def __init__(self,

num_generations,

num_parents_mating,

fitness_func,

initial_population=None,

sol_per_pop=None,

num_genes=None,

init_range_low=-4,

init_range_high=4,

parent_selection_type="sss",

keep_parents=-1,

K_tournament=3,

crossover_type="single_point",

crossover_probability=None,

mutation_type="random",

mutation_probability=None,

mutation_by_replacement=False,

mutation_percent_genes=10,

mutation_num_genes=None,

random_mutation_min_val=-1.0,

random_mutation_max_val=1.0,

gene_space=None,

callback_generation=None,

delay_after_gen=0.0):

"""

self.gene_space_nested = False

if type(gene_space) is type(None):

pass

elif type(gene_space) in [list, tuple, range]:

if len(gene_space) == 0:

self.valid_parameters = False

raise TypeError("'gene_space' cannot be empty (i.e. its length must be >= 0).")

else:

for index, el in enumerate(gene_space):

if type(el) in [list, tuple, range]:

if len(el) == 0:

self.valid_parameters = False

raise TypeError("The element indexed {index} of 'gene_space' with type {el_type} cannot be empty (i.e. its length must be >= 0).".format(index=index, el_type=type(el)))

else:

for val in el:

if not (type(val) in [int, float]):

raise TypeError("All values in the sublists inside the 'gene_space' attribute must be numeric of type int/float but the value ({val}) of type {typ} found.".format(val=val, typ=type(val)))

self.gene_space_nested = True

elif type(el) == type(None):

self.gene_space_nested = True

elif not (type(el) in [int, float]):

self.valid_parameters = False

raise TypeError("Unexpected type {el_type} for the element indexed {index} of 'gene_space'. The accepted types are list/tuple/range of numbers, a single number (int/float), or None.".format(index=index, el_type=type(el)))

else:

self.valid_parameters = False

raise TypeError("The expected type of 'gene_space' is list, tuple, or range but {gene_space_type} found.".format(gene_space_type=type(gene_space)))

if self.gene_space_nested:

if len(gene_space) != num_genes:

self.valid_parameters = False

raise TypeError("When the parameter 'gene_space' is nested, then its length must be equal to the value passed to the 'num_genes' parameter. Instead, length of gene_space ({len_gene_space}) != num_genes ({len_num_genes})".format(len_gene_space=len(gene_space), len_num_genes=num_genes))

self.gene_space = gene_space

self.init_range_low = init_range_low

self.init_range_high = init_range_high

if initial_population is None:

if (sol_per_pop is None) or (num_genes is None):

raise ValueError("Error creating the initail population\n\nWhen the parameter initial_population is None, then neither of the 2 parameters sol_per_pop and num_genes can be None at the same time.\nThere are 2 options to prepare the initial population:\n1) Create an initial population and assign it to the initial_population parameter. In this case, the values of the 2 parameters sol_per_pop and num_genes will be deduced.\n2) Allow the genetic algorithm to create the initial population automatically by passing valid integer values to the sol_per_pop and num_genes parameters.")

elif (type(sol_per_pop) is int) and (type(num_genes) is int):

# Validating the number of solutions in the population (sol_per_pop)

if sol_per_pop <= 0:

self.valid_parameters = False

raise ValueError("The number of solutions in the population (sol_per_pop) must be > 0 but {sol_per_pop} found. \nThe following parameters must be > 0: \n1) Population size (i.e. number of solutions per population) (sol_per_pop).\n2) Number of selected parents in the mating pool (num_parents_mating).\n".format(sol_per_pop=sol_per_pop))

# Validating the number of gene.

if (num_genes <= 0):

self.valid_parameters = False

raise ValueError("Number of genes cannot be <= 0 but {num_genes} found.\n".format(num_genes=num_genes))

# When initial_population=None and the 2 parameters sol_per_pop and num_genes have valid integer values, then the initial population is created.

# Inside the initialize_population() method, the initial_population attribute is assigned to keep the initial population accessible.

self.num_genes = num_genes # Number of genes in the solution.

self.sol_per_pop = sol_per_pop # Number of solutions in the population.

self.initialize_population(self.init_range_low, self.init_range_high)

else:

raise TypeError("The expected type of both the sol_per_pop and num_genes parameters is int but {sol_per_pop_type} and {num_genes_type} found.".format(sol_per_pop_type=type(sol_per_pop), num_genes_type=type(num_genes)))

elif numpy.array(initial_population).ndim != 2:

raise ValueError("A 2D list is expected to the initail_population parameter but a {initial_population_ndim}-D list found.".format(initial_population_ndim=numpy.array(initial_population).ndim))

else:

self.initial_population = numpy.array(initial_population)

self.population = self.initial_population # A NumPy array holding the initial population.

self.num_genes = self.initial_population.shape[1] # Number of genes in the solution.

self.sol_per_pop = self.initial_population.shape[0] # Number of solutions in the population.

self.pop_size = (self.sol_per_pop,self.num_genes) # The population size.

