コード例 #1
0
    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 = laplacian_spectral_embedding(graph,
                                               dimensions=dimensions,
                                               elbow_cut=2,
                                               svd_seed=1234)
        self.assertTrue(results.embeddings().shape[1] < dimensions)
        results = laplacian_spectral_embedding(digraph,
                                               dimensions=dimensions,
                                               elbow_cut=2,
                                               svd_seed=1234)
        self.assertTrue(results.embeddings().shape[1] < dimensions * 2)
        results = laplacian_spectral_embedding(digraph,
                                               dimensions=dimensions,
                                               elbow_cut=None,
                                               svd_seed=1234)
        self.assertTrue(results.embeddings().shape[1] == dimensions * 2)
コード例 #2
0
    def test_dimensions(self):
        graph = self.graph.copy()
        sparse = nx.to_scipy_sparse_matrix(graph)
        loopless = graspologic.utils.remove_loops(sparse)
        ranked = graspologic.utils.pass_to_ranks(loopless)
        lse = LaplacianSpectralEmbed(n_components=100,
                                     n_elbows=None,
                                     svd_seed=1234,
                                     form="R-DAD")
        core_response = lse.fit_transform(ranked)

        embedding = laplacian_spectral_embedding(self.graph.copy(),
                                                 svd_seed=1234)
        np.testing.assert_array_almost_equal(core_response,
                                             embedding.embeddings())
コード例 #3
0
    def test_argument_validation(self):
        # graph types
        with self.assertRaises(BeartypeCallHintPepParamException):
            params = TestLaplacianSpectralEmbedding.parameters()
            params["graph"] = np.array([[1, 2], [2, 1]])
            laplacian_spectral_embedding(**params)
        with self.assertRaises(ValueError):
            params = TestLaplacianSpectralEmbedding.parameters()
            params["graph"] = nx.MultiDiGraph()
            laplacian_spectral_embedding(**params)

        # dimensions
        dimensions = [None, 1.3, "1"]
        for dimension in dimensions:
            with self.assertRaises(BeartypeCallHintPepParamException):
                params = TestLaplacianSpectralEmbedding.parameters(self.graph)
                params["dimensions"] = dimension
                laplacian_spectral_embedding(**params)

        # elbow_cuts
        elbow_cuts = ["3", 1.3]
        for elbow_cut in elbow_cuts:
            with self.assertRaises(BeartypeCallHintPepParamException):
                params = TestLaplacianSpectralEmbedding.parameters(self.graph)
                params["elbow_cut"] = elbow_cut
                laplacian_spectral_embedding(**params)

        with self.assertRaises(BeartypeCallHintPepParamException):
            params = TestLaplacianSpectralEmbedding.parameters(self.graph)
            params["svd_solver_algorithm"] = 1
            laplacian_spectral_embedding(**params)

        with self.assertRaises(BeartypeCallHintPepParamException):
            params = TestLaplacianSpectralEmbedding.parameters(self.graph)
            params["svd_solver_algorithm"] = "sandwich"
            laplacian_spectral_embedding(**params)

        # svd_solver_iterations
        svd_solver_iterations = [None, "5", 5.1]
        for ssi in svd_solver_iterations:
            with self.assertRaises(BeartypeCallHintPepParamException):
                params = TestLaplacianSpectralEmbedding.parameters(self.graph)
                params["svd_solver_iterations"] = ssi
                laplacian_spectral_embedding(**params)

        # svd_seed
        svd_seeds = ["5", 5.1]
        for svd_seed in svd_seeds:
            with self.assertRaises(BeartypeCallHintPepParamException):
                params = TestLaplacianSpectralEmbedding.parameters(self.graph)
                params["svd_seed"] = svd_seed
                laplacian_spectral_embedding(**params)

        # form
        forms = [0, None]
        for form in forms:
            with self.assertRaises(BeartypeCallHintPepParamException):
                params = TestLaplacianSpectralEmbedding.parameters(self.graph)
                params["form"] = form
                laplacian_spectral_embedding(**params)
        with self.assertRaises(BeartypeCallHintPepParamException):
            params = TestLaplacianSpectralEmbedding.parameters(self.graph)
            params["form"] = "formless"
            laplacian_spectral_embedding(**params)

        # regularizer
        with self.assertRaises(BeartypeCallHintPepParamException):
            params = TestLaplacianSpectralEmbedding.parameters(self.graph)
            params["regularizer"] = "1"
            laplacian_spectral_embedding(**params)
        with self.assertRaises(ValueError):
            params = TestLaplacianSpectralEmbedding.parameters(self.graph)
            params["regularizer"] = -1
            laplacian_spectral_embedding(**params)
コード例 #4
0
 def test_unweighted_graph_warning(self):
     graph = self.graph
     with pytest.warns(UserWarning):
         laplacian_spectral_embedding(graph)