update mypy config

Author: Samoed

autointent/_callbacks/tensorboard.py

@@ -57,16 +57,12 @@ def start_module(self, module_name: str, num: int, module_kwargs: dict[str, Any]
         for key, value in module_kwargs.items():
             self.module_writer.add_text(f"module_params/{key}", str(value))  # type: ignore[no-untyped-call]
 
-    def log_value(self, **kwargs: dict[str, Any]) -> None:
+    def log_value(self, **kwargs: dict[str, int | float | Any]) -> None:
         """
         Log data.
 
         :param kwargs: Data to log.
         """
-        if self.module_writer is None:
-            msg = "start_run must be called before log_value."
-            raise RuntimeError(msg)
-
         for key, value in kwargs.items():
             if isinstance(value, int | float):
                 self.module_writer.add_scalar(key, value)
@@ -79,10 +75,6 @@ def log_metrics(self, metrics: dict[str, Any]) -> None:
 
         :param metrics: Metrics to log.
         """
-        if self.module_writer is None:
-            msg = "start_run must be called before log_value."
-            raise RuntimeError(msg)
-
         for key, value in metrics.items():
             if isinstance(value, int | float):
                 self.module_writer.add_scalar(key, value)  # type: ignore[no-untyped-call]

autointent/_pipeline/_pipeline.py

@@ -27,7 +27,7 @@ class Pipeline:
 
     def __init__(
         self,
-        nodes: list[NodeOptimizer] | list[InferenceNode],
+        nodes: list[NodeOptimizer] | list[InferenceNode] | list[Any],
         seed: int = 42,
     ) -> None:
         """
@@ -48,7 +48,7 @@ def __init__(
             msg = "Pipeline should be initialized with list of NodeOptimizers or InferenceNodes"
             raise TypeError(msg)
 
-    def set_config(self, config: LoggingConfig | VectorIndexConfig | DataConfig) -> None:
+    def set_config(self, config: LoggingConfig | VectorIndexConfig | DataConfig | Any) -> None:
         """
         Set configuration for the optimizer.
 

autointent/configs/_inference_node.py

@@ -18,7 +18,3 @@ class InferenceNodeConfig:
     """Configuration of the module"""
     load_path: str | None = None
     """Path to the module dump. If None, the module will be trained from scratch"""
-
-    def __post_init__(self) -> None:
-        if not isinstance(self.node_type, NodeType):
-            self.node_type = NodeType(self.node_type)

autointent/context/_context.py

@@ -94,10 +94,6 @@ def dump(self) -> None:
         optimization_results = self.optimization_info.dump_evaluation_results()
 
         logs_dir = self.logging_config.dirpath
-        if logs_dir is None:
-            msg = "something's wrong with LoggingConfig"
-            raise ValueError(msg)
-
         logs_dir.mkdir(parents=True, exist_ok=True)
 
         logs_path = logs_dir / "logs.json"

autointent/metrics/_converter.py

@@ -22,8 +22,6 @@ def transform(
     :param y_pred: Y_pred values
     :return:
     """
-    if isinstance(y_true, np.ndarray) and isinstance(y_pred, np.ndarray):
-        return y_true, y_pred
     y_pred_ = np.array(y_pred)
     y_true_ = np.array(y_true)
     return y_true_, y_pred_

autointent/modules/abc/_base.py

@@ -40,15 +40,13 @@ def score(self, context: Context, metrics: list[str]) -> dict[str, float]:
         Calculate metric on test set and return metric value.
 
         :param context: Context to score
-        :param split: Split to score on
+        :param metrics: Metrics to score
         :return: Computed metrics value for the test set or error code of metrics
         """
         if context.data_handler.config.scheme == "ho":
             return self.score_ho(context, metrics)
         if context.data_handler.config.scheme == "cv":
             return self.score_cv(context, metrics)
-        msg = "Something's wrong with validation schemas"
-        raise RuntimeError(msg)
 
     @abstractmethod
     def score_cv(self, context: Context, metrics: list[str]) -> dict[str, float]: ...

autointent/nodes/_optimization/_node_optimizer.py

@@ -65,11 +65,12 @@ def fit(self, context: Context, sampler: SamplerType = "brute") -> None:
         Fit the node optimizer.
 
         :param context: Context
+        :param sampler: Sampler to use for optimization
         """
         self._logger.info("starting %s node optimization...", self.node_info.node_type)
 
         for search_space in deepcopy(self.modules_search_spaces):
-            self._counter = 0
+            self._counter: int = 0
             module_name = search_space.pop("module_name")
             n_trials = None
             if "n_trials" in search_space:
@@ -83,9 +84,6 @@ def fit(self, context: Context, sampler: SamplerType = "brute") -> None:
             elif sampler == "random":
                 sampler_instance = optuna.samplers.RandomSampler(seed=context.seed)  # type: ignore[assignment]
                 n_trials = n_trials or 10
-            else:
-                msg = f"Unexpected sampler: {sampler}"
-                raise ValueError(msg)
             study = optuna.create_study(direction="maximize", sampler=sampler_instance)
             optuna.logging.set_verbosity(optuna.logging.WARNING)
             obj = partial(self.objective, module_name=module_name, search_space=search_space, context=context)

autointent/schemas/_schemas.py

@@ -218,7 +218,6 @@ def get_prompt_type(self, prompt_type: TaskTypeEnum | None) -> str | None:  # no
             return self.sts_prompt
         if prompt_type == TaskTypeEnum.default:
             return self.default_prompt
-        return None
 
     use_cache: bool = Field(False, description="Whether to use embeddings caching.")
 

pyproject.toml

@@ -166,6 +166,9 @@ skip_empty = true
 python_version = "3.10"
 strict = true
 warn_redundant_casts = true
+# align with mypy 2.0 release
+warn_unreachable = true
+local_partial_types = true
 plugins = [
     "pydantic.mypy",
     "numpy.typing.mypy_plugin",
@@ -193,3 +196,10 @@ module = [
     "wandb",
 ]
 ignore_missing_imports = true
+
+[[tool.mypy.overrides]]
+module = [
+    "autointent._callbacks.*",
+    "autointent.modules.abc.*",
+]
+warn_unreachable = false