def main():
pop = Population(popSize=args.popSize,
mProb=args.mProb,
xProb=args.xProb,
tournament_size=args.tournamentSize,
verbose=args.verbose)
pop.calc_fit_stats()
# main loop
for i in range(args.nGenerations):
print('Generation {}'.format(i))
# apply genetic steps
pop.tournament_selection()
pop.CrossOver()
pop.Mutation
pop.evaluate_fitnesses()
# print statistics to console
if i % args.print_every == 0:
max_fitness, avg_fitness, std_fitness = pop.calc_fit_stats()
best = pop.find_best_fit_chromosome
best_cluster_assignment = best.cluster_assignment
best_pupil_assignment = \
dataHandler.derive_pupil_assignment_from_cluster_assignment(best_cluster_assignment)
num_pupils = dataHandler.calc_numbers_of_pupils_in_classes(
best_pupil_assignment)
friends_violations = dataHandler.calc_num_friends_violations(
best_pupil_assignment)
anti_friends_violations = dataHandler.calc_num_anti_friends_violations(
best_pupil_assignment)
print('Generation {}/{}, Avg fitmess: {:1.3f}({:1.3f})), best fitness: {:1.3f}, \
num pupils: {} \
friends violations: {}, anti-friends violations: {}'\
.format(i,
args.nGenerations,
avg_fitness,
std_fitness,
max_fitness,
num_pupils,
friends_violations,
anti_friends_violations
))
print('Finished searching')
best = pop.find_best_fit_chromosome
best_cluster_assignment = best.cluster_assignment
best_pupil_assignment = \
dataHandler.derive_pupil_assignment_from_cluster_assignment(best_cluster_assignment)
