use ruff for linting (#291)

Author: LegrandNico

.pre-commit-config.yaml

@@ -1,28 +1,18 @@
 exclude: |
     (?x)^(
-        setup.py|
         docs/.*|
     )$
+    
 repos:
--   repo: https://github.com/pre-commit/mirrors-isort
-    rev: v5.10.1
-    hooks:
-    - id: isort
--   repo: https://github.com/ambv/black
-    rev: 25.1.0
-    hooks:
-    - id: black
-      language_version: python3
--   repo: https://github.com/pycqa/flake8
-    rev: 7.2.0
-    hooks:
-    - id: flake8
--   repo: https://github.com/pycqa/pydocstyle
-    rev: 6.3.0
-    hooks:
-    -   id: pydocstyle
-        args: ['--ignore', 'D213,D100,D203,D104']
-        files: ^pyhgf/
+- repo: https://github.com/astral-sh/ruff-pre-commit
+  # Ruff version.
+  rev: v0.11.4
+  hooks:
+    # Run the linter.
+    - id: ruff
+      args: [ --fix ]
+    # Run the formatter.
+    - id: ruff-format
 -   repo: https://github.com/pre-commit/mirrors-mypy
     rev: 'v1.15.0'
     hooks:

pyhgf/distribution.py

@@ -133,7 +133,6 @@ def logp(
 
     """
     if hgf.model_type == "continuous":
-
         # update this network's attributes
         hgf.attributes[0]["precision"] = input_precision
 
@@ -161,7 +160,6 @@ def logp(
             hgf.attributes[3]["volatility_coupling_children"] = (volatility_coupling_2,)
 
     elif hgf.model_type == "binary":
-
         # update this network's attributes
         hgf.attributes[0]["mean"] = mean_1
         hgf.attributes[1]["mean"] = mean_2
@@ -486,23 +484,26 @@ def perform(
     ):
         """Perform node operations."""
         (
-            grad_mean_1,
-            grad_mean_2,
-            grad_mean_3,
-            grad_precision_1,
-            grad_precision_2,
-            grad_precision_3,
-            grad_tonic_volatility_1,
-            grad_tonic_volatility_2,
-            grad_tonic_volatility_3,
-            grad_tonic_drift_1,
-            grad_tonic_drift_2,
-            grad_tonic_drift_3,
-            grad_volatility_coupling_1,
-            grad_volatility_coupling_2,
-            grad_input_precision,
-            grad_response_function_parameters,
-        ), _ = self.grad_logp(*inputs)
+            (
+                grad_mean_1,
+                grad_mean_2,
+                grad_mean_3,
+                grad_precision_1,
+                grad_precision_2,
+                grad_precision_3,
+                grad_tonic_volatility_1,
+                grad_tonic_volatility_2,
+                grad_tonic_volatility_3,
+                grad_tonic_drift_1,
+                grad_tonic_drift_2,
+                grad_tonic_drift_3,
+                grad_volatility_coupling_1,
+                grad_volatility_coupling_2,
+                grad_input_precision,
+                grad_response_function_parameters,
+            ),
+            _,
+        ) = self.grad_logp(*inputs)
 
         outputs[0][0] = np.asarray(grad_mean_1, dtype=node.outputs[0].dtype)
         outputs[1][0] = np.asarray(grad_mean_2, dtype=node.outputs[1].dtype)

pyhgf/model/add_nodes.py

@@ -159,7 +159,6 @@ def add_ef_state(
 
     # loop over the indexes of nodes created in the previous step
     for node_idx in range(network.n_nodes - 1, network.n_nodes - n_nodes - 1, -1):
-
         # create the sufficient statistic function and store in the side parameters
         if network.attributes[node_idx]["distribution"] == "normal":
             sufficient_stats_fn = Normal().sufficient_statistics_from_observations
@@ -178,7 +177,6 @@ def add_ef_state(
         ] = sufficient_stats_fn
 