# Validating the number of parents to be selected for mating (num_parents_mating)

if num_parents_mating <= 0:

self.valid_parameters = False

raise ValueError("The number of parents mating (num_parents_mating) parameter must be > 0 but {num_parents_mating} found. \nThe following parameters must be > 0: \n1) Population size (i.e. number of solutions per population) (sol_per_pop).\n2) Number of selected parents in the mating pool (num_parents_mating).\n".format(num_parents_mating=num_parents_mating))

# Validating the number of parents to be selected for mating: num_parents_mating

if (num_parents_mating > self.sol_per_pop):

self.valid_parameters = False

raise ValueError("The number of parents to select for mating ({num_parents_mating}) cannot be greater than the number of solutions in the population ({sol_per_pop}) (i.e., num_parents_mating must always be <= sol_per_pop).\n".format(num_parents_mating=num_parents_mating, sol_per_pop=self.sol_per_pop))

self.num_parents_mating = num_parents_mating

# crossover: Refers to the method that applies the crossover operator based on the selected type of crossover in the crossover_type property.

# Validating the crossover type: crossover_type

if (crossover_type == "single_point"):

self.crossover = self.single_point_crossover

elif (crossover_type == "two_points"):

self.crossover = self.two_points_crossover

elif (crossover_type == "uniform"):

self.crossover = self.uniform_crossover

elif (crossover_type is None):

self.crossover = None

else:

self.valid_parameters = False

raise ValueError("Undefined crossover type. \nThe assigned value to the crossover_type ({crossover_type}) argument does not refer to one of the supported crossover types which are: \n-single_point (for single point crossover)\n-two_points (for two points crossover)\n-uniform (for uniform crossover).\n".format(crossover_type=crossover_type))

self.crossover_type = crossover_type

if crossover_probability == None:

self.crossover_probability = None

elif type(crossover_probability) in [int, float]:

if crossover_probability >= 0 and crossover_probability <= 1:

self.crossover_probability = crossover_probability

else:

self.valid_parameters = False

raise ValueError("The value assigned to the 'crossover_probability' parameter must be between 0 and 1 inclusive but {crossover_probability_value} found.".format(crossover_probability_value=crossover_probability))

else:

self.valid_parameters = False

raise ValueError("Unexpected type for the 'crossover_probability' parameter. Float is expected by {crossover_probability_type} found.".format(crossover_probability_type=type(crossover_probability)))

# mutation: Refers to the method that applies the mutation operator based on the selected type of mutation in the mutation_type property.

# Validating the mutation type: mutation_type

if (mutation_type == "random"):

self.mutation = self.random_mutation

elif (mutation_type == "swap"):

self.mutation = self.swap_mutation

elif (mutation_type == "scramble"):

self.mutation = self.scramble_mutation

elif (mutation_type == "inversion"):

self.mutation = self.inversion_mutation

elif (mutation_type is None):

self.mutation = None

else:

self.valid_parameters = False

raise ValueError("Undefined mutation type. \nThe assigned value to the mutation_type argument ({mutation_type}) does not refer to one of the supported mutation types which are: \n-random (for random mutation)\n-swap (for swap mutation)\n-inversion (for inversion mutation)\n-scramble (for scramble mutation).\n".format(mutation_type=mutation_type))

self.mutation_type = mutation_type

if mutation_probability == None:

self.mutation_probability = None

elif type(mutation_probability) in [int, float]:

if mutation_probability >= 0 and mutation_probability <= 1:

self.mutation_probability = mutation_probability

else:

self.valid_parameters = False

raise ValueError("The value assigned to the 'mutation_probability' parameter must be between 0 and 1 inclusive but {mutation_probability_value} found.".format(mutation_probability_value=crossover_probability))

else:

self.valid_parameters = False

raise ValueError("Unexpected type for the 'mutation_probability' parameter. Float is expected by {mutation_probability_type} found.".format(mutation_probability_type=type(mutation_probability)))

if not (self.mutation_type is None):

if (mutation_num_genes == None):

if (mutation_percent_genes < 0 or mutation_percent_genes > 100):

self.valid_parameters = False

raise ValueError("The percentage of selected genes for mutation (mutation_percent_genes) must be >= 0 and <= 100 inclusive but {mutation_percent_genes=mutation_percent_genes} found.\n".format(mutation_percent_genes=mutation_percent_genes))

else:

# Based on the mutation percentage in the 'mutation_percent_genes' parameter, the number of genes to mutate is calculated.

if mutation_num_genes == None:

mutation_num_genes = numpy.uint32((mutation_percent_genes*self.num_genes)/100)

