Fixed some pre-commit hook issues

.flake8

@@ -6,5 +6,7 @@ extend-exclude =
     .venv
     build
 extend-ignore =
-    E203  # No whitespace before ':' in [x : y]
-    E731  # No lambdas — too strict
+    # No whitespace before ':' in [x : y]
+    E203,
+    # No lambdas — too strict
+    E731,  

CHANGELOG.md

@@ -1,3 +1,8 @@
+# Version 1.1.4
+
+## Bug-Fixes
+-
+
 # Version 1.1.3
 
 ## Bug-Fixes

deepcave/evaluators/epm/random_forest.py

@@ -432,10 +432,9 @@ def predict_marginalized(self, X: np.ndarray) -> Tuple[np.ndarray, np.ndarray]:
         # marginalized predictions for each tree
         dat_ = np.zeros((X.shape[0], self._model_options.num_trees))
         for i, x in enumerate(X):
-
             # marginalize over instances
             # 1. get all leaf values for each tree
-            # type: list[list[float]]
+            # type= list[list[float]]
             preds_trees = [[] for i in range(self._model_options.num_trees)]
 
             for feat in self.instance_features:

deepcave/layouts/sidebar.py

@@ -43,7 +43,7 @@ def update_navigation_items(pathname):  # type: ignore
                 ]
 
                 point_layouts = []
-                for (id, name, icon) in points:
+                for id, name, icon in points:
                     href = f"/plugins/{id}"
                     point_layouts += [
                         html.Li(
@@ -106,7 +106,6 @@ def update_queue_info() -> List[Component]:
 
                 collect = []
                 for jobs, status in zip(all_jobs, job_stati):
-
                     for job in jobs:
                         name = job.meta["display_name"]
                         job_id = job.id

deepcave/plugins/__init__.py

@@ -314,7 +314,7 @@ def plugin_input_update(pathname: str, *inputs_list: str) -> List[str]:
                             update_dict(inputs, new_inputs)
 
                         # Set not used inputs
-                        for (id, attribute, _, _) in self.inputs:
+                        for id, attribute, _, _ in self.inputs:
                             if id not in inputs:
                                 inputs[id] = {}
 
@@ -421,7 +421,7 @@ def toggle_help_modal(n: Optional[int], is_open: bool) -> Tuple[bool, str]:
             return is_open
 
         # Register callback to click on configurations
-        for (id, *_) in self.outputs:
+        for id, *_ in self.outputs:
             internal_id = self.get_internal_output_id(id)
 
             @app.callback(
@@ -470,8 +470,7 @@ def _inputs_changed(
         # If only filters changed, then we don't need to
         # calculate the results again.
         if last_inputs is not None:
-            for (id, attribute, filter, _) in self.inputs:
-
+            for id, attribute, filter, _ in self.inputs:
                 if self.activate_run_selection:
                     if id == "run":
                         continue
@@ -524,7 +523,7 @@ def _process_raw_outputs(
         # We have to add no_updates here for the mode we don't want
         count_outputs = 0
         count_mpl_outputs = 0
-        for (_, _, mpl_mode) in self.outputs:
+        for _, _, mpl_mode in self.outputs:
             if mpl_mode:
                 count_mpl_outputs += 1
             else:
@@ -600,7 +599,7 @@ def _dict_to_list(
             order = self.outputs  # type: ignore
 
         result: List[Optional[str]] = []
-        for (id, attribute, instance, *_) in order:
+        for id, attribute, instance, *_ in order:
             if not input:
                 # Instance is mlp_mode in case of outputs
                 # Simply ignore other outputs.
@@ -637,7 +636,7 @@ def _dict_as_key(self, d: Dict[str, Any], remove_filters: bool = False) -> Optio
 
         new_d = copy.deepcopy(d)
         if remove_filters:
-            for (id, _, filter, _) in self.inputs:
+            for id, _, filter, _ in self.inputs:
                 if filter:
                     if id in new_d:
                         del new_d[id]
@@ -662,10 +661,9 @@ def _cast_inputs(self, inputs: Dict[str, Dict[str, str]]) -> Dict[str, Dict[str,
         casted_inputs: Dict[str, Dict[str, str]] = defaultdict(dict)
         for id, attributes in inputs.items():
             for attribute in attributes:
-
                 # Find corresponding input
                 type = None
-                for (id_, attribute_, _, type_) in self.inputs:
+                for id_, attribute_, _, type_ in self.inputs:
                     if id == id_ and attribute == attribute_:
                         type = type_
                         break
@@ -702,7 +700,7 @@ def _clean_inputs(self, inputs: Dict[str, Any]) -> Dict[str, Any]:
         """
         used_ids = []
         cleaned_inputs = {}
-        for (id, attribute, *_) in self.inputs:
+        for id, attribute, *_ in self.inputs:
             # Since self.inputs is ordered, we use the first occuring attribute and add
             # the id so it is not used again.
             if id not in used_ids:
@@ -1332,7 +1330,7 @@ def generate_inputs(self, **kwargs: Any) -> Dict[str, Any]:
             The inputs for the run.
         """
         mapping = {}
-        for (id, attribute, *_) in self.inputs:
+        for id, attribute, *_ in self.inputs:
             # Since `self.inputs` is ordered, we use the first occuring attribute and add
             # the id so it is not used again.
             if id not in mapping:

deepcave/plugins/objective/cost_over_time.py

@@ -32,12 +32,14 @@ def check_runs_compatibility(self, runs: List[AbstractRun]) -> None:
             if run_inequality == RunInequality.INEQ_BUDGET:
                 notification.update("The budgets of the runs are not equal.", color="warning")
             elif run_inequality == RunInequality.INEQ_CONFIGSPACE:
-                notification.update("The configuration spaces of the runs are not equal.", color="warning")
+                notification.update(
+                    "The configuration spaces of the runs are not equal.", color="warning"
+                )
             elif run_inequality == RunInequality.INEQ_META:
                 notification.update("The meta data of the runs is not equal.", color="warning")
             elif run_inequality == RunInequality.INEQ_OBJECTIVE:
                 notification.update("The objectives of the runs are not equal.", color="warning")
-        
+
         # Set some attributes here
         run = runs[0]
 

deepcave/plugins/objective/pareto_front.py

@@ -32,7 +32,9 @@ def check_runs_compatibility(self, runs):
             if run_inequality == RunInequality.INEQ_BUDGET:
                 notification.update("The budgets of the runs are not equal.", color="warning")
             elif run_inequality == RunInequality.INEQ_CONFIGSPACE:
-                notification.update("The configuration spaces of the runs are not equal.", color="warning")
+                notification.update(
+                    "The configuration spaces of the runs are not equal.", color="warning"
+                )
             elif run_inequality == RunInequality.INEQ_META:
                 notification.update("The meta data of the runs is not equal.", color="warning")
             elif run_inequality == RunInequality.INEQ_OBJECTIVE:
@@ -186,7 +188,6 @@ def process(run, inputs):
 
         is_front: Union[List, np.ndarray] = np.ones(points.shape[0], dtype=bool)
         for point_idx, costs in enumerate(points):
-
             if is_front[point_idx]:
                 # Keep any point with a lower/upper cost
                 # This loop is a little bit complicated than

deepcave/plugins/summary/overview.py

@@ -291,7 +291,6 @@ def load_outputs(run, *_):
         }
 
         for hp_name, hp in run.configspace.get_hyperparameters_dict().items():
-
             log = False
             value = None
             if (

deepcave/runs/__init__.py

@@ -939,7 +939,9 @@ def check_equality(
                     continue
 
                 if k not in m2 or m2[k] != v:
-                    raise NotMergeableError("Meta data of runs are not equal.", RunInequality.INEQ_META)
+                    raise NotMergeableError(
+                        "Meta data of runs are not equal.", RunInequality.INEQ_META
+                    )
 
         result["meta"] = m1
 
@@ -951,7 +953,9 @@ def check_equality(
         for run in runs:
             cs2 = run.configspace
             if cs1 != cs2:
-                raise NotMergeableError("Configspace of runs are not equal.", RunInequality.INEQ_CONFIGSPACE)
+                raise NotMergeableError(
+                    "Configspace of runs are not equal.", RunInequality.INEQ_CONFIGSPACE
+                )
 
         result["configspace"] = cs1
 
@@ -978,7 +982,9 @@ def check_equality(
                 continue
 
             if len(o1) != len(o2):
-                raise NotMergeableError("Objectives of runs are not equal.", RunInequality.INEQ_OBJECTIVE)
+                raise NotMergeableError(
+                    "Objectives of runs are not equal.", RunInequality.INEQ_OBJECTIVE
+                )
 
             for o1_, o2_ in zip(o1, o2):
                 o1_.merge(o2_)
@@ -988,4 +994,4 @@ def check_equality(
         if meta:
             result["meta"]["objectives"] = serialized_objectives
 
-    return result
+    return result

deepcave/runs/converters/bohb.py

@@ -49,7 +49,6 @@ def from_path(cls, path):
 
         first_starttime = None
         for bohb_run in bohb.get_all_runs():
-
             times = bohb_run.time_stamps
             starttime = times["started"]
             endtime = times["finished"]

deepcave/runs/exceptions.py

@@ -1,4 +1,6 @@
 from enum import Enum
+
+
 class NotValidRunError(Exception):
     """Raised if directory is not a valid run."""
 
@@ -10,12 +12,11 @@ class NotMergeableError(Exception):
 
     pass
 
+
 class RunInequality(Enum):
     """Check why runs were not compatible."""
+
     INEQ_META = 1
     INEQ_OBJECTIVE = 2
     INEQ_BUDGET = 3
     INEQ_CONFIGSPACE = 4
-
-
-

examples/record/mnist_pytorch.py

@@ -71,7 +71,6 @@ def prepare_data(self):
         MNIST(self.data_dir, train=False, download=True)
 
     def setup(self, stage=None):
-
         # Assign train/val datasets for use in dataloaders
         if stage == "fit" or stage is None:
             mnist_full = MNIST(self.data_dir, train=True, transform=self.transform)