def test_valid_input(self):
self.assertEqual(soft_tfidf(['a', 'b', 'a'], ['a', 'c'], [['a', 'b', 'a'], ['a', 'c'], ['a']], sim_func=jaro,
threshold=0.8), 0.17541160386140586)
self.assertEqual(soft_tfidf(['a', 'b', 'a'], ['a'], [['a', 'b', 'a'], ['a', 'c'], ['a']],
threshold=0.9), 0.5547001962252291)
self.assertEqual(soft_tfidf(['a', 'b', 'a'], ['a'], [['x', 'y'], ['w'], ['q']]), 0.0)
self.assertEqual(soft_tfidf(['aa', 'bb', 'a'], ['ab', 'ba'], sim_func=affine, threshold=0.6),
0.81649658092772592)
def test_valid_input(self):
self.assertEqual(
soft_tfidf(['a', 'b', 'a'], ['a', 'c'],
[['a', 'b', 'a'], ['a', 'c'], ['a']],
sim_func=jaro,
threshold=0.8), 0.17541160386140586)
self.assertEqual(
soft_tfidf(['a', 'b', 'a'], ['a'],
[['a', 'b', 'a'], ['a', 'c'], ['a']],
threshold=0.9), 0.5547001962252291)
self.assertEqual(
soft_tfidf(['a', 'b', 'a'], ['a'], [['x', 'y'], ['w'], ['q']]),
0.0)
self.assertEqual(
soft_tfidf(['aa', 'bb', 'a'], ['ab', 'ba'],
sim_func=affine,
threshold=0.6), 0.81649658092772592)
def test_invalid_input3(self):
soft_tfidf(None, None)
def test_invalid_input2(self):
soft_tfidf(None, ['b'])
def test_invalid_input4(self):
soft_tfidf(['a'], None)
def test_invalid_input1(self):
soft_tfidf(1, 1)
def test_invalid_input3(self):
soft_tfidf(None, None)
def test_invalid_input2(self):
soft_tfidf(None, ['b'])
def test_invalid_input4(self):
soft_tfidf(['a'], None)
def test_invalid_input1(self):
soft_tfidf(1, 1)
def time_medium_large_wi_rep(self):
simfunctions.soft_tfidf(_med_num_tokens_wo_rep, _large_num_tokens_wo_rep, corpus_list=self.corpus_list)
def time_small_large_wi_rep_no_corpus(self):
simfunctions.soft_tfidf(_small_num_tokens_wi_rep, _large_num_tokens_wi_rep)
def time_small_medium_wi_rep_no_corpus(self):
simfunctions.soft_tfidf(_small_num_tokens_wi_rep, _med_num_tokens_wi_rep)
def time_small_large_wi_rep(self):
simfunctions.soft_tfidf(_small_num_tokens_wi_rep, _large_num_tokens_wi_rep, corpus_list=self.corpus_list)
def time_small_medium_wo_rep(self):
simfunctions.soft_tfidf(_small_num_tokens_wo_rep, _med_num_tokens_wo_rep, corpus_list=self.corpus_list)
def time_medium_large_wi_rep_no_corpus(self):
simfunctions.soft_tfidf(_med_num_tokens_wo_rep, _large_num_tokens_wo_rep)
