Corrected deprecation warnings (#1019)

* Editing contributing guidelines 

My edits were based on my experience with getting set up on GitHub. I added some minor changes of things I wish I had read while I was trying to set up my account.

* fix typo

* Resolving two deprecation warnings

* Removing IDAD tests

* Correcting some deprecation errors

* Correcting more deprecation warnings

* Revert "Removing IDAD tests"
Moving commit to the correct location

This reverts commit 00a2e76.

* Correcting formatting errors:

---------

Co-authored-by: Benjamin Pedigo <[email protected]>
Co-authored-by: hugwuoke <[email protected]>

graspologic/align/orthogonal_procrustes.py

@@ -55,7 +55,6 @@ class OrthogonalProcrustes(BaseAlign):
     def __init__(
         self,
     ) -> None:
-
         super().__init__()
 
     def fit(self, X: np.ndarray, Y: np.ndarray) -> "OrthogonalProcrustes":

graspologic/cluster/divisive_cluster.py

@@ -122,7 +122,6 @@ def __init__(
         max_level: int = 4,
         delta_criter: float = 0,
     ):
-
         _check_common_inputs(min_components, max_components, cluster_kws)
 
         if cluster_method not in ["gmm", "kmeans"]:

graspologic/match/solver.py

@@ -694,7 +694,6 @@ def _compare_dimensions(
 def _check_partial_match(
     partial_match: Optional[Union[np.ndarray, Tuple]], n1: int, n2: int
 ) -> np.ndarray:
-
     _partial_match: np.ndarray
     if partial_match is None:
         _partial_match = np.array([[], []]).T

graspologic/models/edge_swaps.py

@@ -46,7 +46,6 @@ class EdgeSwapper:
 
     @beartype
     def __init__(self, adjacency: AdjacencyMatrix, seed: Optional[int] = None):
-
         weight_check = is_unweighted(adjacency)
         check_argument(weight_check, "adjacency must be unweighted")
 

graspologic/nominate/spectralVN.py

@@ -121,7 +121,6 @@ def _check_y(self, y: np.ndarray) -> None:
             )
 
     def _check_params(self) -> None:
-
         if self.n_neighbors is not None and type(self.n_neighbors) is not int:
             raise TypeError("k must be an integer")
         elif self.n_neighbors is not None and self.n_neighbors <= 0:

graspologic/partition/leiden.py

@@ -138,7 +138,7 @@ def _community_python_to_native(
     if starting_communities is None:
         return None
     native_safe: Dict[str, int] = {}
-    for (node_id, partition) in starting_communities.items():
+    for node_id, partition in starting_communities.items():
         node_id_as_str = identity(node_id)
         native_safe[node_id_as_str] = partition
     return native_safe
@@ -399,7 +399,6 @@ def _from_native(
     native_cluster: gn.HierarchicalCluster,
     identifier: _IdentityMapper,
 ) -> HierarchicalCluster:
-
     if not isinstance(native_cluster, gn.HierarchicalCluster):
         raise TypeError(
             "This class method is only valid for graspologic_native.HierarchicalCluster"
@@ -420,7 +419,7 @@ def hierarchical_leiden(
         List[Tuple[Any, Any, Union[int, float]]],
         nx.Graph,
         np.ndarray,
-        scipy.sparse.csr.csr_matrix,
+        scipy.sparse.csr_matrix,
     ],
     max_cluster_size: int = 1000,
     starting_communities: Optional[Dict[str, int]] = None,

graspologic/pipeline/embed/adjacency_spectral_embedding.py

@@ -180,7 +180,7 @@ def adjacency_spectral_embedding(
 
     node_labels = np.array(list(graph.nodes()))
 
-    graph_as_csr = nx.to_scipy_sparse_matrix(
+    graph_as_csr = nx.to_scipy_sparse_array(
         graph, weight=used_weight_attribute, nodelist=node_labels
     )
 

graspologic/pipeline/embed/laplacian_spectral_embedding.py

@@ -196,7 +196,7 @@ def laplacian_spectral_embedding(
 
     node_labels = np.array(list(graph.nodes()))
 
-    graph_as_csr = nx.to_scipy_sparse_matrix(
+    graph_as_csr = nx.to_scipy_sparse_array(
         graph, weight=used_weight_attribute, nodelist=node_labels
     )
 

graspologic/pipeline/embed/omnibus_embedding.py

@@ -296,7 +296,7 @@ def _augment_graph(
     weight_attribute: Optional[str],
     perform_augment_diagonal: bool = True,
 ) -> np.ndarray:
-    graph_sparse = nx.to_scipy_sparse_matrix(
+    graph_sparse = nx.to_scipy_sparse_array(
         graph, weight=weight_attribute, nodelist=node_ids
     )
 

graspologic/plot/plot.py

@@ -996,7 +996,6 @@ def _distplot(
     xlabel: str = "",
     ylabel: str = "Density",
 ) -> matplotlib.pyplot.Axes:
-
     plt.figure(figsize=figsize)
     ax = plt.gca()
     palette = sns.color_palette(palette)

graspologic/plot/plot_matrix.py

@@ -812,7 +812,6 @@ def matrixplot(  # type: ignore
         # in the opposite order
         rev_group = list(col_group[::-1])
         for i, sc in enumerate(rev_group):
-
             # Add a new axis when needed
             tick_ax = divider.append_axes(
                 "top", size="1%", pad=col_tick_pad[i], sharex=ax

tests/embed/test_omni.py

@@ -39,7 +39,6 @@ def generate_data(n, seed=1, symetric=True):
 
 
 class TestOmni(unittest.TestCase):
-
     # Below tests omni matrix generation code
     def test_omni_matrix_ones_zeros(self):
         # Should get all ones

tests/layouts/nooverlap/test_grid.py

@@ -8,7 +8,6 @@
 
 
 class TestGrid(unittest.TestCase):
-
     # def setUp(self):
     # 	self.g = _GridBuckets(10)
 

tests/layouts/nooverlap/test_overlap_check.py

@@ -32,7 +32,6 @@ def test_grid_is_overlap(self):
         self.assertFalse(overlaps)
 
     def test_overlap_check_list(self):
-
         to_check = _Node(0, 2, 2, 1, -1, "blue")
         others = [
             _Node(1, 5, 5, 1, -1, "blue"),

tests/partition/test_leiden.py

@@ -35,7 +35,6 @@ def test_from_native(self):
         # test from_native indirectly through calling graspologic.partition.hierarchical_leiden()
 
     def test_final_hierarchical_clustering(self):
-
         hierarchical_clusters = HierarchicalClusters(
             [
                 HierarchicalCluster("1", 0, None, 0, False),
@@ -189,7 +188,7 @@ def test_correct_types(self):
 
         cleared_partitions = good_args.copy()
         del cleared_partitions["starting_communities"]
-        as_csr = nx.to_scipy_sparse_matrix(graph)
+        as_csr = nx.to_scipy_sparse_array(graph)
         partitions = leiden(graph=as_csr, **cleared_partitions)
         node_ids = partitions.keys()
         for node_id in node_ids:
@@ -310,7 +309,7 @@ def test_isolate_nodes_in_ndarray_are_not_returned(self):
         self.assert_isolate_not_in_hierarchical_result(hierarchical_partitions)
 
     def test_isolate_nodes_in_csr_matrix_are_not_returned(self):
-        sparse_adj_matrix = nx.to_scipy_sparse_matrix(self.graph)
+        sparse_adj_matrix = nx.to_scipy_sparse_array(self.graph)
 
         self.assertEqual(
             10,
@@ -475,8 +474,8 @@ def test_matrices(self):
         dense_undirected = nx.to_numpy_array(graph)
         dense_directed = nx.to_numpy_array(di_graph)
 
-        sparse_undirected = nx.to_scipy_sparse_matrix(graph)
-        sparse_directed = nx.to_scipy_sparse_matrix(di_graph)
+        sparse_undirected = nx.to_scipy_sparse_array(graph)
+        sparse_directed = nx.to_scipy_sparse_array(di_graph)
 
         expected = [("0", "1", 2.2), ("1", "2", 0.001)]
         _, edges = _adjacency_matrix_to_edge_list(

tests/pipeline/embed/test_adjacency_spectral_embedding.py

@@ -108,7 +108,7 @@ def test_unweighted_graph_warning(self):
 
     def test_dimensions(self):
         graph = self.graph.copy()
-        sparse = nx.to_scipy_sparse_matrix(graph)
+        sparse = nx.to_scipy_sparse_array(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)

tests/pipeline/embed/test_laplacian_spectral_embedding.py

@@ -123,7 +123,7 @@ def test_unweighted_graph_warning(self):
 
     def test_dimensions(self):
         graph = self.graph.copy()
-        sparse = nx.to_scipy_sparse_matrix(graph)
+        sparse = nx.to_scipy_sparse_array(graph)
         loopless = graspologic.utils.remove_loops(sparse)
         ranked = graspologic.utils.pass_to_ranks(loopless)
         lse = LaplacianSpectralEmbed(

tests/test_casc.py

@@ -13,6 +13,7 @@
 
 np.random.seed(5)
 
