diff --git a/sklearn/conftest.py b/sklearn/conftest.py
index d2f44f6912b62..d14afddc3773d 100644
--- a/sklearn/conftest.py
+++ b/sklearn/conftest.py
@@ -134,10 +134,16 @@ def pytest_collection_modifyitems(config, items):
datasets_to_download = set()
for item in items:
- if not hasattr(item, "fixturenames"):
+ if isinstance(item, DoctestItem) and "fetch_" in item.name:
+ fetcher_function_name = item.name.split(".")[-1]
+ dataset_fetchers_key = f"{fetcher_function_name}_fxt"
+ dataset_to_fetch = set([dataset_fetchers_key]) & dataset_features_set
+ elif not hasattr(item, "fixturenames"):
continue
- item_fixtures = set(item.fixturenames)
- dataset_to_fetch = item_fixtures & dataset_features_set
+ else:
+ item_fixtures = set(item.fixturenames)
+ dataset_to_fetch = item_fixtures & dataset_features_set
+
if not dataset_to_fetch:
continue
diff --git a/sklearn/datasets/_base.py b/sklearn/datasets/_base.py
index e062bf381b393..ab2b8bd3f5110 100644
--- a/sklearn/datasets/_base.py
+++ b/sklearn/datasets/_base.py
@@ -85,6 +85,11 @@ def clear_data_home(data_home=None):
data_home : str or path-like, default=None
The path to scikit-learn data directory. If `None`, the default path
is `~/scikit_learn_data`.
+
+ Examples
+ ----------
+ >>> from sklearn.datasets import clear_data_home
+ >>> clear_data_home() # doctest: +SKIP
"""
data_home = get_data_home(data_home)
shutil.rmtree(data_home)
diff --git a/sklearn/datasets/_covtype.py b/sklearn/datasets/_covtype.py
index 7620e08c5ec92..4e1b1d7961f2e 100644
--- a/sklearn/datasets/_covtype.py
+++ b/sklearn/datasets/_covtype.py
@@ -156,6 +156,18 @@ def fetch_covtype(
ndarray of shape (n_samples,) containing the target samples.
.. versionadded:: 0.20
+
+ Examples
+ --------
+ >>> from sklearn.datasets import fetch_covtype
+ >>> cov_type = fetch_covtype()
+ >>> cov_type.data.shape
+ (581012, 54)
+ >>> cov_type.target.shape
+ (581012,)
+ >>> # Let's check the 4 first feature names
+ >>> cov_type.feature_names[:4]
+ ['Elevation', 'Aspect', 'Slope', 'Horizontal_Distance_To_Hydrology']
"""
data_home = get_data_home(data_home=data_home)
covtype_dir = join(data_home, "covertype")