def test_grow_rule_1(self):
df = pd.read_csv('tests/test_files/testfile_4.csv',
encoding='utf-8',
delimiter=';')
ds = DictDataset(0, df)
rule = ds.grow_rule()
self.assertEqual("Age < 8 and Sex in ['Female']", rule.to_string())
def test_split_into_growset_pruneset_1(self):
df = pd.read_csv('tests/test_files/testfile_8.csv',
encoding='utf-8',
delimiter=';')
ds = DictDataset(1, df)
grow, prune = ds.split_into_growset_pruneset()
self.assertEqual(21, grow.length())
self.assertEqual(11, prune.length())
def test_find_best_num_literal_3(self):
df = pd.read_csv('tests/test_files/testfile_3.csv',
encoding='utf-8',
delimiter=';')
ds = DictDataset(1, df)
best_l, best_foil = ds.find_best_literal(0, 0, df['Age'].unique(),
'Age')
self.assertEqual(10, round(best_foil, 5))
self.assertEqual('Age > 88', best_l.to_string())
def test_grow_rule_2(self):
df = pd.read_csv('tests/test_files/mushroom.csv',
encoding='utf-8',
delimiter=';')
ds = DictDataset(0, df)
rule = ds.grow_rule()
rule = ds.make_rule(rule)
self.assertEqual("a5 in ['c', 'f', 'm', 'p', 's', 'y']",
rule.to_string())
def test_literal_count_p_n2(self):
df = pd.read_csv('tests/test_files/testfile_8.csv',
encoding='utf-8',
delimiter=';')
l = Literal('ClassOfSeat', 'in', ['xab', 'asdas'])
ds = DictDataset(1, df)
p, n = ds.count_p_n_literal(l)
self.assertEqual(0, p)
self.assertEqual(0, n)
def test_literal_count_p_n3(self):
df = pd.read_csv('tests/test_files/testfile_8.csv',
encoding='utf-8',
delimiter=';')
l = Literal('Age', '>', 1000)
ds = DictDataset(1, df)
p, n = ds.count_p_n_literal(l)
self.assertEqual(0, p)
self.assertEqual(4, n)
def test_literal_count_p_n4(self):
df = pd.read_csv('tests/test_files/testfile_8.csv',
encoding='utf-8',
delimiter=';')
l = Literal('Sex', 'in', 'Male')
ds = DictDataset(1, df)
p, n = ds.count_p_n_literal(l)
self.assertEqual(8, p)
self.assertEqual(8, n)
def test_find_best_char_literal_3(self):
df = pd.read_csv('tests/test_files/testfile_3.csv',
encoding='utf-8',
delimiter=';')
ds = DictDataset(1, df)
best_l, best_foil = ds.find_best_literal(0, 0, df['Sex'].unique(),
'Sex')
self.assertEqual(9.38889, round(best_foil, 5))
self.assertEqual("Sex in ['Female']", best_l.to_string())
def test_rule_count_p_n4(self):
df = pd.read_csv('tests/test_files/testfile_8.csv',
encoding='utf-8',
delimiter=';')
rule = Rule()
ds = DictDataset(1, df)
p, n = ds.count_p_n_rule(rule)
self.assertEqual(16, p)
self.assertEqual(16, n)
def test_find_best_char_literal_2(self):
df = pd.read_csv('tests/test_files/testfile_2.csv',
encoding='utf-8',
delimiter=';')
ds = DictDataset(1, df)
best_l, best_foil = ds.find_best_literal(0, 0,
df['ClassOfSeat'].unique(),
'ClassOfSeat')
self.assertEqual(9, round(best_foil, 5))
self.assertEqual("ClassOfSeat in ['1st', '3rd', 'Crew']",
best_l.to_string())
def test_prune_rule_2(self):
df = pd.read_csv('tests/test_files/mushroom.csv',
encoding='utf-8',
delimiter=';')
ds = DictDataset(0, df)
l = Literal('a5', 'in', ['c', 'f', 'm', 'p', 's', 'y'])
rule = Rule()
rule.add_literal(l)
rule = ds.prune_rule(rule)
self.assertEqual("a5 in ['c', 'f', 'm', 'p', 's', 'y']",
rule.to_string())
def test_prune_rule_1(self):
df = pd.read_csv('tests/test_files/testfile_3.csv',
encoding='utf-8',
delimiter=';')
l = Literal('Sex', 'in', 'Female')
l2 = Literal('Sex', 'in', 'Male')
rule = Rule()
rule.add_literal(l)
rule.add_literal(l2)
ds = DictDataset(1, df)
rule = ds.prune_rule(rule)
self.assertEqual("Sex in Female", rule.to_string())
def test_rule_count_p_n1(self):
df = pd.read_csv('tests/test_files/testfile_8.csv',
encoding='utf-8',
delimiter=';')
l = Literal('ClassOfSeat', 'in', '1st')
l2 = Literal('Age', '<', 20)
rule = Rule()
rule.add_literal(l)
rule.add_literal(l2)
ds = DictDataset(1, df)
p, n = ds.count_p_n_rule(rule)
self.assertEqual(1, p)
self.assertEqual(1, n)
def test_delete_not_covered_1(self):
df = pd.read_csv('tests/test_files/testfile_8.csv',
encoding='utf-8',
delimiter=';')
rule = Rule()
ds = DictDataset(1, df)
len_before = ds.length()
ds.delete_not_covered(rule)
len_after = ds.length()
self.assertEqual(0, len_after)
def test_delete_not_covered_3(self):
df = pd.read_csv('tests/test_files/testfile_8.csv',
encoding='utf-8',
delimiter=';')
l = Literal('ClassOfSeat', 'in', '1st')
l2 = Literal('Age', '<', 20)
rule = Rule()
rule.add_literal(l)
rule.add_literal(l2)
ds = DictDataset(1, df)
len_before = ds.length()
ds.delete_not_covered(rule)
len_after = ds.length()
self.assertEqual(len_before - 30, len_after)
def test_is_any_pos_example_false_all_n(self):
df = pd.read_csv('tests/test_files/testfile_6_all_n.csv',
encoding='utf-8',
delimiter=';')
ds = DictDataset(1, df)
self.assertEqual(False, ds.is_any_pos_example())
def test_length_dataset_2_no_rows(self):
df = pd.read_csv('tests/test_files/testfile_11_no_rows.csv',
encoding='utf-8',
delimiter=';')
ds = DictDataset(1, df)
self.assertEqual(0, ds.length())
def test_is_any_pos_example_true_all_p(self):
df = pd.read_csv('tests/test_files/testfile_10_all_p.csv',
encoding='utf-8',
delimiter=';')
ds = DictDataset(1, df)
self.assertEqual(True, ds.is_any_pos_example())
def test_length_dataset_4_all_n(self):
df = pd.read_csv('tests/test_files/testfile_6_all_n.csv',
encoding='utf-8',
delimiter=';')
ds = DictDataset(1, df)
self.assertEqual(32, ds.length())