tiktoken做lazyload处理

This commit is contained in:
Your Name
2023-04-19 14:27:34 +08:00
parent 28aa6d1dc0
commit b0409b929b
12 changed files with 83 additions and 35 deletions

View File

@@ -11,9 +11,8 @@ class PaperFileGroup():
self.sp_file_tag = []
# count_token
import tiktoken
from toolbox import get_conf
enc = tiktoken.encoding_for_model("gpt-3.5-turbo")
from request_llm.bridge_all import model_info
enc = model_info["gpt-3.5-turbo"]['tokenizer']
def get_token_num(txt): return len(enc.encode(txt, disallowed_special=()))
self.get_token_num = get_token_num