def test_create_cache_files_after_first_set(self):
self.assertTrue(
os.path.isfile(
os.path.join(bayes.BASE_DIR,
'all_data/Chinese/cache/vector.cache')))
folder = os.path.join(bayes.BASE_DIR, 'all_data/Chinese/cache')
for the_file in os.listdir(folder):
file_path = os.path.join(folder, the_file)
try:
if os.path.isfile(file_path):
os.unlink(file_path)
except Exception as e:
pass
test_bayes = bayes.Classify(cache=False)
self.assertTrue(
os.path.isfile(
os.path.join(bayes.BASE_DIR,
'all_data/Chinese/cache/vector.cache')))
self.assertTrue(
os.path.isfile(
os.path.join(bayes.BASE_DIR,
'all_data/Chinese/cache/vocab.cache')))
self.assertTrue(
os.path.isfile(
os.path.join(bayes.BASE_DIR,
'all_data/Chinese/cache/classify.cache')))
def test_cache_work(self):
test_bayes = bayes.Classify(cache=False)
percentage_list, word_list = test_bayes.bayes_classify(
'美联储当天结束货币政策例会后发表声明说，自2017年12月以来，' +
'美国就业市场和经济活动继续保持稳健增长，失业率继续维持在低水平。')
test_result = sorted(percentage_list, key=lambda x: x[1],
reverse=True)[0][0]
self.assertEqual(test_result, 'normal.dat')
def test_have_default_cache_files(self):
bayes.Classify(cache=False)
self.assertTrue(
os.path.isfile(
os.path.join(bayes.BASE_DIR,
'all_data/Chinese/cache/vector.cache')))
self.assertTrue(
os.path.isfile(
os.path.join(bayes.BASE_DIR,
'all_data/Chinese/cache/classify.cache')))
self.assertTrue(
os.path.isfile(
os.path.join(bayes.BASE_DIR,
'all_data/Chinese/cache/vocab.cache')))
def test_error_rate(self):
'''
test error rate
'''
def error_rate(instance):
classify_results = []
for i in range(len(instance.test_data)):
percentage_list, word_list = (instance.bayes_classify(
instance.test_data[i]))
test_result = sorted(percentage_list,
key=lambda x: x[1],
reverse=True)[0][0]
classify_results.append(test_result)
# Uncomment to see which sentence was classified wrong.
# if test_result != instance.test_classify[i]:
# print('-'*20)
# print(instance.test_data[i])
# print('test_result is %s' % test_result)
# print('true is %s' % instance.test_classify[i])
# print('percentage_list is %s' % percentage_list)
# print('-'*20)
wrong_results = [
i for i, j in zip(instance.test_classify, classify_results)
if i != j
]
return len(wrong_results) / len(instance.test_data)
a = []
test_times = 20
print('\nThis may takes some time')
for i in range(test_times):
if i % 5 == 0:
print('Completed %s tasks, %s tasks left.' %
(i, test_times - i))
test_bayes = bayes.Classify(cache=False)
a.append(error_rate(test_bayes))
print('The error rate is %s' %
"{0:.2f}".format(sum(a) / test_times * 100) + '%')
def test_data_num_correct(self):
test_bayes = bayes.Classify(test_num=80, cache=False)
self.assertTrue(len(test_bayes.test_data), 80)