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/
< akfluffy>
zoq: what's the difference?
akfluffy has left #mlpack []
Masstran_ has joined #mlpack
Masstran has quit [Ping timeout: 245 seconds]
akfluffy has joined #mlpack
Yashwants19 has joined #mlpack
< Yashwants19>
Thank You rcurtin for your time and valuable feedback. :)
Yashwants19 has quit [Client Quit]
pd09041999 has joined #mlpack
akfluffy has left #mlpack []
sreenik has joined #mlpack
pd09041999 has quit [Ping timeout: 250 seconds]
pd09041999 has joined #mlpack
seewishnew has joined #mlpack
seewishnew has quit [Remote host closed the connection]
seewishnew has joined #mlpack
seewishnew has quit [Ping timeout: 252 seconds]
seewishnew has joined #mlpack
seewishnew has quit [Remote host closed the connection]
seewishnew has joined #mlpack
seewishnew has quit [Remote host closed the connection]
seewishnew has joined #mlpack
seewishnew has quit [Ping timeout: 246 seconds]
seewishnew has joined #mlpack
seewishnew has quit [Remote host closed the connection]
seewishnew has joined #mlpack
seewishnew has quit [Remote host closed the connection]
seewishnew has joined #mlpack
seewishnew has quit [Remote host closed the connection]
seewishnew has joined #mlpack
seewishnew has quit [Remote host closed the connection]
seewishnew has joined #mlpack
seewishnew has quit [Ping timeout: 268 seconds]
seewishnew has joined #mlpack
mulx10 has joined #mlpack
seewishnew has quit [Client Quit]
< mulx10>
akfluffy, zoq: akfluffy you are correct. The parameter size is set in Reset().
< mulx10>
I guess that's the reason only Predict was not working. I faced this issue when trying to add cell param in the LSTM layer.
< mulx10>
By only Predict I mean calling Predict() without Train().
< mulx10>
ShikharJ: Sorry for the delay. I have rectified the PR.
< mulx10>
Sumedh, zoq, rcurtin: Thank you for taking out the time to review my proposal.
< jenkins-mlpack2>
noreply: Merge pull request #1800 from MuLx10/LSTMCell
rf_sust2018 has quit [Quit: Leaving.]
rf_sust2018 has joined #mlpack
rf_sust2018 has quit [Quit: Leaving.]
pd09041999 has quit [Ping timeout: 264 seconds]
pd09041999 has joined #mlpack
rf_sust2018 has joined #mlpack
sreenik has quit [Ping timeout: 256 seconds]
rf_sust2018 has quit [Ping timeout: 264 seconds]
rf_sust2018 has joined #mlpack
rf_sust2018 has quit [Quit: Leaving.]
rf_sust2018 has joined #mlpack
rf_sust2018 has quit [Quit: Leaving.]
sreenik has joined #mlpack
< KimSangYeon-DGU>
rcurtin: Thanks for reviewing my proposal, and I answered the comments. Can you take a look, if you have a chance?
favre49 has joined #mlpack
favre49 has quit [Client Quit]
favre49 has joined #mlpack
favre49 has quit [Client Quit]
< jeffin>
What is the equivalent of dense layer in mlpack..?
mulx10 has joined #mlpack
< mulx10>
jeffin: You can use `Linear`.
rf_sust2018 has joined #mlpack
sreenik has quit [Ping timeout: 256 seconds]
rf_sust2018 has quit [Ping timeout: 268 seconds]
Suryo has joined #mlpack
rf_sust2018 has joined #mlpack
sreenik has joined #mlpack
< Suryo>
zoq, rcurtin: hey, I know that you guys are really busy right now but if you could take another look at my proposal and leave some final comments, it would be great... I've updated some aspects of the deliverables and also the tentative plan of action.
< zoq>
Suryo: I'll take another look over some applications later today.
< Suryo>
Thanks a lot zoq!
favre49 has joined #mlpack
< favre49>
zoq: I was also hoping you could look at the timeline of my project and tell me whether I have packed too much into too little time.
< zoq>
favre49: Sure
< favre49>
I am considering adding certain features like the tests for multi-objective optimization before the start of the coding period (if my proposal is accepted of course)
< favre49>
thanks :)
mulx10 has quit [Ping timeout: 256 seconds]
Suryo has quit [Quit: Page closed]
< jeffin>
Mulx10 : yup , thanks
mulx10 has joined #mlpack
rf_sust2018 has quit [Ping timeout: 268 seconds]
saksham189 has joined #mlpack
chandramouli_r has joined #mlpack
chandramouli_r has quit [Client Quit]
Sabeel has joined #mlpack
pd_09041999 has joined #mlpack
pd09041999 has quit [Ping timeout: 264 seconds]
akfluffy has joined #mlpack
< akfluffy>
mulx10: I think rnn_impl.hpp already checks if the matrix is empty and calls Reset(), no?
pd_09041999 has quit [Ping timeout: 246 seconds]
Yashwants19 has joined #mlpack
< Yashwants19>
Hi rcurtin: I have answered the comments and updated the proposal as suggested by you.Please take a look on my updated proposal and leave some final comments.
< Yashwants19>
Thank You :)
Yashwants19 has quit [Client Quit]
< mulx10>
akfluffy: Yes it does but that doesn't explain why matrix shape is 0x0.
< mulx10>
akfluffy: I was wondering if what could possibly go wrong in reset(), any ideas?
akfluffy` has joined #mlpack
akfluffy has quit [Ping timeout: 268 seconds]
mulx10 has quit [Quit: Page closed]
favre49 has quit [Ping timeout: 256 seconds]
< akfluffy`>
mulx10: I can't figure it out. It looks like it should work but it obviously does not. I'm going to see if there are any inconsistencies between FFN Reset and RNN Reset
< zoq>
Suryo: For some reason I can't comment on the pdf, maybe something wrong with the permissions?
Sabeel has quit [Ping timeout: 256 seconds]
pd_09041999 has joined #mlpack
yashMustak has joined #mlpack
Toshal has joined #mlpack
< Toshal>
zoq ShikharJ rcurtin: Hi sir, this is just as a reminder as only three days are left. Please, take a look at the proposal I submitted. Even a quick glance will be helpful.
< zoq>
Toshal: Hello there, what's the title of the application?
< Toshal>
zoq : Implementing Essential Deep Learning Modules. Thanks for the prompt reply.
favre49 has joined #mlpack
< favre49>
zoq: I saw your comment on my proposal's, and thought it would be simpler to have the discussion on IRC. I saw you advised me to keep NSGA as something to do only if I have the time left.
< favre49>
However, I need NSGA to implement MM-NEAT
< favre49>
Would you advise I take a different direction with the project, given the time constraints?
< zoq>
favre49: I see, mazbe it works the other way around implement MM-NEAT if time allows?
favre49 has quit [Ping timeout: 256 seconds]
< zoq>
Toshal: Can you comment on the pdf?
< abhinavsagar>
@zoq: Could you please take a look at my proposal and tell me what changes do I need to make in it?
< zoq>
abhinavsagar: Sure, What's the title? Makes it easier for me to find the application.
< abhinavsagar>
@zoq: Reinforcement learning. Thanks for the reply.
favre49 has joined #mlpack
< favre49>
zoq: Yup, thats my plan. I've mentioned that in the body of the proposal, but left the timetable rather idealized. Should i keep the timetable more realistic?
< zoq>
favre49: Probably a good idea, or make that point more clear.
< Toshal>
zoq: Actually No. The fact is I didn't made my proposal on docs it's alignment is getting messed up in google docs.
< favre49>
okay, thanks for the advice :)
favre49 has quit [Client Quit]
< zoq>
hm, I was able to comment on another application; are you able to share this in another way just to make comments?
< zoq>
another application that was a pdf as well
< Toshal>
zoq: yes we can comment, sorry. There will be chat symbol in the top right corner of the pdf alongside printer symbol. With that we can comment.
< Toshal>
Okay I found what is the issue I will upload new link.
< Toshal>
zoq: Reopen the link it's fixed. Sorry for the delay.
< ShikharJ>
Hmm, I think it should be a general advice to everyone to include as much detail as you can in your proposal. We'll be judging your application seriously on your proposal, and the better the re-assurance it gives us about your expertise, the better it is.
< ShikharJ>
So don't shy away from including what, how and why of the features you're planning to implement.
< ShikharJ>
A simple "I plan to implement X. X is used for blah blah blah" doesn't provide as much confidence as "I plan to implement X. This should be introduced in mlpack because ... I plan to implement X through Y. A prototype is Z..."
< dhawgupta>
Hey, thanks for the details I will keep this in kind. ;D
typoerfvs has quit [Ping timeout: 256 seconds]
dhawgupta has quit [Quit: dhawgupta]
sreenik has quit [Quit: Page closed]
< jeffin>
ShikarJ : thanks for the input, would be happy if you could see my proposal and suggest something to add upon
< jeffin>
Thanks
saksham189 has quit [Quit: Page closed]
< akfluffy`>
mulx10: let me know if you figure it out
< akfluffy`>
Hey, so I have a question about RL (specifically q_learning)
< akfluffy`>
Basically the method of DQN is to approximate the Q function right?