weiweiw
/
Langchain-Chatchat
1
0
Fork 0
Code Issues Pull Requests Packages Projects Releases Wiki Activity

manually split

This commit is contained in:
wvivi2023 2023-12-15 08:59:13 +08:00
parent bf21b8f116
commit 77bc5891c8
3 changed files with 5 additions and 4 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

BIN
.DS_Store vendored
View File

Binary file not shown.

2
server/chat/knowledge_base_chat.py
View File

@ -39,7 +39,7 @@ async def knowledge_base_chat(query: str = Body(..., description="用户输入",
return BaseResponse(code=404, msg=f"未找到知识库 {knowledge_base_name}") return BaseResponse(code=404, msg=f"未找到知识库 {knowledge_base_name}")
history = [History.from_data(h) for h in history] history = [History.from_data(h) for h in history]
print(f"server/chat/knowledge_base_chat function, history:{history}") print(f"******server/chat/knowledge_base_chat function, history:{history}")
async def knowledge_base_chat_iterator(query: str, async def knowledge_base_chat_iterator(query: str,
top_k: int, top_k: int,
history: Optional[List[History]], history: Optional[List[History]],

7
text_splitter/chinese_recursive_text_splitter.py
View File

@ -57,12 +57,13 @@ class ChineseRecursiveTextSplitter(RecursiveCharacterTextSplitter):
new_separators = [] new_separators = []
text = re.sub(r'(\n+前\s+言\n+)', r"\n\n\n\n\n\n\n\n\n\n\1", text) #通过前言分块 text = re.sub(r'(\n+前\s+言\n+)', r"\n\n\n\n\n\n\n\n\n\n\1", text) #通过前言分块
text = re.sub(r'(\n+\d+[^\S\n]+[^\s\.]+)', r"\n\n\n\n\n\n\n\n\n\n\1", text) #通过1 这样的 text = re.sub(r'(\n+\d+[^\S\n]+[^\s\.]+)', r"\n\n\n\n\n\n\n\n\n\n\1", text) #通过1 这样的
text = re.sub(r'(\n+[a-zA-Z1-9]+\s*\.\s*[a-zA-Z1-9]+\s+(?!\.|[a-zA-Z1-9]))', r"\n\n\n\n\n\n\n\n\n\n\1", text) # 通过1.2 这样的章和节来分块 text = re.sub(r'(\n+[a-zA-Z0-9]+\s*\.\s*[a-zA-Z0-9]+\s+(?!\.|[a-zA-Z0-9]))', r"\n\n\n\n\n\n\n\n\n\n\1", text) # 通过\n1.2 这样的章和节来分块
text = re.sub(r'(\n+表\s*[A-Za-z0-9]+(\.[A-Za-z0-9]+)+\s+)', r"\n\n\n\n\n\n\n\n\n\n\1", text) # 通过表 A.4.a text = re.sub(r'(\n+表\s*[A-Za-z0-9]+(\.[A-Za-z0-9]+)+\s+)', r"\n\n\n\n\n\n\n\n\n\n\1", text) # 通过表 A.4.a
text = re.sub(r'(\n+第\s*\S+\s*条\s+)', r"\n\n\n\n\n\n\n\n\n\n\1", text) # 通过第 条 text = re.sub(r'(\n+第\s*\S+\s*条\s+)', r"\n\n\n\n\n\n\n\n\n\n\1", text) # 通过第 条
text = re.sub(r'(\n+第\s*\S+\s*章\s+)', r"\n\n\n\n\n\n\n\n\n\n\1", text) # 通过第 条 text = re.sub(r'(\n+第\s*\S+\s*章\s+)', r"\n\n\n\n\n\n\n\n\n\n\1", text) # 通过第 条
text = re.sub(r'(\n+(一、|二、|三、|四、|五、|六、|七、|八、|九、|十、|十一、|十二、|十三、|十四、|十五、|十六、|十七、|十八、|十九、|二十、))', r"\n\n\n\n\n\n\n\n\n\n\1", text) # 通过第 条 text = re.sub(r'(\n+(一、|二、|三、|四、|五、|六、|七、|八、|九、|十、|十一、|十二、|十三、|十四、|十五、|十六、|十七、|十八、|十九、|二十、))', r"\n\n\n\n\n\n\n\n\n\n\1", text) # 通过第 条
text = re.sub(r'(\n+[a-zA-Z1-9]+\s*\.\s*[a-zA-Z1-9]+\s+)', r"\n\n\n\n\n\n\n\n\n\n\1", text) # 再通过 1.2 来分块 text = re.sub(r'(手工分段\*\*\s+)', r"\n\n\n\n\n\n\n\n\n\n", text) # 通过“手工分段**”
#text = re.sub(r'(\n+[a-zA-Z0-9]+\s*\.\s*[a-zA-Z0-9]+\s+)', r"\n\n\n\n\n\n\n\n\n\n\1", text) # 再通过 1.2 来分块
text = text.rstrip() # 段尾如果有多余的\n就去掉它 text = text.rstrip() # 段尾如果有多余的\n就去掉它
for i, _s in enumerate(separators): for i, _s in enumerate(separators):
_separator = _s if self._is_separator_regex else re.escape(_s) _separator = _s if self._is_separator_regex else re.escape(_s)
@ -91,7 +92,7 @@ class ChineseRecursiveTextSplitter(RecursiveCharacterTextSplitter):
if not new_separators: if not new_separators:
final_chunks.append(s) final_chunks.append(s)
else: else:
s = re.sub(r'(\n+[a-zA-Z1-9]+\s*\.\s*[a-zA-Z1-9]+\s*\.\s*[a-zA-Z1-9]+\s+)', r"\n\n\n\n\n\n\n\n\n\n\1", s) # 再通过 1.2.3 来分块 s = re.sub(r'(\n+[a-zA-Z0-9]+\s*\.\s*[a-zA-Z0-9]+\s*\.\s*[a-zA-Z0-9]+\s+)', r"\n\n\n\n\n\n\n\n\n\n\1", s) # 再通过 1.2.3 来分块
other_info = self._split_text(s, new_separators) other_info = self._split_text(s, new_separators)
final_chunks.extend(other_info) final_chunks.extend(other_info)
if _good_splits: if _good_splits: