Lightrag/support import existing kg

[SuperEver]

Below is an example to use GraphDocument

reader_conf = {
    "chunk_size" : 1200,
    "chunk_overlap" : 100,
}
store_conf = {
    "root_path": "/home/mnt/zhangyongchao/workspace/gitlab/LightRAG",
    "name_space": "dickens",
    "chunk_store_type": "JsonChunkStore",
    "chunk_store_config": {},
    "er_store_type": "NanoDBGraphERStore",
    "er_store_config": {"embedding_dim": 1024},
    "network_store_type": "NetworkXStore",
    "network_store_config": {},
}

llm = lazyllm.OnlineChatModule(source="qwen")
embed_module = lazyllm.OnlineEmbeddingModule(source="openai", embed_url="http://***")

md_dir = "/home/mnt/zhangyongchao/workspace/gitlab/lazyllm-jinan/LazyLLM/books"
documents = GraphDocument(dataset_path=md_dir, llm=llm, embed=embed_module, reader_conf=reader_conf, store_conf=store_conf)
retriever = lazyllm.Retriever(doc=documents, topk=50)

query = "Who is Bob Cratchit"
doc_node_list = retriever(query=query)  # retrieving process, returns the collected chunks
content_str = "".join([node.get_content() for node in doc_node_list])

llm_answer = llm.share()
prompt = 'You will act as an AI question-answering assistant and complete a dialogue task. In this task, \
      you need to provide your answers based on the given context and questions.'
llm_answer.prompt(lazyllm.ChatPrompter(instruction=prompt, extra_keys=['context_str']))  # prompting

res = llm_answer({"query": query, "context_str": content_str})  # get answer from llm
print("RES:\n", res)