# Based on the mutation percentage of genes, if the number of selected genes for mutation is less than the least possible value which is 1, then the number will be set to 1.

if mutation_num_genes == 0:

mutation_num_genes = 1

elif (mutation_num_genes <= 0):

self.valid_parameters = False

raise ValueError("The number of selected genes for mutation (mutation_num_genes) cannot be <= 0 but {mutation_num_genes} found.\n".format(mutation_num_genes=mutation_num_genes))

elif (mutation_num_genes > self.num_genes):

self.valid_parameters = False

raise ValueError("The number of selected genes for mutation (mutation_num_genes) ({mutation_num_genes}) cannot be greater than the number of genes ({num_genes}).\n".format(mutation_num_genes=mutation_num_genes, num_genes=self.num_genes))

elif (type(mutation_num_genes) is not int):

self.valid_parameters = False

raise ValueError("The number of selected genes for mutation (mutation_num_genes) must be a positive integer >= 1 but {mutation_num_genes} found.\n".format(mutation_num_genes=mutation_num_genes))

else:

pass

if not (type(mutation_by_replacement) is bool):

self.valid_parameters = False

raise TypeError("The expected type of the 'mutation_by_replacement' parameter is bool but {mutation_by_replacement_type} found.".format(mutation_by_replacement_type=type(mutation_by_replacement)))

self.mutation_by_replacement = mutation_by_replacement

if self.mutation_type != "random" and self.mutation_by_replacement:

print("Warning: The mutation_by_replacement parameter is set to True while the mutation_type parameter is not set to random but {mut_type}. Note that the mutation_by_replacement parameter has an effect only when mutation_type='random'.".format(mut_type=mutation_type))

if (self.mutation_type is None) and (self.crossover_type is None):

print("Warning: the 2 parameters mutation_type and crossover_type are None. This disables any type of evolution the genetic algorithm can make. As a result, the genetic algorithm cannot find a better solution that the best solution in the initial population.")

# select_parents: Refers to a method that selects the parents based on the parent selection type specified in the parent_selection_type attribute.

# Validating the selected type of parent selection: parent_selection_type

if (parent_selection_type == "sss"):

self.select_parents = self.steady_state_selection

elif (parent_selection_type == "rws"):

self.select_parents = self.roulette_wheel_selection

elif (parent_selection_type == "sus"):

self.select_parents = self.stochastic_universal_selection

elif (parent_selection_type == "random"):

self.select_parents = self.random_selection

elif (parent_selection_type == "tournament"):

self.select_parents = self.tournament_selection

elif (parent_selection_type == "rank"):

self.select_parents = self.rank_selection

else:

self.valid_parameters = False

raise ValueError("Undefined parent selection type: {parent_selection_type}. \nThe assigned value to the parent_selection_type argument does not refer to one of the supported parent selection techniques which are: \n-sss (for steady state selection)\n-rws (for roulette wheel selection)\n-sus (for stochastic universal selection)\n-rank (for rank selection)\n-random (for random selection)\n-tournament (for tournament selection).\n".format(parent_selection_type))

if(parent_selection_type == "tournament"):

if (K_tournament > self.sol_per_pop):

K_tournament = self.sol_per_pop

print("Warining: K of the tournament selection ({K_tournament}) should not be greater than the number of solutions within the population ({sol_per_pop}).\nK will be clipped to be equal to the number of solutions in the population (sol_per_pop).\n".format(K_tournament=K_tournament, sol_per_pop=self.sol_per_pop))

elif (K_tournament <= 0):

self.valid_parameters = False

raise ValueError("K of the tournament selection cannot be <=0 but {K_tournament} found.\n".format(K_tournament=K_tournament))

self.K_tournament = K_tournament

# Validating the number of parents to keep in the next population: keep_parents

if (keep_parents > self.sol_per_pop or keep_parents > self.num_parents_mating or keep_parents < -1):

self.valid_parameters = False

raise ValueError("Incorrect value to the keep_parents parameter: {keep_parents}. \nThe assigned value to the keep_parent parameter must satisfy the following conditions: \n1) Less than or equal to sol_per_pop\n2) Less than or equal to num_parents_mating\n3) Greater than or equal to -1.".format(keep_parents=keep_parents))

self.keep_parents = keep_parents

if (self.keep_parents == -1): # Keep all parents in the next population.

self.num_offspring = self.sol_per_pop - self.num_parents_mating

elif (self.keep_parents == 0): # Keep no parents in the next population.

self.num_offspring = self.sol_per_pop

elif (self.keep_parents > 0): # Keep the specified number of parents in the next population.

self.num_offspring = self.sol_per_pop - self.keep_parents

# Check if the fitness_func is a function.

if callable(fitness_func):

