replace black, isort, pylint with ruff

.github/workflows/build.yml

@@ -68,7 +68,7 @@ jobs:
       - name: Install dependencies
         run: poetry install
       - name: Run Format Check
-        run: poetry run poe format_check
+        run: poetry run poe static_checks
   test-coverage:
     runs-on: ubuntu-latest
     strategy:

docs/conf.py

@@ -15,6 +15,11 @@
 import os
 import sys
 
+# Use RTD Theme
+import sphinx_rtd_theme
+
+import graspologic
+
 sys.path.append(os.path.abspath("./sphinx-ext/"))
 sys.path.insert(0, os.path.abspath(".."))
 
@@ -28,7 +33,6 @@
 dir_realpath = os.path.dirname(realpath)
 sys.path.append(dir_realpath)
 
-import graspologic
 
 version = graspologic.__version__
 # Append "dev" and the github run to the version when on the dev branch
@@ -117,9 +121,6 @@
 pygments_style = "sphinx"
 smartquotes = False
 
-# Use RTD Theme
-import sphinx_rtd_theme
-
 html_theme = "sphinx_rtd_theme"
 html_theme_path = [sphinx_rtd_theme.get_html_theme_path()]
 html_theme_options = {

graspologic/cluster/autogmm.py

@@ -574,7 +574,7 @@ def fit(self, X: np.ndarray, y: Optional[np.ndarray] = None) -> "AutoGMMCluster"
                     agg = AgglomerativeClustering(
                         n_clusters=self.min_components,
                         metric=affinity,
-                        **p_ag_without_affinity
+                        **p_ag_without_affinity,
                     )
                     agg.fit(X_subset)
                     hierarchical_labels = _hierarchical_labels(
@@ -762,9 +762,10 @@ def _hierarchical_labels(
         inds = np.where(np.isin(hierarchical_labels[:, n], children[n, :]))[0]
         hierarchical_labels[inds, -1] = n_samples + n
         if n < merge_end:
-            hierarchical_labels = np.hstack(
-                (hierarchical_labels, hierarchical_labels[:, -1].reshape((-1, 1)))
-            )
+            hierarchical_labels = np.hstack((
+                hierarchical_labels,
+                hierarchical_labels[:, -1].reshape((-1, 1)),
+            ))
 
     hierarchical_labels = hierarchical_labels[:, merge_start:]
     for i in range(hierarchical_labels.shape[1]):

graspologic/cluster/divisive_cluster.py

@@ -214,7 +214,7 @@ def _cluster_and_decide(self, X: np.ndarray) -> np.ndarray:
             cluster = AutoGMMCluster(
                 min_components=min_components,
                 max_components=self.max_components,
-                **self.cluster_kws
+                **self.cluster_kws,
             )
             cluster.fit(X)
             model = cluster.model_
@@ -275,9 +275,10 @@ def _fit(self, X: np.ndarray) -> np.ndarray:
                 ):
                     child_labels = dc._fit(new_X)
                     while labels.shape[1] <= child_labels.shape[1]:
-                        labels = np.column_stack(
-                            (labels, np.zeros((len(X), 1), dtype=int))
-                        )
+                        labels = np.column_stack((
+                            labels,
+                            np.zeros((len(X), 1), dtype=int),
+                        ))
                     labels[inds, 1 : child_labels.shape[1] + 1] = child_labels
                 else:
                     # make a "GaussianMixture" model for clusters

graspologic/cluster/kclust.py

@@ -88,9 +88,7 @@ def fit(self, X: np.ndarray, y: Optional[np.ndarray] = None) -> "KMeansCluster":
         # Deal with number of clusters
         if self.max_clusters > X.shape[0]:
             msg = "n_components must be >= n_samples, but got \
-                n_components = {}, n_samples = {}".format(
-                self.max_clusters, X.shape[0]
-            )
+                n_components = {}, n_samples = {}".format(self.max_clusters, X.shape[0])
             raise ValueError(msg)
         else:
             max_clusters = self.max_clusters

graspologic/embed/ase.py

@@ -142,7 +142,7 @@ def fit(
         graph: GraphRepresentation,
         y: Optional[Any] = None,
         *args: Any,
-        **kwargs: Any
+        **kwargs: Any,
     ) -> "AdjacencySpectralEmbed":
         """
         Fit ASE model to input graph

graspologic/embed/base.py

@@ -144,7 +144,7 @@ def fit(
         graph: GraphRepresentation,
         y: Optional[Any] = None,
         *args: Any,
-        **kwargs: Any
+        **kwargs: Any,
     ) -> "BaseSpectralEmbed":
         """
         A method for embedding.
@@ -221,7 +221,7 @@ def fit_transform(
         graph: GraphRepresentation,
         y: Optional[Any] = None,
         *args: Any,
-        **kwargs: Any
+        **kwargs: Any,
     ) -> Union[np.ndarray, Tuple[np.ndarray, np.ndarray]]:
         """
         Fit the model with graphs and apply the transformation.
@@ -312,8 +312,8 @@ def transform(self, X):  # type: ignore
         # correct types?
         if directed and not isinstance(X, tuple):
             if X.shape[0] == X.shape[1]:  # in case original matrix was passed
-                msg = """A square matrix A was passed to ``transform`` in the directed case. 
-                If this was the original in-sample matrix, either use ``fit_transform`` 
+                msg = """A square matrix A was passed to ``transform`` in the directed case.
+                If this was the original in-sample matrix, either use ``fit_transform``
                 or pass a tuple (A.T, A). If this was an out-of-sample matrix, directed
                 graphs require a tuple (X_out, X_in)."""
                 raise TypeError(msg)

graspologic/embed/lse.py

@@ -148,7 +148,7 @@ def fit(
         graph: GraphRepresentation,
         y: Optional[Any] = None,
         *args: Any,
-        **kwargs: Any
+        **kwargs: Any,
     ) -> "LaplacianSpectralEmbed":
         """
         Fit LSE model to input graph

graspologic/embed/mase.py

@@ -172,18 +172,14 @@ def _reduce_dim(self, graphs):  # type: ignore
             Vs = np.hstack([V.T[:, :best_dimension] for V in Vs])
         else:
             # Equivalent to ASE
-            Us = np.hstack(
-                [
-                    U[:, :best_dimension] @ np.diag(np.sqrt(D[:best_dimension]))
-                    for U, D in zip(Us, Ds)
-                ]
-            )
-            Vs = np.hstack(
-                [
-                    V.T[:, :best_dimension] @ np.diag(np.sqrt(D[:best_dimension]))
-                    for V, D in zip(Vs, Ds)
-                ]
-            )
+            Us = np.hstack([
+                U[:, :best_dimension] @ np.diag(np.sqrt(D[:best_dimension]))
+                for U, D in zip(Us, Ds)
+            ])
+            Vs = np.hstack([
+                V.T[:, :best_dimension] @ np.diag(np.sqrt(D[:best_dimension]))
+                for V, D in zip(Vs, Ds)
+            ])
 
         # Second SVD for vertices
         # The notation is slightly different than the paper

graspologic/embed/mds.py

@@ -1,18 +1,19 @@
 # Copyright (c) Microsoft Corporation and contributors.
 # Licensed under the MIT License.
 
-from typing import Any, Optional, Union
+from typing import TYPE_CHECKING, Any, Optional, Union
 
 import numpy as np
 from sklearn.base import BaseEstimator
 from sklearn.utils import check_array
 from typing_extensions import Literal
 
-from graspologic.types import Tuple
-
 from ..utils import is_symmetric
 from .svd import SvdAlgorithmType, select_svd
 
+if TYPE_CHECKING:
+    from graspologic.types import Tuple
+
 
 def _get_centering_matrix(n: int) -> np.ndarray:
     """

graspologic/embed/omni.py

@@ -56,7 +56,7 @@ def _get_omnibus_matrix_sparse(matrices: List[csr_array]) -> csr_array:
 
 
 def _get_laplacian_matrices(
-    graphs: Union[np.ndarray, List[GraphRepresentation]]
+    graphs: Union[np.ndarray, List[GraphRepresentation]],
 ) -> Union[np.ndarray, List[np.ndarray]]:
     """
     Helper function to convert graph adjacency matrices to graph Laplacian
@@ -82,7 +82,7 @@ def _get_laplacian_matrices(
 
 
 def _get_omni_matrix(
-    graphs: Union[AdjacencyMatrix, List[AdjacencyMatrix]]
+    graphs: Union[AdjacencyMatrix, List[AdjacencyMatrix]],
 ) -> np.ndarray:
     """
     Helper function for creating the omnibus matrix.

