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

pre-commit-ci[bot] · pre-commit-ci[bot] · commit 669e847972e4 · 2026-03-16T18:55:44.000Z
for more information, see https://pre-commit.ci
diff --git a/README.md b/README.md
@@ -83,7 +83,7 @@ Additionally, `UNIQUE` also generates graphical outputs in the form of tables an
 
 For more hands-on examples and detailed usage, check out some of the examples in the [docs](https://opensource.nibr.com/UNIQUE/examples/index.html#examples).
 
-In addition, `example_usage/ADME_pulic/` contains files, scripts, and notebooks to reproduce the models and results for the ADME public dataset presented in "Uncertainty quantification in molecular machine learning for property predictions under data shifts" by Parrondo-Pizarro et al. 
+In addition, `example_usage/ADME_pulic/` contains files, scripts, and notebooks to reproduce the models and results for the ADME public dataset presented in "Uncertainty quantification in molecular machine learning for property predictions under data shifts" by Parrondo-Pizarro et al.
 
 ## Deep Dive
 
diff --git a/example_usage/ADME_public/scripts/model_building/chemprop_fingerprint.sh b/example_usage/ADME_public/scripts/model_building/chemprop_fingerprint.sh
@@ -26,6 +26,3 @@ chemprop fingerprint \
     --output ../MDR1_model/latent_fps.csv \
     --smiles-columns "Structure" \
     --ffn-block-index 0
-
-
-
diff --git a/example_usage/ADME_public/scripts/model_building/chemprop_predict.sh b/example_usage/ADME_public/scripts/model_building/chemprop_predict.sh
@@ -26,4 +26,3 @@ chemprop predict \
     --preds-path ../MDR1_model/predictions.csv \
     --smiles-columns "Structure" \
     --uncertainty-method ensemble
-
diff --git a/example_usage/ADME_public/scripts/run_unique.py b/example_usage/ADME_public/scripts/run_unique.py
@@ -1,32 +1,53 @@
-from unique import Pipeline
 import pandas as pd
 
+from unique import Pipeline
+
 # CLint Rat
-pipeline = Pipeline.from_config("../unique_config_files/rLM LogCLint_unique_config_file.yaml")
+pipeline = Pipeline.from_config(
+    "../unique_config_files/rLM LogCLint_unique_config_file.yaml"
+)
 uq_methods_outputs, uq_evaluation_outputs = pipeline.fit()
 
-pd.DataFrame.from_dict(uq_methods_outputs).to_csv(pipeline.output_path / "clint_rat_uq_metrics_values.csv", index=False)
+pd.DataFrame.from_dict(uq_methods_outputs).to_csv(
+    pipeline.output_path / "clint_rat_uq_metrics_values.csv", index=False
+)
 
 # CLint Human
-pipeline = Pipeline.from_config("../unique_config_files/hLM LogCLint_unique_config_file.yaml")
+pipeline = Pipeline.from_config(
+    "../unique_config_files/hLM LogCLint_unique_config_file.yaml"
+)
 uq_methods_outputs, uq_evaluation_outputs = pipeline.fit()
 
-pd.DataFrame.from_dict(uq_methods_outputs).to_csv(pipeline.output_path / "clint_human_uq_metrics_values.csv", index=False)
+pd.DataFrame.from_dict(uq_methods_outputs).to_csv(
+    pipeline.output_path / "clint_human_uq_metrics_values.csv", index=False
+)
 
 # PPB Rat
-pipeline = Pipeline.from_config("../unique_config_files/LogFu-Rat_unique_config_file.yaml")
+pipeline = Pipeline.from_config(
+    "../unique_config_files/LogFu-Rat_unique_config_file.yaml"
+)
 uq_methods_outputs, uq_evaluation_outputs = pipeline.fit()
 
-pd.DataFrame.from_dict(uq_methods_outputs).to_csv(pipeline.output_path / "ppb_rat_uq_metrics_values.csv", index=False)
+pd.DataFrame.from_dict(uq_methods_outputs).to_csv(
+    pipeline.output_path / "ppb_rat_uq_metrics_values.csv", index=False
+)
 
 # PPB Human
-pipeline = Pipeline.from_config("../unique_config_files/LogFu-Human_unique_config_file.yaml")
+pipeline = Pipeline.from_config(
+    "../unique_config_files/LogFu-Human_unique_config_file.yaml"
+)
 uq_methods_outputs, uq_evaluation_outputs = pipeline.fit()
 
-pd.DataFrame.from_dict(uq_methods_outputs).to_csv(pipeline.output_path / "ppb_human_uq_metrics_values.csv", index=False)
+pd.DataFrame.from_dict(uq_methods_outputs).to_csv(
+    pipeline.output_path / "ppb_human_uq_metrics_values.csv", index=False
+)
 
 # MDCK-MDR1 ER
-pipeline = Pipeline.from_config("../unique_config_files/MDCK-MDR1_LogER_unique_config_file.yaml")
+pipeline = Pipeline.from_config(
+    "../unique_config_files/MDCK-MDR1_LogER_unique_config_file.yaml"
+)
 uq_methods_outputs, uq_evaluation_outputs = pipeline.fit()
 
-pd.DataFrame.from_dict(uq_methods_outputs).to_csv(pipeline.output_path / "mdck_mdr1_er_uq_metrics_values.csv", index=False)
+pd.DataFrame.from_dict(uq_methods_outputs).to_csv(
+    pipeline.output_path / "mdck_mdr1_er_uq_metrics_values.csv", index=False
+)
diff --git a/example_usage/ADME_public/unique-env.yml b/example_usage/ADME_public/unique-env.yml
@@ -225,4 +225,4 @@ dependencies:
   - zipp=3.21.0=pyhd8ed1ab_1
   - zlib=1.3.1=hb9d3cd8_2
   - zstandard=0.23.0=py312h66e93f0_1
-  - zstd=1.5.7=hb8e6e7a_2
+  - zstd=1.5.7=hb8e6e7a_2
diff --git a/example_usage/ADME_public/unique_config_files/LogFu-Human_unique_config_file.yaml b/example_usage/ADME_public/unique_config_files/LogFu-Human_unique_config_file.yaml
@@ -35,7 +35,7 @@ inputs_list:
     column_name: latent_fp_scaled
     metrics:
     - manhattan_distance
-    
+
 - ModelInputType:
       # Name of the column containing the variance
     column_name: variance
@@ -75,4 +75,4 @@ display_outputs: true
 # Number of bootstrapping samples to run. Note: bootstrapping to determine the best UQ metric is ALWAYS run unless the private attribute `Pipeline._bootstrap` is set to False.
 n_bootstrap: 100
 # Logging messages levels. If True, logger will output DEBUG level messages.
-verbose: false
+verbose: false
diff --git a/example_usage/ADME_public/unique_config_files/LogFu-Rat_unique_config_file.yaml b/example_usage/ADME_public/unique_config_files/LogFu-Rat_unique_config_file.yaml
@@ -35,7 +35,7 @@ inputs_list:
     column_name: latent_fp_scaled
     metrics:
     - manhattan_distance
-    
+
 - ModelInputType:
       # Name of the column containing the variance
     column_name: variance
@@ -75,4 +75,4 @@ display_outputs: true
 # Number of bootstrapping samples to run. Note: bootstrapping to determine the best UQ metric is ALWAYS run unless the private attribute `Pipeline._bootstrap` is set to False.
 n_bootstrap: 100
 # Logging messages levels. If True, logger will output DEBUG level messages.
-verbose: false
+verbose: false
diff --git a/example_usage/ADME_public/unique_config_files/MDCK-MDR1_LogER_unique_config_file.yaml b/example_usage/ADME_public/unique_config_files/MDCK-MDR1_LogER_unique_config_file.yaml
@@ -35,7 +35,7 @@ inputs_list:
     column_name: latent_fp_scaled
     metrics:
     - manhattan_distance
-    
+
 - ModelInputType:
       # Name of the column containing the variance
     column_name: variance
@@ -75,4 +75,4 @@ display_outputs: true
 # Number of bootstrapping samples to run. Note: bootstrapping to determine the best UQ metric is ALWAYS run unless the private attribute `Pipeline._bootstrap` is set to False.
 n_bootstrap: 100
 # Logging messages levels. If True, logger will output DEBUG level messages.
-verbose: false
+verbose: false
diff --git a/example_usage/ADME_public/unique_config_files/hLM LogCLint_unique_config_file.yaml b/example_usage/ADME_public/unique_config_files/hLM LogCLint_unique_config_file.yaml
@@ -35,7 +35,7 @@ inputs_list:
     column_name: latent_fp_scaled
     metrics:
     - manhattan_distance
-    
+
 - ModelInputType:
       # Name of the column containing the variance
     column_name: variance
@@ -75,4 +75,4 @@ display_outputs: true
 # Number of bootstrapping samples to run. Note: bootstrapping to determine the best UQ metric is ALWAYS run unless the private attribute `Pipeline._bootstrap` is set to False.
 n_bootstrap: 100
 # Logging messages levels. If True, logger will output DEBUG level messages.
-verbose: false
+verbose: false
diff --git a/example_usage/ADME_public/unique_config_files/rLM LogCLint_unique_config_file.yaml b/example_usage/ADME_public/unique_config_files/rLM LogCLint_unique_config_file.yaml
@@ -75,4 +75,4 @@ display_outputs: true
 # Number of bootstrapping samples to run. Note: bootstrapping to determine the best UQ metric is ALWAYS run unless the private attribute `Pipeline._bootstrap` is set to False.
 n_bootstrap: 100
 # Logging messages levels. If True, logger will output DEBUG level messages.
-verbose: false
+verbose: false
diff --git a/unique/evaluation/base.py b/unique/evaluation/base.py
@@ -302,9 +302,7 @@ def __post_init__(self):
         )
 
         if self.perf_metric is not None:
-            assert (
-                self.perf_metric in self.supported_performance_metrics
-            ), f"""
+            assert self.perf_metric in self.supported_performance_metrics, f"""
                 Unrecognized evaluation metric. Supported evaluation metrics:
                 {self.supported_performance_metrics}.
                 Got: {self.perf_metric}.
@@ -594,9 +592,7 @@ class CalibrationBasedEvaluation(UniqueUncertaintyEvaluation):
     def __post_init__(self):
         super().__post_init__()
 
