class TestPopulation(unittest.TestCase):
def setUp(self):
self.population = Population()
self.population.fitness(lambda c: sum([g.value for g in c.genes]))
def test_init(self):
self.assertIsInstance(self.population, Population)
def test_evolve_failure(self):
with self.assertRaises(PopulationException):
for g in self.population.evolve(10):
pass
def test_evolve(self):
blueprint = {'test': range(10)}
self.population.populate(10, blueprint)
results = list(self.population.evolve(10))
self.assertTrue(len(results), 10)
def test_populate(self):
blueprint = {'test': range(10)}
self.population.populate(10, blueprint)
self.assertTrue(self.population.size == 10)
self.assertTrue(len(self.population.chromosomes) == 10)
def test_select(self):
blueprint = {'test': range(100)}
self.population.populate(100, blueprint)
self.population.select()
self.assertIsNotNone(self.population.best_fit)
def test_multiply(self):
# Populate
blueprint = {'test': range(10)}
self.population.populate(10, blueprint)
# Reduce the population down to 2
self.population.chromosomes = self.population.chromosomes[:2]
# Call multiply and make sure we're back to the right population size.
self.population.multiply()
self.assertTrue(len(self.population.chromosomes) == 10)
def test_fitness(self):
self.population.fitness(lambda: 0)
self.assertIsInstance(self.population.fitness_function, types.FunctionType)
4: values,
5: values,
6: values,
7: values,
8: values,
9: values,
10: values,
11: values,
12: values
})
# Decorate a function to use test fitness.
@population.fitness
def max(chromosome):
string = "".join([g.value for g in chromosome.genes])
target = "Hello, World!"
# Use fuzzywuzzy to return the % match of the string to the target string.
return fuzz.QRatio(string, target)
# Evolve for 10000 generations.
# The method is a generator to make it easier to do work after each generation.
for g in population.evolve(10000):
# The 'fittest()' method will return the best chromosome as determined by
# the fitness method.
chromosome, score = population.fittest()
if chromosome:
string = "".join([g.value for g in chromosome.genes])
print string, score