graspologic/embed/svd.py

@@ -44,7 +44,7 @@ def _compute_likelihood(arr: np.ndarray) -> np.ndarray:
             mu2 = -np.inf
 
         # compute pooled variance
-        variance = ((np.sum((s1 - mu1) ** 2) + np.sum((s2 - mu2) ** 2))) / (
+        variance = (np.sum((s1 - mu1) ** 2) + np.sum((s2 - mu2) ** 2)) / (
             n_elements - 1 - (idx < n_elements)
         )
         std = np.sqrt(variance)

graspologic/inference/binomial.py

@@ -79,6 +79,6 @@ def binom_2samp(
             value=null_ratio,
         )
     else:
-        raise ValueError()
+        raise ValueError
 
     return BinomialResult(stat, pvalue)

graspologic/inference/group_connection_test.py

@@ -378,18 +378,18 @@ def group_connection_test(
     B1, n_observed1, n_possible1, group_counts1 = fit_sbm(A1, labels1)
     B2, n_observed2, n_possible2, group_counts2 = fit_sbm(A2, labels2)
     if not n_observed1.index.equals(n_observed2.index):
-        raise ValueError()
+        raise ValueError
     elif not n_observed1.columns.equals(n_observed2.columns):
-        raise ValueError()
+        raise ValueError
     elif not n_possible1.index.equals(n_possible2.index):
-        raise ValueError()
+        raise ValueError
     elif not n_observed1.columns.equals(n_observed2.columns):
-        raise ValueError()
+        raise ValueError
 
     index = n_observed1.index.copy()
 
     if n_observed1.shape[0] != n_observed2.shape[0]:
-        raise ValueError()
+        raise ValueError
 
     K = n_observed1.shape[0]
 

graspologic/inference/latent_position_test.py

@@ -124,9 +124,9 @@ def latent_position_test(
     if type(embedding) is not str:
         raise TypeError("embedding must be str")
     if type(n_bootstraps) is not int:
-        raise TypeError()
+        raise TypeError
     if type(test_case) is not str:
-        raise TypeError()
+        raise TypeError
     if n_bootstraps < 1:
         raise ValueError(
             "{} is invalid number of bootstraps, must be greater than 1".format(
@@ -148,7 +148,7 @@ def latent_position_test(
     A1 = import_graph(A1)
     A2 = import_graph(A2)
     if not is_symmetric(A1) or not is_symmetric(A2):
-        raise NotImplementedError()  # TODO asymmetric case
+        raise NotImplementedError  # TODO asymmetric case
     if A1.shape != A2.shape:
         raise ValueError("Input matrices do not have matching dimensions")
     num_components: int

graspologic/layouts/nooverlap/_grid.py

@@ -180,7 +180,7 @@ def get_all_grid_cells(self) -> List[List[int]]:
 
     def print_stats(self) -> None:
         print(
-            f"cell size: {self.cell_size}, area: {self.cell_size*self.cell_size}, "
+            f"cell size: {self.cell_size}, area: {self.cell_size * self.cell_size}, "
             f"rows: {self._get_y_cells()}, cols: {self._get_x_cells()}"
         )
 

graspologic/layouts/nooverlap/nooverlap.py

@@ -23,7 +23,7 @@ def remove_overlaps(node_positions: List[NodePosition]) -> List[NodePosition]:
     qt = _QuadTree(local_nodes, 50)
     qt.layout_dense_first(first_color=None)
     stop = time.time()
-    logger.info(f"removed overlap in {stop-start} seconds")
+    logger.info(f"removed overlap in {stop - start} seconds")
 
     new_positions = [
         NodePosition(

graspologic/match/solver.py

@@ -430,9 +430,10 @@ def finalize(self, P: np.ndarray, rng: np.random.Generator) -> None:
             permutation = np.array([], dtype=int)
 