# Check if the fitness function accepts 2 paramaters.

if (fitness_func.__code__.co_argcount == 2):

self.fitness_func = fitness_func

else:

self.valid_parameters = False

raise ValueError("The fitness function must accept 2 parameters representing the solution to which the fitness value is calculated and the solution index within the population.\nThe passed fitness function named '{funcname}' accepts {argcount} argument(s).".format(funcname=fitness_func.__code__.co_name, argcount=fitness_func.__code__.co_argcount))

else:

self.valid_parameters = False

raise ValueError("The value assigned to the 'fitness_func' parameter is expected to be of type function but {fitness_func_type} found.".format(fitness_func_type=type(fitness_func)))

# Check if the callback_generation exists.

if not (callback_generation is None):

# Check if the callback_generation is a function.

if callable(callback_generation):

# Check if the callback_generation function accepts only a single paramater.

if (callback_generation.__code__.co_argcount == 1):

self.callback_generation = callback_generation

else:

self.valid_parameters = False

raise ValueError("The callback_generation function must accept only 1 parameter representing the instance of the genetic algorithm.\nThe passed callback_generation function named '{funcname}' accepts {argcount} argument(s).".format(funcname=callback_generation.__code__.co_name, argcount=callback_generation.__code__.co_argcount))

else:

self.valid_parameters = False

raise ValueError("The value assigned to the 'callback_generation' parameter is expected to be of type function but {callback_generation_type} found.".format(callback_generation_type=type(callback_generation)))

else:

self.callback_generation = None

if delay_after_gen >= 0.0:

self.delay_after_gen = delay_after_gen

else:

self.valid_parameters = False

raise ValueError("The value passed to the 'delay_after_gen' parameter must be a non-negative number. The value passed is {delay_after_gen} of type {delay_after_gen_type}.".format(delay_after_gen=delay_after_gen, delay_after_gen_type=type(delay_after_gen)))

# The number of completed generations.

self.generations_completed = 0

# At this point, all necessary parameters validation is done successfully and we are sure that the parameters are valid.

self.valid_parameters = True # Set to True when all the parameters passed in the GA class constructor are valid.

# Parameters of the genetic algorithm.

self.num_generations = abs(num_generations)

self.parent_selection_type = parent_selection_type

# Parameters of the mutation operation.

self.mutation_percent_genes = mutation_percent_genes

self.mutation_num_genes = mutation_num_genes

self.random_mutation_min_val = random_mutation_min_val

self.random_mutation_max_val = random_mutation_max_val

# Even such this parameter is declared in the class header, it is assigned to the object here to access it after saving the object.

self.best_solutions_fitness = [] # A list holding the fitness value of the best solution for each generation.

self.best_solution_generation = -1 # The generation number at which the best fitness value is reached. It is only assigned the generation number after the `run()` method completes. Otherwise, its value is -1.

def initialize_population(self, low, high):

"""

# Population size = (number of chromosomes, number of genes per chromosome)

self.pop_size = (self.sol_per_pop,self.num_genes) # The population will have sol_per_pop chromosome where each chromosome has num_genes genes.

if self.gene_space == None:

# Creating the initial population randomly.

self.population = numpy.random.uniform(low=low,

high=high,

size=self.pop_size) # A NumPy array holding the initial population.

elif self.gene_space_nested:

self.population = numpy.zeros(shape=self.pop_size)

for sol_idx in range(self.sol_per_pop):

for gene_idx in range(self.num_genes):

curr_gene_space = self.gene_space[gene_idx]

if type(curr_gene_space) in [list, tuple, range]:

self.population[sol_idx, gene_idx] = random.choice(curr_gene_space)

elif type(curr_gene_space) == type(None):

self.population[sol_idx, gene_idx] = numpy.random.uniform(low=low,

high=high,

size=1)

elif type(curr_gene_space) in [int, float]:

self.population[sol_idx, gene_idx] = curr_gene_space

else:

# Creating the initial population by randomly selecting the genes' values from the values inside the 'gene_space' parameter.

self.population = numpy.random.choice(self.gene_space,

size=self.pop_size) # A NumPy array holding the initial population.

