def test_KNN_wrong_type(self):
"""
Negative test
data: array (unsupported type)
Checks that the function raises a TypeError if the data is passed as
an array.
"""
# 1. Arrange
data = [2, 4, np.nan, 1]
# 2. Act & 3. Assert
with self.assertRaises(TypeError):
knn(data)
def test_KNN_inplace(self):
"""
Positive test
data: Correct data frame (example_df)
The data frame (example_df) contains 2 NA values.
knn() should impute both of them.
Checks that the data frame contains no NA values after the operation.
"""
# 1. Arrange
df = generate_example_df()
# 2. Act
knn(df, inplace=True)
# 3. Assert
self.assertEqual(df.isna().sum().sum(), 0)
def test_KNN_wrong_column(self):
"""
Negative test
data: Correct data frame (example_df)
columns: ['x', 'a'] ('a' is not a column of example_df)
Checks that the function raises a ValueError if one of the specified
columns doesn't exist in the data.
"""
# 1. Arrange
df = generate_example_df()
# 2. Act
df2 = knn(df, columns=['x'])
# 2. Act & 3. Assert
with self.assertRaises(ValueError):
knn(df, columns=['x', 'a'])
def test_KNN_returning(self):
"""
Positive test
data: Correct data frame (example_df)
The data frame (example_df) contains 2 NA values.
knn() should impute all of them.
Checks that the original data frame remains unmodified and that the
returned data frame contains no NA values.
"""
# 1. Arrange
df = generate_example_df()
# 2. Act
df2 = knn(df)
# 3. Assert
self.assertEqual(df.isna().sum().sum(), 2)
self.assertEqual(df2.isna().sum().sum(), 0)