def operate(population, crossover, mutation, pc, pm):
length = len(population)
for i in range(length):
if random() < pc:
samp = sample(list(population), 2)
bin_reprs = [i.bin_value for i in samp]
crossover(bin_reprs[0], bin_reprs[1])
if random() < pm:
bin_rep = choice(population).bin_value
mutation(bin_rep)
def basicSelection(population, referenceCromossome, evaluatedPopulation, crossoverType, crossoverProbability, crossoverTasksNumPerc, mutationType, mutationTasksNumPerc, tasksMutationProbability, operatorsMutationProbability, drivenMutation, drivenMutationPart, limitBestFitnessRepetionCount, bestFitnessRepetionCount, drivenMutatedIndividuals, drivenMutatedGenerations, TPweight, precisenessWeight, simplicityWeight, completenessWeight, elitismPerc, sortedEvaluatedPopulation, selectionOp, currentGeneration, completenessAttemptFactor1, completenessAttemptFactor2, numberOfcyclesAfterDrivenMutation, alphabet, log):
drivenMutationCycle = 0
auxPopulation = copy.deepcopy(population)
auxDrivenMutatedIndividuals = copy.deepcopy(drivenMutatedIndividuals)
for j in range(len(auxDrivenMutatedIndividuals)):
auxDrivenMutatedIndividuals[j] = 0
if (drivenMutation == 1) and (currentGeneration > 0) and (bestFitnessRepetionCount > 0) and (divmod(bestFitnessRepetionCount, limitBestFitnessRepetionCount)[1] == 0):
drivenMutationCycle = 1
drivenMutatedEvaluatedPopulation = [[], []]
if drivenMutatedGenerations >= 1:
drivenMutatedEvaluatedPopulation[0] = 0
drivenMutatedEvaluatedPopulation[1] = [0 for _ in range(len(drivenMutatedIndividuals))]
for j in range(len(drivenMutatedIndividuals)):
if drivenMutatedIndividuals[j] == 1:
drivenMutatedEvaluatedPopulation[1][j] = 1
drivenMutatedEvaluatedPopulation[0] = drivenMutatedEvaluatedPopulation[0] + 1
else:
drivenMutatedEvaluatedPopulation[1][j] = evaluatedPopulation[1][j][0]
drivenMutatedEvaluatedPopulation[0] = drivenMutatedEvaluatedPopulation[0] + evaluatedPopulation[1][j][0]
i = 0
while i < len(population):
chosenIndividual1 = op.parentSelection(evaluatedPopulation, sortedEvaluatedPopulation, selectionOp, drivenMutatedIndividuals, drivenMutatedEvaluatedPopulation, drivenMutatedGenerations)
auxPopulation[i] = copy.deepcopy(population[chosenIndividual1[0]])
chosenIndividual2 = op.parentSelection(evaluatedPopulation, sortedEvaluatedPopulation, selectionOp, drivenMutatedIndividuals, drivenMutatedEvaluatedPopulation, drivenMutatedGenerations)
auxPopulation[i + 1] = copy.deepcopy(population[chosenIndividual2[0]])
if (chosenIndividual1[1] == 1) or (chosenIndividual2[1] == 1):
auxDrivenMutatedIndividuals[i] = 1
auxDrivenMutatedIndividuals[i + 1] = 1
(auxPopulation[i], auxPopulation[i + 1]) = op.crossoverPerProcess(crossoverType, crossoverProbability, crossoverTasksNumPerc, auxPopulation[i], auxPopulation[i+1], i, evaluatedPopulation, alphabet)
if drivenMutationCycle == 0:
op.mutation(auxPopulation[i], tasksMutationProbability, operatorsMutationProbability, mutationType, mutationTasksNumPerc, alphabet)
op.mutation(auxPopulation[i+1], tasksMutationProbability, operatorsMutationProbability, mutationType, mutationTasksNumPerc, alphabet)
i = i + 2
if (i + 1 == len(population)) and (len(population) % 2 == 1):
i = i - 1
if (elitismPerc > 0) or (drivenMutationCycle == 1):
evaluatedAuxPopulation = fitn.evaluationPopulation(auxPopulation, referenceCromossome, TPweight, precisenessWeight, simplicityWeight, completenessWeight, completenessAttemptFactor1, completenessAttemptFactor2, selectionOp, alphabet, log)
sortedEvaluatedAuxPopulation = sorted(evaluatedAuxPopulation[1], reverse=True, key=takeFirst)
if drivenMutatedGenerations >= 1:
drivenMutatedGenerations = drivenMutatedGenerations - 1
if drivenMutatedGenerations == 0:
for j in range(len(auxDrivenMutatedIndividuals)):
auxDrivenMutatedIndividuals[j] = 0
drivenMutatedIndividuals = copy.deepcopy(auxDrivenMutatedIndividuals)
if drivenMutationCycle == 1:
drivenMutatedIndividuals = op.drivenMutation(auxPopulation, sortedEvaluatedAuxPopulation, drivenMutationPart, drivenMutatedIndividuals)
drivenMutatedGenerations = numberOfcyclesAfterDrivenMutation
if elitismPerc > 0:
op.elitism(population, elitismPerc, sortedEvaluatedAuxPopulation, sortedEvaluatedPopulation, auxPopulation, drivenMutatedIndividuals)
evaluatedNewPopulation = fitn.evaluationPopulation(auxPopulation, referenceCromossome, TPweight, precisenessWeight, simplicityWeight, completenessWeight, completenessAttemptFactor1, completenessAttemptFactor2, selectionOp, alphabet, log)
return (auxPopulation, evaluatedNewPopulation, drivenMutatedIndividuals, drivenMutatedGenerations)
def __mutation(pop, p_mutation_pop, p_mutation_specimen):
new_pop = []
for specimen in pop:
if random.random() < p_mutation_pop:
mutated = operators.mutation(specimen, p_mutation_specimen)
new_pop.append(mutated)
else:
new_pop.append(specimen)
return new_pop