Ran precommit hooks

Author: HashirA123

data/catalog/_data_main/process_data/process_german_data.py

@@ -16,7 +16,7 @@
 def load_german_data(modified=False):
     # input vars
     this_files_directory = os.path.dirname(os.path.realpath(__file__))
-    if modified == False:
+    if modified is False:
         raw_data_file = os.path.join(
             this_files_directory, "..", "raw_data", "german_v1.csv"
         )

data/catalog/_data_main/process_data/process_sba_data.py

@@ -1,10 +1,7 @@
 import os
-from random import seed, shuffle
+from random import seed
 
-import numpy as np
 import pandas as pd
-import process_data.process_utils_data as ut
-from sklearn.preprocessing import StandardScaler
 
 RANDOM_SEED = 54321
 seed(

data/catalog/loadData.py

@@ -74,10 +74,7 @@
     print(f"[ENV WARNING] process_boston_housing_data not available. Error: {e}")
 
 try:
-    from data.catalog._data_main.process_data.process_sba_data import (
-        load_sba_data,
-        load_sba_data_modified,
-    )
+    from data.catalog._data_main.process_data.process_sba_data import load_sba_data
 except Exception as e:
     print(f"[ENV WARNING] process_sba_data not available. Error: {e}")
 

experiments/run_experiment.py

@@ -284,7 +284,7 @@ def create_parser():
             "mace",
             "revise",
             "wachter",
-            'roar',
+            "roar",
         ],
         help="Recourse methods for experiment",
     )

methods/catalog/roar/model.py

@@ -203,7 +203,9 @@ def get_counterfactuals(self, factuals: pd.DataFrame) -> pd.DataFrame:
 
                 coeffs_neg = (
                     # self._mlmodel.raw_model.output.weight.cpu().detach()[0].numpy()
-                    self._mlmodel.raw_model.linear.weight.cpu().detach()[0].numpy()
+                    self._mlmodel.raw_model.linear.weight.cpu()
+                    .detach()[0]
+                    .numpy()
                 )
                 coeffs_pos = (
                     self._mlmodel.raw_model.linear.weight.cpu().detach()[1].numpy()

methods/catalog/roar/reproduce.py

@@ -1,7 +1,4 @@
-from pathlib import Path
-
 import numpy as np
-import pandas as pd
 import pytest
 
 from data.catalog.online_catalog import DataCatalog
@@ -10,6 +7,7 @@
 
 RANDOM_SEED = 54321
 
+
 # Find indices where recourse is needed
 def recourse_needed(predict_fn, X, target=1):
     return np.where(predict_fn(X) == 1 - target)[0]
@@ -86,17 +84,17 @@ def test_roar(dataset_name, model_type, backend):
     m2._test_accuracy()
 
     print("Using %s cost" % args["cost"])
-    if args["cost"] == "l1":
-        feature_costs = None
+    # if args["cost"] == "l1":
+    #     feature_costs = None
 
     coefficients = intercept = None
 
     roar = Roar(mlmodel=m1, hyperparams={}, coeffs=coefficients, intercepts=intercept)
 
-    lamb = args["lamb"]
+    # lamb = args["lamb"]
 
     recourses = []
-    deltas = []
+    # deltas = []
 
     factuals = (data._df_test).sample(n=10, random_state=RANDOM_SEED)