Hello, im getting a `'LLMPredictor'

File "D:\LLM_Work\llm-server-webapp\bring-your-own-documents\app-bring-your-own-docs3.py", line 319, in response_stream
    yield from (line for line in query_engine.query(user_question).response_gen)
  File "D:\LLM_Work\llm-server-webapp\.venv\lib\site-packages\llama_index\core\base_query_engine.py", line 30, in query
    return self._query(str_or_query_bundle)
  File "D:\LLM_Work\llm-server-webapp\.venv\lib\site-packages\llama_index\query_engine\retriever_query_engine.py", line 171, in _query
    response = self._response_synthesizer.synthesize(
  File "D:\LLM_Work\llm-server-webapp\.venv\lib\site-packages\llama_index\response_synthesizers\base.py", line 146, in synthesize
    response_str = self.get_response(
  File "D:\LLM_Work\llm-server-webapp\.venv\lib\site-packages\llama_index\response_synthesizers\compact_and_refine.py", line 38, in get_response
    return super().get_response(
  File "D:\LLM_Work\llm-server-webapp\.venv\lib\site-packages\llama_index\response_synthesizers\refine.py", line 146, in get_response
    response = self._give_response_single(
  File "D:\LLM_Work\llm-server-webapp\.venv\lib\site-packages\llama_index\response_synthesizers\refine.py", line 194, in _give_response_single
    program = self._program_factory(text_qa_template)
  File "D:\LLM_Work\llm-server-webapp\.venv\lib\site-packages\llama_index\response_synthesizers\refine.py", line 177, in _default_program_factory
    llm=self._service_context.llm,
  File "D:\LLM_Work\llm-server-webapp\.venv\lib\site-packages\llama_index\service_context.py", line 322, in llm
    return self.llm_predictor.llm
  File "D:\LLM_Work\llm-server-webapp\.venv\lib\site-packages\llama_index\llm_predictor\base.py", line 143, in llm
    return self._llm
AttributeError: 'LLMPredictor' object has no attribute '_llm'

@app.server.route("/esic-rag/streaming-chat", methods=["POST"])
def streaming_chat():
    sys_prompt = request.json["sys_prompt"]
    user_prompt = request.json["prompt"]
    user_question = request.json["question"]
    sim_top_k = request.json["sim_top_k"]
    session_id = request.json["session_id"]

    llm = OpenAILike(
        model="local:llama-2-13b-chat.Q4_K_S",
        api_base="http://localhost:8000/v1",
        api_key="fake",
        api_type="fake",
        max_tokens=3900,
        is_chat_model=True
    )
    service_context = ServiceContext.from_defaults(llm=llm, embed_model="local:BAAI/bge-small-en-v1.5", chunk_size=256, num_output=256)
    set_global_service_context(service_context)

    index = pickle.loads(cache.get(session_id))
    cache.set(session_id, pickle.dumps(index))

    # Create a system message
    system_message = ChatMessage(role=MessageRole.SYSTEM, content=sys_prompt)
    user_prompt = ChatMessage(role=MessageRole.USER, content=user_prompt)
    text_qa_template = ChatPromptTemplate(message_templates=[system_message, user_prompt])

    ### QUERY ENGINE SECTION
    query_engine = index.as_query_engine(streaming=True, text_qa_template=text_qa_template, similarity_top_k=sim_top_k)
    def response_stream():
        yield from (line for line in query_engine.query(user_question).response_gen)

    return Response(response_stream(), mimetype="text/response-stream")

@app.callback(
    [Output('context-results', 'children'),
     Output('search-result-header', 'children')],
    Input("submit-prompt", "n_clicks"),
    [State('text-question', 'value'),
     State("num-excerpts", "value"),
     State('session-id', 'data')]
)
def query_vector_db(clicks, question, sim_top_k, session_id):
    if clicks == None:
        raise PreventUpdate

    # get previously created cached index from filesystem
    index = pickle.loads(cache.get(session_id))

    retriever = index.as_retriever(similarity_top_k=sim_top_k)
    nodes = retriever.retrieve(question)

    search_results = []

    for node in nodes:
        doc = node.metadata['file_name']
        page = node.metadata['page_label']
        url = 'https://www.hello.com/00000000' + doc.split('_')[0]
        excerpt_text = node.text

        search_results.append(
            dbc.Card(
                [
                    dbc.CardBody(
                        [
                            html.H4(doc, className="card-title"),
                            html.H6('Page ' + page, className="card-title"),
                            html.P(excerpt_text, className="card-text"),
                            dbc.Button("Get the Report", color="primary", href=url, target='_blank'),
                        ]
                    ),
                ],
            )
        )

    return [search_results, "Top %d Search Results (input context for LLM):" % sim_top_k]

import json

# "save"
storage_string = json.dumps(index.storage_context.to_dict())

# "load"
from llama_index import load_index_from_storage, StorageContext
storage_dict = json.loads(storage_string)
storage_context = StorageContext.from_dict(storage_dict)

# optional service context
index = load_index_from_storage(storage_context, service_context=service_context)