Skip to content

Commit

Permalink
[Tokenizer] Support adding special tokens to Qwen tokenizer (#9344)
Browse files Browse the repository at this point in the history
* update qwen tokenizer

* add test case
  • Loading branch information
DrownFish19 authored Nov 1, 2024
1 parent 149cfa8 commit 71dafa6
Show file tree
Hide file tree
Showing 3 changed files with 76 additions and 5 deletions.
33 changes: 30 additions & 3 deletions paddlenlp/transformers/qwen/tokenizer.py
Original file line number Diff line number Diff line change
Expand Up @@ -125,14 +125,41 @@ def convert_tokens_to_ids(self, tokens: Union[bytes, str, List[Union[bytes, str]
ids.append(self.mergeable_ranks.get(token))
return ids

def _update_tiktoken(self, tokens: List[str], special_tokens: bool = False) -> int:
if special_tokens:
added_tokens = []
for token in tokens:
if token in self.special_tokens:
continue

token_id = len(self.mergeable_ranks) + len(self.special_tokens)
self.special_tokens[token] = token_id
self.decoder[token_id] = token

added_tokens.append(token)

import tiktoken

self.tokenizer = tiktoken.Encoding(
"Qwen",
pat_str=PAT_STR,
mergeable_ranks=self.mergeable_ranks,
special_tokens=self.special_tokens,
)

return len(added_tokens)
else:
raise ValueError("Adding regular tokens is not supported")

def _add_tokens(self, new_tokens: Union[List[str], List[AddedToken]], special_tokens: bool = False) -> int:
if not special_tokens and new_tokens:
raise ValueError("Adding regular tokens is not supported")
new_tokens_str = []
for token in new_tokens:
surface_form = token.content if isinstance(token, AddedToken) else token
if surface_form not in SPECIAL_TOKENS:
raise ValueError("Adding unknown special tokens is not supported")
return 0
new_tokens_str.append(surface_form)

return self._update_tiktoken(new_tokens_str, special_tokens)

def save_vocabulary(self, save_directory: str, **kwargs) -> Tuple[str]:
"""
Expand Down
2 changes: 0 additions & 2 deletions paddlenlp/utils/import_utils.py
Original file line number Diff line number Diff line change
Expand Up @@ -52,7 +52,6 @@ def _is_package_available(pkg_name: str, return_version: bool = False) -> Union[
else:
# For packages other than "torch", don't attempt the fallback and set as not available
package_exists = False
logger.debug(f"Detected {pkg_name} version: {package_version}")
if return_version:
return package_exists, package_version
else:
Expand Down Expand Up @@ -96,7 +95,6 @@ def _is_package_available(pkg_name: str, return_version: bool = False) -> Union[
else:
# For packages other than "torch", don't attempt the fallback and set as not available
package_exists = False
logger.debug(f"Detected {pkg_name} version: {package_version}")
if return_version:
return package_exists, package_version
else:
Expand Down
46 changes: 46 additions & 0 deletions tests/transformers/qwen/test_tokenizer.py
Original file line number Diff line number Diff line change
@@ -0,0 +1,46 @@
# Copyright (c) 2024 PaddlePaddle Authors. All Rights Reserved.
# Copyright 2024 The Qwen team, Alibaba Group and the HuggingFace Team. All rights reserved.
#
# Licensed under the Apache License, Version 2.0 (the "License");
# you may not use this file except in compliance with the License.
# You may obtain a copy of the License at
#
# http://www.apache.org/licenses/LICENSE-2.0
#
# Unless required by applicable law or agreed to in writing, software
# distributed under the License is distributed on an "AS IS" BASIS,
# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
# See the License for the specific language governing permissions and
# limitations under the License.


import unittest

from paddlenlp.transformers import QWenTokenizer


class Qwen2TokenizationTest(unittest.TestCase):
from_pretrained_id = "qwen/qwen-7b"
tokenizer_class = QWenTokenizer
test_slow_tokenizer = True
space_between_special_tokens = False
from_pretrained_kwargs = None
test_seq2seq = False

def setUp(self):
super().setUp()

def get_tokenizer(self, **kwargs):
return QWenTokenizer.from_pretrained(self.from_pretrained_id, **kwargs)

def test_add_special_tokens(self):
tokenizer = self.get_tokenizer()
origin_tokens_len = len(tokenizer)

add_tokens_num = tokenizer.add_special_tokens({"additional_special_tokens": ["<img>"]})
assert add_tokens_num == 1
assert len(tokenizer) == origin_tokens_len + 1

add_tokens_num = tokenizer.add_special_tokens({"unk_token": "<unk>"})
assert add_tokens_num == 1
assert len(tokenizer) == origin_tokens_len + 2

0 comments on commit 71dafa6

Please sign in to comment.