         # deal with seed-nonseed sorting from the initialization
-        permutation = np.concatenate(
-            (np.arange(self.n_seeds), permutation + self.n_seeds)
-        )
+        permutation = np.concatenate((
+            np.arange(self.n_seeds),
+            permutation + self.n_seeds,
+        ))
         final_permutation = np.empty(self.n, dtype=int)
         final_permutation[self.perm_A] = self.perm_B[permutation]
 

graspologic/match/wrappers.py

@@ -30,7 +30,7 @@ class MatchResult(NamedTuple):
     indices_B: np.ndarray
     """
     Indices in ``B`` which were matched. Element ``indices_B[i]`` was matched
-    to element ``indices_A[i]``. ``indices_B`` can also be thought of as a 
+    to element ``indices_A[i]``. ``indices_B`` can also be thought of as a
     permutation of the nodes of ``B`` with respect to ``A``.
     """
 

graspologic/models/sbm_estimators.py

@@ -166,7 +166,7 @@ def _estimate_assignments(self, graph: GraphRepresentation) -> None:
         gc = GaussianCluster(
             min_components=self.min_comm,
             max_components=self.max_comm,
-            **self.cluster_kws
+            **self.cluster_kws,
         )
         vertex_assignments = gc.fit_predict(latent)  # type: ignore
         self.vertex_assignments_ = vertex_assignments
@@ -374,7 +374,7 @@ def _estimate_assignments(self, graph: GraphRepresentation) -> None:
         gc = GaussianCluster(
             min_components=self.min_comm,
             max_components=self.max_comm,
-            **self.cluster_kws
+            **self.cluster_kws,
         )
         self.vertex_assignments_ = gc.fit_predict(latent)  # type: ignore
 

graspologic/nominate/VNviaSGM.py

@@ -259,9 +259,7 @@ def fit(
         if len(close_seeds) <= 0:
             warnings.warn(
                 'Voi {} was not a member of the induced subgraph A[{}], \
-                Try increasing "order_voi_subgraph"'.format(
-                    voi, seedsA
-                )
+                Try increasing "order_voi_subgraph"'.format(voi, seedsA)
             )
             self.n_seeds_ = None
             self.nomination_list_ = None

graspologic/partition/leiden.py

@@ -352,8 +352,8 @@ class HierarchicalCluster(NamedTuple):
     """Only used when level != 0, but will indicate the previous cluster id that this node was in"""
     level: int
     """
-    Each time a community has a higher population than we would like, we create a subnetwork 
-    of that community and process it again to break it into smaller chunks. Each time we 
+    Each time a community has a higher population than we would like, we create a subnetwork
+    of that community and process it again to break it into smaller chunks. Each time we
     detect this, the level increases by 1
     """
     is_final_cluster: bool

graspologic/pipeline/embed/__init__.py

@@ -1,6 +1,6 @@
 # Copyright (c) Microsoft Corporation.
 # Licensed under the MIT license.
-
+# ruff: noqa: E402 SVD_SOLVER_TYPES needs to be first
 """
 The embed module of ``graspologic.pipeline.embed`` is intended to provide faster
 application development support. The functions provided in it reflect common call
@@ -9,8 +9,8 @@
 """
 
 __SVD_SOLVER_TYPES = ["randomized", "full", "truncated"]
-
 from .adjacency_spectral_embedding import adjacency_spectral_embedding
 from .embeddings import Embeddings, EmbeddingsView
 from .laplacian_spectral_embedding import laplacian_spectral_embedding
 from .omnibus_embedding import omnibus_embedding_pairwise
+

graspologic/pipeline/embed/_elbow.py

@@ -38,7 +38,7 @@ def _compute_likelihood(arr: np.ndarray) -> np.ndarray:
             mu2 = -np.inf
 
         # compute pooled variance
-        variance = ((np.sum((s1 - mu1) ** 2) + np.sum((s2 - mu2) ** 2))) / (
+        variance = (np.sum((s1 - mu1) ** 2) + np.sum((s2 - mu2) ** 2)) / (
             n_elements - 1 - (idx < n_elements)
         )
         std = np.sqrt(variance)

graspologic/pipeline/embed/omnibus_embedding.py

@@ -220,12 +220,10 @@ def omnibus_embedding_pairwise(
             elbow_cut, graph.is_directed(), model.singular_values_, current_embedding
         )
 
-        graph_embeddings.append(
-            (
-                Embeddings(union_node_ids, previous_embedding_cut),
-                Embeddings(union_node_ids, current_embedding_cut),
-            )
-        )
+        graph_embeddings.append((
+            Embeddings(union_node_ids, previous_embedding_cut),
+            Embeddings(union_node_ids, current_embedding_cut),
+        ))
 
     return graph_embeddings