app.py

import os
from openai import OpenAI
from langchain_community.document_loaders import PyPDFLoader
from langchain.text_splitter import RecursiveCharacterTextSplitter
from langchain_community.vectorstores import Chroma
from langchain_openai import OpenAIEmbeddings
from langchain_openai import ChatOpenAI
from langchain_core.prompts import ChatPromptTemplate
from langchain_core.runnables import RunnablePassthrough
from langchain_core.output_parsers import StrOutputParser
from langchain.schema import AIMessage, HumanMessage
import gradio as gr

# 1. PDF 로드 및 전처리
pdf_filepath = '3.pdf'
loader = PyPDFLoader(pdf_filepath)
pages = loader.load()

text_splitter = RecursiveCharacterTextSplitter(chunk_size=4000, chunk_overlap=400)
splits = text_splitter.split_documents(pages)

# 2. 벡터 스토어 생성 및 임베딩
vectorstore = Chroma.from_documents(documents=splits,
                                    embedding=OpenAIEmbeddings())

# 3. RAG 체인 설정
template = '''Answer the question based only on the following context:
{context}

Question: {question}
'''
prompt = ChatPromptTemplate.from_template(template)

model = ChatOpenAI(model='gpt-4o-mini', temperature=0)

retriever = vectorstore.as_retriever()

def format_docs(docs):
    return '\n\n'.join(doc.page_content for doc in docs)

rag_chain = (
    {'context': retriever | format_docs, 'question': RunnablePassthrough()}
    | prompt
    | model
    | StrOutputParser()
)

# 4. Gradio 인터페이스 predict 함수 수정
def predict(message, history):
    # RAG 체인을 사용하여 답변 생성
    response = rag_chain.invoke(message)
    return response

# 5. Gradio 인터페이스 설정 및 실행
demo = gr.ChatInterface(
    predict,
    title="영어 학습 챗봇 (Powered by RAG & LangChain)",
    description="영어 교재(1.pdf) 내용을 기반으로 질문에 답변합니다."
)

demo.launch()