From 33d78065c2296266d4650b4219239a7ddde24c2f Mon Sep 17 00:00:00 2001 From: ch-liuzhide Date: Tue, 4 Mar 2025 18:40:55 +0800 Subject: [PATCH 1/2] refactor(rag): update rag params --- server/agent/tools/knowledge.py | 2 +- server/rag/router.py | 4 ++-- 2 files changed, 3 insertions(+), 3 deletions(-) diff --git a/server/agent/tools/knowledge.py b/server/agent/tools/knowledge.py index 332c7f70..bdecf8c5 100644 --- a/server/agent/tools/knowledge.py +++ b/server/agent/tools/knowledge.py @@ -32,7 +32,7 @@ async def search_knowledge( space_id_list=[bot_id, repo_name], question=query, embedding_model_name=EmbeddingModelEnum.OPENAI, - similarity_threshold=0.65, + similarity_threshold=0.6, top=10, metadata_filter={}, ) diff --git a/server/rag/router.py b/server/rag/router.py index f282f1bd..ea9bc4f2 100644 --- a/server/rag/router.py +++ b/server/rag/router.py @@ -60,8 +60,8 @@ async def reload_repo( repo_name=request.repo_name, auth_token=user.access_token ), split_config=KnowledgeSplitConfig( - chunk_size=500, - chunk_overlap=100, + chunk_size=1000, + chunk_overlap=200, ), ) ] From 450fa580c853cba4c896e536434ea7074b24efa3 Mon Sep 17 00:00:00 2001 From: ch-liuzhide Date: Tue, 4 Mar 2025 18:49:02 +0800 Subject: [PATCH 2/2] refactor(api): add user Depends for list api --- server/rag/router.py | 27 ++++++++++++++++++++++++--- 1 file changed, 24 insertions(+), 3 deletions(-) diff --git a/server/rag/router.py b/server/rag/router.py index ea9bc4f2..7ba0006a 100644 --- a/server/rag/router.py +++ b/server/rag/router.py @@ -75,7 +75,14 @@ async def reload_repo( @router.post("/knowledge/list", dependencies=[Depends(verify_rate_limit)]) -async def get_knowledge_list(params: PageParams[Knowledge]): +async def get_knowledge_list( + params: PageParams[Knowledge], + user: Annotated[User | None, Depends(get_user)] = None, +): + if user is None: + raise HTTPException( + status_code=status.HTTP_401_UNAUTHORIZED, detail="Github Login needed" + ) try: api_client = APIClient( base_url=get_env_variable("WHISKER_API_URL"), @@ -101,7 +108,14 @@ async def get_chunk_list(params: PageParams[Chunk]): @router.post("/task/list", dependencies=[Depends(verify_rate_limit)]) -async def get_rag_task(params: PageParams[Task]): +async def get_rag_task( + params: PageParams[Task], + user: Annotated[User | None, Depends(get_user)] = None, +): + if user is None: + raise HTTPException( + status_code=status.HTTP_401_UNAUTHORIZED, detail="Github Login needed" + ) try: api_client = APIClient( base_url=get_env_variable("WHISKER_API_URL"), @@ -114,7 +128,14 @@ async def get_rag_task(params: PageParams[Task]): @router.post("/task/restart", dependencies=[Depends(verify_rate_limit)]) -async def restart_rag_task(params: RestartTaskRequest): +async def restart_rag_task( + params: RestartTaskRequest, + user: Annotated[User | None, Depends(get_user)] = None, +): + if user is None: + raise HTTPException( + status_code=status.HTTP_401_UNAUTHORIZED, detail="Github Login needed" + ) try: api_client = APIClient( base_url=get_env_variable("WHISKER_API_URL"),