langchain_qna_with_azure_openai_service.py

import os
import openai
from dotenv import load_dotenv
from langchain.llms import AzureOpenAI
from langchain.embeddings import OpenAIEmbeddings

# Load environment variables (set OPENAI_API_KEY and OPENAI_API_BASE in .env)
load_dotenv()

# Configure OpenAI API
openai.api_type = "azure"
openai.api_version = "2022-12-01"
openai.api_base = os.getenv('OPENAI_API_BASE')
openai.api_key = os.getenv("OPENAI_API_KEY")

# Init LLM and embeddings model
llm = AzureOpenAI(deployment_name="text-davinci-003", temperature=0.1)
embeddings = OpenAIEmbeddings(model="text-embedding-ada-002", chunk_size=1)

from langchain.vectorstores import Chroma
from langchain.document_loaders import DirectoryLoader
from langchain.document_loaders import TextLoader
from langchain.text_splitter import CharacterTextSplitter
from langchain.chains import RetrievalQAWithSourcesChain

loader = DirectoryLoader('../data/qna/', glob="*.txt", loader_cls=TextLoader)

documents = loader.load()
text_splitter = CharacterTextSplitter(chunk_size=1000, chunk_overlap=50)
docs = text_splitter.split_documents(documents)

print(docs)

db = Chroma.from_documents(documents=docs, embedding=embeddings, persist_directory="db/")
db.persist()

chain = RetrievalQAWithSourcesChain.from_chain_type(llm=llm, chain_type="stuff", retriever=db.as_retriever())

query = "what is azure openai service?"
answer = chain({"question": query}, return_only_outputs=True)

print(answer)