-        assert (
-            self.problem_type != "classification"
-        ), """
+        assert self.problem_type != "classification", """
             Calibration-based evaluation metrics for classification problems have not
             been implemented yet.
         """
diff --git a/unique/evaluation/evaluation_metrics.py b/unique/evaluation/evaluation_metrics.py
@@ -121,9 +121,7 @@ def get_indices_bin(
         A list of arrays containing each bin's indices.
     """
     ORDERS = ("increasing", "decreasing", "within")
-    assert (
-        order in ORDERS
-    ), f"""
+    assert order in ORDERS, f"""
         Unsupported ordering criteria. Supported values: {ORDERS}.
         Got: order={order}.
     """
@@ -207,9 +205,7 @@ def get_performance_per_bin(
     """
 
     if average:
-        assert (
-            problem_type == "regression"
-        ), f"""
+        assert problem_type == "regression", f"""
             Returning the average value for the performance metrics is only supported
             for regression problems (``problem_type="regression"``).
             Got problem_type={problem_type}.
diff --git a/unique/input_type/base.py b/unique/input_type/base.py
@@ -130,11 +130,9 @@ def initialize_metrics(self):
             )  # List[str]
             for metric in self.metrics:
                 if metric not in self.supported_metrics.keys():
-                    raise ValueError(
-                        f"""Please, provide a valid (list of) metric(s).
+                    raise ValueError(f"""Please, provide a valid (list of) metric(s).
                         Supported metrics: {list(self.supported_metrics.keys())}.
-                        Got: "{metric}"."""
-                    )
+                        Got: "{metric}".""")
                 metrics_to_compute[metric] = self.supported_metrics[metric]
         else:
             # If no metrics are specified, compute all supported metrics
diff --git a/unique/uncertainty/data_based/distance_to_training.py b/unique/uncertainty/data_based/distance_to_training.py
@@ -56,9 +56,7 @@ class DistanceToTraining(DataBasedUncertaintyMetric):
 
     def __post_init__(self):
         super().__post_init__()
-        assert (
-            self.distance is not None
-        ), """
+        assert self.distance is not None, """
             Please, specify a distance metric to compute the k-nearest
             neighbors to each sample.
         """
diff --git a/unique/uncertainty/data_based/kernel_density.py b/unique/uncertainty/data_based/kernel_density.py
@@ -57,14 +57,10 @@ class KernelDensityEstimation(DataBasedUncertaintyMetric):
 
     def __post_init__(self):
         super().__post_init__()
-        assert (
-            self.kernel is not None
-        ), """
+        assert self.kernel is not None, """
             Please, specify a kernel to use for the KDE computation.
         """
-        assert (
-            self.distance is not None
-        ), """
+        assert self.distance is not None, """
             Please, specify a distance metric to use for the KDE computation.
         """
 
diff --git a/unique/utils/error_model_utils.py b/unique/utils/error_model_utils.py
@@ -44,9 +44,7 @@ def compute_error(
 ) -> np.ndarray:
     """Compute the specified error type given the predictions and the corresponding true labels."""
 
-    assert (
-        error_type in SUPPORTED_ERROR_TYPES
-    ), f"""
+    assert error_type in SUPPORTED_ERROR_TYPES, f"""
         Supported error types: {SUPPORTED_ERROR_TYPES}. Got: {error_type}.
     """
     error = labels - preds
diff --git a/unique/utils/evaluation_utils.py b/unique/utils/evaluation_utils.py
@@ -67,9 +67,7 @@ def calculate_proportions(
     y_std = np.sqrt(y_variance)
 
     strategies = ["uniform", "quantile"]
-    assert (
-        strategy in strategies
-    ), f"""
+    assert strategy in strategies, f"""
         Unsupported strategy. Supported strategies: {strategies}.
         Got: strategy={strategy}.
     """
diff --git a/unique/utils/uncertainty_utils.py b/unique/utils/uncertainty_utils.py
@@ -58,12 +58,10 @@ class AnalyticsMode:
     def __init__(self, mode: str):
         self.mode = mode.lower()
         if self.mode not in [self.COMPACT, self.EXTENDED, self.FULL]:
-            raise ValueError(
-                f"""
+            raise ValueError(f"""
                     Allowed analysis modalities: [{self.COMPACT}, {self.EXTENDED}, {self.FULL}].
                     Got: {self.mode}.
-                """
-            )
+                """)
 
         self.explaination = self.EXPLAINATION[self.mode]
 
@@ -173,9 +171,7 @@ def convert_distances_to_variances(
         A np.ndarray object containing the corresponding distances converted to variances.
     """
     # Check that calibration set exists
-    assert "CALIBRATION" in np.unique(
-        which_set
-    ), f""""
+    assert "CALIBRATION" in np.unique(which_set), f""""
         Please, provide a calibration set on which to compute the Calibrated NLL.
         You can specify it in the ``which_set`` parameter.
         Got: {np.unique(which_set)}.
