def test_make_child_returns_individual(self): """ call the make child method, and assert that the returned type is of type individual. """ myPair = Pair( self.I1, self.I2 ) child = myPair.make_child(delta = 0.5) self.assertIsInstance(child, Individual)
def test_fitness_is_mean_of_individual(self): """ Define a payoff matrix, and check that the fitness returned is the correct value simply by computing by hand. """ myPair = Pair( self.I1, self.I2 ) fitness_matrix = np.array( [ [ 1.1, 0.1 ],[0.1, 0 ] ] ) pair_fitness = myPair.fitness( fitness_matrix ) expected = 0.6 self.assertAlmostEqual( pair_fitness, expected )
def test_make_child_return_sensible_genes(self): """ Use an extreme case to test that make child is not doing something silly. """ I1 = Individual(1,1,1,1) I2 = Individual(0,0,0,0) myPair = Pair(I1,I2) ##Repeat a few time to be safe for _ in range(64): child = myPair.make_child(delta = 0.5, mu_strat = 0, mu_assort = 0) ##The child must have an a value of 1 and one of 0 and likewise for m self.assertEqual( child.a1 + child.a2, 1 ) self.assertEqual( child.m1 + child.m2, 1 )
def test_can_be_constructed_from_random(self): """ Call the random constructor and check that it returns the right type. """ myPair = Pair.from_random() self.assertIsInstance( myPair, Pair )
