On 1/17/12 5:22 PM, blcooley wrote:
On Jan 17, 5:43 pm, Sam Ritchie<sritchi...@gmail.com> wrote:
Update -- I wrapped the JBLAS library and ended up with stellar
performance, about 3x faster than numpy.
https://gist.github.com/264a2756fc657140fdb8
You might not be interested at this point, but you can get better
performance than using the Incanter wrappers by just using Parallel
Colt directly (I'm just using the Parallel Colt library from within
Incanter by listing Incanter as a dependency in my project file).
https://gist.github.com/1629970
I don't know what your machine specs are, but the times are within
reason (3871 ms) on my MBP.
Best regards,
Brian Cooley
I shared my own Parallel Colt wrappers with Sam yesterday and decided to
release them as a library. The nascent project can be found on github:
https://github.com/bmabey/claw/blob/master/src/claw/core.clj
Sam and I (and anyone else who wants to!) are going to flesh out the API
a bit more and explore other libraries (such as UJMP). I'll make an
announcement on this list when the API settles down a bit and it is
ready for 1.0.
-Ben
--
You received this message because you are subscribed to the Google
Groups "Clojure" group.
To post to this group, send email to clojure@googlegroups.com
Note that posts from new members are moderated - please be patient with your
first post.
To unsubscribe from this group, send email to
clojure+unsubscr...@googlegroups.com
For more options, visit this group at
http://groups.google.com/group/clojure?hl=en