# Keeping the initial population in the initial_population attribute.

self.initial_population = self.population.copy()

def cal_pop_fitness(self):

"""

if self.valid_parameters == False:

raise ValueError("ERROR calling the cal_pop_fitness() method: \nPlease check the parameters passed while creating an instance of the GA class.\n")

pop_fitness = []

# Calculating the fitness value of each solution in the current population.

for sol_idx, sol in enumerate(self.population):

fitness = self.fitness_func(sol, sol_idx)

pop_fitness.append(fitness)

pop_fitness = numpy.array(pop_fitness)

return pop_fitness

def run(self):

"""

if self.valid_parameters == False:

raise ValueError("ERROR calling the run() method: \nThe run() method cannot be executed with invalid parameters. Please check the parameters passed while creating an instance of the GA class.\n")

for generation in range(self.num_generations):

# Measuring the fitness of each chromosome in the population.

fitness = self.cal_pop_fitness()

# Appending the fitness value of the best solution in the current generation to the best_solutions_fitness attribute.

self.best_solutions_fitness.append(numpy.max(fitness))

# Selecting the best parents in the population for mating.

parents = self.select_parents(fitness, num_parents=self.num_parents_mating)

# If self.crossover_type=None, then no crossover is applied and thus no offspring will be created in the next generations. The next generation will use the solutions in the current population.

if self.crossover_type is None:

if self.num_offspring <= self.keep_parents:

offspring_crossover = parents[0:self.num_offspring]

else:

offspring_crossover = numpy.concatenate((parents, self.population[0:(self.num_offspring - parents.shape[0])]))

else:

# Generating offspring using crossover.

offspring_crossover = self.crossover(parents,

offspring_size=(self.num_offspring, self.num_genes))

# If self.mutation_type=None, then no mutation is applied and thus no changes are applied to the offspring created using the crossover operation. The offspring will be used unchanged in the next generation.

if self.mutation_type is None:

offspring_mutation = offspring_crossover

else:

# Adding some variations to the offspring using mutation.

offspring_mutation = self.mutation(offspring_crossover)

if (self.keep_parents == 0):

self.population = offspring_mutation

elif (self.keep_parents == -1):

# Creating the new population based on the parents and offspring.

self.population[0:parents.shape[0], :] = parents

self.population[parents.shape[0]:, :] = offspring_mutation

elif (self.keep_parents > 0):

parents_to_keep = self.steady_state_selection(fitness, num_parents=self.keep_parents)

self.population[0:parents_to_keep.shape[0], :] = parents_to_keep

self.population[parents_to_keep.shape[0]:, :] = offspring_mutation

self.generations_completed = generation + 1 # The generations_completed attribute holds the number of the last completed generation.

# If the callback_generation attribute is not None, then cal the callback function after the generation.

if not (self.callback_generation is None):

r = self.callback_generation(self)

if type(r) is str and r.lower() == "stop":

break

time.sleep(self.delay_after_gen)

self.best_solution_generation = numpy.where(numpy.array(self.best_solutions_fitness) == numpy.max(numpy.array(self.best_solutions_fitness)))[0][0]

# After the run() method completes, the run_completed flag is changed from False to True.

self.run_completed = True # Set to True only after the run() method completes gracefully.

def steady_state_selection(self, fitness, num_parents):

"""

fitness_sorted = sorted(range(len(fitness)), key=lambda k: fitness[k])

fitness_sorted.reverse()

# Selecting the best individuals in the current generation as parents for producing the offspring of the next generation.

parents = numpy.empty((num_parents, self.population.shape[1]))

for parent_num in range(num_parents):

parents[parent_num, :] = self.population[fitness_sorted[parent_num], :]

return parents

def rank_selection(self, fitness, num_parents):

"""

fitness_sorted = sorted(range(len(fitness)), key=lambda k: fitness[k])

fitness_sorted.reverse()

# Selecting the best individuals in the current generation as parents for producing the offspring of the next generation.

parents = numpy.empty((num_parents, self.population.shape[1]))

for parent_num in range(num_parents):

parents[parent_num, :] = self.population[fitness_sorted[parent_num], :]

return parents

def random_selection(self, fitness, num_parents):

"""

parents = numpy.empty((num_parents, self.population.shape[1]))

rand_indices = numpy.random.randint(low=0.0, high=fitness.shape[0], size=num_parents)

for parent_num in range(num_parents):

parents[parent_num, :] = self.population[rand_indices[parent_num], :]

return parents

def tournament_selection(self, fitness, num_parents):

"""

parents = numpy.empty((num_parents, self.population.shape[1]))

for parent_num in range(num_parents):

rand_indices = numpy.random.randint(low=0.0, high=len(fitness), size=self.K_tournament)

K_fitnesses = fitness[rand_indices]

selected_parent_idx = numpy.where(K_fitnesses == numpy.max(K_fitnesses))[0][0]

parents[parent_num, :] = self.population[rand_indices[selected_parent_idx], :]

return parents

def roulette_wheel_selection(self, fitness, num_parents):

"""

fitness_sum = numpy.sum(fitness)

probs = fitness / fitness_sum

probs_start = numpy.zeros(probs.shape, dtype=numpy.float) # An array holding the start values of the ranges of probabilities.

probs_end = numpy.zeros(probs.shape, dtype=numpy.float) # An array holding the end values of the ranges of probabilities.

