Spaces:
Runtime error
Runtime error
File size: 2,688 Bytes
b72c08a a2c1b0b b72c08a e4e59cb b72c08a 9227993 b72c08a a2c1b0b 9227993 b72c08a 9227993 b72c08a a2c1b0b b72c08a a2c1b0b b72c08a a2c1b0b b72c08a 9227993 |
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 |
import os
import openai
import logging
from llama_index.query_engine.retriever_query_engine import RetrieverQueryEngine
from llama_index.callbacks.base import CallbackManager
from llama_index import (
LLMPredictor,
ServiceContext,
StorageContext,
load_index_from_storage,
)
from langchain.chat_models import ChatOpenAI
import chainlit as cl
# Set up logging for debugging and monitoring
logging.basicConfig(level=logging.INFO)
logger = logging.getLogger(__name__)
# Load OpenAI API key
openai.api_key = os.environ.get("OPENAI_API_KEY")
try:
# Attempt to rebuild storage context and load index
logger.info("Attempting to load index from storage.")
storage_context = StorageContext.from_defaults(persist_dir="./storage")
index = load_index_from_storage(storage_context)
except Exception as e:
# If index loading fails, create a new index
logger.warning(f"Failed to load index from storage: {e}. Creating a new index.")
from llama_index import GPTVectorStoreIndex, SimpleDirectoryReader
documents = SimpleDirectoryReader("./data").load_data()
index = GPTVectorStoreIndex.from_documents(documents)
index.storage_context.persist()
logger.info("New index created and persisted.")
@cl.on_chat_start
async def factory():
#embed_model = OpenAIEmbedding()
chunk_size = 1000
llm_predictor = LLMPredictor(
llm=ChatOpenAI(
temperature=0,
model_name="gpt-4-32k",
streaming=True,
),
)
service_context = ServiceContext.from_defaults(
llm_predictor=llm_predictor,
chunk_size=chunk_size,
callback_manager=CallbackManager([cl.LlamaIndexCallbackHandler()]),
)
query_engine = index.as_query_engine(
service_context=service_context,
streaming=True,
)
logger.info("Query engine initialized.") # to facilitate debugging and monitoring
cl.user_session.set("query_engine", query_engine)
@cl.on_message
async def main(message):
try:
query_engine = cl.user_session.get("query_engine") # type: RetrieverQueryEngine
logger.info(f"Received message: {message}")
response = await cl.make_async(query_engine.query)(message)
response_message = cl.Message(content="")
for token in response.response_gen:
await response_message.stream_token(token=token)
if response.response_txt:
response_message.content = response.response_txt
await response_message.send()
logger.info(f"Response sent: {response.response_txt}")
except Exception as e:
logger.error(f"An error occurred while processing the message: {e}")
|