DOC update remaining gallery examples to load datasets from paths

Author: MuditAtrey

CHANGES.rst

@@ -66,12 +66,15 @@ Bug Fixes
 - :class:`CheckInputDataFrame` no longer collects Polars LazyFrames automatically;
   a ``TypeError`` is now raised instead, consistent with the rest of the library.
   :pr:`1941` by :user:`Mudit Atrey <MuditAtrey>`.
+- :func:`fetch_employee_salaries` now correctly writes the train and test
+  split CSV files to their respective paths when ``split`` is specified.
+  :pr:`1964` by :user:`MuditAtrey <MuditAtrey>`.
 
 Documentation
 -------------
 - Updated gallery examples to load datasets from their file paths using
   ``pd.read_csv()``, following the pattern established in :pr:`1852`.
-  :pr:`1940` by :user:`MuditAtrey <MuditAtrey>`.
+  :pr:`1940` and :pr:`1964` by :user:`MuditAtrey <MuditAtrey>`.
 
 Release 0.7.2
 =============

examples/FIXME/07_grid_searching_with_the_tablevectorizer.py

@@ -32,11 +32,14 @@
 #
 # Throughout this example, we will use the employee salaries dataset.
 
+import pandas as pd
+
 from skrub.datasets import fetch_employee_salaries
 
 dataset = fetch_employee_salaries()
-X = dataset.X
-y = dataset.y
+df = pd.read_csv(dataset.employee_salaries_path)
+X = df.drop(columns=["current_annual_salary"])
+y = df["current_annual_salary"]
 
 X.head(10)
 

examples/data_ops/1110_data_ops_intro.py

@@ -46,9 +46,13 @@
 # By default, the |fetch_employee_salaries| function returns the training set.
 # We will load the test set later, to evaluate our model on unseen data.
 
+import pandas as pd
+
 from skrub.datasets import fetch_employee_salaries
 
-training_data = fetch_employee_salaries(split="train").employee_salaries
+training_data = pd.read_csv(
+    fetch_employee_salaries(split="train").employee_salaries_path
+)
 
 # %%
 # We can take a look at the dataset using the |TableReport|.
@@ -160,7 +164,7 @@
 # case, "data").
 #
 # We can now get the test set of the employee salaries dataset:
-unseen_data = fetch_employee_salaries(split="test").employee_salaries
+unseen_data = pd.read_csv(fetch_employee_salaries(split="test").employee_salaries_path)
 
 # %%
 # Then, we can use the loaded model to make predictions on the unseen data by

skrub/datasets/_fetching.py

@@ -64,7 +64,7 @@ def fetch_employee_salaries(data_home=None, split="all"):
         )
         dataset["employee_salaries_path"] = str(train_path)
         dataset["path"] = str(train_path)
-        dataset["employee_salaries"][:id_split].to_csv(str(train_path), index=False)
+        dataset["employee_salaries"].to_csv(str(train_path), index=False)
         dataset["X"] = dataset["X"][:id_split]
         dataset["y"] = dataset["y"][:id_split]
     elif split == "test":
@@ -74,7 +74,7 @@ def fetch_employee_salaries(data_home=None, split="all"):
         )
         dataset["employee_salaries_path"] = str(test_path)
         dataset["path"] = str(test_path)
-        dataset["employee_salaries"][id_split:].to_csv(str(test_path), index=False)
+        dataset["employee_salaries"].to_csv(str(test_path), index=False)
         dataset["X"] = dataset["X"][id_split:]
         dataset["y"] = dataset["y"][id_split:]
     return dataset