def test_calc_first_layer_out(self):
uut = n_network.NNetwork(1, 0.5, 0.2)
when(genotype.Genotype).init_genes().thenReturn(np.ones(54, dtype=str))
out = uut.calc_first_layer_out([0, 0], 1, [0, 0, 0], [0, 0, 0])
expected = np.array([0.5, 0.5])
unstub(genotype.Genotype)
assert np.array_equal(expected, out)
def test_calc_out(self):
uut = n_network.NNetwork(1, 0.5, 0.2)
when(genotype.Genotype).init_genes().thenReturn(np.zeros(54,
dtype=str))
genotype_var = genotype.Genotype()
input = [0, 0]
expected = 0.5
assert uut.calc_out(input, 1, genotype_var) == expected
unstub(genotype.Genotype)
def test_selection(self):
uut = n_network.NNetwork(2, 1, 0.2)
first_contender = uut.genotypes[0]
second_contender = uut.genotypes[1]
when(first_contender).get_fitness_indicator().thenReturn(0.5)
when(second_contender).get_fitness_indicator().thenReturn(0.7)
uut.selection()
unstub(first_contender)
unstub(second_contender)
assert np.array_equal(first_contender.genes, uut.genotypes[0].genes)
assert np.size(uut.genotypes) == 1
def test_calc_fitness_func(self):
uut = n_network.NNetwork(5, 0.5, 0.2)
uut.calc_fitness_function()
def test_cross_population_size(self):
uut = n_network.NNetwork(4, 0.5, 0.2)
uut.genotypes = uut.genotypes[0:2]
uut.cross_population()
assert uut.genotypes.size == 4