def test_bad_inputs():
np.random.seed(1)
single_graph = er_np(100, 0.2)
different_size_graphs = [er_np(100, 0.2)] + [er_np(200, 0.2)]
with pytest.raises(TypeError):
"Invalid unscaled"
mase = MultipleASE(scaled="1")
with pytest.raises(TypeError):
wrong_diag_aug = "True"
mase = MultipleASE(diag_aug=wrong_diag_aug)
with pytest.raises(ValueError):
"Test single graph input"
MultipleASE().fit(single_graph)
with pytest.raises(ValueError):
"Test 3-d tensor with 1 graph"
single_graph_tensor = single_graph.reshape(1, 100, -1)
MultipleASE().fit(single_graph_tensor)
with pytest.raises(ValueError):
"Empty list"
MultipleASE().fit([])
with pytest.raises(ValueError):
"Test graphs with different sizes"
MultipleASE().fit(different_size_graphs)
def test_gridplot_outputs():
"""
simple function to see if plot is made without errors
"""
X = [er_np(10, 0.5) for _ in range(2)]
labels = ["Random A", "Random B"]
fig = gridplot(X, labels)
fig = gridplot(X, labels, transform="zero-boost")
fig = gridplot(X, labels, "simple-all", title="Test", font_scale=0.9)
def test_omni_matrix_symmetric():
np.random.seed(3)
n = 15
p = 0.4
n_graphs = [2, 5, 10]
for n in n_graphs:
graphs = [er_np(n, p) for _ in range(n)]
output = _get_omni_matrix(graphs)
assert is_symmetric(output)
def test_diag_aug():
np.random.seed(5)
n = 100
p = 0.25
graphs_list = [er_np(n, p) for _ in range(2)]
graphs_arr = np.array(graphs_list)
# Test that array and list inputs results in same embeddings
mase_arr = MultipleASE(diag_aug=True).fit_transform(graphs_arr)
mase_list = MultipleASE(diag_aug=True).fit_transform(graphs_list)
assert array_equal(mase_list, mase_arr)
def test_diag_aug():
np.random.seed(5)
n = 100
p = 0.25
graphs_list = [er_np(n, p) for _ in range(2)]
graphs_arr = np.array(graphs_list)
# Test that array and list inputs results in same embeddings
omni_arr = OmnibusEmbed(diag_aug=True).fit_transform(graphs_arr)
omni_list = OmnibusEmbed(diag_aug=True).fit_transform(graphs_list)
assert array_equal(omni_list, omni_arr)
def test_gridplot_inputs():
X = [er_np(10, 0.5)]
labels = ["ER(10, 0.5)"]
with pytest.raises(TypeError):
gridplot(X="input", labels=labels)
with pytest.raises(ValueError):
gridplot(X, labels=["a", "b"])
# transform
with pytest.raises(ValueError):
transform = "bad transform"
gridplot(X, labels=labels, transform=transform)
def test_omni_matrix_random():
expected_output = np.array([
[0.0, 1.0, 1.0, 0.0, 0.5, 0.5],
[1.0, 0.0, 1.0, 0.5, 0.0, 1.0],
[1.0, 1.0, 0.0, 0.5, 1.0, 0.0],
[0.0, 0.5, 0.5, 0.0, 0.0, 0.0],
[0.5, 0.0, 1.0, 0.0, 0.0, 1.0],
[0.5, 1.0, 0.0, 0.0, 1.0, 0.0],
])
np.random.seed(4)
graphs = [er_np(3, 0.5) for _ in range(2)]
A = _get_omni_matrix(graphs)
assert_allclose(A, expected_output)
def test_heatmap_output():
"""
simple function to see if plot is made without errors
"""
X = er_np(10, 0.5)
xticklabels = ["Dimension {}".format(i) for i in range(10)]
yticklabels = ["Dimension {}".format(i) for i in range(10)]
fig = heatmap(X,
transform="log",
xticklabels=xticklabels,
yticklabels=yticklabels)
fig = heatmap(X, transform="zero-boost")
fig = heatmap(X, transform="simple-all")
fig = heatmap(X, transform="simple-nonzero")
fig = heatmap(X, cmap="gist_rainbow")
def _test_sbm_er_binary_undirected(self, method, P, *args, **kwargs):
np.random.seed(8888)
num_sims = 50
verts = 200
communities = 2
verts_per_community = [100, 100]
sbm_wins = 0
er_wins = 0
for sim in range(0, num_sims):
sbm_sample = sbm(verts_per_community, P)
er = er_np(verts, 0.5)
embed_sbm = method(n_components=2)
embed_er = method(n_components=2)
labels_sbm = np.zeros((verts), dtype=np.int8)
labels_er = np.zeros((verts), dtype=np.int8)
labels_sbm[100:] = 1
labels_er[100:] = 1
embed_sbm.fit(sbm_sample)
embed_er.fit(er)
X_sbm = embed_sbm.latent_left_
X_er = embed_er.latent_left_
self.assertEqual(X_sbm.shape, (verts, communities))
self.assertEqual(X_er.shape, (verts, communities))
aris = _kmeans_comparison((X_sbm, X_er), (labels_sbm, labels_er),
communities)
sbm_wins = sbm_wins + (aris[0] > aris[1])
er_wins = er_wins + (aris[0] < aris[1])
self.assertTrue(sbm_wins > er_wins)
def test_common_inputs():
X = er_np(100, 0.5)
grid_labels = ["Test1"]
# test figsize
with pytest.raises(TypeError):
figsize = "bad figsize"
heatmap(X, figsize=figsize)
# test height
height = "1"
with pytest.raises(TypeError):
gridplot([X], grid_labels, height=height)
with pytest.raises(TypeError):
pairplot(X, height=height)
# test title
title = 1
with pytest.raises(TypeError):
heatmap(X, title=title)
with pytest.raises(TypeError):
gridplot([X], grid_labels, title=title)
with pytest.raises(TypeError):
pairplot(X, title=title)
# test context
context = 123
with pytest.raises(TypeError):
heatmap(X, context=context)
with pytest.raises(TypeError):
gridplot([X], grid_labels, context=context)
with pytest.raises(TypeError):
pairplot(X, context=context)
context = "journal"
with pytest.raises(ValueError):
heatmap(X, context=context)
with pytest.raises(ValueError):
gridplot([X], grid_labels, context=context)
with pytest.raises(ValueError):
pairplot(X, context=context)
# test font scales
font_scales = ["1", []]
for font_scale in font_scales:
with pytest.raises(TypeError):
heatmap(X, font_scale=font_scale)
with pytest.raises(TypeError):
gridplot([X], grid_labels, font_scale=font_scale)
with pytest.raises(TypeError):
pairplot(X, cont_scale=font_scale)
# ticklabels
with pytest.raises(TypeError):
xticklabels = "labels"
yticklabels = "labels"
heatmap(X, xticklabels=xticklabels, yticklabels=yticklabels)
with pytest.raises(ValueError):
xticklabels = ["{}".format(i) for i in range(5)]
yticklabels = ["{}".format(i) for i in range(5)]
heatmap(X, xticklabels=xticklabels, yticklabels=yticklabels)
def setup_class(cls):
# simple ERxN graph
cls.n = 20
cls.p = 0.5
cls.A = er_np(cls.n, cls.p, directed=True, loops=False)
