def test1_smallset(self):
path_input = "test/data/schema_matching/default_matches_cities_input.csv"
df = pd.read_csv(path_input)
pairs_relational = relational_matching(df)[["uri_1","uri_2"]].sort_values(by=["uri_1","uri_2"]).reset_index(drop=True)
pairs_string = string_similarity_matching(df)[["uri_1","uri_2"]].sort_values(by=["uri_1","uri_2"]).reset_index(drop=True)
pairs_schema = label_schema_matching(df)[["uri_1","uri_2"]].sort_values(by=["uri_1","uri_2"]).reset_index(drop=True)
pairs_overlap = value_overlap_matching(df)[["uri_1","uri_2"]].sort_values(by=["uri_1","uri_2"]).reset_index(drop=True)
assert all([pairs_relational.equals(x) for x in [pairs_string, pairs_schema, pairs_overlap]])
def test4_uri_querier_no_matches(self):
path_input = "test/data/schema_matching/no_matches_cities_boolean_input.csv"
df = pd.read_csv(path_input)
path_expected = "test/data/schema_matching/no_matches_cities_boolean_expected.csv"
expected_matches = pd.read_csv(path_expected)
output_matches = label_schema_matching(df, uri_data_model=True)
output_matches['same_label'] = pd.to_numeric(output_matches['same_label'])
pd.testing.assert_frame_equal(
output_matches, expected_matches, check_like=True)
def test1_default(self):
path_input = "test/data/schema_matching/default_matches_cities_boolean_input.csv"
df = pd.read_csv(path_input)
path_expected = "test/data/schema_matching/default_matches_cities_boolean_expected.csv"
expected_matches = pd.read_csv(path_expected)
output_matches = label_schema_matching(df)
output_matches['same_label'] = pd.to_numeric(output_matches['same_label'])
pd.testing.assert_frame_equal(
output_matches, expected_matches, check_like=True)
def test2_bigset(self):
#WARNING: Takes long to run!
path_input = "test/data/schema_matching/pair_equality_test2_bigset.csv"
df = pd.read_csv(path_input)
pairs_relational = relational_matching(df)[["uri_1","uri_2"]].sort_values(by=["uri_1","uri_2"]).reset_index(drop=True)
pairs_string = string_similarity_matching(df)[["uri_1","uri_2"]].sort_values(by=["uri_1","uri_2"]).reset_index(drop=True)
pairs_schema = label_schema_matching(df)[["uri_1","uri_2"]].sort_values(by=["uri_1","uri_2"]).reset_index(drop=True)
pairs_overlap = value_overlap_matching(df)[["uri_1","uri_2"]].sort_values(by=["uri_1","uri_2"]).reset_index(drop=True)
assert all([pairs_relational.equals(x) for x in [pairs_string, pairs_schema, pairs_overlap]])