preprocessor = ColumnTransformer(
    transformers=[
        ('scaler', StandardScaler(), numerical_features) # масштабируем численные признаки
    ],
    remainder='passthrough'  # остальные признаки остаются без изменений
)

# Создание пайплайна
cbc_pipeline = Pipeline([
    ('preprocessor', preprocessor),
    #('imputer', SimpleImputer(strategy='median')),
    ('clf', CatBoostClassifier(random_seed=STATE,
                                     cat_features=categorical_features,
                                     eval_metric='F1',
                                     auto_class_weights='Balanced',
                                     verbose=100))
  ]
)

# Обучение модели
cbc_pipeline.fit(X_train, y_train)


Ошибка:
CatBoostError                             Traceback (most recent call last)
<ipython-input-51-4f605af09d07> in <cell line: 22>()
     20
     21 # Обучение модели
---> 22 cbc_pipeline.fit(X_train, y_train)

7 frames
/usr/local/lib/python3.10/dist-packages/catboost/core.py in _get_features_indices(features, feature_names)
    323         for f in features:
    324             if isinstance(f, STRING_TYPES):
--> 325                 raise CatBoostError("features parameter contains string value '{}' but feature names "
    326                                     "for a dataset are not specified".format(f))
    327     return features

CatBoostError: features parameter contains string value 'пол' but feature names for a dataset are not specified