has joined #mlpack
daniel is now known as Guest88812
travis-ci has joined #mlpack
< travis-ci>
ShikharJ/mlpack#106 (ConvolutionalLayer - 4128d92 : Shikhar Jaiswal): The build has errored.
sumedhghaisas has quit [Ping timeout: 256 seconds]
sumedhghaisas has joined #mlpack
csoni has joined #mlpack
csoni has quit [Read error: Connection reset by peer]
csoni2 has joined #mlpack
csoni2 has quit [Ping timeout: 260 seconds]
sumedhghaisas has quit [Ping timeout: 260 seconds]
csoni has joined #mlpack
< yashsharan>
Hello.Is it possible to add Atari game environment from openAI gym? I am doubtful about this because in the source code for atari game environment there is an import atari statement which i think isnt currently present in ml pack
rf_sust2018 has left #mlpack []
sumedhghaisas has joined #mlpack
yashsharan has quit [Ping timeout: 260 seconds]
sumedhghaisas2 has joined #mlpack
sumedhghaisas has quit [Ping timeout: 276 seconds]
sumedhghaisas2 has quit [Read error: Connection reset by peer]
sumedhghaisas has joined #mlpack
thepro has quit [Ping timeout: 260 seconds]
thepro has joined #mlpack
thepro has quit [Client Quit]
thepro has joined #mlpack
bhuwnesh has joined #mlpack
sumedhghaisas has quit [Read error: Connection reset by peer]
sumedhghaisas has joined #mlpack
sumedhghaisas2 has joined #mlpack
sumedhghaisas has quit [Read error: Connection reset by peer]
csoni has quit [Ping timeout: 246 seconds]
dan_ has joined #mlpack
dan_ has quit [Client Quit]
dan_ has joined #mlpack
dan_ is now known as Guest37764
sumedhghaisas2 has quit [Read error: Connection reset by peer]
< Guest37764>
excuse me but how should I identify my username??
sumedhghaisas has joined #mlpack
sumedhghaisas has quit [Read error: Connection reset by peer]
Guest37764 has quit [Quit: Page closed]
csoni has joined #mlpack
daniel has joined #mlpack
sumedhghaisas has joined #mlpack
daniel is now known as Guest51721
sumedhghaisas has quit [Ping timeout: 240 seconds]
Guest51721 has quit [Quit: Page closed]
thepro has quit [Ping timeout: 246 seconds]
yashsharan has joined #mlpack
sourabhvarshney1 has joined #mlpack
< sourabhvarshney1>
@zoq I am not getting why are there two implementation of rnn in ann codebase? Is this the case that one is due to addition of a layer in a network and other is the implementation of a general rnn?
csoni has quit [Ping timeout: 264 seconds]
csoni has joined #mlpack
csoni2 has joined #mlpack
csoni has quit [Read error: Connection reset by peer]
< zoq>
yashsharan: You could use https://github.com/zoq/gym_tcp_api, the main purpose of the environments in the codebase is to have something easy to run tests against.
< zoq>
sourabhvars: Hello, you are right, besides adding a layer, the class also preprocesses the input/output pairs for the optimization step.
< sourabhvarshney1>
@zoq I am looking forward to adding brnn and blstm to the codebase this summer. Should I add it like adding blstm in the layer only and brnn both in the layer and as a general method also?
< zoq>
I think RNN class is just fine, that way it will effect any other layer.
< sourabhvarshney1>
I think I can also add one more deep learning module. But I am confused between RBFN and DBN. Can you suggest me one which can be collaborated with these?
< zoq>
collaborated with brnn?
yashsharan has quit [Quit: Page closed]
< sourabhvarshney1>
@zoq Regarding the RNN class. This resolves to just addition of two types of layes of blstm and brnn in ann codebase of mlpack.
< sourabhvarshney1>
@zoq Regarding the collabration. Sorry for the unclear words. I just wanted to know which of these two can be done with brnn and blstm during the time/
< zoq>
sourabhvars: blstm is brrn + lstm
< zoq>
sourabhvars: Ah I see, both.
< sourabhvarshney1>
@zoq Yes. That is what I wanted to say.
< sourabhvarshney1>
I think once brnn is done, blstm can be done quickly. So, I think I can add one more algo. What say @zoq
< zoq>
Agreed, I don't think brnn is enough work for the summer.
< sourabhvarshney1>
So, should I go with brnn with brfn or brnn with dbn. And later can continue with blstm after summer.
qwebirc819267 has joined #mlpack
sourabhvarshney1 has quit [Ping timeout: 260 seconds]
< qwebirc819267>
@zoq I am going with this research paper for brnn: Bidirectional Recurrent Neural Networks by Mike Schuster and Kuldip K. Paliwal. Will this be fine?
qwebirc819267 has quit [Client Quit]
sourabhvarshney1 has joined #mlpack
< sourabhvarshney1>
@zoq I am going with this research paper for brnn: Bidirectional Recurrent Neural Networks by Mike Schuster and Kuldip K. Paliwal. Will this be fine?
swetha_ has joined #mlpack
fsub has joined #mlpack
< fsub>
hello
fsub has quit [Read error: Connection reset by peer]
fsub has joined #mlpack
< sourabhvarshney1>
@zoq Searched for rbfn research paper and found: Multivariable Functional Interpolation and Adaptive Networks By Broomhead and Lowe. Should this be used?
sourabhvarshney1 has quit [Ping timeout: 260 seconds]
csoni2 has quit [Read error: Connection reset by peer]
< zoq>
sourabhvars: Yes, both are good to dive in.
< zoq>
fsub: Hello there!
sourabhvarshney1 has joined #mlpack
< zoq>
sourabhvars: Go with the method you like the most, I think both are interesting.
csoni has joined #mlpack
< sourabhvarshney1>
@zoq I would go with the RBFN. Can you check the papers which I just listed out. Are they fine?
< zoq>
sourabhvars: yes, both are good to dive into the topic.
csoni has quit [Read error: Connection reset by peer]
fsub has quit [Quit: Leaving]
csoni has joined #mlpack
< sourabhvarshney1>
Thanks @zoq for help. I would submit the proposal asap.
bhuwnesh has quit [Quit: Connection closed for inactivity]
ImQ009 has quit [Quit: Leaving]
csoni has quit [Ping timeout: 240 seconds]
csoni has joined #mlpack
sourabhvarshney1 has quit [Quit: Page closed]
rf_sust2018 has joined #mlpack
< rf_sust2018>
rcurtin: can you please tell me why Cython files are to be generated for python bindings for mlpack befire using SWIG
< rf_sust2018>
*before
rf_sust2018 has quit [Quit: Leaving.]
< rcurtin>
rf_sust2018: swig is not being used for mlpack bindingd