def test_extract_feature_list(): res = ig.expand_data(day_lists, ig.min_val_tie, ig.min_val_tie, False) label_list, feature_list = ig.extract_feature_list(res) # check size for i in feature_list: assert len(feature_list[i]) == 2 # check labels assert label_list[1] == 0.5 assert label_list[2] == 0.4 assert label_list[3] == 0.35 assert label_list[4] == 0.21 assert label_list[5] == -0.79 assert label_list[6] == -0.79 # check features assert feature_list[1][0] == 0.31 # node '1' has score 0.31 on previous day and 0.11 on the day before the previous day assert feature_list[1][1] == 0.11 assert feature_list[2][0] == 0.6 assert feature_list[2][1] == 0.44 assert feature_list[3][0] == -0.89 assert feature_list[3][1] == -0.95 assert feature_list[4][0] == 0.35 assert feature_list[4][1] == 0.65 assert feature_list[5][0] == 0.11 assert feature_list[5][1] == 0.25 assert feature_list[6][0] == -0.89 assert feature_list[6][1] == 0.05
def test_extract_feature_list(): res = ig.expand_data(day_lists, ig.min_val_tie, ig.min_val_tie, False) label_list, feature_list = ig.extract_feature_list(res) # check size for i in feature_list: assert len(feature_list[i]) == 2 # check labels assert label_list[1] == 0.5 assert label_list[2] == 0.4 assert label_list[3] == 0.35 assert label_list[4] == 0.21 assert label_list[5] == -0.79 assert label_list[6] == -0.79 # check features assert feature_list[1][0] == 0.31 assert feature_list[1][1] == 0.11 assert feature_list[2][0] == 0.6 assert feature_list[2][1] == 0.44 assert feature_list[3][0] == -0.89 assert feature_list[3][1] == -0.95 assert feature_list[4][0] == 0.35 assert feature_list[4][1] == 0.65 assert feature_list[5][0] == 0.11 assert feature_list[5][1] == 0.25 assert feature_list[6][0] == -0.89 assert feature_list[6][1] == 0.05
def test_expand_data_avg_for_train(): res = ig.expand_data(day_lists, ig.average_tie, ig.average_tie, True) # check size for l in res: assert 5 == len(l) # check appended values assert abs(res[0][5] - 4.5) < epsilon assert abs(res[0][6] - 4.5) < epsilon assert False == (3 in res[1]) assert abs(res[1][6] - 5.0) < epsilon assert False == (3 in res[2])
def test_expand_data_avg(): res = ig.expand_data(day_lists, ig.average_tie, ig.average_tie, False) # check size for l in res: assert 6 == len(l) # check appended values assert abs(res[0][5] - 5.5) < epsilon # (5 + 6) / 2 = 5.5 assert abs(res[0][6] - 5.5) < epsilon assert abs(res[1][3] - 5.5) < epsilon assert abs(res[1][6] - 5.5) < epsilon assert abs(res[2][3] - 6) < epsilon # 6 / 1 = 6
def test_expand_data_minval(): res = ig.expand_data(day_lists, ig.min_val_tie, ig.min_val_tie, False) # check size for l in res: assert 6 == len(l) # check appended values assert abs(res[0][5] + 0.79) < epsilon # 0.21 - 1 = -0.79 assert abs(res[0][6] + 0.79) < epsilon assert abs(res[1][3] + 0.89) < epsilon # 0.11 - 1 = -0.89 assert abs(res[1][6] + 0.89) < epsilon assert abs(res[2][3] + 0.95) < epsilon
def test_expand_data_avg(): res = ig.expand_data(day_lists, ig.average_tie, ig.average_tie, False) # check size for l in res: assert 6 == len(l) # check appended values assert abs(res[0][5] - 5.5) < epsilon assert abs(res[0][6] - 5.5) < epsilon assert abs(res[1][3] - 5.5) < epsilon assert abs(res[1][6] - 5.5) < epsilon assert abs(res[2][3] - 6) < epsilon
def test_expand_data_minval(): res = ig.expand_data(day_lists, ig.min_val_tie, ig.min_val_tie, False) # check size for l in res: assert 6 == len(l) # check appended values assert abs(res[0][5] + 0.79) < epsilon assert abs(res[0][6] + 0.79) < epsilon assert abs(res[1][3] + 0.89) < epsilon assert abs(res[1][6] + 0.89) < epsilon assert abs(res[2][3] + 0.95) < epsilon
def test_expand_data_minval_for_train(): res = ig.expand_data(day_lists, ig.min_val_tie, ig.min_val_tie, True) assert False == (3 in res[1]) assert False == (3 in res[2])