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/
sumedhghaisas has joined #mlpack
mikeling has quit [Quit: Connection closed for inactivity]
abhinav has joined #mlpack
sumedhghaisas has quit [Ping timeout: 256 seconds]
abhinav has quit [Remote host closed the connection]
mikeling has joined #mlpack
jenkins-mlpack has quit [Ping timeout: 245 seconds]
cult- has joined #mlpack
< cult->
hi again.
< cult->
rcurtin: i tried with the real data now. and without the transition redefining hack, the model was still unable to pick up the separate guassian dimensions.
< cult->
my question is: will you push this fix into master anytime soon?!
< cult->
alternatively, should i normalize my data before training it?
< cult->
rcurtin: is this c.col(i) /= arma::accu(c.col(i)); to make sure that the states are incrementing/decrementing by 1 states only at a time?
< mikeling>
my local system is os x 10.12 and my gcc version is 6.2.0
< mikeling>
after I google it, I found it's a unlink issue, but how could I link to mlpack when I compile it by g++? I'm a totally freshman in C/C++, sorry
< zoq>
or g++ cov_example.cpp -std=c++11 -lmlpack -larmadillo -lboost_serialization -lboost_program_options
< mikeling>
zoq: hello, thank you for your helping! :) But it told me "ld: library not found for -l/usr/local/lib/libarmadillo.dylib" this time when I try to use g++ cov_example.cpp -std=c++11 `pkg-config --libs mlpack`
< mikeling>
but actually I do found libarmadillo.dylib in /usr/local/lib