fix codestyle

Author: voorhs

autointent/context/context.py

@@ -90,7 +90,9 @@ def set_dataset(self, dataset: Dataset, force_multilabel: bool = False) -> None:
         :param force_multilabel: Whether to force multilabel classification.
         """
         self.data_handler = DataHandler(
-            dataset=dataset, force_multilabel=force_multilabel, random_seed=self.seed,
+            dataset=dataset,
+            force_multilabel=force_multilabel,
+            random_seed=self.seed,
         )
 
     def get_best_index(self) -> VectorIndex:

autointent/context/data_handler/dataset.py

@@ -29,7 +29,7 @@ class Dataset(dict[str, HFDataset]):
     label_feature = "label"
     utterance_feature = "utterance"
 
-    def __init__(self, *args: Any, intents: list[Intent], **kwargs: Any) -> None:
+    def __init__(self, *args: Any, intents: list[Intent], **kwargs: Any) -> None:  # noqa: ANN401
         super().__init__(*args, **kwargs)
 
         self.intents = intents
@@ -51,11 +51,13 @@ def n_classes(self) -> int:
     @classmethod
     def from_json(cls, filepath: str | Path) -> "Dataset":
         from .reader import JsonReader
+
         return JsonReader().read(filepath)
 
     @classmethod
     def from_dict(cls, mapping: dict[str, Any]) -> "Dataset":
         from .reader import DictReader
+
         return DictReader().read(mapping)
 
     @classmethod
@@ -82,7 +84,7 @@ def to_multilabel(self) -> Self:
             self[split_name] = split.map(self._to_multilabel)
         return self
 
-    def push_to_hub(self, repo_id: str)-> None:
+    def push_to_hub(self, repo_id: str) -> None:
         for split_name, split in self.items():
             split.push_to_hub(repo_id, split=split_name)
 
@@ -95,10 +97,7 @@ def get_tags(self) -> list[Tag]:
         for intent in self.intents:
             for tag in intent.tags:
                 tag_mapping[tag].append(intent.id)
-        return [
-            Tag(name=tag, intent_ids=intent_ids)
-            for tag, intent_ids in tag_mapping.items()
-        ]
+        return [Tag(name=tag, intent_ids=intent_ids) for tag, intent_ids in tag_mapping.items()]
 
     def get_n_classes(self, split: str) -> int:
         classes = set()

autointent/context/data_handler/multilabel_generation.py

@@ -10,7 +10,7 @@
 
 
 class BaseReader(ABC):
-    def read(self, *args: Any, **kwargs: Any) -> Dataset:
+    def read(self, *args: Any, **kwargs: Any) -> Dataset:  # noqa: ANN401
         dataset_reader = DatasetValidator.validate(self._read(*args, **kwargs))
         splits = dataset_reader.model_dump(exclude={"intents"}, exclude_defaults=True)
         return Dataset(
@@ -19,8 +19,7 @@ def read(self, *args: Any, **kwargs: Any) -> Dataset:
         )
 
     @abstractmethod
-    def _read(self, *args: Any, **kwargs: Any) -> DatasetReader:
-        ...
+    def _read(self, *args: Any, **kwargs: Any) -> DatasetReader: ...  # noqa: ANN401
 
 
 class DictReader(BaseReader):

autointent/context/data_handler/reader.py

@@ -37,8 +37,7 @@ def _validate_label(self) -> Self:
         label = [self.label] if isinstance(self.label, int) else self.label
         if not label:
             message = (
-                "The `label` field cannot be empty for a multilabel sample. "
-                "Please provide at least one valid label."
+                "The `label` field cannot be empty for a multilabel sample. " "Please provide at least one valid label."
             )
             raise ValueError(message)
         if any(label_ < 0 for label_ in label):

autointent/context/data_handler/sampling.py

@@ -25,7 +25,11 @@ class StratifiedSplitter:
     """
 
     def __init__(
-        self, test_size: float, label_feature: str, random_seed: int, shuffle: bool = True,
+        self,
+        test_size: float,
+        label_feature: str,
+        random_seed: int,
+        shuffle: bool = True,
     ) -> None:
         """
         Initialize the StratifiedSplitter.
@@ -52,7 +56,7 @@ def __call__(self, dataset: HFDataset, multilabel: bool) -> tuple[Dataset, Datas
         return dataset.select(splits[0]), dataset.select(splits[1])
 
     def _split(self, dataset: HFDataset) -> Sequence[npt.NDArray[np.int_]]:
-        return train_test_split( # type: ignore[no-any-return]
+        return train_test_split(  # type: ignore[no-any-return]
             np.arange(len(dataset)),
             test_size=self.test_size,
             random_state=self.random_seed,
@@ -81,7 +85,9 @@ def split_dataset(dataset: Dataset, random_seed: int) -> Dataset:
     :return: The input dataset with training and testing splits.
     """
     splitter = StratifiedSplitter(
-        test_size=0.25, label_feature=dataset.label_feature, random_seed=random_seed,
+        test_size=0.25,
+        label_feature=dataset.label_feature,
+        random_seed=random_seed,
     )
     dataset[Split.TRAIN], dataset[Split.TEST] = splitter(dataset[Split.TRAIN], dataset.multilabel)
     return dataset

autointent/context/data_handler/schemas.py

@@ -7,7 +7,7 @@
 from collections import defaultdict
 from dataclasses import dataclass, field
 
-from .schemas import Dataset
+from .dataset import Dataset
 
 
 @dataclass

autointent/context/data_handler/stratification.py

@@ -36,8 +36,7 @@ def _validate_split(self, split: list[Sample]) -> Self:
         intent_ids = {intent.id for intent in self.intents}
         for sample in split:
             message = (
-                f"Sample with label {sample.label} references a non-existent intent ID. "
-                f"Valid IDs are {intent_ids}."
+                f"Sample with label {sample.label} references a non-existent intent ID. " f"Valid IDs are {intent_ids}."
             )
             if isinstance(sample.label, int) and sample.label not in intent_ids:
                 raise ValueError(message)

autointent/context/data_handler/tags.py

@@ -74,7 +74,8 @@ def delete(self) -> None:
         """Delete the embedding model and its associated directory."""
         self.clear_ram()
         shutil.rmtree(
-            self.dump_dir, ignore_errors=True,
+            self.dump_dir,
+            ignore_errors=True,
         )  # TODO: `ignore_errors=True` is workaround for PermissionError: [WinError 5] Access is denied
 
     def dump(self, path: Path) -> None: