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/
kris1 has joined #mlpack
partobs-mdp has joined #mlpack
kris1 has quit [Quit: kris1]
< partobs-mdp>
zoq: Checked both manually and with some ad-hoc Python + Numpy code, didn't really find anything nasty. What about you?
< partobs-mdp>
By the way, do we have some type that would be inherited by both FFN and RNN? It would be nice since then we could test both MLP and LSTM controllers. (But if there is no such type, I think we'd better bite the bullet and do it immediately with RNN controller, just as described in the paper)
< partobs-mdp>
(Oops, sorry, forgot about templates - we could do typename C = FFN<CrossEntropyError<>> and run whatever methods we need)
kris1 has joined #mlpack
< partobs-mdp>
zoq: rcurtin: Finished working on controller - the complete blind HAM is also there. If I didn't mess up with the numbers, then it should be okay :)
partobs-mdp has quit [Remote host closed the connection]
kris1 has quit [Quit: kris1]
kris1 has joined #mlpack
partobs-mdp has joined #mlpack
kris1 has quit [Quit: kris1]
kris1 has joined #mlpack
shikhar has joined #mlpack
shikhar has quit [Ping timeout: 255 seconds]
< rcurtin>
partobs-mdp: great! :) I will try to give a review in the next couple of days
govg has joined #mlpack
shikhar has joined #mlpack
govg has quit [Ping timeout: 240 seconds]
govg has joined #mlpack
partobs-mdp has quit [Read error: Connection reset by peer]
shikhar has quit [Quit: WeeChat 1.7]
vivek___ has joined #mlpack
< vivek___>
Hello
vivek___ has quit [Client Quit]
< zoq>
vivek___: Hello there.
< zoq>
too late
kris1 has quit [Quit: kris1]
kris1 has joined #mlpack
kris1 has quit [Ping timeout: 246 seconds]
< ironstark>
rcurtin: zoq: How do I install new packages in R. for e.g on doing install.packages("mlr") the Rscript halts
< ironstark>
Also on running libraries/bin/R
< ironstark>
and then running install.packages("mlr") I am getting package ‘mlr’ is not available (for R version 3.4.1)
< ironstark>
I am getting the same for e1071 too
kris1 has joined #mlpack
< zoq>
ironstark: Did you use R from the jenkins build?
< ironstark>
zoq: yes
< ironstark>
I ran the script which you had mentioned in the PR
< ironstark>
and then R started working
< ironstark>
but I am unable to install/use the packages which are not already installed
< zoq>
ironstark: Ah okay, so if you use the binary from the PR, only the jenkins user can install packages in the build folder.
< zoq>
ironstark: I've installed the package for you, but once the benchmark job runs again it will be deleted, what you could do is to install everything in your home folder.
< rcurtin>
yeah, agreed, I always keep a local copy of the benchmark repo on slake with the libraries built for testing
< ironstark>
zoq: okat
< ironstark>
*okay
< zoq>
If you write the R script the first step should install the necessary package: e.g. run install.packages('mlr', repos='http://cran.us.r-project.org') first
< zoq>
rcurtin: Right it's super simple just call 'make setup' and you are good to go :)