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/
marcosirc has quit [Quit: WeeChat 1.4]
nilay has quit [Ping timeout: 250 seconds]
Mathnerd314 has quit [Ping timeout: 250 seconds]
mentekid has joined #mlpack
mentekid has quit [Ping timeout: 244 seconds]
govg has quit [Ping timeout: 272 seconds]
govg has joined #mlpack
mentekid has joined #mlpack
nilay has joined #mlpack
marcosirc has joined #mlpack
< rcurtin>
marcosirc: what did you use to generate the figures in the pdf in #728? tikz?
< marcosirc>
rcurtin: Hi! I use autocad :)
< marcosirc>
because I am accustomed.
< marcosirc>
In the past, I used to work with my mother in Autocad.
< marcosirc>
Many years ago.
< rcurtin>
ah, okay
< rcurtin>
I was wondering if there was a quick easy tool to make diagrams like the ones you made; I think Autocad is quick but not easy :)
< rcurtin>
usually I just write the tikz commands to draw the figures I want (or use graphviz then dot2tex or something), but drawing a figure is hard when what you are actually doing is just specifying the locations to draw points... it's hard to get it to look exactly right
< marcosirc>
Yeah, I agree this could be uselful. I mean, everything related to metric trees becomes far clearer when we see an image.
< marcosirc>
I can imagine..
< marcosirc>
Autocad is really useful, but I agree that maybe it is not too easy to start with.
< rcurtin>
yeah; I am not already familiar with it and I think it would be very time-consuming to figure it out :)
travis-ci has joined #mlpack
< travis-ci>
mlpack/mlpack#1232 (master - 59bc0b3 : Ryan Curtin): The build is still failing.
< rcurtin>
with you adding backport support for trigamma, I think the right thing to do here is to create a boost_backport.hpp file in core/boost_backport/
< rcurtin>
and inside that file, take all of the boost backporting stuff (including the BOOST_PFTO bit) and put it in boost_backport.hpp
< rcurtin>
along with something like #if BOOST_VERSION < 104800 #include "trigamma.hpp" #endif or whatever is needed
< rcurtin>
I am not sure 104800 is the right minimal version, you'll have to check
govg has joined #mlpack
< mentekid>
you mean instead of having many .hpp files from boost, copy all the file contents into one file?
< rcurtin>
no, I mean, use boost_backport.hpp as a file that you can include which in turn includes all of the backported code
< rcurtin>
sort of like core/arma_extend/arma_extend.hpp
< mentekid>
aha
< mentekid>
I see
< mentekid>
so instead of manually including trigamma from the backported code, I simply include boost_backport.hpp and that manages which file gets included
< mentekid>
cool
< rcurtin>
yeah, just like that
< mentekid>
but I get another error as well from travis:
< mentekid>
return boost::math::trigamma(x, pol);
< mentekid>
../mlpack/core/boost_backport/polygamma.hpp:29:16: error: ‘trigamma’ is not a member of ‘boost::math
< rcurtin>
do you need to include trigamma.hpp first?
< mentekid>
I'm not sure (to be honest I'm a bit lost) but I replaced the boost include directive with my own file
< rcurtin>
can you show me what boost_backport.hpp has in it?
< mentekid>
everything it had before + polygamma.hpp, trigamma.hpp
< mentekid>
oh
< mentekid>
wait I need to copy the detail/ folder too probably
< rcurtin>
yeah, you'll have to be sure that you get whatever polygamma.hpp and trigamma.hpp depend upon
< rcurtin>
and hopefully the structure of the old digamma function is the same
< mentekid>
argh this is the worst dependency hell
< mentekid>
Have to go now :/
< mentekid>
I'll keep working on this tomorrow
mentekid has quit [Ping timeout: 276 seconds]
govg has quit [Quit: leaving]
govg has joined #mlpack
travis-ci has joined #mlpack
< travis-ci>
mlpack/mlpack#1237 (master - 0d9a0e2 : Ryan Curtin): The build is still failing.