refactor: python lint (new black version)

demo/cats/code_submission/sanitize.py

@@ -4,7 +4,6 @@
 
 
 class Sanitize:
-
     S_CODE = 0
     S_COMMENT = 1
     SANITIZE_WORDS = [

demo/cats/utility/acp_regression_data_join.py

@@ -4,11 +4,10 @@
 
 class PredictDataJoiner_acp:
     def join(self, data_file_name, predict_file_name):
-
         data_file = open(data_file_name, "r")
         predict_file = open(predict_file_name, "r")
 
-        for (data_line, predict_line) in zip(data_file, predict_file):
+        for data_line, predict_line in zip(data_file, predict_file):
             data_line = self.strip_label(data_line)
             print("ca", predict_line.strip(), "|", data_line.strip())
 

demo/cats/utility/ap_regression_data_join.py

@@ -21,7 +21,6 @@ def __init__(
         self.loss_type_return = loss_type_return
 
     def join(self):
-
         data_file = open(self.data_file_name, "r")
         predict_file = open(self.predict_file_name, "r")
 
@@ -35,7 +34,7 @@ def join(self):
         sqr_loss_max = float("-inf")
         max_found = float("-inf")
         min_found = float("inf")
-        for (data_line, predict_line) in zip(data_file, predict_file):
+        for data_line, predict_line in zip(data_file, predict_file):
             # Get data
             act = self.get_action(predict_line)
             reg = self.get_regression_val(data_line)

demo/cats/utility/get_offline_results.py

@@ -47,7 +47,6 @@ def __init__(self, srm_file_name, test_file_name, delta, alpha, quiet):
         self.pmin = 0.05  # epsilon = 5%
 
     def eval(self):
-
         data_file = open(self.srm_file_name, "r")
         line = data_file.readline()
 

demo/cats/utility/get_online_results.py

@@ -40,7 +40,6 @@ def __init__(self, file_name, alpha, quiet):
         self.quiet = quiet
 
     def eval(self):
-
         data_file = open(self.file_name, "r")
         line = data_file.readline()
 

demo/cats/utility/srm.py

@@ -12,14 +12,13 @@ def __init__(self, data_file_name, predict_file_name, min_val, max_val, k, h):
         self.hh = h * (max_val - min_val) / k
 
     def calc(self):
-
         data_file = open(self.data_file_name, "r")
         predict_file = open(self.predict_file_name, "r")
 
         N = 0
         _loss_acc = 0.0
 
-        for (data_line, predict_line) in zip(data_file, predict_file):
+        for data_line, predict_line in zip(data_file, predict_file):
             # Get data
             pred = self.get_predict_action(predict_line)
             logd, L_s, P_s = self.get_logged_data(data_line)

demo/recall_tree/wikipara/WikiExtractor.py

@@ -2163,7 +2163,6 @@ def compact(text):
     listLevel = ""  # nesting of lists
 
     for line in text.split("\n"):
-
         if not line:
             continue
         # Handle section titles
@@ -2235,7 +2234,7 @@ def compact(text):
             if not Extractor.keepSections:
                 items = headers.items()
                 items.sort()
-                for (i, v) in items:
+                for i, v in items:
                     page.append(v)
             headers.clear()
             page.append(line)  # first line

python/docs/source/conf.py

@@ -149,7 +149,6 @@ def add_binder_url_for_page(
     context,
     doctree,
 ):
-
     # First decide if we'll insert any links
     path = app.env.doc2path(pagename)
     extension = Path(path).suffix

python/docs/source/tutorials/python_cats.ipynb

@@ -229,7 +229,6 @@
     "    max_value,\n",
     "    do_learn=True,\n",
     "):\n",
-    "\n",
     "    reward_rate = []\n",
     "    hits = 0\n",
     "    cost_sum = 0.0\n",
@@ -616,7 +615,6 @@
     "    max_value,\n",
     "    do_learn=True,\n",
     "):\n",
-    "\n",
     "    reward_rate = []\n",
     "    hits = 0\n",
     "    cost_sum = 0.0\n",

python/tests/test_ccb.py

@@ -1,5 +1,6 @@
 import vowpalwabbit
 
+
 # Named specifically as the delimiter used is specific for the number of actions
 # used in this test case.
 def count_weights_from_readable_model_file_for_equiv_test(file_name):

python/tests/test_distributionally_robust.py

@@ -26,7 +26,7 @@ def test_recompute_duals_lower():
 
     np.testing.assert_almost_equal(list(zip(ws, rs)), ws_rs, 5)
 
-    for (w, r) in zip(ws, rs):
+    for w, r in zip(ws, rs):
         ocrl.update(1, w, r)
         ocrl.recomputeduals()
         if ocrl.duals[1][0] is None:
@@ -105,7 +105,7 @@ def test_recompute_duals_upper():
     duals = []
     bounds = []
 
-    for (w, r) in zip(ws, rs):
+    for w, r in zip(ws, rs):
         ocrl.update(1, w, r)
         ocrl.recomputeduals(is_upper=True)
         if ocrl.duals[1][0] is None:
@@ -159,7 +159,7 @@ def test_qlb():
 
     np.testing.assert_almost_equal(list(zip(ws, rs)), ws_rs, 5)
 
