fix typing

Author: Samoed

autointent/modules/scoring/_catboost/catboost_scorer.py

@@ -106,6 +106,7 @@ def __init__(
         self.loss_function = loss_function
         self.verbose = verbose
         self.catboost_kwargs = catboost_kwargs or {}
+        self.catboost_kwargs |= self.get_extra_params()
 
     @classmethod
     def from_context(
@@ -125,6 +126,7 @@ def from_context(
             loss_function=loss_function,
             verbose=verbose,
             features_type=features_type,
+            use_embedding_features=use_embedding_features,
             **catboost_kwargs,
         )
 
@@ -144,12 +146,26 @@ def _prepare_data_for_fit(
             if self.use_embedding_features:
                 data = pd.DataFrame({"embedding": encoded_utterances})
             else:
-                data = pd.DataFrame(encoded_utterances)
+                data = pd.DataFrame(np.array(encoded_utterances))
             if self.features_type == FeaturesType.BOTH:
                 data["text"] = utterances
             return data
         return pd.DataFrame({"text": utterances})
 
+    def get_extra_params(self) -> dict[str, Any]:
+        extra_params = {}
+        if self.features_type == FeaturesType.EMBEDDING:
+            if self.use_embedding_features:  # to not raise error if embedding witout embedding_features
+                extra_params["embedding_features"] = ["embedding"]
+        elif self.features_type in {FeaturesType.TEXT, FeaturesType.BOTH}:
+            extra_params["text_features"] = ["text"]
+            if self.features_type == FeaturesType.BOTH and self.use_embedding_features:
+                extra_params["embedding_features"] = ["embedding"]
+        else:
+            msg = f"Unsupported features type: {self.features_type}"
+            raise ValueError(msg)
+        return extra_params
+
     def fit(
         self,
         utterances: list[str],
@@ -167,19 +183,6 @@ def fit(
             else ("MultiClass" if self._n_classes > BINARY_CLASS_THRESHOLD else "Logloss")
         )
 
-        extra_params = {}
-        if self.features_type == FeaturesType.EMBEDDING:
-            if self.use_embedding_features:  # to not raise error if embedding witout embedding_features
-                extra_params["embedding_features"] = ["embedding"]
-        elif self.features_type in {FeaturesType.TEXT, FeaturesType.BOTH}:
-            extra_params["text_features"] = ["text"]
-            if self.features_type == FeaturesType.BOTH and self.use_embedding_features:
-                extra_params["embedding_features"] = ["embedding"]
-        else:
-            msg = f"Unsupported features type: {self.features_type}"
-            raise ValueError(msg)
-        self.catboost_kwargs.update(extra_params)
-
         self._model = CatBoostClassifier(
             loss_function=self.loss_function or default_loss,
             verbose=self.verbose,

pyproject.toml

@@ -70,6 +70,7 @@ typing = [
     "types-pygments (>=2.18.0.20240506,<3.0.0)",
     "types-setuptools (>=75.2.0.20241019,<76.0.0)",
     "joblib-stubs (>=1.4.2.5.20240918,<2.0.0)",
+    "pandas-stubs (>= 2.2.3.250527, <3.0.0)",
 ]
 docs = [
     "sphinx (>=8.1.3,<9.0.0)",