def step(self, count=1):
for i in range(count):
self.age += 1
self.evaluate()
if len(self.children_before_eval) == 0:
new_generation = self.survival_selection()
self.population.extend(new_generation)
parents = self.select_parents()
self.children_before_eval = crossoverer.rex(parents, self.nchi)
for i in self.children_before_eval:
i.birth_year = self.age
def until_goal(self, goal=10e-7, max_count=200000):
for i in range(max_count):
self.age += 1
self.evaluate()
if len(self.children_before_eval) == 0:
if self.get_best_individual().raw_fitness < goal:
# print("goal")
return True
new_generation = self.survival_selection()
self.population.extend(new_generation)
parents = self.select_parents()
self.children_before_eval = crossoverer.rex(parents, self.nchi)
for i in self.children_before_eval:
i.birth_year = self.age
# print("time is up")
return False
def generate_first_children(self):
parents = self.select_parents()
self.children_before_eval = crossoverer.rex(parents, self.nchi)
self.children_after_eval = []
def op(x):
x.sort(key=lambda i: i.fitness)
return crossoverer.rex(x[:npar])
def gaq_op_plain_origopt(x):
x.sort(key=lambda i: i.fitness)
n = x[0].n
return crossoverer.rex(x[:n + 1])
def gaq_op_plain_origopt(x):
x.sort(key=lambda i: i.fitness)
parents = x[:n + 1]
for p in parents:
p.state = State.USED_IN_GAQ
return crossoverer.rex(parents)