DarwinPy

Introduction

DarwinPy is a Python-based evolutionary computation module that makes it easy to implement evolutionary methods such as genetic algorithms and evolutionary strategies with only a few lines of code.

Installing DarwinPy

DarwinPy can be installed with the command pip install DarwinPy.

Classes in DarwinPy

Genetics Classes (DarwinPy.Genetics.Genetics)

Methods:

• setSearchSpace(search_space): Sets the search_space attribute of the DarwinPy object.
• setChromosomeLength(chromosome_length): Sets the chromosome_length attribute of the DarwinPy object.
• setPopulationSize(population_size): Sets the population_size attribute of the DarwinPy object.
• setChromosomeMatrix(search_space): Sets the chromosome_matrix attribute of the DarwinPy object.
• getChromosomeMatrix(): Returns the chromosome_matrix attribute, which is a NumPy array.
• getSearchSpace(): Returns the search_space attribute, which is a tuple.
• populate(): Initializes the chromosome_matrix attribute with initial values or gene population.
• select(fitness_vector): Selects mating pairs based on the fitness.
• crossover(crossover_rate): Mates selected pairs and updates the chromosome_matrix attribute.
• mutate(mutation_rate): Mutates the chromosome_matrix attribute of the DarwinPy object.
• evolve(fitness_vector, mutation_rate, crossover_rate): Performs a complete genetic algorithm cycle, which includes selection, crossover, and mutation.

Identifier	Type
setSearchSpace	None
setChromosomeLength	None
setPopulationSize	None
setChromosomeMatrix	None
getChromosomeMatrix	numpy.array
getSearchSpace	tuple
populate	None
select	None
crossover	None
mutate	None
evolve	None

Attributes:

• chromosome_length: The length of each chromosome in the DarwinPy object.
• population_size: The size of the population in the DarwinPy object.
• search_space: The search space for the genetic algorithm, which is a tuple with an upper and lower bound.
• chromosome_matrix: The array of chromosomes in the DarwinPy object.
• pair_list: The list of mating pairs.
• data_type: The data type in which the genetic algorithm is bounded.

Identifier	Type
chromosome_length	int
population_size	int
search_space	tuple
chromosome_matrix	numpy.array
pair_list	list
data_type	type

A Sample DarwinPy implementation

import DarwinPy
import numpy as np

def hammingDist(goal, matrix):
    result = []
    for i in range(len(matrix)):
        temp = 0.
        for j in range(len(goal)):
            if goal[j] == matrix[i][j]:
                temp += 1.
        result.append(temp)
    return result


if __name__ == "__main__":
    mouse_population = 5
    mutation_rate = 0.5
    search_space = (0,1)
    goal = np.array([1,0,1,1,0,1,1, 0,1,1,0],int)
    chromosome_length = len(goal)

    mouse_species = DarwinPy.Genetics.Genetics(chromosome_length,
    mouse_population, (0,1), int)
    print(f"mouse species instantiated:\n {mouse_species}")

    print("the goal:\n {}".format(goal))

    mouse_species.populate()
    print(f"get mouse matrix(GA):\n {mouse_species.getChromosomeMatrix()}")


    fitness_vector = np.array(
    hammingDist(goal,mouse_species.getChromosomeMatrix()),
    float)

    print(f"get fitness vector: {fitness_vector}")

    is_goal = False
    gen = 1
    while is_goal == False:
        print(f"Generation #{gen}\n")
        gen += 1
        mouse_species.evolve(fitness_vector,
        mutation_rate, 0.5)
        print(f"get mouse matrix(GA):\n {mouse_species.getChromosomeMatrix()}")
        fitness_vector = np.array(
        hammingDist(goal,mouse_species.getChromosomeMatrix()),
        float)

        print(f"get fitness vector: {fitness_vector}")
        if chromosome_length in fitness_vector:
            is_goal = True