Automated autopep8 fixes

benchmark/automl_forecasting.py

@@ -443,13 +443,14 @@
 # Parameters
 # full_file_path_and_name - complete .tsf file path
 # replace_missing_vals_with - a term to indicate the missing values in series in the returning dataframe
-# value_column_name - Any name that is preferred to have as the name of the column containing series values in the returning dataframe
+# value_column_name - Any name that is preferred to have as the name of
+# the column containing series values in the returning dataframe
 def convert_tsf_to_dataframe(full_file_path_and_name, replace_missing_vals_with='NaN',
                              value_column_name='series_value'):
     try:
         with open(full_file_path_and_name, 'r', encoding='utf-8') as file:
             return parse_file(file, replace_missing_vals_with, value_column_name)
-    except:
+    except BaseException:
         with open(full_file_path_and_name, 'r', encoding='cp1252') as file:
             return parse_file(file, replace_missing_vals_with, value_column_name)
 
@@ -564,9 +565,8 @@ def parse_file(file, replace_missing_vals_with, value_column_name):
                                 full_info[i], '%Y-%m-%d %H-%M-%S'
                             )
                         else:
-                            raise Exception(
-                                'Invalid attribute type.'
-                            )  # Currently, the code supports only numeric, string and date types. Extend this as required.
+                            # Currently, the code supports only numeric, string and date types. Extend this as required.
+                            raise Exception('Invalid attribute type.')
 
                         if att_val is None:
                             raise Exception('Invalid attribute value.')
@@ -672,8 +672,8 @@ def regression_scores(actual, predicted, y_train,
         if 'duration' in kwargs.keys():
             results['duration'] = kwargs['duration']
 
-        if scores_dir != None:
-            if forecaster_name == None:
+        if scores_dir is not None:
+            if forecaster_name is None:
                 raise TypeError('Forecaster name required to save scores')
             os.makedirs(scores_dir, exist_ok=True)
 
@@ -732,7 +732,7 @@ def correlation(actual, predicted, method='pearson'):
         try:
             correlation = result.correlation
             pvalue = result.pvalue
-        except:  # older scipy versions returned a tuple instead of an object
+        except BaseException:  # older scipy versions returned a tuple instead of an object
             correlation = result[0]
             pvalue = result[1]
 
@@ -779,12 +779,12 @@ def write_to_csv(path, results):
                 HEADERS.insert(0, HEADERS.pop(HEADERS.index('model')))
 
             for key, value in results.items():
-                if value == None or value == '':
+                if value is None or value == '':
                     results[key] = 'None'
 
             try:
                 Utils._write_to_csv(path, results, HEADERS)
-            except OSError as _:
+            except OSError:
                 # try a second time: permission error can be due to Python not
                 # having closed the file fast enough after the previous write
                 time.sleep(1)  # in seconds
@@ -847,25 +847,25 @@ def save_plot(title,
         :param yscale: Y-Scale ('linear' or 'log'), defaults to 'linear'
         """
 
-        if xlabel != None:
+        if xlabel is not None:
             plt.xlabel(xlabel)
 
-        if ylabel != None:
+        if ylabel is not None:
             plt.ylabel(ylabel)
 
         plt.yscale(yscale)
 
         plt.title(title)
         plt.suptitle(suptitle)
 
-        if legend != None:
+        if legend is not None:
             plt.legend(legend, loc='upper left')
 
         # Show plot
         if show:
             plt.show()
         # Show plot as file
-        if save_path != None:
+        if save_path is not None:
             plt.savefig(save_path, bbox_inches='tight')
 
         # Clear for next plot
@@ -910,7 +910,7 @@ def split_test_set(test_df, horizon):
         for _ in range(0, len(test_df) - 1, horizon):  # The -1 is because the last split may be less than horizon
             try:
                 test_splits.append(test_df.iloc[total:total + horizon, :])
-            except:  # If 1D (series)
+            except BaseException:  # If 1D (series)
                 test_splits.append(test_df.iloc[total:total + horizon])
             total += horizon
 
@@ -1160,11 +1160,11 @@ def save_heatmap(df, csv_path, png_path):
         heatmap.to_csv(csv_path)  # Save correlations as CSV
         heatmap.to_latex(csv_path.replace('.csv', '.tex'))  # Save correlations as .tex
         try:
-            calculate_pvalues = lambda x, y: pearsonr(x, y).pvalue
+            def calculate_pvalues(x, y): return pearsonr(x, y).pvalue
             df[columns].corr(method=calculate_pvalues).to_csv(csv_path.replace('.csv', '_pvalues.csv'))
         # older scipy versions return a tuple instead of an object
-        except:
-            calculate_pvalues = lambda x, y: pearsonr(x, y)[1]
+        except BaseException:
+            def calculate_pvalues(x, y): return pearsonr(x, y)[1]
             df[columns].corr(method=calculate_pvalues).to_csv(csv_path.replace('.csv', '_pvalues.csv'))
 
         # Save correlation heatmap as image

benchmark/benchmark_TSF.py

@@ -70,9 +70,9 @@ def evaluate_loop(self, dataset, experiment_setup: dict = None):
             experiment_setup['task_params'] = TsForecastingParams(
                 forecast_length=M4_FORECASTING_LENGTH[dataset[0]])
         target = train_data.iloc[-experiment_setup['task_params']
-        .forecast_length:, :].values.ravel()
+                                 .forecast_length:, :].values.ravel()
         train_data = train_data.iloc[:-
-        experiment_setup['task_params'].forecast_length, :]
+                                     experiment_setup['task_params'].forecast_length, :]
         model = FedotIndustrial(**experiment_setup)
         model.fit(train_data)
         prediction = model.predict(train_data)

benchmark/feature_utils.py

@@ -429,7 +429,7 @@ def format_univariate_forecasting_data(data_dir, return_df: bool = False):
             meta_data['step_size'] = []
         try:
             files = os.listdir(data_dir)
-        except:
+        except BaseException:
             files = os.listdir(os.path.join(PROJECT_PATH, data_dir))
 
         csv_files = [f for f in files if '0_metadata.csv' not in f and f.endswith('csv')]
@@ -510,7 +510,7 @@ def format_global_forecasting_data(data_dir, gather_metadata=False):
                 data, freq, horizon, has_nans, equal_length = convert_tsf_to_dataframe(
                     os.path.join(data_dir, tsf_file), 'NaN', 'value')
 
-                if horizon == None:
+                if horizon is None:
                     horizon = DatasetFormatting.select_horizon(freq, csv_path)
 
                 if gather_metadata:
@@ -568,7 +568,7 @@ def select_horizon(freq, csv_path):
         # The following horizons are suggested by Godahewa et al. (2021)
         elif 'solar_weekly_dataset' in csv_path:
             horizon = 5
-        elif freq == None:
+        elif freq is None:
             raise ValueError('No frequency or horizon found in file')
         elif freq == 'monthly':
             horizon = 12
@@ -684,7 +684,7 @@ def format_3W_data(data_dir):
 
         :param data_dir: Path to directory of datasets
         """
-        subdir = os.path.join(data_dir, '3W')
+        os.path.join(data_dir, '3W')
 
     @staticmethod
     def format_falling_data(data_dir):