Visualizing the evolution
Let's see how to visualize the evolution process. In DEAP, there is a method called Covariance Matrix Adaptation Evolution Strategy (CMA-ES) to visualize evolutions. It is an evolutionary algorithm that's used to solve non-linear problems in the continuous domain. The CMA-ES technique is robust, well studied, and is considered "state-of-the-art" in evolutionary algorithms. Let's see how it works by delving into the source code. The following code is a slight variation of the example shown in the DEAP library.
Create a new Python file and import the following:
import numpy as np
import matplotlib.pyplot as plt
from deap import algorithms, base, benchmarks, \
cma, creator, tools
Define a function to create the toolbox. We will define a FitnessMin function using negative weights:
# Function to create a toolbox
def create_toolbox(strategy):
creator.create("FitnessMin", base.Fitness, weights...