def hybrid(n_geracoes, graph):
"""
:param n_geracoes: igual a numero de vertices no grapho
:param graph: tuple dos dois mapas retirados de read file
:return:
O Algoritmo hibrido e exactamente igual ao genetico, no entanto, introduz a solucao optima
conseguida pela pesquisa local na populacao inicial.
"""
sol_bin = sol_in_bin(n_geracoes, pesquisa_local(graph))
full_graph = {
**graph[0],
**graph[1]
} # syntax python 3.5 (junta os 2 dicts num so)
runs = int(input("N. runs: "))
settings = get_settings()
mbf = 0.0
best_ever = None
for r in range(runs):
pop = Pop(settings, n_geracoes)
pop.pop[0].sol = sol_bin
pop.evaluate(full_graph)
gen_actual = 1
best_run = pop.get_best()
while gen_actual < n_geracoes:
parents = pop.tournament()
pop = pop.genetic_operators(parents)
pop.evaluate(full_graph)
best_run = pop.get_best()
gen_actual += 1
invalids = 0
for s in pop.pop:
if not s.valido:
invalids += 1
print("\nRepeticao", r)
print(best_run)
print("\nPercentagem Invalidos:",
invalids / 20 * 100) # pop_size / 100
mbf += best_run.fitness
print("MBF = ", mbf)
if r == 0 or best_run.fitness > best_ever.fitness:
best_ever = best_run
print("MBF: ", mbf)
print("MELHOR SOLUCAO ENCONTRADA:")
print(best_ever.sol)
def hybrid(n_geracoes, graph):
"""
:param n_geracoes: igual a numero de vertices no grapho
:param graph: tuple dos dois mapas retirados de read file
:return:
O Algoritmo hibrido e exactamente igual ao genetico, no entanto, introduz a solucao optima
conseguida pela pesquisa local na populacao inicial.
"""
sol_bin = sol_in_bin(n_geracoes, pesquisa_local(graph))
full_graph = {**graph[0], **graph[1]} # syntax python 3.5 (junta os 2 dicts num so)
runs = int(input("N. runs: "))
settings = get_settings()
mbf = 0.0
best_ever = None
for r in range(runs):
pop = Pop(settings, n_geracoes)
pop.pop[0].sol = sol_bin
pop.evaluate(full_graph)
gen_actual = 1
best_run = pop.get_best()
while gen_actual < n_geracoes:
parents = pop.tournament()
pop = pop.genetic_operators(parents)
pop.evaluate(full_graph)
best_run = pop.get_best()
gen_actual += 1
invalids = 0
for s in pop.pop:
if not s.valido:
invalids += 1
print("\nRepeticao", r)
print(best_run)
print("\nPercentagem Invalidos:", invalids/20*100) # pop_size / 100
mbf += best_run.fitness
print("MBF = ", mbf)
if r == 0 or best_run.fitness > best_ever.fitness:
best_ever = best_run
print("MBF: ", mbf)
print("MELHOR SOLUCAO ENCONTRADA:")
print(best_ever.sol)
def alg_genetico(n_geracoes, graph):
runs = int(input("N. runs: "))
settings = get_settings()
mbf = 0.0
best_ever = None
for r in range(runs):
pop = Pop(settings, n_geracoes)
pop.evaluate(graph)
gen_actual = 1
best_run = pop.get_best()
while gen_actual < n_geracoes:
parents = pop.tournament()
pop = pop.genetic_operators(parents)
pop.evaluate(graph)
best_run = pop.get_best()
gen_actual += 1
invalids = 0
for s in pop.pop:
if not s.valido:
invalids += 1
print("\nRepeticao", r)
print(best_run)
print("\nPercentagem Invalidos:",
invalids / 20 * 100) # pop_size / 100
mbf += best_run.fitness
print("MBF = ", mbf)
if r == 0 or best_run.fitness > best_ever.fitness:
best_ever = best_run
print("MBF: ", mbf)
print("MELHOR SOLUCAO ENCONTRADA:")
print(best_ever.sol)
def alg_genetico(n_geracoes, graph):
runs = int(input("N. runs: "))
settings = get_settings()
mbf = 0.0
best_ever = None
for r in range(runs):
pop = Pop(settings, n_geracoes)
pop.evaluate(graph)
gen_actual = 1
best_run = pop.get_best()
while gen_actual < n_geracoes:
parents = pop.tournament()
pop = pop.genetic_operators(parents)
pop.evaluate(graph)
best_run = pop.get_best()
gen_actual += 1
invalids = 0
for s in pop.pop:
if not s.valido:
invalids += 1
print("\nRepeticao", r)
print(best_run)
print("\nPercentagem Invalidos:", invalids/20*100) # pop_size / 100
mbf += best_run.fitness
print("MBF = ", mbf)
if r == 0 or best_run.fitness > best_ever.fitness:
best_ever = best_run
print("MBF: ", mbf)
print("MELHOR SOLUCAO ENCONTRADA:")
print(best_ever.sol)
