richardsheridan

# Rough draft of a Queue object that can be used simultaneously from
# sync threads + *multiple* trio and asyncio threads, all at once.
#
# If you don't have multiple threads each doing their own separate calls to Xio.run,
# then don't use this; there are simpler solutions. This was mostly an exercise to
# figure out if and how this could be done.
#
# Currently, the test will provide 94% coverage given sufficient timeout. The
# remaining are (apparently rare) races and the durable aio shielding.
import random

richardsheridan

import queue
import random
from functools import partial
from time import sleep, perf_counter

import trio

CONCURRENCY_LIMIT = 8
limiter = trio.CapacityLimiter(CONCURRENCY_LIMIT)

richardsheridan

N-dimensional test functions for optimization, in Python.

These are the n-dim Matlab functions by A. Hedar (2005), translated to Python-numpy.
http://www-optima.amp.i.kyoto-u.ac.jp/member/student/hedar/Hedar_files/TestGO.htm
ackley dp griew levy mich perm powell power rast rosen schw sphere sum2 trid zakh .m + ellipse nesterov powellsincos

http://imgur.com/Iu9H0jk is a plot of 3 no-derivative algorithms, NLOpt LN_BOBYQA LN_PRAXIS LN_SBPLX,