+
 # UTILITY FUNCTIONS
 def gen_covariates(m1, m2, labels, type, ndim=3):
     n = len(labels)
@@ -107,7 +108,6 @@ def test_labels_match(A, labels, M):
 
 
 def test_wrong_inputs(A, X):
-
     with pytest.raises(ValueError):
         "wrong assortative type"
         CASE(assortative="1")

tests/test_models.py

@@ -565,7 +565,7 @@ def setUpClass(cls):
         cls.A = er_np(20, 0.5)
         cls.B = csr_matrix(cls.A)
         cls.C = nx.from_numpy_array(cls.A)
-        cls.D = nx.from_scipy_sparse_matrix(cls.B)
+        cls.D = nx.from_scipy_sparse_array(cls.B)
 
     def test_numpy_edge_swap(self):
         Swapper = EdgeSwapper(self.A)

tests/test_spectral_nomination.py

@@ -14,7 +14,7 @@
 n_verts = 50
 p = np.array([[0.7, 0.25, 0.2], [0.25, 0.8, 0.3], [0.2, 0.3, 0.85]])
 labels = np.array([0] * n_verts + [1] * n_verts + [2] * n_verts)
-adj = np.array(sbm(3 * [n_verts], p), dtype=np.int)
+adj = np.array(sbm(3 * [n_verts], p), dtype=int)
 embeder = AdjacencySpectralEmbed()
 pre_embeded = embeder.fit_transform(adj)
 
@@ -32,34 +32,34 @@ def _nominate(self, X, seed, nominator=None, k=None):
 
     def test_seed_inputs(self):
         with self.assertRaises(IndexError):
-            self._nominate(adj, np.zeros((1, 50), dtype=np.int))
+            self._nominate(adj, np.zeros((1, 50), dtype=int))
         with self.assertRaises(TypeError):
             self._nominate(adj, np.random.random((10, 2)))
 
     def test_X_inputs(self):
         with self.assertRaises(IndexError):
-            self._nominate(np.zeros((5, 5, 5), dtype=np.int), np.zeros(3, dtype=np.int))
+            self._nominate(np.zeros((5, 5, 5), dtype=int), np.zeros(3, dtype=int))
         with self.assertRaises(TypeError):
-            self._nominate([[0] * 10] * 10, np.zeros(3, dtype=np.int))
+            self._nominate([[0] * 10] * 10, np.zeros(3, dtype=int))
         # embedding should have fewer cols than rows.
         svn = SpectralVertexNomination(input_graph=False)
         with self.assertRaises(IndexError):
             self._nominate(
-                np.zeros((10, 20), dtype=np.int),
-                np.zeros(3, dtype=np.int),
+                np.zeros((10, 20), dtype=int),
+                np.zeros(3, dtype=int),
                 nominator=svn,
             )
         # adj matrix should be square
         with self.assertRaises(IndexError):
-            self._nominate(np.zeros((3, 4), dtype=np.int), np.zeros(3, dtype=np.int))
+            self._nominate(np.zeros((3, 4), dtype=int), np.zeros(3, dtype=int))
 
     def _test_k(self):
         # k should be > 0
         with self.assertRaises(ValueError):
-            self._nominate(adj, np.zeros(3, dtype=np.int), k=0)
+            self._nominate(adj, np.zeros(3, dtype=int), k=0)
         # k of wrong type
         with self.assertRaises(TypeError):
-            self._nominate(adj, np.zeros(3, dtype=np.int), k="hello world")
+            self._nominate(adj, np.zeros(3, dtype=int), k="hello world")
 
     def test_constructor_inputs(self):
         with self.assertRaises(ValueError):
@@ -90,7 +90,7 @@ def test_basic_unattributed(self):
         seeds = [
             np.array([8]),
             np.array([2, 6, 9, 15, 25]),
-            np.arange(n_verts - 1, dtype=np.int),
+            np.arange(n_verts - 1, dtype=int),
         ]
         for nominator, seed in itertools.product(nominators, seeds):
             self._nominate(adj, seed, nominator)
@@ -99,7 +99,7 @@ def test_pre_embedded(self):
         seeds = [
             np.array([8]),
             np.array([2, 6, 9, 15, 25]),
-            np.arange(n_verts - 1, dtype=np.int),
+            np.arange(n_verts - 1, dtype=int),
         ]
         for seed in seeds:
             svn = SpectralVertexNomination(input_graph=False)

tests/test_svd.py

@@ -36,7 +36,7 @@ def test_outputs(self):
             ]
         )
         P = X @ X.T
-        A = np.random.binomial(1, P).astype(np.float)
+        A = np.random.binomial(1, P).astype(float)
 
         n_components = 3
 
@@ -75,7 +75,7 @@ def test_eigsh(self):
             ]
         )
         P = X @ X.T
-        A = np.random.binomial(1, P).astype(np.float)
+        A = np.random.binomial(1, P).astype(float)
         A = symmetrize(A, method="triu")
         n_components = 3