mirror of
https://github.com/HKUDS/LightRAG.git
synced 2025-07-22 16:31:19 +00:00
102 lines
2.7 KiB
Python
102 lines
2.7 KiB
Python
import os
|
|
from lightrag import LightRAG, QueryParam
|
|
from lightrag.llm.openai import openai_complete_if_cache, openai_embed
|
|
from lightrag.utils import EmbeddingFunc
|
|
import numpy as np
|
|
import asyncio
|
|
import nest_asyncio
|
|
|
|
# Apply nest_asyncio to solve event loop issues
|
|
nest_asyncio.apply()
|
|
|
|
DEFAULT_RAG_DIR = "index_default"
|
|
|
|
# Configure working directory
|
|
WORKING_DIR = os.environ.get("RAG_DIR", f"{DEFAULT_RAG_DIR}")
|
|
print(f"WORKING_DIR: {WORKING_DIR}")
|
|
LLM_MODEL = os.environ.get("LLM_MODEL", "gpt-4o-mini")
|
|
print(f"LLM_MODEL: {LLM_MODEL}")
|
|
EMBEDDING_MODEL = os.environ.get("EMBEDDING_MODEL", "text-embedding-3-small")
|
|
print(f"EMBEDDING_MODEL: {EMBEDDING_MODEL}")
|
|
EMBEDDING_MAX_TOKEN_SIZE = int(os.environ.get("EMBEDDING_MAX_TOKEN_SIZE", 8192))
|
|
print(f"EMBEDDING_MAX_TOKEN_SIZE: {EMBEDDING_MAX_TOKEN_SIZE}")
|
|
BASE_URL = os.environ.get("BASE_URL", "https://api.openai.com/v1")
|
|
print(f"BASE_URL: {BASE_URL}")
|
|
API_KEY = os.environ.get("API_KEY", "xxxxxxxx")
|
|
print(f"API_KEY: {API_KEY}")
|
|
|
|
if not os.path.exists(WORKING_DIR):
|
|
os.mkdir(WORKING_DIR)
|
|
|
|
|
|
# LLM model function
|
|
|
|
|
|
async def llm_model_func(
|
|
prompt, system_prompt=None, history_messages=[], keyword_extraction=False, **kwargs
|
|
) -> str:
|
|
return await openai_complete_if_cache(
|
|
model=LLM_MODEL,
|
|
prompt=prompt,
|
|
system_prompt=system_prompt,
|
|
history_messages=history_messages,
|
|
base_url=BASE_URL,
|
|
api_key=API_KEY,
|
|
**kwargs,
|
|
)
|
|
|
|
|
|
# Embedding function
|
|
|
|
|
|
async def embedding_func(texts: list[str]) -> np.ndarray:
|
|
return await openai_embed(
|
|
texts=texts,
|
|
model=EMBEDDING_MODEL,
|
|
base_url=BASE_URL,
|
|
api_key=API_KEY,
|
|
)
|
|
|
|
|
|
async def get_embedding_dim():
|
|
test_text = ["This is a test sentence."]
|
|
embedding = await embedding_func(test_text)
|
|
embedding_dim = embedding.shape[1]
|
|
print(f"{embedding_dim=}")
|
|
return embedding_dim
|
|
|
|
|
|
# Initialize RAG instance
|
|
rag = LightRAG(
|
|
working_dir=WORKING_DIR,
|
|
llm_model_func=llm_model_func,
|
|
embedding_func=EmbeddingFunc(
|
|
embedding_dim=asyncio.run(get_embedding_dim()),
|
|
max_token_size=EMBEDDING_MAX_TOKEN_SIZE,
|
|
func=embedding_func,
|
|
),
|
|
)
|
|
|
|
with open("./book.txt", "r", encoding="utf-8") as f:
|
|
rag.insert(f.read())
|
|
|
|
# Perform naive search
|
|
print(
|
|
rag.query("What are the top themes in this story?", param=QueryParam(mode="naive"))
|
|
)
|
|
|
|
# Perform local search
|
|
print(
|
|
rag.query("What are the top themes in this story?", param=QueryParam(mode="local"))
|
|
)
|
|
|
|
# Perform global search
|
|
print(
|
|
rag.query("What are the top themes in this story?", param=QueryParam(mode="global"))
|
|
)
|
|
|
|
# Perform hybrid search
|
|
print(
|
|
rag.query("What are the top themes in this story?", param=QueryParam(mode="hybrid"))
|
|
)
|