[erlang-questions] Why so slow kmeans implementation
Tue Feb 24 15:04:48 CET 2015
I second that.
BTW the process dictionary version is 50 times faster on mine computer.
Anyway I would definitely not use Erlang for this sort of task in the
production. I would write NIF or use port or C-node for this sort of task.
On Tue, Feb 24, 2015 at 11:36 AM, Loïc Hoguin <essen@REDACTED> wrote:
> On 02/24/2015 10:53 AM, Björn Gustavsson wrote:
>> On Tue, Feb 24, 2015 at 9:40 AM, Andrea Peruffo
>> <andrea.peruffo1982@REDACTED> wrote:
>>> is a benchmark of different languages on a kmeans clustering algo.
>>> I made an implementation of it but is terribly slow...
>>> A question is about the data structure "dict" I used, is that the proper
>>> case or there is anything better?
>>> I have checked with "ets" but looks even slower...
>> Using sofs was 10 times faster on my computer:
> If only a kind soul could improve the sofs documentation so it is
> understandable by lesser human beings... :-)
> Loïc Hoguin
> erlang-questions mailing list
-------------- next part --------------
An HTML attachment was scrubbed...
More information about the erlang-questions