scikit-learn
diff --git a/‎sklearn/compose/_target.py
Lines changed: 20 additions & 1 deletion b/‎sklearn/compose/_target.py
Lines changed: 20 additions & 1 deletion
diff --git a/‎sklearn/tests/test_common.py
Lines changed: 1 addition & 21 deletions b/‎sklearn/tests/test_common.py
Lines changed: 1 addition & 21 deletions
@@ -8,6 +8,7 @@
 
 from ..base import BaseEstimator, RegressorMixin, clone
 from ..utils.validation import check_is_fitted
+from ..utils._tags import _safe_tags
 from ..utils import check_array, _safe_indexing
 from ..preprocessing import FunctionTransformer
 from ..exceptions import NotFittedError
@@ -88,6 +89,12 @@ class TransformedTargetRegressor(RegressorMixin, BaseEstimator):
 
         .. versionadded:: 0.24
 
+    feature_names_in_ : ndarray of shape (`n_features_in_`,)
+        Names of features seen during :term:`fit`. Defined only when `X`
+        has feature names that are all strings.
+
+        .. versionadded:: 1.0
+
     Examples
     --------
     >>> import numpy as np
@@ -233,6 +240,9 @@ def fit(self, X, y, **fit_params):
 
         self.regressor_.fit(X, y_trans, **fit_params)
 
+        if hasattr(self.regressor_, "feature_names_in_"):
+            self.feature_names_in_ = self.regressor_.feature_names_in_
+
         return self
 
     def predict(self, X, **predict_params):
@@ -272,7 +282,16 @@ def predict(self, X, **predict_params):
         return pred_trans
 
     def _more_tags(self):
-        return {"poor_score": True, "no_validation": True}
+        regressor = self.regressor
+        if regressor is None:
+            from ..linear_model import LinearRegression
+
+            regressor = LinearRegression()
+
+        return {
+            "poor_score": True,
+            "multioutput": _safe_tags(regressor, key="multioutput"),
+        }
 
     @property
     def n_features_in_(self):
 
@@ -316,17 +316,6 @@ def test_check_n_features_in_after_fitting(estimator):
     check_n_features_in_after_fitting(estimator.__class__.__name__, estimator)
 
 
-# TODO: When more modules get added, we can remove it from this list to make
-# sure it gets tested. After we finish each module we can move the checks
-# into check_estimator.
-# NOTE: When running `check_dataframe_column_names_consistency` on a meta-estimator that
-# delegates validation to a base estimator, the check is testing that the base estimator
-# is checking for column name consistency.
-
-COLUMN_NAME_MODULES_TO_IGNORE = {
-    "compose",
-}
-
 _estimators_to_test = list(
     chain(
         _tested_estimators(),
@@ -336,16 +325,7 @@ def test_check_n_features_in_after_fitting(estimator):
 )
 
 
-column_name_estimators = [
-    est
-    for est in _estimators_to_test
-    if est.__module__.split(".")[1] not in COLUMN_NAME_MODULES_TO_IGNORE
-]
-
-
-@pytest.mark.parametrize(
-    "estimator", column_name_estimators, ids=_get_check_estimator_ids
-)
+@pytest.mark.parametrize("estimator", _estimators_to_test, ids=_get_check_estimator_ids)
 def test_pandas_column_name_consistency(estimator):
     _set_checking_parameters(estimator)
     with ignore_warnings(category=(FutureWarning)):