def test_elbow_cuts(self):
# smoke test, unsure how to validate the elbow is cut at the right spot
# for this data
graph = nx.Graph()
digraph = nx.DiGraph()
with open(data_file("large-graph.csv"), "r") as graph_io:
for line in graph_io:
source, target, weight = line.strip().split(",")
prev_weight = graph.get_edge_data(source, target, default={}).get(
"weight", 0.0
)
graph.add_edge(source, target, weight=float(weight) + prev_weight)
digraph.add_edge(source, target, weight=float(weight))
dimensions = 100
results = adjacency_spectral_embedding(
graph, dimensions=dimensions, elbow_cut=2, svd_seed=1234
)
self.assertTrue(results.embeddings().shape[1] < dimensions)
results = adjacency_spectral_embedding(
digraph, dimensions=dimensions, elbow_cut=2, svd_seed=1234
)
self.assertTrue(results.embeddings().shape[1] < dimensions * 2)
results = adjacency_spectral_embedding(
digraph, dimensions=dimensions, elbow_cut=None, svd_seed=1234
)
self.assertTrue(results.embeddings().shape[1] == dimensions * 2)
def test_dimensions(self):
graph = self.graph.copy()
sparse = nx.to_scipy_sparse_matrix(graph)
ranked = graspologic.utils.pass_to_ranks(sparse)
ase = AdjacencySpectralEmbed(n_components=100, n_elbows=None, svd_seed=1234)
core_response = ase.fit_transform(ranked)
embedding = adjacency_spectral_embedding(self.graph.copy(), svd_seed=1234)
np.testing.assert_array_almost_equal(core_response, embedding.embeddings())
def test_argument_validation(self):
# graph types
with self.assertRaises(BeartypeCallHintPepParamException):
adjacency_spectral_embedding(
graph=np.array([[1, 2], [2, 1]]), **self.default_parameters
)
with self.assertRaises(ValueError):
adjacency_spectral_embedding(
graph=nx.MultiDiGraph(), **self.default_parameters
)
# dimensions
dimensions = [None, 1.3, "1"]
for dimension in dimensions:
with self.assertRaises(BeartypeCallHintPepParamException):
params = TestAdjacencySpectralEmbedding.parameters()
params["dimensions"] = dimension
params["graph"] = self.graph
adjacency_spectral_embedding(**params)
# elbow_cuts
elbow_cuts = ["3", 1.3]
for elbow_cut in elbow_cuts:
with self.assertRaises(BeartypeCallHintPepParamException):
params = TestAdjacencySpectralEmbedding.parameters()
params["elbow_cut"] = elbow_cut
params["graph"] = self.graph
adjacency_spectral_embedding(**params)
with self.assertRaises(BeartypeCallHintPepParamException):
params = TestAdjacencySpectralEmbedding.parameters()
params["svd_solver_algorithm"] = 1
params["graph"] = self.graph
adjacency_spectral_embedding(**params)
with self.assertRaises(BeartypeCallHintPepParamException):
params = TestAdjacencySpectralEmbedding.parameters()
params["svd_solver_algorithm"] = "sandwich"
params["graph"] = self.graph
adjacency_spectral_embedding(**params)
# svd_solver_iterations
svd_solver_iterations = [None, "5", 5.1]
for ssi in svd_solver_iterations:
with self.assertRaises(BeartypeCallHintPepParamException):
params = TestAdjacencySpectralEmbedding.parameters()
params["svd_solver_iterations"] = ssi
params["graph"] = self.graph
adjacency_spectral_embedding(**params)
# svd_seed
svd_seeds = ["5", 5.1]
for svd_seed in svd_seeds:
with self.assertRaises(BeartypeCallHintPepParamException):
params = TestAdjacencySpectralEmbedding.parameters()
params["svd_seed"] = svd_seed
params["graph"] = self.graph
adjacency_spectral_embedding(**params)
def test_unweighted_graph_warning(self):
graph = self.graph
with pytest.warns(UserWarning):
adjacency_spectral_embedding(graph)