def test_fit_attributes(df_duplicate_features):
transformer = DropDuplicateFeatures()
transformer.fit(df_duplicate_features)
assert transformer.features_to_drop_ == {"dob", "dob3", "City2", "Age2"}
assert transformer.duplicated_feature_sets_ == [
{"dob", "dob2", "dob3"},
{"City", "City2"},
{"Age", "Age2"},
]
def test_variables_assigned_correctly(df_duplicate_features):
transformer = DropDuplicateFeatures()
assert transformer.variables is None
transformer.fit(df_duplicate_features)
assert transformer.variables == (list(df_duplicate_features.columns))
def fake_columns(var_list,df):
dupis = DropDuplicateFeatures()
dupis_train = dupis.fit(df[var_list])
duplicates_train = list(dupis_train.features_to_drop_)
return duplicates_train