gardel/lightrag
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

fix: unexpected keyword argument error

Browse Source
This commit is contained in:
Larfii
2024-12-05 11:47:56 +08:00
parent dff3fe18e9
commit a2072a055a
1 changed files with 7 additions and 0 deletions
Download Patch File Download Diff File Expand all files Collapse all files

7
lightrag/llm.py
View File

@@ -478,6 +478,7 @@ class GPTKeywordExtractionFormat(BaseModel):
async def gpt_4o_complete( async def gpt_4o_complete(
prompt, system_prompt=None, history_messages=[], keyword_extraction=False, **kwargs prompt, system_prompt=None, history_messages=[], keyword_extraction=False, **kwargs
) -> str: ) -> str:
keyword_extraction = kwargs.pop("keyword_extraction", None)
if keyword_extraction: if keyword_extraction:
kwargs["response_format"] = GPTKeywordExtractionFormat kwargs["response_format"] = GPTKeywordExtractionFormat
return await openai_complete_if_cache( return await openai_complete_if_cache(
@@ -492,6 +493,7 @@ async def gpt_4o_complete(
async def gpt_4o_mini_complete( async def gpt_4o_mini_complete(
prompt, system_prompt=None, history_messages=[], keyword_extraction=False, **kwargs prompt, system_prompt=None, history_messages=[], keyword_extraction=False, **kwargs
) -> str: ) -> str:
keyword_extraction = kwargs.pop("keyword_extraction", None)
if keyword_extraction: if keyword_extraction:
kwargs["response_format"] = GPTKeywordExtractionFormat kwargs["response_format"] = GPTKeywordExtractionFormat
return await openai_complete_if_cache( return await openai_complete_if_cache(
@@ -506,6 +508,7 @@ async def gpt_4o_mini_complete(
async def nvidia_openai_complete( async def nvidia_openai_complete(
prompt, system_prompt=None, history_messages=[], keyword_extraction=False, **kwargs prompt, system_prompt=None, history_messages=[], keyword_extraction=False, **kwargs
) -> str: ) -> str:
keyword_extraction = kwargs.pop("keyword_extraction", None)
result = await openai_complete_if_cache( result = await openai_complete_if_cache(
"nvidia/llama-3.1-nemotron-70b-instruct", # context length 128k "nvidia/llama-3.1-nemotron-70b-instruct", # context length 128k
prompt, prompt,
@@ -522,6 +525,7 @@ async def nvidia_openai_complete(
async def azure_openai_complete( async def azure_openai_complete(
prompt, system_prompt=None, history_messages=[], keyword_extraction=False, **kwargs prompt, system_prompt=None, history_messages=[], keyword_extraction=False, **kwargs
) -> str: ) -> str:
keyword_extraction = kwargs.pop("keyword_extraction", None)
result = await azure_openai_complete_if_cache( result = await azure_openai_complete_if_cache(
"conversation-4o-mini", "conversation-4o-mini",
prompt, prompt,
@@ -537,6 +541,7 @@ async def azure_openai_complete(
async def bedrock_complete( async def bedrock_complete(
prompt, system_prompt=None, history_messages=[], keyword_extraction=False, **kwargs prompt, system_prompt=None, history_messages=[], keyword_extraction=False, **kwargs
) -> str: ) -> str:
keyword_extraction = kwargs.pop("keyword_extraction", None)
result = await bedrock_complete_if_cache( result = await bedrock_complete_if_cache(
"anthropic.claude-3-haiku-20240307-v1:0", "anthropic.claude-3-haiku-20240307-v1:0",
prompt, prompt,
@@ -552,6 +557,7 @@ async def bedrock_complete(
async def hf_model_complete( async def hf_model_complete(
prompt, system_prompt=None, history_messages=[], keyword_extraction=False, **kwargs prompt, system_prompt=None, history_messages=[], keyword_extraction=False, **kwargs
) -> str: ) -> str:
keyword_extraction = kwargs.pop("keyword_extraction", None)
model_name = kwargs["hashing_kv"].global_config["llm_model_name"] model_name = kwargs["hashing_kv"].global_config["llm_model_name"]
result = await hf_model_if_cache( result = await hf_model_if_cache(
model_name, model_name,
@@ -568,6 +574,7 @@ async def hf_model_complete(
async def ollama_model_complete( async def ollama_model_complete(
prompt, system_prompt=None, history_messages=[], keyword_extraction=False, **kwargs prompt, system_prompt=None, history_messages=[], keyword_extraction=False, **kwargs
) -> str: ) -> str:
keyword_extraction = kwargs.pop("keyword_extraction", None)
if keyword_extraction: if keyword_extraction:
kwargs["format"] = "json" kwargs["format"] = "json"
model_name = kwargs["hashing_kv"].global_config["llm_model_name"] model_name = kwargs["hashing_kv"].global_config["llm_model_name"]