lecrapaud 0.19.1__tar.gz → 0.19.3__tar.gz

{lecrapaud-0.19.1 → lecrapaud-0.19.3}/PKG-INFO

@@ -1,8 +1,9 @@
-Metadata-Version: 2.3
+Metadata-Version: 2.4
 Name: lecrapaud
-Version: 0.19.1
+Version: 0.19.3
 Summary: Framework for machine and deep learning, with regression, classification and time series analysis
 License: Apache License
+License-File: LICENSE
 Author: Pierre H. Gallet
 Requires-Python: ==3.12.*
 Classifier: License :: Other/Proprietary License

{lecrapaud-0.19.1 → lecrapaud-0.19.3}/lecrapaud/api.py

@@ -475,6 +475,9 @@ class ExperimentEngine:
             # For lightgbm models
             importances = model.feature_importance(importance_type="split")
             importance_type = "Split"
+        elif hasattr(model, "get_feature_importance"):
+            importances = model.get_feature_importance()
+            importance_type = "Feature importance"
         elif hasattr(model, "coef_"):
             # For linear models
             importances = np.abs(model.coef_.flatten())

{lecrapaud-0.19.1 → lecrapaud-0.19.3}/lecrapaud/experiment.py

@@ -35,7 +35,7 @@ def create_experiment(
     groups = {}
     if group_column:
         groups["number_of_groups"] = data[group_column].nunique()
-        groups["list_of_groups"] = data[group_column].unique().tolist().sort()
+        groups["list_of_groups"] = sorted(data[group_column].unique().tolist())
 
     with get_db() as db:
         all_targets = Target.get_all(db=db)

{lecrapaud-0.19.1 → lecrapaud-0.19.3}/lecrapaud/feature_engineering.py

@@ -629,6 +629,7 @@ class PreprocessFeature:
         """
 
         pcas_dict = {}
+        index_saved = df.index
 
         for pca_cross_sectional in self.pca_cross_sectional:
             name, index_col, columns_col, value_col = (
@@ -659,6 +660,7 @@ class PreprocessFeature:
             scores_df = pd.DataFrame(scores, index=pivot.index, columns=cols)
 
             df = df.merge(scores_df.reset_index(), on=index_col, how="left")
+            df.index = index_saved
             pcas_dict.update({name: pipe})
 
         return df, pcas_dict

{lecrapaud-0.19.1 → lecrapaud-0.19.3}/lecrapaud/model_selection.py

@@ -110,6 +110,63 @@ def test_hardware():
 warnings.filterwarnings("ignore", category=UserWarning, module="pydantic")
 
 
+class CatBoostWrapper:
+    """
+    Transparent proxy for a CatBoost model that accepts arbitrary keyword arguments
+    as direct attributes, while forwarding all method calls and properties.
+    """
+
+    __slots__ = ("_model", "_extra_attrs")
+
+    def __init__(self, model, **kwargs):
+        object.__setattr__(self, "_model", model)
+        object.__setattr__(self, "_extra_attrs", {})
+        # Register kwargs as direct attributes
+        for key, value in kwargs.items():
+            setattr(self, key, value)
+
+    # ---- Transparent access ----
+    def __getattr__(self, name):
+        """Forward attribute access to the underlying model if not found."""
+        model = object.__getattribute__(self, "_model")
+        if hasattr(model, name):
+            return getattr(model, name)
+        extra_attrs = object.__getattribute__(self, "_extra_attrs")
+        if name in extra_attrs:
+            return extra_attrs[name]
+        raise AttributeError(f"{type(self).__name__!r} has no attribute {name!r}")
+
+    def __setattr__(self, name, value):
+        """Set to wrapper or forward to model when appropriate."""
+        if name in CatBoostWrapper.__slots__:
+            object.__setattr__(self, name, value)
+            return
+
+        model = object.__getattribute__(self, "_model")
+        if hasattr(model, name):
+            setattr(model, name, value)
+        else:
+            extra_attrs = object.__getattribute__(self, "_extra_attrs")
+            extra_attrs[name] = value
+
+    def __dir__(self):
+        """Merge dir() from wrapper, model, and custom attributes."""
+        base = set(super().__dir__())
+        model_attrs = set(dir(object.__getattribute__(self, "_model")))
+        extra_attrs = set(object.__getattribute__(self, "_extra_attrs").keys())
+        return sorted(base | model_attrs | extra_attrs)
+
+    def __repr__(self):
+        model = object.__getattribute__(self, "_model")
+        extras = object.__getattribute__(self, "_extra_attrs")
+        return f"CatBoostWrapper(model={model.__class__.__name__}, extras={extras})"
+
+    @property
+    def model(self):
+        """Access the raw CatBoost model."""
+        return object.__getattribute__(self, "_model")
+
+
 class ModelEngine:
 
     def __init__(
@@ -296,12 +353,15 @@ class ModelEngine:
             )
 
         # Attach metadata for consistency with sklearn path
-        model.model_name = self.model_name
-        model.target_type = self.target_type
-        logger.info(f"Successfully created a {model.model_name} at {datetime.now()}")
+        model_wrapped = CatBoostWrapper(
+            model, model_name=self.model_name, target_type=self.target_type
+        )
+        logger.info(
+            f"Successfully created a {model_wrapped.model_name} at {datetime.now()}"
+        )
 
-        self._model = model
-        return model
+        self._model = model_wrapped
+        return model_wrapped
 
     def fit_boosting(self, x_train, y_train, x_val, y_val, params):
         """

{lecrapaud-0.19.1 → lecrapaud-0.19.3}/pyproject.toml

@@ -1,6 +1,6 @@
 [project]
 name = "lecrapaud"
-version = "0.19.1"
+version = "0.19.3"
 description = "Framework for machine and deep learning, with regression, classification and time series analysis"
 authors = [
     {name = "Pierre H. Gallet"}