def test_initialize(self):
p = Population(20, objective_function)
p.add_param(0, 10)
p.add_param(0, 10)
p.initialize()
self.assertEqual(len(p._members), 20)
def test_kwargs(self):
p = Population(10, objective_function_kwargs, {'my_kwarg': 2})
p.add_param(0, 0)
p.add_param(0, 0)
p.initialize()
self.assertEqual(p.best_ret_val, 2)
def test_multiprocessing(self):
p = Population(10, objective_function, num_processes=4)
self.assertEqual(p._num_processes, 4)
p.add_param(0, 10)
p.add_param(0, 10)
p.initialize()
p.next_generation()
def test_get_stats(self):
p = Population(10, objective_function)
p.add_param(0, 0)
p.add_param(0, 0)
p.initialize()
self.assertEqual(p.best_fitness, 1)
self.assertEqual(p.best_ret_val, 0)
self.assertEqual(p.best_params, [0, 0])
self.assertEqual(p.average_fitness, 1)
self.assertEqual(p.average_ret_val, 0)
def test_next_generation(self):
p = Population(100, objective_function)
p.add_param(0, 10)
p.add_param(0, 10)
p.add_param(0, 10)
p.initialize()
for _ in range(100):
p.next_generation()
self.assertEqual(p.best_fitness, 1)
self.assertEqual(p.best_ret_val, 0)
self.assertEqual(p.best_params, [0, 0, 0])
def test_exceptions(self):
with self.assertRaises(ReferenceError):
p = Population(10, None)
with self.assertRaises(ValueError):
p = Population(1, objective_function)
p = Population(10, objective_function)
with self.assertRaises(RuntimeError):
p.initialize()
with self.assertRaises(RuntimeError):
p.next_generation()
p.add_param(0, 10)
p.initialize()
with self.assertRaises(RuntimeError):
p.add_param(0, 10)
with self.assertRaises(ValueError):
p.next_generation(p_crossover=1.1)
with self.assertRaises(ValueError):
p.next_generation(p_crossover=-0.1)
with self.assertRaises(ValueError):
p.next_generation(p_mutation=1.1)
with self.assertRaises(ValueError):
p.next_generation(p_mutation=-0.1)
from pygenetics import Population
def minimize_integers(integers):
return sum(integers)
pop = Population(10, minimize_integers)
pop.add_param(0, 10)
pop.add_param(0, 10)
pop.add_param(0, 10)
pop.initialize()
for _ in range(10):
pop.next_generation()
print('Average fitness: {}'.format(pop.average_fitness))
print('Average obj. fn. return value: {}'.format(pop.average_ret_val))
print('Best fitness score: {}'.format(pop.best_fitness))
print('Best obj. fn. return value: {}'.format(pop.best_ret_val))
print('Best parameters: {}\n'.format(pop.best_params))