def test_differential_operator(self):
G = graphs.StochasticBlockModel(N=100, directed=False)
L = G.D.T.dot(G.D)
np.testing.assert_allclose(L.toarray(), G.L.toarray())
G = graphs.StochasticBlockModel(N=100, directed=True)
self.assertRaises(NotImplementedError, G.compute_differential_operator)
def test_edge_list(self):
G = graphs.StochasticBlockModel(N=100, directed=False)
v_in, v_out, weights = G.get_edge_list()
self.assertEqual(G.W[v_in[42], v_out[42]], weights[42])
G = graphs.StochasticBlockModel(N=100, directed=True)
self.assertRaises(NotImplementedError, G.get_edge_list)
def test_laplacian(self):
# TODO: should test correctness.
G = graphs.StochasticBlockModel(N=100, directed=False)
self.assertFalse(G.is_directed())
G.compute_laplacian(lap_type='combinatorial')
G.compute_laplacian(lap_type='normalized')
G = graphs.StochasticBlockModel(N=100, directed=True)
self.assertTrue(G.is_directed())
G.compute_laplacian(lap_type='combinatorial')
self.assertRaises(NotImplementedError, G.compute_laplacian,
lap_type='normalized')
def test_stochasticblockmodel(self):
graphs.StochasticBlockModel(N=100, directed=True)
graphs.StochasticBlockModel(N=100, directed=False)
graphs.StochasticBlockModel(N=100, self_loops=True)
graphs.StochasticBlockModel(N=100, self_loops=False)
graphs.StochasticBlockModel(N=100, connected=True)
graphs.StochasticBlockModel(N=100, connected=False)
self.assertRaises(ValueError, graphs.StochasticBlockModel,
N=100, p=0, q=0, connected=True)
