Skip to content

Commit 56a22e0

Browse files
committed
feat: Specify the directory for calculating tokens models
1 parent f457588 commit 56a22e0

File tree

2 files changed

+3
-4
lines changed

2 files changed

+3
-4
lines changed

apps/common/config/tokenizer_manage_config.py

Lines changed: 2 additions & 3 deletions
Original file line numberDiff line numberDiff line change
@@ -6,7 +6,6 @@
66
@date:2024/4/28 10:17
77
@desc:
88
"""
9-
import os
109
from pathlib import Path
1110

1211
BASE_DIR = Path(__file__).resolve().parent.parent.parent
@@ -27,6 +26,6 @@ class TokenizerManage:
2726
def get_tokenizer():
2827
from tokenizers import Tokenizer
2928
# 创建Tokenizer
30-
s = os.path.join(BASE_DIR.parent, 'tokenizer', 'bert-base-cased', 'tokenizer.json')
31-
TokenizerManage.tokenizer = Tokenizer.from_file(s)
29+
TokenizerManage.tokenizer = Tokenizer.from_file(
30+
'/opt/maxkb-app/model/tokenizer/models--bert-base-cased/snapshots/cd5ef92a9fb2f889e972770a36d4ed042daf221e')
3231
return MKTokenizer(TokenizerManage.tokenizer)

apps/maxkb/wsgi/web.py

Lines changed: 1 addition & 1 deletion
Original file line numberDiff line numberDiff line change
@@ -32,7 +32,7 @@ def __call__(self, name, *args, **kwargs):
3232
builtins.__import__ = TorchBlocker()
3333

3434
os.environ.setdefault('DJANGO_SETTINGS_MODULE', 'maxkb.settings')
35-
35+
os.environ['TIKTOKEN_CACHE_DIR'] = '/opt/maxkb-app/model/tokenizer/openai-tiktoken-cl100k-base'
3636
application = get_wsgi_application()
3737

3838

0 commit comments

Comments
 (0)