Merge pull request #121 from yarikoptic/enh-codespell

codespell: action, config + typos fixed

Author: kunwuz

.codespellrc

@@ -0,0 +1,5 @@
+[codespell]
+skip = .git,*.pdf,*.svg,*.ipynb,*.js,*.css,build
+# interm,te - quick variables used
+# alph - skip as might be not desired change at this stage
+ignore-words-list = ges,interm,te,alph

.github/workflows/codespell.yml

@@ -0,0 +1,22 @@
+---
+name: Codespell
+
+on:
+  push:
+    branches: [main]
+  pull_request:
+    branches: [main]
+
+permissions:
+  contents: read
+
+jobs:
+  codespell:
+    name: Check for spelling errors
+    runs-on: ubuntu-latest
+
+    steps:
+      - name: Checkout
+        uses: actions/checkout@v3
+      - name: Codespell
+        uses: codespell-project/actions-codespell@v2

causallearn/search/ConstraintBased/FCI.py

@@ -387,7 +387,7 @@ def doDdpOrientation(node_d: Node, node_a: Node, node_b: Node, node_c: Node, pre
                      independence_test_method, alpha: float, sep_sets: Dict[Tuple[int, int], Set[int]],
                      change_flag: bool, bk, verbose: bool = False) -> (bool, bool):
     """
-    Orients the edges inside the definte discriminating path triangle. Takes
+    Orients the edges inside the definite discriminating path triangle. Takes
     the left endpoint, and a,b,c as arguments.
     """
     if graph.is_adjacent_to(node_d, node_c):

causallearn/search/ConstraintBased/PC.py

@@ -394,7 +394,7 @@ def detect_parent(r: int, data_: ndarray, alpha: float, indep_test, stable: bool
 def get_parent(r: int, cg_skel_adj: ndarray) -> ndarray:
     """Get the neighbors of missingness indicators which are the parents
     :param r: the missingness indicator index
-    :param cg_skel_adj: adjacancy matrix of a causal skeleton
+    :param cg_skel_adj: adjacency matrix of a causal skeleton
     :return:
     prt: list, parents of the missingness indicator r
     """

causallearn/search/FCMBased/lingam/causal_effect.py

@@ -80,7 +80,7 @@ def _get_propagated_effects(self, En, intervention_index, intervention_value):
 
         Parameters
         ----------
-        En : array-like, shpae (n_features)
+        En : array-like, shape (n_features)
             Expectations of each noise variable.
         intervention_index : int
             Index of variable to apply intervention.
@@ -89,7 +89,7 @@ def _get_propagated_effects(self, En, intervention_index, intervention_value):
 
         Returns
         -------
-        propagated_effects : array-like, shpae (n_features)
+        propagated_effects : array-like, shape (n_features)
             Propagated effects, where ``n_features`` is the number of features.
         """
         effects = np.zeros(len(self._causal_order))
@@ -106,7 +106,7 @@ def _predict(self, X, pred_model):
 
         Parameters
         ----------
-        X : array-like, shpae (n_predictors)
+        X : array-like, shape (n_predictors)
             Predictors, where ``n_predictors`` is the number of variables.
         pred_model : model object implementing 'predict' or 'predict_proba'
             Model to predict the expectation. For linear regression or non-linear reggression, model object must have ``predict`` method.

causallearn/search/FCMBased/lingam/direct_lingam.py

@@ -146,7 +146,7 @@ def _entropy(self, u):
                k2 * (np.mean(u * np.exp((-u ** 2) / 2))) ** 2
 
     def _diff_mutual_info(self, xi_std, xj_std, ri_j, rj_i):
-        """Calculate the difference of the mutual informations."""
+        """Calculate the difference of the mutual information."""
         return (self._entropy(xj_std) + self._entropy(ri_j / np.std(ri_j))) - \
                (self._entropy(xi_std) + self._entropy(rj_i / np.std(rj_i)))
 