curr = 0.0

# Calculating the probabilities of the solutions to form a roulette wheel.

for _ in range(probs.shape[0]):

min_probs_idx = numpy.where(probs == numpy.min(probs))[0][0]

probs_start[min_probs_idx] = curr

curr = curr + probs[min_probs_idx]

probs_end[min_probs_idx] = curr

probs[min_probs_idx] = 99999999999

# Selecting the best individuals in the current generation as parents for producing the offspring of the next generation.

parents = numpy.empty((num_parents, self.population.shape[1]))

for parent_num in range(num_parents):

rand_prob = numpy.random.rand()

for idx in range(probs.shape[0]):

if (rand_prob >= probs_start[idx] and rand_prob < probs_end[idx]):

parents[parent_num, :] = self.population[idx, :]

break

return parents

def stochastic_universal_selection(self, fitness, num_parents):

"""

fitness_sum = numpy.sum(fitness)

probs = fitness / fitness_sum

probs_start = numpy.zeros(probs.shape, dtype=numpy.float) # An array holding the start values of the ranges of probabilities.

probs_end = numpy.zeros(probs.shape, dtype=numpy.float) # An array holding the end values of the ranges of probabilities.

curr = 0.0

# Calculating the probabilities of the solutions to form a roulette wheel.

for _ in range(probs.shape[0]):

min_probs_idx = numpy.where(probs == numpy.min(probs))[0][0]

probs_start[min_probs_idx] = curr

curr = curr + probs[min_probs_idx]

probs_end[min_probs_idx] = curr

probs[min_probs_idx] = 99999999999

pointers_distance = 1.0 / self.num_parents_mating # Distance between different pointers.

first_pointer = numpy.random.uniform(low=0.0, high=pointers_distance, size=1) # Location of the first pointer.

# Selecting the best individuals in the current generation as parents for producing the offspring of the next generation.

parents = numpy.empty((num_parents, self.population.shape[1]))

for parent_num in range(num_parents):

rand_pointer = first_pointer + parent_num*pointers_distance

for idx in range(probs.shape[0]):

if (rand_pointer >= probs_start[idx] and rand_pointer < probs_end[idx]):

parents[parent_num, :] = self.population[idx, :]

break

return parents

def single_point_crossover(self, parents, offspring_size):

"""

offspring = numpy.empty(offspring_size)

for k in range(offspring_size[0]):

# The point at which crossover takes place between two parents. Usually, it is at the center.

crossover_point = numpy.random.randint(low=0, high=parents.shape[1], size=1)[0]

if self.crossover_probability != None:

probs = numpy.random.random(size=parents.shape[1])

indices = numpy.where(probs <= self.crossover_probability)[0]

# If no parent satisfied the probability, no crossover is applied and a parent is selected.

if len(indices) == 0:

offspring[k, :] = parents[k % parents.shape[0], :]

continue

elif len(indices) == 1:

parent1_idx = indices[0]

parent2_idx = parent1_idx

else:

indices = random.sample(set(indices), 2)

parent1_idx = indices[0]

parent2_idx = indices[1]

else:

# Index of the first parent to mate.

parent1_idx = k % parents.shape[0]

# Index of the second parent to mate.

parent2_idx = (k+1) % parents.shape[0]

# The new offspring has its first half of its genes from the first parent.

offspring[k, 0:crossover_point] = parents[parent1_idx, 0:crossover_point]

# The new offspring has its second half of its genes from the second parent.

offspring[k, crossover_point:] = parents[parent2_idx, crossover_point:]

return offspring

def two_points_crossover(self, parents, offspring_size):

"""

offspring = numpy.empty(offspring_size)

for k in range(offspring_size[0]):

if (parents.shape[1] == 1): # If the chromosome has only a single gene. In this case, this gene is copied from the second parent.

crossover_point1 = 0

else:

crossover_point1 = numpy.random.randint(low=0, high=numpy.ceil(parents.shape[1]/2 + 1), size=1)[0]

crossover_point2 = crossover_point1 + int(parents.shape[1]/2) # The second point must always be greater than the first point.

if self.crossover_probability != None:

probs = numpy.random.random(size=parents.shape[1])

indices = numpy.where(probs <= self.crossover_probability)[0]

# If no parent satisfied the probability, no crossover is applied and a parent is selected.

if len(indices) == 0:

offspring[k, :] = parents[k % parents.shape[0], :]

continue

elif len(indices) == 1:

parent1_idx = indices[0]

parent2_idx = parent1_idx

else:

indices = random.sample(set(indices), 2)

parent1_idx = indices[0]

parent2_idx = indices[1]

else:

# Index of the first parent to mate.

parent1_idx = k % parents.shape[0]

# Index of the second parent to mate.

parent2_idx = (k+1) % parents.shape[0]

# The genes from the beginning of the chromosome up to the first point are copied from the first parent.

offspring[k, 0:crossover_point1] = parents[parent1_idx, 0:crossover_point1]

# The genes from the second point up to the end of the chromosome are copied from the first parent.

offspring[k, crossover_point2:] = parents[parent1_idx, crossover_point2:]

# The genes between the 2 points are copied from the second parent.

offspring[k, crossover_point1:crossover_point2] = parents[parent2_idx, crossover_point1:crossover_point2]

return offspring

def uniform_crossover(self, parents, offspring_size):

"""

offspring = numpy.empty(offspring_size)

for k in range(offspring_size[0]):

if self.crossover_probability != None:

probs = numpy.random.random(size=parents.shape[1])

indices = numpy.where(probs <= self.crossover_probability)[0]

# If no parent satisfied the probability, no crossover is applied and a parent is selected.

if len(indices) == 0:

offspring[k, :] = parents[k % parents.shape[0], :]

continue

elif len(indices) == 1:

parent1_idx = indices[0]

parent2_idx = parent1_idx

else:

indices = random.sample(set(indices), 2)

parent1_idx = indices[0]

parent2_idx = indices[1]

else:

# Index of the first parent to mate.

parent1_idx = k % parents.shape[0]

# Index of the second parent to mate.

parent2_idx = (k+1) % parents.shape[0]

genes_source = numpy.random.randint(low=0, high=2, size=offspring_size[1])

for gene_idx in range(offspring_size[1]):

if (genes_source[gene_idx] == 0):

# The gene will be copied from the first parent if the current gene index is 0.

offspring[k, gene_idx] = parents[parent1_idx, gene_idx]

elif (genes_source[gene_idx] == 1):

# The gene will be copied from the second parent if the current gene index is 1.

offspring[k, gene_idx] = parents[parent2_idx, gene_idx]

return offspring

def random_mutation(self, offspring):

"""

# If the mutation values are selected from the mutation space, the attribute 'gene_space' is True. Otherwise, it is set to False.

# When the attribute 'gene_space' is False, the mutation values are selected randomly from the mutation space.

if self.mutation_probability == None:

if self.gene_space != None:

offspring = self.mutation_by_space(offspring)

# When the attribute 'gene_space' is False, the mutation values are selected randomly based on the continuous range specified by the 2 attributes 'random_mutation_min_val' and 'random_mutation_max_val'.

else:

offspring = self.mutation_randomly(offspring)

else:

if self.gene_space != None:

offspring = self.mutation_probs_by_space(offspring)

# When the attribute 'gene_space' is False, the mutation values are selected randomly based on the continuous range specified by the 2 attributes 'random_mutation_min_val' and 'random_mutation_max_val'.

else:

offspring = self.mutation_probs_randomly(offspring)

return offspring

def mutation_by_space(self, offspring):

"""

# For each offspring, a value from the gene space is selected randomly and assigned to the selected mutated gene.

for offspring_idx in range(offspring.shape[0]):

mutation_indices = numpy.array(random.sample(range(0, self.num_genes), self.mutation_num_genes))

for gene_idx in mutation_indices:

if self.gene_space_nested:

# Returning the current gene space from the 'gene_space' attribute.

curr_gene_space = self.gene_space[gene_idx]

# If the gene space has only a single value, use it as the new gene value.

if type(curr_gene_space) in [int, float]:

value_from_space = curr_gene_space

# Keep the gene unchanged if the gene space is None.

elif curr_gene_space == None:

rand_val = numpy.random.uniform(low=self.random_mutation_min_val,

high=self.random_mutation_max_val,

size=1)

if self.mutation_by_replacement:

value_from_space = rand_val

else:

value_from_space = offspring[offspring_idx, gene_idx] + rand_val

else:

# Selecting a value randomly from the current gene's space in the 'gene_space' attribute.

value_from_space = random.choice(curr_gene_space)

else:

# Selecting a value randomly from the global gene space in the 'gene_space' attribute.

value_from_space = random.choice(self.gene_space)

# Assinging the selected value from the space to the gene.

offspring[offspring_idx, gene_idx] = value_from_space

return offspring

def mutation_probs_by_space(self, offspring):

"""

# For each offspring, a value from the gene space is selected randomly and assigned to the selected mutated gene.

for offspring_idx in range(offspring.shape[0]):

probs = numpy.random.random(size=offspring.shape[1])

for gene_idx in range(offspring.shape[1]):

if probs[gene_idx] <= self.mutation_probability:

if self.gene_space_nested:

# Returning the current gene space from the 'gene_space' attribute.

curr_gene_space = self.gene_space[gene_idx]

# If the gene space has only a single value, use it as the new gene value.

if type(curr_gene_space) in [int, float]:

value_from_space = curr_gene_space

# Keep the gene unchanged if the gene space is None.

elif curr_gene_space == None:

rand_val = numpy.random.uniform(low=self.random_mutation_min_val,

high=self.random_mutation_max_val,

size=1)

if self.mutation_by_replacement:

value_from_space = rand_val

else:

value_from_space = offspring[offspring_idx, gene_idx] + rand_val

else:

# Selecting a value randomly from the current gene's space in the 'gene_space' attribute.

value_from_space = random.choice(curr_gene_space)

else:

# Selecting a value randomly from the global gene space in the 'gene_space' attribute.

value_from_space = random.choice(self.gene_space)

# Assinging the selected value from the space to the gene.

offspring[offspring_idx, gene_idx] = value_from_space

return offspring

def mutation_randomly(self, offspring):

"""

# Random mutation changes a single gene in each offspring randomly.

for offspring_idx in range(offspring.shape[0]):

mutation_indices = numpy.array(random.sample(range(0, self.num_genes), self.mutation_num_genes))

for gene_idx in mutation_indices:

# Generating a random value.

random_value = numpy.random.uniform(self.random_mutation_min_val, self.random_mutation_max_val, 1)

# If the mutation_by_replacement attribute is True, then the random value replaces the current gene value.

if self.mutation_by_replacement:

offspring[offspring_idx, gene_idx] = random_value

# If the mutation_by_replacement attribute is False, then the random value is added to the gene value.

else:

offspring[offspring_idx, gene_idx] = offspring[offspring_idx, gene_idx] + random_value

return offspring

def mutation_probs_randomly(self, offspring):

"""

# Random mutation changes a single gene in each offspring randomly.

for offspring_idx in range(offspring.shape[0]):

probs = numpy.random.random(size=offspring.shape[1])

for gene_idx in range(offspring.shape[1]):

if probs[gene_idx] <= self.mutation_probability:

# Generating a random value.

random_value = numpy.random.uniform(self.random_mutation_min_val, self.random_mutation_max_val, 1)

# If the mutation_by_replacement attribute is True, then the random value replaces the current gene value.

if self.mutation_by_replacement:

offspring[offspring_idx, gene_idx] = random_value

# If the mutation_by_replacement attribute is False, then the random value is added to the gene value.

else:

offspring[offspring_idx, gene_idx] = offspring[offspring_idx, gene_idx] + random_value

return offspring

def swap_mutation(self, offspring):

"""

for idx in range(offspring.shape[0]):

mutation_gene1 = numpy.random.randint(low=0, high=offspring.shape[1]/2, size=1)[0]

mutation_gene2 = mutation_gene1 + int(offspring.shape[1]/2)

temp = offspring[idx, mutation_gene1]

offspring[idx, mutation_gene1] = offspring[idx, mutation_gene2]

offspring[idx, mutation_gene2] = temp

return offspring

def inversion_mutation(self, offspring):

"""

for idx in range(offspring.shape[0]):

mutation_gene1 = numpy.random.randint(low=0, high=numpy.ceil(offspring.shape[1]/2 + 1), size=1)[0]

mutation_gene2 = mutation_gene1 + int(offspring.shape[1]/2)

genes_to_scramble = numpy.flip(offspring[idx, mutation_gene1:mutation_gene2])

offspring[idx, mutation_gene1:mutation_gene2] = genes_to_scramble

return offspring

def scramble_mutation(self, offspring):

"""

for idx in range(offspring.shape[0]):

mutation_gene1 = numpy.random.randint(low=0, high=numpy.ceil(offspring.shape[1]/2 + 1), size=1)[0]

mutation_gene2 = mutation_gene1 + int(offspring.shape[1]/2)

genes_range = numpy.arange(start=mutation_gene1, stop=mutation_gene2)

numpy.random.shuffle(genes_range)

genes_to_scramble = numpy.flip(offspring[idx, genes_range])

offspring[idx, genes_range] = genes_to_scramble

return offspring

def best_solution(self):

"""

if self.generations_completed < 1:

raise RuntimeError("The best_solution() method can only be called after completing at least 1 generation but {generations_completed} is completed.".format(generations_completed=self.generations_completed))

# Getting the best solution after finishing all generations.

# At first, the fitness is calculated for each solution in the final generation.

fitness = self.cal_pop_fitness()

# Then return the index of that solution corresponding to the best fitness.

best_match_idx = numpy.where(fitness == numpy.max(fitness))[0][0]

best_solution = self.population[best_match_idx, :]

best_solution_fitness = fitness[best_match_idx]

return best_solution, best_solution_fitness, best_match_idx

def plot_result(self, title="PyGAD - Iteration vs. Fitness", xlabel="Generation", ylabel="Fitness", linewidth=3):

"""

if self.generations_completed < 1:

raise RuntimeError("The plot_result() method can only be called after completing at least 1 generation but {generations_completed} is completed.".format(generations_completed=self.generations_completed))