def __init__(self, learningData, testData, size):
self.chromosomes = []
self.learningData = learningData
self.testData = testData
for index in range(size):
chromosome = Chromosome()
chromosome.initialise()
self.chromosomes.append(chromosome)
def solve(self):
graph = read_graph_from_file(self.__file)
self.__probParameters = ProblemParams(graph, get_route_length)
ga = GA(self.__gaParameters, self.__probParameters)
ga.initialisation()
ga.evaluation()
best = Chromosome(self.__probParameters)
for generation in range(self.__gaParameters.noOfGenerations):
ga.one_generation_steady_state()
bestChromo = ga.best_chromosome()
print('Generatia ' + str(generation) +
' are cel mai bun cromozom:' + str(best.representation) +
' cu fitness ' + str(bestChromo.fitness))
if bestChromo.fitness < best.fitness:
best = bestChromo
print(best.representation)
print(
get_route_length(best.representation,
self.__probParameters.network))
def one_generation_steady_state(self):
bestOff = Chromosome(self.problParams)
for _ in range(self.params.populationSize):
p1 = self.population[self.selection()]
p2 = self.population[self.selection()]
off = p1.crossover(p2, self.params.crossoverProb)
off.mutation()
off.fitness = self.problParams.function(off.representation,
self.problParams.network)
if off.fitness < bestOff.fitness:
bestOff = off
self.population.remove(self.worst_chromosome())
self.population.append(bestOff)
def initialisation(self):
for _ in range(0, self.__param['popSize']):
c = Chromosome(self.__problParam)
self.__population.append(c)
def initialization(self):
for _ in range(0, self.__populationSize):
c = Chromosome(self.__problParam)
self.__population.append(c)
self.evaluation(self.__population)
def crossover(chromosome1, chromosome2):
offspring1 = Chromosome()
offspring2 = Chromosome()
stop = True
while stop:
startChromosome1 = randint(0, chromosome1.size - 1)
endChromosome1 = chromosome1.traverse(startChromosome1)
startChromosome2 = randint(0, chromosome2.size - 1)
endChromosome2 = chromosome2.traverse(startChromosome2)
if len(offspring1.representation) > endChromosome1 + (endChromosome2 - startChromosome2 - 1) + (
chromosome1.size - endChromosome1 - 1) and \
len(offspring2.representation) > endChromosome2 + (endChromosome1 - startChromosome1 - 1) + (
chromosome2.size - endChromosome2 - 1):
stop = False
# startChromosome1 = randint(0, chromosome1.size - 1)
# endChromosome1 = chromosome1.traverse(startChromosome1)
# startChromosome2 = randint(0, chromosome2.size - 1)
# endChromosome2 = chromosome2.traverse(startChromosome2)
# print("start1: " + str(startChromosome1))
# print("end1: " + str(endChromosome1))
# print("size1: " + str(chromosome1.size))
# print("start2: " + str(startChromosome2))
# print("end2: " + str(endChromosome2))
# print("size2: " + str(chromosome2.size))
#
# print("off1 size: " + str(len(offspring1.representation)))
# print("off2 size: " + str(len(offspring2.representation)))
i = -1
for i in range(startChromosome1):
offspring1.representation[i] = chromosome1.representation[i]
# print("current i value: " + str(i))
for j in range(startChromosome2, endChromosome2):
i += 1
# print("current i value: " + str(i))
offspring1.representation[i] = chromosome2.representation[j]
for j in range(endChromosome1, chromosome1.size):
i += 1
# print("current i value: " + str(i))
offspring1.representation[i] = chromosome1.representation[j]
offspring1.size = i + 1
i = -1
for i in range(startChromosome2):
# print("current i value: " + str(i))
offspring2.representation[i] = chromosome2.representation[i]
for j in range(startChromosome1, endChromosome1):
i += 1
# print("current i value: " + str(i))
offspring2.representation[i] = chromosome1.representation[j]
for j in range(endChromosome2, chromosome2.size):
i += 1
# print("current i value: " + str(i))
offspring2.representation[i] = chromosome2.representation[j]
offspring2.size = i + 1
return [offspring1, offspring2]
def initialisation(self):
for i in range(self.params.populationSize):
chromo = Chromosome(self.problParams)
self.population.append(chromo)