def test_add_special_tokens(self, test_data_dir):
tokenizer = SentencePieceTokenizer(test_data_dir + self.model_name,
legacy=True)
special_tokens = MODEL_SPECIAL_TOKENS
tokenizer.add_special_tokens(special_tokens)
assert tokenizer.vocab_size == tokenizer.original_vocab_size + len(
set(special_tokens.values()))
def test_ids_to_text(self, test_data_dir):
tokenizer = SentencePieceTokenizer(test_data_dir + self.model_name)
special_tokens = MODEL_SPECIAL_TOKENS
tokenizer.add_special_tokens(special_tokens)
text = "[CLS] a b c [MASK] e f [SEP] g h i [SEP]"
ids = tokenizer.text_to_ids(text)
result = tokenizer.ids_to_text(ids)
assert text == result
def test_text_to_ids(self, test_data_dir):
tokenizer = SentencePieceTokenizer(test_data_dir + self.model_name)
special_tokens = MODEL_SPECIAL_TOKENS
tokenizer.add_special_tokens(special_tokens)
text = "[CLS] a b c [MASK] e f [SEP] g h i [SEP]"
ids = tokenizer.text_to_ids(text)
assert len(ids) == len(text.split())
assert ids.count(tokenizer.token_to_id("[CLS]")) == 1
assert ids.count(tokenizer.token_to_id("[MASK]")) == 1
assert ids.count(tokenizer.token_to_id("[SEP]")) == 2
def test_ids_to_tokens(self, test_data_dir):
tokenizer = SentencePieceTokenizer(test_data_dir + self.model_name)
special_tokens = MODEL_SPECIAL_TOKENS
tokenizer.add_special_tokens(special_tokens)
text = "[CLS] a b c [MASK] e f [SEP] g h i [SEP]"
tokens = tokenizer.text_to_tokens(text)
ids = tokenizer.tokens_to_ids(tokens)
result = tokenizer.ids_to_tokens(ids)
assert len(result) == len(tokens)
for i in range(len(result)):
assert result[i] == tokens[i]