         if "hgf" in network.attributes[node_idx]["learning"]:
-
             # create a collection of continuous state nodes
             # to track the sufficient statistics of the implied distribution
             for i in range(n_suff_stats):

pyhgf/model/hgf.py

@@ -156,7 +156,6 @@ def __init__(
                     )
 
             elif model_type == "binary":
-
                 if binary_precision == jnp.inf:
                     # X - 0
                     self.add_nodes(

pyhgf/model/network.py

@@ -142,7 +142,6 @@ def create_belief_propagation_fn(
 
         # Create the generative scan function if it doesn't exist, and if requested.
         if (self.scan_fn_sample is None) and sampling_fn:
-
             self.sample_scan_fn = Partial(
                 beliefs_propagation,
                 update_sequence=self.update_sequence,

pyhgf/plots/graphviz/plot_network.py

@@ -38,7 +38,6 @@ def plot_network(network: "Network") -> "Source":
 
     # create the rest of nodes
     for idx in range(len(network.edges)):
-
         style = "filled" if idx in network.input_idxs else ""
 
         if network.edges[idx].node_type == 1:
@@ -89,7 +88,6 @@ def plot_network(network: "Network") -> "Source":
 
         if value_parents is not None:
             for value_parents_idx in value_parents:
-
                 # get the coupling function from the value parent
                 child_idx = network.edges[value_parents_idx].value_children.index(i)
                 coupling_fn = network.edges[value_parents_idx].coupling_fn[child_idx]

pyhgf/plots/matplotlib/plot_nodes.py

@@ -148,11 +148,9 @@ def plot_nodes(
                 # plotting standard deviation - in the case of a binary input node, the
                 # CI should be read from the value parent using the sigmoid transform
                 if ci is True:
-
                     # get parent nodes and sum predictions
                     mean_parent, precision_parent = 0.0, 0.0
                     for idx in network.edges[node_idx].value_parents:  # type: ignore
-
                         # compute  mu +/- sd at time t-1
                         # and use the sigmoid transform before plotting
                         mean_parent += trajectories_df[f"x_{idx}_expected_mean"]
@@ -164,7 +162,6 @@ def plot_nodes(
                     y2 = 1 / (1 + np.exp(-mean_parent - sd))
 
             if ci is True:
-
                 axs[i].fill_between(
                     x=trajectories_df["time"],
                     y1=y1,
@@ -180,7 +177,6 @@ def plot_nodes(
         # plotting state nodes
         # --------------------
         else:
-
             axs[i].set_title(
                 f"State Node {node_idx}",
                 loc="left",

pyhgf/plots/matplotlib/plot_trajectories.py

@@ -8,7 +8,6 @@
 from pyhgf.plots.matplotlib.plot_nodes import plot_nodes
 
 if TYPE_CHECKING:
-
     from pyhgf.model import Network
 
 
@@ -128,7 +127,6 @@ def plot_trajectories(
     # --------------
     ax_i = n_nodes - 1
     for node_idx in range(n_nodes):
-
         if node_idx in network.input_idxs:
             _show_posterior = True
             color = "#4c72b0"

pyhgf/plots/networkx/plot_network.py

@@ -6,7 +6,6 @@
 import matplotlib.pyplot as plt
 
 if TYPE_CHECKING:
-
     from pyhgf.model import Network
 
 

pyhgf/updates/posterior/continuous/posterior_update_mean_continuous_node.py

@@ -147,13 +147,11 @@ def posterior_update_mean_continuous_node(
             # sum the precision weigthed prediction errors over all children
             value_precision_weigthed_prediction_error += (
                 (
-                    (
-                        value_coupling
-                        * attributes[value_child_idx]["expected_precision"]
-                        * coupling_fn_prime
-                    )
-                    / node_precision
+                    value_coupling
+                    * attributes[value_child_idx]["expected_precision"]
+                    * coupling_fn_prime
                 )
+                / node_precision
             ) * value_prediction_error
 
     # Volatility coupling updates - update the mean of a volatility parent