@@ -215,7 +215,7 @@ def _search_causal_order(self, X, U):
         return Uc[np.argmax(M_list)]
 
     def _mutual_information(self, x1, x2, param):
-        """Calculate the mutual informations."""
+        """Calculate the mutual information."""
         kappa, sigma = param
         n = len(x1)
         X1 = np.tile(x1, (n, 1))

causallearn/search/FCMBased/lingam/utils/__init__.py

@@ -24,7 +24,7 @@ def print_causal_directions(cdc, n_sampling, labels=None):
     n_sampling : int
         Number of bootstrapping samples.
     labels : array-like, optional (default=None)
-        List of feature lables.
+        List of feature labels.
         If set labels, the output feature name will be the specified label.
     """
     for i, (fr, to, co) in enumerate(zip(cdc['from'], cdc['to'], cdc['count'])):
@@ -47,7 +47,7 @@ def print_dagc(dagc, n_sampling, labels=None):
     n_sampling : int
         Number of bootstrapping samples.
     labels : array-like, optional (default=None)
-        List of feature lables.
+        List of feature labels.
         If set labels, the output feature name will be the specified label.
     """
     for i, (dag, co) in enumerate(zip(dagc['dag'], dagc['count'])):

causallearn/search/ScoreBased/ExactSearch.py

@@ -60,7 +60,7 @@ def bic_exact_search(X, super_graph=None, search_method='astar',
     dag_est :  numpy.ndarray, shape=(d, d)
         Estimated DAG.
     search_stats :  dict
-        Some statistics related to the seach procedure.
+        Some statistics related to the search procedure.
     """
     n, d = X.shape
     if super_graph is None:
@@ -76,7 +76,7 @@ def bic_exact_search(X, super_graph=None, search_method='astar',
     if max_parents is None:
         max_parents = d
 
-    # To store statistics related to the seach procedure
+    # To store statistics related to the search procedure
     search_stats = {}
 
     # Generate parent graphs (without parallel computing)
@@ -102,7 +102,7 @@ def bic_exact_search(X, super_graph=None, search_method='astar',
     if verbose:
         _logger.info("Finished searching for shortest path.")
 
-    # Covnert structures to adjacency matrix
+    # Convert structures to adjacency matrix
     dag_est = np.zeros((d, d))
     for i, parents in enumerate(structures):
         dag_est[parents, i] = 1
@@ -135,14 +135,14 @@ def astar_shortest_path(parent_graphs, use_path_extension=True,
     structures :  tuple, shape=(d,)
         Optimal parents for each variable.
     shortest_path_stats :  dict
-        Some statistics related to the shortest path seach.
+        Some statistics related to the shortest path search.
     """
     d = len(parent_graphs)
     opened = PriorityQueue()
     closed = set()
 
     if use_k_cycle_heuristic:
-        # Create pattern databse
+        # Create pattern database
         PD = create_dynamic_pd(parent_graphs, k)
         if verbose:
             _logger.info('Finished creating pattern database.')
@@ -234,7 +234,7 @@ def dp_shortest_path(parent_graphs, use_path_extension=True, verbose=False):
     structures :  tuple, shape=(d,)
         Optimal parents for each variable.
     shortest_path_stats :  dict
-        Some statistics related to the shortest path seach.
+        Some statistics related to the shortest path search.
     """
     d = len(parent_graphs)
     order_graph = nx.DiGraph()
@@ -320,7 +320,7 @@ def generate_parent_graph(X, i, max_parents=None, parent_set=None, include_paren
         parent_set = tuple(set(parent_set))
 
     if include_parents is None:
-        include_parents = ()  # Emptry tuple
+        include_parents = ()  # Empty tuple
 
     parent_graph = []
     for j in range(len(parent_set) + 1):

causallearn/utils/DAG2CPDAG.py

@@ -8,7 +8,7 @@
 
 def dag2cpdag(G: Dag) -> GeneralGraph:
     """
-    Covert a DAG to its corresponding PDAG
+    Convert a DAG to its corresponding PDAG
 
     Parameters
     ----------
@@ -99,7 +99,7 @@ def dag2cpdag(G: Dag) -> GeneralGraph:
             id2 = np.intersect1d(np.where(sign_edges == 0)[0], id1)
             sign_edges[id2] = -1  # label all "unknown" edges incident into y with "reversible"
 
-    # create CPDAG accoring the labelled edge
+    # create CPDAG according the labelled edge
     nodes = G.get_nodes()
     CPDAG = GeneralGraph(nodes)
     for m in range(M):

causallearn/utils/DAG2PAG.py

@@ -15,7 +15,7 @@
 
 def dag2pag(dag: Dag, islatent: List[Node]) -> GeneralGraph:
     """
-    Covert a DAG to its corresponding PAG
+    Convert a DAG to its corresponding PAG
     Parameters
     ----------
     dag : Direct Acyclic Graph