mirror of
https://github.com/QuivrHQ/quivr.git
synced 2024-12-13 11:02:33 +03:00
285fe5b960
# Description This PR includes far too many new features: - detection of user intent (closes CORE-211) - treating multiple questions in parallel (closes CORE-212) - using the chat history when answering a question (closes CORE-213) - filtering of retrieved chunks by relevance threshold (closes CORE-217) - dynamic retrieval of chunks (closes CORE-218) - enabling web search via Tavily (closes CORE-220) - enabling agent / assistant to activate tools when relevant to complete the user task (closes CORE-224) Also closes CORE-205 ## Checklist before requesting a review Please delete options that are not relevant. - [ ] My code follows the style guidelines of this project - [ ] I have performed a self-review of my code - [ ] I have commented hard-to-understand areas - [ ] I have ideally added tests that prove my fix is effective or that my feature works - [ ] New and existing unit tests pass locally with my changes - [ ] Any dependent changes have been merged ## Screenshots (if appropriate): --------- Co-authored-by: Stan Girard <stan@quivr.app>
53 lines
1.8 KiB
Python
53 lines
1.8 KiB
Python
import asyncio
|
|
import json
|
|
from uuid import uuid4
|
|
|
|
from langchain_core.embeddings import DeterministicFakeEmbedding
|
|
from langchain_core.messages.ai import AIMessageChunk
|
|
from langchain_core.vectorstores import InMemoryVectorStore
|
|
from quivr_core.rag.entities.chat import ChatHistory
|
|
from quivr_core.rag.entities.config import LLMEndpointConfig, RetrievalConfig
|
|
from quivr_core.llm import LLMEndpoint
|
|
from quivr_core.rag.quivr_rag_langgraph import QuivrQARAGLangGraph
|
|
|
|
|
|
async def main():
|
|
retrieval_config = RetrievalConfig(llm_config=LLMEndpointConfig(model="gpt-4o"))
|
|
embedder = DeterministicFakeEmbedding(size=20)
|
|
vec = InMemoryVectorStore(embedder)
|
|
|
|
llm = LLMEndpoint.from_config(retrieval_config.llm_config)
|
|
chat_history = ChatHistory(uuid4(), uuid4())
|
|
rag_pipeline = QuivrQARAGLangGraph(
|
|
retrieval_config=retrieval_config, llm=llm, vector_store=vec
|
|
)
|
|
|
|
conversational_qa_chain = rag_pipeline.build_chain()
|
|
|
|
with open("response.jsonl", "w") as f:
|
|
async for event in conversational_qa_chain.astream_events(
|
|
{
|
|
"messages": [
|
|
("user", "What is NLP, give a very long detailed answer"),
|
|
],
|
|
"chat_history": chat_history,
|
|
"custom_personality": None,
|
|
},
|
|
version="v1",
|
|
config={"metadata": {}},
|
|
):
|
|
kind = event["event"]
|
|
if (
|
|
kind == "on_chat_model_stream"
|
|
and event["metadata"]["langgraph_node"] == "generate"
|
|
):
|
|
chunk = event["data"]["chunk"]
|
|
dict_chunk = {
|
|
k: v.dict() if isinstance(v, AIMessageChunk) else v
|
|
for k, v in chunk.items()
|
|
}
|
|
f.write(json.dumps(dict_chunk) + "\n")
|
|
|
|
|
|
asyncio.run(main())
|