def test_parse_dataset(self):
itemsets = parse.parse_dataset()
assert itemsets and type(itemsets) is list
itemset = itemsets[0]
assert itemset and type(itemset) is frozenset
item = random.choice(tuple(itemset))
assert item and type(item) is tuple
def test_apriori_gen(self):
dataset = parse.parse_dataset()
frequent_itemsets = generate.frequent_itemsets(0.05, dataset)
l = frequent_itemsets[0]
k = 2
C = generate.apriori_gen(l, k)
assert type(C) is dict
assert not C or random.choice(list(C.values())) is 0
def test_association_rules(self):
dataset = parse.parse_dataset()
frequent_itemsets = generate.frequent_itemsets(0.05, dataset)
association_rules = generate.association_rules(dataset, frequent_itemsets, utils.Metric.Confidence, 0.3)
assert association_rules and type(association_rules) is list
rule = association_rules[0]
assert type(rule[0]) is tuple
assert type(rule[1]) is float
assert type(rule[2]) is float
def test_frequent_itemsets(self):
dataset = parse.parse_dataset()
frequent_itemsets = generate.frequent_itemsets(0.05, dataset)
assert frequent_itemsets
assert len(frequent_itemsets) is len(dataset[0])
assert type(frequent_itemsets) is list
one_itemsets = frequent_itemsets[0]
assert type(one_itemsets) is set
assert len(one_itemsets) is 0 or type(random.sample(one_itemsets, 1)[0]) is frozenset
def main():
support_threshold, metric, metric_threshold, filepath = interact.get_input()
dataset = parse.parse_dataset(filepath)
frequent_itemsets = generate.frequent_itemsets(support_threshold, dataset)
interact.print_frequent_itemsets(frequent_itemsets, LOG_FILE)
association_rules = generate.association_rules(dataset, frequent_itemsets, metric, metric_threshold)
interact.print_association_rules(association_rules, LOG_FILE, metric)
print("Log file written at {}".format(os.path.join(os.getcwd(), LOG_FILE)))