def test_toy_dataset_frame_dtype(loader_func, data_dtype, target_dtype):
default_result = loader_func()
check_as_frame(
default_result,
loader_func,
expected_data_dtype=data_dtype,
expected_target_dtype=target_dtype,
)
def test_20news_as_frame(fetch_20newsgroups_vectorized_fxt):
pd = pytest.importorskip('pandas')
bunch = fetch_20newsgroups_vectorized_fxt(as_frame=True)
check_as_frame(bunch, fetch_20newsgroups_vectorized_fxt)
frame = bunch.frame
assert frame.shape == (11314, 130108)
assert all([isinstance(col, pd.SparseDtype) for col in bunch.data.dtypes])
# Check a small subset of features
for expected_feature in [
"beginner",
"beginners",
"beginning",
"beginnings",
"begins",
"begley",
"begone",
]:
assert expected_feature in frame.keys()
assert "category_class" in frame.keys()
assert bunch.target.name == "category_class"
def test_fetch_kddcup99_as_frame(fetch_kddcup99_fxt):
bunch = fetch_kddcup99_fxt()
check_as_frame(bunch, fetch_kddcup99_fxt)
