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/
atulim has joined #mlpack
< atulim>
@rcurtin Sir. Did you take a look at my Gsoc proposal? I submitted it on the second last day and wasn't able to get your feedback
atulim has quit [Remote host closed the connection]
abhinavsagar has quit [Quit: Connection closed for inactivity]
< chandramouli_r>
zoq: I am trying to test some models against atari games. I am not able to follow the c++ way of implementation. I am trying to solve them using mlpack in python, but the c++ version I need some help from you. I don't know where to execute the code to perform the actions.
jeffin143 has joined #mlpack
< jeffin143>
rcurtin , zoq : would it be apt to implement sinusoidal activation function in mlpack code base..??
< zoq>
jeffin143: Do you have any model in mind that uses sinc?
< jeffin143>
Zoq : just had a glance over this paper
< jeffin143>
But yes, I don't have a specific model in mind*
< zoq>
jeffin143: I haven't seen a model that used the activation function, so I was curious.
< zoq>
jeffin143: Personally I think we shouldn't add a function that does not provide any benefit; will take a look at the paper.
< jeffin143>
Sure .
< jeffin143>
Also any take on bent identity activation, or gaussian activation function ..??
< zoq>
jeffin143: I think ReLU outperforms gaussian activation function.
< jeffin143>
I do agree with that point, yeah no point in increasing code base*
< zoq>
jeffin143: jeffin143: Not sure about bent identity.
< jeffin143>
I would go through some paper or models to have strong conclusion* , I have seen it being used, but long ago , not so much now.
< jeffin143>
Will get back to you on that *
< zoq>
Sounds good, will go over that paper.
mulx10 has joined #mlpack
< mulx10>
zoq: The PR #1847 is still failing . It is the timelimit I guess. Is it odd that the build crosses the time limit?
< zoq>
mulx10: Strange let me restart the job again. Perhaps it was just temporary.
mulx10 has quit [Ping timeout: 256 seconds]
vivekp has quit [Ping timeout: 255 seconds]
vivekp has joined #mlpack
riaash04 has joined #mlpack
< riaash04>
rcurtin: I have pushed some changes suggested by you in the Isomap pr. Sorry, this took so long. I was unable to work on that for a long time. I have posted some doubts too. Please review whenever you get time. Thanks :)
riaash04 has quit [Client Quit]
xa0 has quit [Ping timeout: 240 seconds]
xa0 has joined #mlpack
riaash04 has joined #mlpack
< riaash04>
I am unable to use the is_symmetric function in armadillo. VSCode refuses to recognise it as part of armadillo and building the project fails. I am using armadillo 8.400 in ubuntu. What could be wrong?