def code(cls, phenotype: float) -> List[int]:
assert phenotype <= GenotypePhenotypeConfigSingleton(
).max_phenotype_value, "Value of phenotype is greater than the possible maximum value"
assert phenotype >= GenotypePhenotypeConfigSingleton(
).min_phenotype_value, "Value of phenotype is lower than the possible minimal value"
sign = Coder.__get_value_of_bit_sign(phenotype)
bits_integer_part = Coder.__get_bits_values_for_integer_parts(
phenotype)
bits_floating_part = Coder.__get_bits_value_for_floating_part(
phenotype)
return sign + bits_integer_part + bits_floating_part
def __generate_list_exponents_for_floating_part(cls) -> List[int]:
list_exponents = []
for index in range(
-1, -(GenotypePhenotypeConfigSingleton().
number_of_bits_on_floating_part + 1), -1):
list_exponents.append(index)
return list_exponents
def mutation_of_genotype(cls, genotype, number_bits_for_mutation):
new_genotype = list(genotype)
indexes_of_bits_for_mutation_for_genotype_x = BitsRandomizer.random_without_returning(
GenotypePhenotypeConfigSingleton().number_of_bits_for_genotype, number_bits_for_mutation)
for index_to_invert in indexes_of_bits_for_mutation_for_genotype_x:
genotype[index_to_invert] = BitInverter.invert(genotype[index_to_invert])
new_phenotype = Decoder.decode(new_genotype)
return {'genotype': new_genotype,
'phenotype': new_phenotype}
def __init__(self, size_of_population: int, probability_of_crossing: float,
number_points_to_crossing: int,
probability_of_mutation: float,
number_points_to_mutation: int):
SizePopulationValidator.validation(size_of_population)
ProbabilityValidator.validation(probability_of_crossing,
'probability_of_crossing')
ProbabilityValidator.validation(probability_of_mutation,
'probability_of_mutation')
NumberOfBitsValidator.validation(
number_points_to_crossing, 'Crossing',
GenotypePhenotypeConfigSingleton().number_of_bits_for_genotype)
NumberOfBitsValidator.validation(
number_points_to_mutation, 'Mutation',
GenotypePhenotypeConfigSingleton().number_of_bits_for_genotype)
self.__size_of_population = size_of_population
self.__probability_of_crossing = probability_of_crossing
self.number_points_to_crossing = number_points_to_crossing
self.probability_of_mutation = probability_of_mutation
self.number_points_to_mutation = number_points_to_mutation
def __crossing_of_genotypes(self, genotype_one, genotype_two,
number_bits_for_crossing) -> Dict:
new_genotype_one = list(genotype_one)
new_genotype_two = list(genotype_two)
indexes_of_bits_for_crossing_for_genotype = BitsRandomizer.random_without_returning(
GenotypePhenotypeConfigSingleton().number_of_bits_for_genotype,
number_bits_for_crossing)
for index_for_crossing in indexes_of_bits_for_crossing_for_genotype:
tmp_bit_value = new_genotype_one[index_for_crossing]
new_genotype_one[index_for_crossing] = new_genotype_two[
index_for_crossing]
new_genotype_two[index_for_crossing] = tmp_bit_value
return {
'genotype_one': new_genotype_one,
'genotype_two': new_genotype_two
}
def generate(cls):
start = GenotypePhenotypeConfigSingleton().min_phenotype_value
stop = GenotypePhenotypeConfigSingleton().max_phenotype_value
phenotype_x = random.uniform(start, stop)
phenotype_y = random.uniform(start, stop)
return Individual(phenotype_x, phenotype_y)
import matplotlib.pyplot as plt
import numpy as np
from geneticAlgorithm.GeneticAlgorithm import GeneticAlgorithm
from geneticAlgorithm.fun.ConcreteFunction import ConcreteFunction
from geneticAlgorithm.lib.config.ConfigGeneticAlgorithm import ConfigGeneticAlgorithm
from geneticAlgorithm.lib.config.GenotypePhenotypeConfigSingleton import GenotypePhenotypeConfigSingleton
from geneticAlgorithm.lib.selection.RankedSelection import RankedSelection
# Skonfigurowanie kodowania genotypu
from geneticAlgorithm.lib.selection.RouletteWheelSelection import RouletteWheelSelection
from geneticAlgorithm.lib.stopconditionals.StopByAchieveValueFunction import StopByAchieveValueFunction
from geneticAlgorithm.lib.stopconditionals.StopByPercentageBestIndividualsInPopulation import \
StopByPercentageBestIndividualsInPopulation
config_genotype_phenotype = GenotypePhenotypeConfigSingleton()
config_genotype_phenotype.min_phenotype_value = -8
config_genotype_phenotype.max_phenotype_value = 8
config_genotype_phenotype.set_and_count_number_of_bits_to_write_integer()
config_genotype_phenotype.number_of_bits_on_floating_part = 28
# Przygotowanie konfiguracji algorytmu genetycznego
config_genetic_algorithm = ConfigGeneticAlgorithm(probability_of_crossing=0.8,
number_points_to_crossing=4,
probability_of_mutation=0.05,
number_points_to_mutation=1,
size_of_population=1000)
# Określenie warunku stopu
# stop_condition = StopByNumberOfGenerations(number_of_generation_to_stop=100)
# stop_condition = StopByPercentageBestIndividualsInPopulation(percentage_the_best_individuals=0.95)