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

Add ability to passadditional parameters to ollama library like host and timeout

Browse Source
This commit is contained in:
Andrii Lazarchuk
2024-10-21 11:53:06 +00:00
parent 08caaa0b53
commit 25a2dd41c1
4 changed files with 13 additions and 5 deletions
Download Patch File Download Diff File Expand all files Collapse all files

9
lightrag/llm.py
View File

@@ -222,8 +222,10 @@ async def ollama_model_if_cache(
) -> str:
kwargs.pop("max_tokens", None)
kwargs.pop("response_format", None)
host = kwargs.pop("host", None)
timeout = kwargs.pop("timeout", None)
ollama_client = ollama.AsyncClient()
ollama_client = ollama.AsyncClient(host=host, timeout=timeout)
messages = []
if system_prompt:
messages.append({"role": "system", "content": system_prompt})
@@ -415,10 +417,11 @@ async def hf_embedding(texts: list[str], tokenizer, embed_model) -> np.ndarray:
return embeddings.detach().numpy()
async def ollama_embedding(texts: list[str], embed_model) -> np.ndarray:
async def ollama_embedding(texts: list[str], embed_model, **kwargs) -> np.ndarray:
embed_text = []
ollama_client = ollama.Client(**kwargs)
for text in texts:
data = ollama.embeddings(model=embed_model, prompt=text)
data = ollama_client.embeddings(model=embed_model, prompt=text)
embed_text.append(data["embedding"])
return embed_text