-    for (w, r) in zip(ws, rs):
+    for w, r in zip(ws, rs):
         ocrl.update(1, w, r)
         qlbs.append(ocrl.qlb(w, r))
 

python/tests/test_pyvw.py

@@ -9,14 +9,14 @@
 
 BIT_SIZE = 18
 
+
 # Since these tests still run with Python 2, this is required.
 # Otherwise we could use math.isclose
 def isclose(a, b, rel_tol=1e-05, abs_tol=0.0):
     return abs(a - b) <= max(rel_tol * max(abs(a), abs(b)), abs_tol)
 
 
 class TestVW:
-
     model = Workspace(quiet=True, b=BIT_SIZE)
 
     def test_constructor(self):

python/tests/test_sklearn_vw.py

@@ -303,7 +303,6 @@ def test_repr(self):
 
 
 class TestVWMultiClassifier(BaseVWTest):
-
     estimator = VWMultiClassifier
 
     def test_predict_proba(self, data):

python/tests/test_vwconfig.py

@@ -6,7 +6,7 @@ def helper_options_to_list_strings(config):
     cmd_str_list = []
 
     for name, config_group in config.items():
-        for (group_name, options) in config_group:
+        for group_name, options in config_group:
             for option in options:
                 temp_str = str(option)
                 if temp_str:

python/vowpalwabbit/dftovw.py

@@ -302,7 +302,7 @@ def process(self, df: pd.DataFrame) -> pd.Series:
             The MultiLabel string representation.
         """
         labels = self.label if isinstance(self.label, list) else [self.label]
-        for (i, label) in enumerate(labels):
+        for i, label in enumerate(labels):
             label_col = label.get_col(df)
             if i == 0:
                 out = label_col
@@ -542,7 +542,6 @@ class _ListLabel(object):
     sep_by_label = dict(ContextualbanditLabel=" ", MultiLabel=",")
 
     def __init__(self, label_list: List[Union[ContextualbanditLabel, MultiLabel]]):
-
         instance_classes = set(
             [type(label_instance).__name__ for label_instance in label_list]
         )
@@ -605,7 +604,7 @@ def process(self, df: pd.DataFrame) -> pd.Series:
         Returns:
             The _ListLabel string representation.
         """
-        for (i, label) in enumerate(self):
+        for i, label in enumerate(self):
             if i == 0:
                 out = label.process(df)
             else:
@@ -988,7 +987,6 @@ def check_instance_columns(self, instance):
         class_name = type(instance).__name__
 
         for attribute_name, attribute_value in vars(instance).items():
-
             if not isinstance(attribute_value, list):
                 attribute_value = [attribute_value]
 
@@ -1032,7 +1030,7 @@ def convert_df(self) -> List[str]:
         if not all([x is None for x in [self.label, self.tag]]):
             self.out += self.process_label_and_tag()
 
-        for (i, namespace) in enumerate(self.namespaces):
+        for i, namespace in enumerate(self.namespaces):
             to_add = namespace.process() + self.process_features(namespace.features)
             self.out += (to_add + " ") if (i < len(self.namespaces) - 1) else to_add
 

python/vowpalwabbit/pyvw.py

@@ -1931,7 +1931,6 @@ def get_prediction(
         Tuple[int, List[int]],
         None,
     ]:
-
         """Get prediction object from this example.
 
         Args:

setup.py

@@ -63,7 +63,6 @@ def run(self):
         _build_ext.run(self)
 
     def build_cmake(self, ext):
-
         # Make build directory
         distutils.dir_util.mkpath(self.build_temp)
 

test/run_tests.py

@@ -385,7 +385,6 @@ def run_command_line_test(
     valgrind=False,
     timeout=100,
 ) -> TestOutcome:
-
     if test.skip:
         completed_tests.report_completion(test.id, False)
         return TestOutcome(test.id, Result.SKIPPED, {}, skip_reason=test.skip_reason)

test/runtest_coverage.py

@@ -21,7 +21,7 @@ def get_all_options():
 def to_json():
     config = get_all_options()
     for name, config_group in config.items():
-        for (group_name, options) in config_group:
+        for group_name, options in config_group:
             for option in options:
                 option._type = str(type(option._default_value).__name__)
 
@@ -40,7 +40,7 @@ def get_config_of_vw_cmd(test):
 
 
 def update_option(config, name, group_name, option_name):
-    for (g_n, options) in config[name]:
+    for g_n, options in config[name]:
         if g_n == group_name:
             for option in options:
                 if option.name == option_name:
@@ -51,7 +51,7 @@ def update_option(config, name, group_name, option_name):
 
 def merge_config(tracker, b):
     for name, config_group in b.items():
-        for (group_name, options) in config_group:
+        for group_name, options in config_group:
             for option in options:
                 if option.value_supplied:
                     tracker = update_option(tracker, name, group_name, option.name)
@@ -64,7 +64,7 @@ def print_non_supplied(config):
     without_default = []
 
     for name, config_group in config.items():
-        for (group_name, options) in config_group:
+        for group_name, options in config_group:
             for option in options:
                 if not option.value_supplied:
                     default_val_str = ""

test/tools/fuzzing/fuzz.py

@@ -11,6 +11,7 @@
 
 # Python 3.8+ required
 
+
 # The code used to find the test binaries are taken from run_tests.py. If any changes occur
 # in one, make sure to update the other
 def find_in_path(paths, file_matcher, debug_file_name):
@@ -71,7 +72,6 @@ def is_test_binary(file_path):
 
 
 def main():
-
     parser = argparse.ArgumentParser(
         formatter_class=argparse.ArgumentDefaultsHelpFormatter
     )

utl/active_interactor.py

@@ -63,7 +63,6 @@ def _getch():
 # The MacOS Carbon API was removed in 10.15, however the
 # standard unix approach should work and so should be tried first.
 def _get_getch_impl_macos() -> Optional[Callable[[], str]]:
-
     try:
         # try OS/X API
         import Carbon  # type: ignore

utl/vw-hyperopt.py

@@ -318,7 +318,6 @@ def _configure_logger(self):
         return logger
 
     def get_hyperparam_string(self, **kwargs):
-
         # print 'KWARGS: ', kwargs
         flags = [key for key in kwargs if key.startswith("-")]
         for flag in flags: