Forums  > Software  > Scalalab Fslab, functional/quasi functional languages  
     
Page 3 of 3Goto to page: 1, 2, [3] Prev
Display using:  

rickyvic


Total Posts: 186
Joined: Jul 2013
 
Posted: 2016-07-22 13:18
Hopefully interest will grow...

"amicus Plato sed magis amica Veritas"

rickyvic


Total Posts: 186
Joined: Jul 2013
 
Posted: 2019-04-03 21:31
News about julia, they mix with C++ in prod it appears
https://discourse.julialang.org/t/maven-securities-hft-systems-developer/17517


"amicus Plato sed magis amica Veritas"

secsy


Total Posts: 10
Joined: Mar 2019
 
Posted: 2019-04-04 18:43
rickyvic, do you know of any good resources for learning julia? I do most of my work in python and drop down to C++ for performance, but would love to add another language if possible; julia does look interesting

rickyvic


Total Posts: 186
Joined: Jul 2013
 
Posted: 2019-04-05 11:11
The risk that I see is that it might die and you are left with a dead language in your toolbox.
I started learning it but I am now refocusing on learning C well, so no time. By the way C interop with julia is easy.
If you type julialang a lot comes out from the language main website too. Should be easy to get started.

"amicus Plato sed magis amica Veritas"

secsy


Total Posts: 10
Joined: Mar 2019
 
Posted: 2019-04-05 22:32
gotcha, yeah I think I would just learn it as an interesting new tool, so wouldn't matter too much if its usage died off. I'll check out some resources

quantie


Total Posts: 889
Joined: Jun 2004
 
Posted: 2019-04-24 02:57
Why not just use R? R came from S and Scheme (lisp family) it has everything you would want anonymous functions, closure etc.

rickyvic


Total Posts: 186
Joined: Jul 2013
 
Posted: 2019-04-24 15:49
R is slow and cannot be used in real time. It is efficient at doing vectorized stuff but not so much at looping.
I use it extensively although it is not the solution to our problem.
I find MATLAB faster at pretty much everything and it can be deployed in production better.

"amicus Plato sed magis amica Veritas"

gmetric_Flow


Total Posts: 13
Joined: Oct 2016
 
Posted: 2019-04-24 16:41
Fiddling around with Mex files you can increase performance. Whether that is easier than Rcpp (which ends up slower regardless) or Cython/Numba depends on what you're trying to do and your familiarity with the language.


https://www.sas.upenn.edu/~jesusfv/Update_March_23_2018.pdf
Previous Thread :: Next Thread 
Page 3 of 3Goto to page: 1, 2, [3] Prev