Comments (4)
I guess my question is how do you change the inputs and outputs to use Lanarky and FastAPI...
from lanarky.
Figured out the issue.
from lanarky.
@Haste171 it would be great if you post the solution too
from lanarky.
@talhaanwarch I managed to get this working with following. I am using Weaviate as a vector store. I struggled quite alot trying to get this work, and im still not 100% sure why it works with this library instead of just using langchain, will need to dig into their codebase abit more.
In the meantime you can refer to this example:
from dotenv import load_dotenv
from fastapi import FastAPI
from fastapi.templating import Jinja2Templates
from langchain.chains import ConversationalRetrievalChain, LLMChain
from langchain.chains.conversational_retrieval.prompts import CONDENSE_QUESTION_PROMPT
from langchain.chains.question_answering import load_qa_chain
from langchain.chat_models import ChatOpenAI
from langchain.embeddings import OpenAIEmbeddings
from langchain.vectorstores import FAISS
from langchain.vectorstores.weaviate import Weaviate
from lanarky import LangchainRouter
import weaviate
from langchain.prompts import (
ChatPromptTemplate,
HumanMessagePromptTemplate,
MessagesPlaceholder,
SystemMessagePromptTemplate,
PromptTemplate,
)
from langchain.memory import ConversationBufferMemory
from langchain.chains.conversational_retrieval.prompts import CONDENSE_QUESTION_PROMPT
load_dotenv()
app = FastAPI(title="ConversationalRetrievalChainDemo")
chatTemplate = """
Answer the question based on the chat history(delimited by <hs></hs>) and context(delimited by <ctx> </ctx>) below.
-----------
<ctx>
{context}
</ctx>
-----------
<hs>
{chat_history}
</hs>
-----------
Question: {question}
Answer:
"""
PROMPT = PromptTemplate(
input_variables=["context", "question", "chat_history"], template=chatTemplate
)
def create_chain():
weaviate_client = weaviate.Client("http://localhost:8080")
vectorstore: Any = Weaviate(weaviate_client, "Idx_664773d4e6", "text")
question_generator = LLMChain(
llm=ChatOpenAI(
temperature=0,
streaming=True,
),
prompt=PROMPT,
)
doc_chain = load_qa_chain(
llm=ChatOpenAI(
temperature=0,
streaming=True,
),
chain_type="stuff",
)
memory = ConversationBufferMemory(
return_messages=True,
memory_key="chat_history",
max_token_limit=20,
prompt=PROMPT,
)
chain = ConversationalRetrievalChain.from_llm(
llm=ChatOpenAI(
temperature=0,
streaming=True,
),
retriever=vectorstore.as_retriever(),
memory=memory,
combine_docs_chain_kwargs={"prompt": PROMPT},
verbose=True,
)
return chain
chain = create_chain()
langchain_router = LangchainRouter(
langchain_url="/chat", langchain_object=chain, streaming_mode=0
)
app.include_router(langchain_router)
if __name__ == "__main__":
import uvicorn
uvicorn.run(host="0.0.0.0", port=8000, app=app)
from lanarky.
Related Issues (20)
- Question: SSE StreamingResponse support for FastAPI HOT 2
- Final answer not returned with using GPTCache, with streaming JSON. HOT 6
- LangchainRouter not using chain's built-in memory HOT 2
- Feature Request: Customized Chain Support HOT 4
- Issue with Gradio Integration in conversation_retrieval in examples/app HOT 4
- StreamingResponse SSE text/event-stream requires "data:" prefix and "\n\n" terminator to work properly HOT 5
- typing-extensions outdated for pydantic v2 HOT 6
- Vercel is not recognizing my LangchainRouter HOT 3
- Create Streaming for Agent types and add Memory HOT 2
- How to custom an AsyncLanarkyCallback in llmchain? HOT 7
- UserWarning: Field name "raise_error" shadows an attribute in parent "AsyncCallbackHandler"; HOT 7
- How to export OpenAI-Compatible API ? HOT 2
- Streaming sometimes has no response on repeated same request HOT 10
- AttributeError: module 'openai' has no attribute 'aiosession' HOT 5
- bug: callback handlers for langchain adapter doesn't work for LLMs without async support HOT 1
- bug: Expected response header Content-Type to contain 'text/event-stream', got 'text/plain' HOT 31
- bug: TokenStreamingCallbackHandler accepts only one key HOT 5
- bug: openai error: list index out of range HOT 1
- feat: LLMChain was deprecated HOT 1
- Allow langchain version > 0.2.0 to be used HOT 2
Recommend Projects
-
React
A declarative, efficient, and flexible JavaScript library for building user interfaces.
-
Vue.js
🖖 Vue.js is a progressive, incrementally-adoptable JavaScript framework for building UI on the web.
-
Typescript
TypeScript is a superset of JavaScript that compiles to clean JavaScript output.
-
TensorFlow
An Open Source Machine Learning Framework for Everyone
-
Django
The Web framework for perfectionists with deadlines.
-
Laravel
A PHP framework for web artisans
-
D3
Bring data to life with SVG, Canvas and HTML. 📊📈🎉
-
Recommend Topics
-
javascript
JavaScript (JS) is a lightweight interpreted programming language with first-class functions.
-
web
Some thing interesting about web. New door for the world.
-
server
A server is a program made to process requests and deliver data to clients.
-
Machine learning
Machine learning is a way of modeling and interpreting data that allows a piece of software to respond intelligently.
-
Visualization
Some thing interesting about visualization, use data art
-
Game
Some thing interesting about game, make everyone happy.
Recommend Org
-
Facebook
We are working to build community through open source technology. NB: members must have two-factor auth.
-
Microsoft
Open source projects and samples from Microsoft.
-
Google
Google ❤️ Open Source for everyone.
-
Alibaba
Alibaba Open Source for everyone
-
D3
Data-Driven Documents codes.
-
Tencent
China tencent open source team.
from lanarky.