Using D and std.ndslice as a Numpy Replacement

Re:Using D and std.ndslice as a Numpy Replacement by snydly on Saturday January 02, 2016 @07:07PM (#10828708)

Am I correct in thinking that this is only reasonable if you're already using D? The switching cost seems too high if you're python everything.

I tried the Armadillo C++ library a while ago (http://arma.sourceforge.net/). The speed up and time spend learning the syntax didn't seem worth it.

[ Parent | on HN ]

Re:Using D and std.ndslice as a Numpy Replacement by bionsuba on Saturday January 02, 2016 @07:15PM (#10828751)

I completely understand that for existing projects it might not make sense to switch, but as I say at the start of the article

    why you should consider D for your next numerical project.

[ Parent | on HN ]

Re:Using D and std.ndslice as a Numpy Replacement by snydly on Saturday January 02, 2016 @09:02PM (#10829141)

I just ran your column means benchmark. It is pretty impressive... maybe my C++ is trash. Never saw that much speedup with Arma. I'll try it out :)

[ Parent | on HN ]