This is a restructured version of the Modular CMA-ES framework, first introduced in ... Its modular design allows for the creation of thousands of variants of the CMA-ES algorithm.
Installation can be done via pip, using:
$ pip install modcma
To optimize a single function, we provide a basic fmin interface, which can be used as follows:
from modcma import modularcmaes
modularcmaes.fmin(func=sum, dim=5, maxfun=100)
Running tests
$ python3 -m unittest discover
Running optimizer
$ python3 -m modcma [-h] [-f FID] [-d DIM] [-i ITERATIONS] [-l] [-c] [-L LABEL] [-s SEED] [-a ARGUMENTS [ARGUMENTS ...]]