Sandbox repo for various genetic algorithms, writing in python. See full documentation/experimentation here.
There are two packages containing problem scripts for both continuous and combinatorial optimisation.
Each package contains a common.py script for the: population generation; selection; crossover; mutation; survive and data visualisation methods can be found (each section seperated by comment headers).
Each problem has a respective script which imports the common methods for use within the main_threaded_loop function (which execute a GA run for each thread defined by the THREADS global param).
Each problem uses a different fitness function defined within each script respectively, with some problems requiring slightly different implementation.
The global parameters for each script are set their respective default value, which can be changed by directly modifying the values within each script.
Each scripts main_threaded_loop uses the 'basic' selection, mutation and crossover methods by default, which can be changed by directly modifying the script to the desired methods.
Each script can be directly run from console, however the scripts for ga-continuous-distrib and ga-continuous-distrib must be in separate packages as each package uses a separate common.py script.\
The following continuous optimisation problems are implemented:
The following combinatorial optimisation problems are implemented: