FIX Ensure that index is correct when global transform_output is pandas (scikit-learn#26454)

Author: thomasjpfan

doc/whats_new/v1.3.rst

@@ -401,6 +401,9 @@ Changelog
 - |Enhancement| Added the parameter `fill_value` to :class:`impute.IterativeImputer`.
   :pr:`25232` by :user:`Thijs van Weezel <ValueInvestorThijs>`.
 
+- |Fix| :class:`impute.IterativeImputer` now correctly preserves the Pandas
+  Index when the `set_config(transform_output="pandas")`. :pr:`26454` by `Thomas Fan`_.
+
 :mod:`sklearn.inspection`
 .........................
 
@@ -444,6 +447,12 @@ Changelog
   on linearly separable problems.
   :pr:`25214` by `Tom Dupre la Tour`_.
 
+:mod:`sklearn.manifold`
+.......................
+
+- |Fix| :class:`manifold.Isomap` now correctly preserves the Pandas
+  Index when the `set_config(transform_output="pandas")`. :pr:`26454` by `Thomas Fan`_.
+
 :mod:`sklearn.metrics`
 ......................
 
@@ -636,6 +645,9 @@ Changelog
   The `sample_interval_` attribute is deprecated and will be removed in 1.5.
   :pr:`25190` by :user:`Vincent Maladière <Vincent-Maladiere>`.
 
+- |Fix| :class:`preprocessing.PowerTransformer` now correctly preserves the Pandas
+  Index when the `set_config(transform_output="pandas")`. :pr:`26454` by `Thomas Fan`_.
+
 - |Fix| :class:`preprocessing.PowerTransformer` now correcly raises error when
   using `method="box-cox"` on data with a constant `np.nan` column.
   :pr:`26400` by :user:`Yao Xiao <Charlie-XIAO>`.

sklearn/impute/_iterative.py

@@ -627,7 +627,7 @@ def _initial_imputation(self, X, in_fit=False):
                 strategy=self.initial_strategy,
                 fill_value=self.fill_value,
                 keep_empty_features=self.keep_empty_features,
-            )
+            ).set_output(transform="default")
             X_filled = self.initial_imputer_.fit_transform(X)
         else:
             X_filled = self.initial_imputer_.transform(X)

sklearn/manifold/_isomap.py

@@ -235,7 +235,7 @@ def _fit_transform(self, X):
             tol=self.tol,
             max_iter=self.max_iter,
             n_jobs=self.n_jobs,
-        )
+        ).set_output(transform="default")
 
         if self.n_neighbors is not None:
             nbg = kneighbors_graph(

sklearn/preprocessing/_data.py

@@ -3180,7 +3180,7 @@ def _fit(self, X, y=None, force_transform=False):
                     X[:, i] = transform_function(X[:, i], self.lambdas_[i])
 
         if self.standardize:
-            self._scaler = StandardScaler(copy=False)
+            self._scaler = StandardScaler(copy=False).set_output(transform="default")
             if force_transform:
                 X = self._scaler.fit_transform(X)
             else:

sklearn/utils/estimator_checks.py

@@ -4424,7 +4424,7 @@ def _output_from_fit_transform(transformer, name, X, df, y):
     return outputs
 
 
-def _check_generated_dataframe(name, case, outputs_default, outputs_pandas):
+def _check_generated_dataframe(name, case, index, outputs_default, outputs_pandas):
     import pandas as pd
 
     X_trans, feature_names_default = outputs_default
@@ -4434,7 +4434,12 @@ def _check_generated_dataframe(name, case, outputs_default, outputs_pandas):
     # We always rely on the output of `get_feature_names_out` of the
     # transformer used to generate the dataframe as a ground-truth of the
     # columns.
-    expected_dataframe = pd.DataFrame(X_trans, columns=feature_names_pandas, copy=False)
+    # If a dataframe is passed into transform, then the output should have the same
+    # index
+    expected_index = index if case.endswith("df") else None
+    expected_dataframe = pd.DataFrame(
+        X_trans, columns=feature_names_pandas, copy=False, index=expected_index
+    )
 
     try:
         pd.testing.assert_frame_equal(df_trans, expected_dataframe)
@@ -4469,7 +4474,8 @@ def check_set_output_transform_pandas(name, transformer_orig):
     set_random_state(transformer)
 
     feature_names_in = [f"col{i}" for i in range(X.shape[1])]
-    df = pd.DataFrame(X, columns=feature_names_in, copy=False)
+    index = [f"index{i}" for i in range(X.shape[0])]
+    df = pd.DataFrame(X, columns=feature_names_in, copy=False, index=index)
 
     transformer_default = clone(transformer).set_output(transform="default")
     outputs_default = _output_from_fit_transform(transformer_default, name, X, df, y)
@@ -4483,7 +4489,7 @@ def check_set_output_transform_pandas(name, transformer_orig):
 
     for case in outputs_default:
         _check_generated_dataframe(
-            name, case, outputs_default[case], outputs_pandas[case]
+            name, case, index, outputs_default[case], outputs_pandas[case]
         )
 
 
@@ -4511,7 +4517,8 @@ def check_global_ouptut_transform_pandas(name, transformer_orig):
     set_random_state(transformer)
 
     feature_names_in = [f"col{i}" for i in range(X.shape[1])]
-    df = pd.DataFrame(X, columns=feature_names_in, copy=False)
+    index = [f"index{i}" for i in range(X.shape[0])]
+    df = pd.DataFrame(X, columns=feature_names_in, copy=False, index=index)
 
     transformer_default = clone(transformer).set_output(transform="default")
     outputs_default = _output_from_fit_transform(transformer_default, name, X, df, y)
@@ -4528,5 +4535,5 @@ def check_global_ouptut_transform_pandas(name, transformer_orig):
 
     for case in outputs_default:
         _check_generated_dataframe(
-            name, case, outputs_default[case], outputs_pandas[case]
+            name, case, index, outputs_default[case], outputs_pandas[case]
         )