def measure(sequences, min_sup=0.5):
time_result = []
for i in range(10):
bitmaps_for_words_ids = generate_words_bitmaps(sequences)
spam_alg = SPAM(min_sup, bitmaps_for_words_ids)
time_result.append(timeit.timeit(spam_alg.spam, number = 1))
mean = sum(time_result)/len(time_result)
return mean
def start(minsup, data_path):
sequences = DataSequence.data_sequence_factory(customers=TEST['customers'],
texts=TEST['texts'],
path=data_path)
bitmaps_for_words_ids = generate_words_bitmaps(sequences)
spam_alg = SPAM(minsup, bitmaps_for_words_ids)
frequent_patterns = spam_alg.spam()
with open("frequent_patterns", 'w') as file:
for pattern in translate_patterns(frequent_patterns):
file.write(str(pattern) + '\n')
def measure_spam(number_of_items, number_of_sequences,
number_of_customers, min_items_in_transaction,
max_items_in_transaction, min_sup):
time_result = []
for i in range(10):
#print("probka {}".format(i))
sequences = generate_test_sequeneces(
number_of_items, number_of_sequences, number_of_customers,
min_items_in_transaction, max_items_in_transaction)
bitmaps_for_words_ids = generate_words_bitmaps(sequences)
spam_alg = SPAM(min_sup, bitmaps_for_words_ids)
# start = time.time()
# spam_alg.spam()
# end = time.time()
# result = end - start
time_result.append(timeit.timeit(spam_alg.spam, number = 1))
mean = sum(time_result)/len(time_result)
return mean
from read_data import generate_test_sequeneces
from bitmap import generate_words_bitmaps
import timeit
import matplotlib.pyplot as plt
def translate_patterns(frequent_patterns):
translated_patterns = []
for pattern in frequent_patterns:
translated_sequence = []
for itemset in pattern[0]:
translated_itemset = []
for item in itemset:
translated_itemset.append(DataSequence.get_words(item))
translated_sequence.append(translated_itemset)
translated_patterns.append(translated_sequence)
return translated_patterns
if __name__ == "__main__":
sequences = generate_simple_sequeneces()
bitmaps_for_words_ids = generate_words_bitmaps(sequences)
#print(bitmaps_for_words_ids)
spam_alg = SPAM(0.5, bitmaps_for_words_ids)
frequent_patterns = spam_alg.spam()
with open("frequent_patterns", 'w') as file:
for pattern in translate_patterns(frequent_patterns):
file.write(str(pattern) + '\n')
def measure_memory(sequences, min_sup=0.5):
bitmaps_for_words_ids = generate_words_bitmaps(sequences)
spam_alg = SPAM(min_sup, bitmaps_for_words_ids)
spam_alg.spam()