[pre-commit.ci] auto fixes from pre-commit.com hooks

for more information, see https://pre-commit.ci

dodiscover/_version.py

@@ -1,4 +1,5 @@
 """Version number."""
+
 from importlib.metadata import version  # type: ignore
 
 __version__ = version(__package__)

dodiscover/constraint/_classes.py

@@ -132,7 +132,7 @@ def _initialize_sep_sets(self, init_graph: nx.Graph) -> SeparatingSet:
         sep_set: SeparatingSet = defaultdict(lambda: defaultdict(list))
 
         # since we are not starting from a complete graph, find the separating sets
-        for (node_i, node_j) in itertools.combinations(init_graph.nodes, 2):
+        for node_i, node_j in itertools.combinations(init_graph.nodes, 2):
             if not init_graph.has_edge(node_i, node_j):
                 sep_set[node_i][node_j] = []
                 sep_set[node_j][node_i] = []

dodiscover/constraint/fcialg.py

@@ -706,7 +706,7 @@ def _apply_rule10(
                     # that:
                     # i) begin the uncovered pd path and
                     # ii) are distinct (done by construction) here
-                    for (m, w) in combinations(graph.neighbors(a), 2):  # type: ignore
+                    for m, w in combinations(graph.neighbors(a), 2):  # type: ignore
                         if m == c or w == c:
                             continue
 
@@ -763,7 +763,7 @@ def _apply_orientation_rules(self, graph: EquivalenceClass, sep_set: SeparatingS
             logger.info(f"Running R1-10 for iteration {idx}")
 
             for u in graph.nodes:
-                for (a, c) in permutations(graph.neighbors(u), 2):
+                for a, c in permutations(graph.neighbors(u), 2):
                     logger.debug(f"Check {u} {a} {c}")
 
                     # apply R1-3 to orient triples and arrowheads

dodiscover/constraint/intervention.py

@@ -307,7 +307,7 @@ def _apply_orientation_rules(self, graph: EquivalenceClass, sep_set: SeparatingS
             logger.info(f"Running R1-10 for iteration {idx}")
 
             for u in graph.nodes:
-                for (a, c) in permutations(graph.neighbors(u), 2):
+                for a, c in permutations(graph.neighbors(u), 2):
                     logger.debug(f"Check {u} {a} {c}")
 
                     # apply R1-3 to orient triples and arrowheads

dodiscover/constraint/pcalg.py

@@ -347,7 +347,7 @@ def _apply_meek_rule3(self, graph: EquivalenceClass, i: Column, j: Column) -> bo
         if graph.has_edge(i, j, graph.undirected_edge_name):
             # For all the pairs of nodes adjacent to i,
             # look for (k, l), such that j -> l and k -> l
-            for (k, l) in combinations(graph.neighbors(i), 2):
+            for k, l in combinations(graph.neighbors(i), 2):
                 # Skip if k and l are adjacent.
                 if l in graph.neighbors(k):
                     continue

dodiscover/constraint/skeleton.py

@@ -1040,7 +1040,7 @@ def _prep_second_stage_skeleton(self, context: Context) -> Context:
         # Update the Context:
         # add the corresponding intermediate PAG now to the context
         # new initialization graph
-        for (_, _, d) in new_init_graph.edges(data=True):
+        for _, _, d in new_init_graph.edges(data=True):
             if "test_stat" in d:
                 d.pop("test_stat")
             if "pvalue" in d:

dodiscover/toporder/cam.py

@@ -162,9 +162,9 @@ def _update_directed_paths(self, parent: int, child: int, directed_paths: NDArra
         directed_paths : np.ndarray
             Existing directed paths in the graph.
         """
-        directed_paths[
-            parent, child
-        ] = 1  # directed_paths[i,j]=1 if there is a directed path from i to j
+        directed_paths[parent, child] = (
+            1  # directed_paths[i,j]=1 if there is a directed path from i to j
+        )
         child_descendants = np.argwhere(directed_paths[child, :])
         parent_ancestors = np.argwhere(directed_paths[:, parent])
         for p in parent_ancestors:  # add path from every ancestor to every descendant
@@ -240,9 +240,9 @@ def _initialize_score(self, X: NDArray, directed_paths: NDArray) -> Tuple[NDArra
 
         # Initialize matrix of score gains and vector of initial scores
         score_gains = np.zeros((d, d))
-        score_gains[
-            np.transpose(directed_paths == 1, (1, 0))
-        ] = self.inf  # avoid cycles setting entries to -Inf according to current directed_paths
+        score_gains[np.transpose(directed_paths == 1, (1, 0))] = (
+            self.inf
+        )  # avoid cycles setting entries to -Inf according to current directed_paths
         nodes_variance = np.var(X, axis=0)
         init_score = -np.log(nodes_variance)
 

examples/plot_pc_alg.py

@@ -21,6 +21,7 @@
 
 .. currentmodule:: dodiscover
 """
+
 # %%
 # Authors: Adam Li <[email protected]>
 #

examples/plot_psifci_alg.py

@@ -22,7 +22,6 @@
 .. currentmodule:: dodiscover
 """
 
-
 # %%
 # Authors: Adam Li <[email protected]>
 #

examples/plot_score_alg.py

@@ -53,6 +53,7 @@
 
 .. currentmodule:: dodiscover
 """
+
 # %%
 # Authors: Francesco Montagna <[email protected]>
 #

tests/unit_tests/replearning/test_gin.py

@@ -1,4 +1,5 @@
 """Tests wrapper for GIN algorithm from causal"""
+
 import networkx as nx
 import numpy as np
 import pandas as pd