yujunjun/LightRAG
1
0
Fork 0
mirror of https://github.com/HKUDS/LightRAG.git synced 2025-11-23 05:26:38 +00:00
Code Issues Packages Projects Releases Wiki Activity

Merge pull request #1996 from danielaskdd/key-work-empty

Browse Source 
feat: Improve Empty Keyword Handling logic
This commit is contained in:
Daniel.y 2025-08-23 11:54:26 +08:00 committed by GitHub
commit df4927f837
No known key found for this signature in database
GPG Key ID: B5690EEEBB952194
2 changed files with 29 additions and 29 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

2
env.example
View File

@ -91,7 +91,7 @@ ENABLE_LLM_CACHE=true
### For rerank model deployed by vLLM use cohere binding
#########################################################
RERANK_BINDING=null
### Enable rerank by default in query params
### Enable rerank by default in query params when RERANK_BINDING is not null
# RERANK_BY_DEFAULT=True
### rerank score chunk filter(set to 0.0 to keep all chunks, 0.6 or above if LLM is not strong enought)
# MIN_RERANK_SCORE=0.0

56
lightrag/operate.py
View File

@ -1727,6 +1727,9 @@ async def kg_query(
system_prompt: str | None = None,
chunks_vdb: BaseVectorStorage = None,
) -> str | AsyncIterator[str]:
if not query:
return PROMPTS["fail_response"]
if query_param.model_func:
use_model_func = query_param.model_func
else:
@ -1763,21 +1766,16 @@ async def kg_query(
logger.debug(f"Low-level keywords: {ll_keywords}")
# Handle empty keywords
if ll_keywords == [] and query_param.mode in ["local", "hybrid", "mix"]:
logger.warning("low_level_keywords is empty")
if hl_keywords == [] and query_param.mode in ["global", "hybrid", "mix"]:
logger.warning("high_level_keywords is empty")
if hl_keywords == [] and ll_keywords == []:
logger.warning("low_level_keywords and high_level_keywords is empty")
return PROMPTS["fail_response"]
if ll_keywords == [] and query_param.mode in ["local", "hybrid"]:
logger.warning(
"low_level_keywords is empty, switching from %s mode to global mode",
query_param.mode,
)
query_param.mode = "global"
if hl_keywords == [] and query_param.mode in ["global", "hybrid"]:
logger.warning(
"high_level_keywords is empty, switching from %s mode to local mode",
query_param.mode,
)
query_param.mode = "local"
if len(query) < 50:
logger.warning(f"Forced low_level_keywords to origin query: {query}")
ll_keywords = [query]
else:
return PROMPTS["fail_response"]
ll_keywords_str = ", ".join(ll_keywords) if ll_keywords else ""
hl_keywords_str = ", ".join(hl_keywords) if hl_keywords else ""
@ -2133,7 +2131,7 @@ async def _build_query_context(
query_embedding = None
# Handle local and global modes
if query_param.mode == "local":
if query_param.mode == "local" and len(ll_keywords) > 0:
local_entities, local_relations = await _get_node_data(
ll_keywords,
knowledge_graph_inst,
@ -2141,7 +2139,7 @@ async def _build_query_context(
query_param,
)
elif query_param.mode == "global":
elif query_param.mode == "global" and len(hl_keywords) > 0:
global_relations, global_entities = await _get_edge_data(
hl_keywords,
knowledge_graph_inst,
@ -2150,18 +2148,20 @@ async def _build_query_context(
)
else: # hybrid or mix mode
local_entities, local_relations = await _get_node_data(
ll_keywords,
knowledge_graph_inst,
entities_vdb,
query_param,
)
global_relations, global_entities = await _get_edge_data(
hl_keywords,
knowledge_graph_inst,
relationships_vdb,
query_param,
)
if len(ll_keywords) > 0:
local_entities, local_relations = await _get_node_data(
ll_keywords,
knowledge_graph_inst,
entities_vdb,
query_param,
)
if len(hl_keywords) > 0:
global_relations, global_entities = await _get_edge_data(
hl_keywords,
knowledge_graph_inst,
relationships_vdb,
query_param,
)
# Get vector chunks first if in mix mode
if query_param.mode == "mix" and chunks_vdb: