def test_masked_encode():
df = cudf.DataFrame({
"filter_col": [1, 1, 2, 3, 1, 1, 1, 1, 6, 5],
"cat_col": ['a', 'b', 'c', 'd', 'a', 'a', 'a', 'c', 'b', 'c']
})
df_filter = df[df["filter_col"] == 1]
df_filter["cat_col"] = LabelEncoder().fit_transform(df_filter["cat_col"])
df["cat_col"] = LabelEncoder().fit_transform(df["cat_col"])
df = df[df["filter_col"] == 1]
assert (df_filter["cat_col"] == df["cat_col"]).all()
def test_empty_input(empty, ord_label):
# prepare LabelEncoder
le = LabelEncoder()
le.fit(empty)
assert (le._fitted is True)
# test if correctly raies ValueError
with pytest.raises(ValueError, match='y contains previously unseen label'):
le.inverse_transform(ord_label)
# check fit_transform()
le = LabelEncoder()
transformed = le.fit_transform(empty)
assert (le._fitted is True)
assert (len(transformed) == 0)
def test_labelencoder_unfitted():
""" Try calling `.transform()` without fitting first
"""
df = cudf.Series(np.random.choice(10, (10, )))
le = LabelEncoder()
assert not le._fitted
with pytest.raises(NotFittedError):
le.transform(df)
def test_labelencoder_unseen():
""" Try encoding a value that was not present during fitting
"""
df = cudf.Series(np.random.choice(10, (10, )))
le = LabelEncoder().fit(df)
assert le._fitted
with pytest.raises(KeyError):
le.transform(cudf.Series([-1]))
def test_labelencoder_fit_transform(length, cardinality):
""" Try encoding the entire df
"""
df = cudf.Series(np.random.choice(cardinality, (length, )))
encoded = LabelEncoder().fit_transform(df)
df_arr = _df_to_similarity_mat(df)
encoded_arr = _df_to_similarity_mat(encoded)
assert ((encoded_arr == encoded_arr.T) == (df_arr == df_arr.T)).all()
def test_unfitted_inverse_transform():
""" Try calling `.inverse_transform()` without fitting first
"""
df = cudf.Series(np.random.choice(10, (10, )))
le = LabelEncoder()
assert (not le._fitted)
with pytest.raises(RuntimeError):
le.transform(df)
def test_masked_encode():
int_values = [3, 1, 1, 2, 1, 1, 1, 1, 6, 5]
cat_values = ['a', 'd', 'b', 'c', 'd', 'd', 'd', 'c', 'b', 'c']
df = cudf.DataFrame({"filter_col": int_values,
"cat_col": cat_values})
df_filter = df[df["filter_col"] == 1]
df_filter["cat_col"] = LabelEncoder().fit_transform(df_filter["cat_col"])
filtered_int_values = [int_values[i] for i in range(len(int_values))
if int_values[i] == 1]
filtered_cat_values = [cat_values[i] for i in range(len(int_values))
if int_values[i] == 1]
df_test = cudf.DataFrame({"filter_col": filtered_int_values,
"cat_col": filtered_cat_values})
df_test["cat_col"] = LabelEncoder().fit_transform(df_test["cat_col"])
assert(df_filter["cat_col"].values == df_test["cat_col"].values).all()
def test_labelencoder_fit_transform_cupy_numpy(length, cardinality, dtype):
""" Try encoding the cupy array
"""
x = cp.random.choice(cardinality, (length,))
if dtype == 'numpy':
x = x.get()
encoded = LabelEncoder().fit_transform(x)
x_arr = _array_to_similarity_mat(x)
encoded_arr = _array_to_similarity_mat(encoded.values)
if dtype == 'numpy':
encoded_arr = encoded_arr.get()
assert ((encoded_arr == encoded_arr.T) == (x == x_arr.T)).all()
def test_labelencoder_transform(length, cardinality):
""" Try fitting and then encoding a small subset of the df
"""
df = cudf.Series(np.random.choice(cardinality, (length, )))
le = LabelEncoder().fit(df)
assert le._fitted
subset = df.iloc[0:df.shape[0] // 2]
encoded = le.transform(subset)
subset_arr = _df_to_similarity_mat(subset)
encoded_arr = _df_to_similarity_mat(encoded)
assert ((encoded_arr == encoded_arr.T) == (
subset_arr == subset_arr.T)).all()
def test_inverse_transform(orig_label, ord_label, expected_reverted,
bad_ord_label):
# prepare LabelEncoder
le = LabelEncoder()
le.fit(orig_label)
assert (le._fitted is True)
# test if inverse_transform is correct
reverted = le.inverse_transform(ord_label)
assert (len(reverted) == len(expected_reverted))
assert (len(reverted) == len(reverted[reverted == expected_reverted]))
# test if correctly raies ValueError
with pytest.raises(ValueError, match='y contains previously unseen label'):
le.inverse_transform(bad_ord_label)