GSK-2625: fix OneHotEncoder

Author: andreybavt

giskard/demo/titanic_classification.py

@@ -57,7 +57,7 @@ def get_model_and_df(model: str = ModelTypes.LOGISTIC_REGRESSION, max_iter: int
     cat_transormer = Pipeline(
         [
             ("imputer", SimpleImputer(strategy="constant", fill_value="missing")),
-            ("onehot", OneHotEncoder(handle_unknown="ignore", sparse=False)),
+            ("onehot", OneHotEncoder(handle_unknown="ignore")),
         ]
     )
 

pyproject.toml

@@ -156,7 +156,7 @@ dependencies = [
     "zstandard>=0.10.0",
     "mlflow-skinny>=2",
     "numpy>=1.22.0",
-    "scikit-learn>=1.0, <1.4.0", # TODO: GSK-2625
+    "scikit-learn>=1.0",
     "scipy>=1.7.3",
     "mixpanel>=4.4.0",
     "requests>=2.19",

tests/fixtures/enron_multilabel_classification.py

@@ -152,7 +152,7 @@ def enron_raw_model(enron_raw_data: pd.DataFrame) -> Tuple[Pipeline, List[str]]:
     categorical_transformer = Pipeline(
         [
             ("imputer", SimpleImputer(strategy="constant", fill_value="missing")),
-            ("onehot", OneHotEncoder(handle_unknown="ignore", sparse=False)),
+            ("onehot", OneHotEncoder(handle_unknown="ignore")),
         ]
     )
 

tests/fixtures/german_credit_scoring.py

@@ -112,7 +112,7 @@ def german_credit_raw_model(german_credit_raw_data):
     categorical_transformer = Pipeline(
         [
             ("imputer", SimpleImputer(strategy="constant", fill_value="missing")),
-            ("onehot", OneHotEncoder(handle_unknown="ignore", sparse=False)),
+            ("onehot", OneHotEncoder(handle_unknown="ignore")),
         ]
     )