app.py

from langchain.chat_models import ChatOpenAI
from langchain.document_loaders import PyPDFLoader
from langchain.embeddings.openai import OpenAIEmbeddings
from langchain.embeddings.cohere import CohereEmbeddings
from langchain.text_splitter import CharacterTextSplitter
from langchain.vectorstores.elastic_vector_search import ElasticVectorSearch
from langchain.vectorstores import Chroma
from PyPDF2 import PdfWriter
import gradio as gr
import os
from dotenv import load_dotenv
import openai

load_dotenv()

os.environ["OPENAI_API_KEY"] = os.environ['my_secret']

loader = PyPDFLoader("/home/user/app/docs.pdf")
documents = loader.load()

text_splitter = CharacterTextSplitter(chunk_size=800, chunk_overlap=0)
texts = text_splitter.split_documents(documents)

#vector embedding
embeddings = OpenAIEmbeddings()
vector_store = Chroma.from_documents(texts, embeddings)
retriever = vector_store.as_retriever(search_kwargs={"k": 2})

from langchain.chat_models import ChatOpenAI
from langchain.chains import RetrievalQAWithSourcesChain

llm = ChatOpenAI(model_name="gpt-4", temperature=0)  # Modify model_name if you have access to GPT-4

chain = RetrievalQAWithSourcesChain.from_chain_type(
    llm=llm,
    chain_type="stuff",
    retriever = retriever,
    return_source_documents=True)

from langchain.prompts.chat import (
    ChatPromptTemplate,
    SystemMessagePromptTemplate,
    HumanMessagePromptTemplate,
)

system_template="""You are a chatbot that advises public officials on how to respond to complaints. Use the following pieces of context to answer the users question shortly.
Given the following summaries of a long document and a question, create a final answer with references ("SOURCES"), use "SOURCES" in capital letters regardless of the number of sources.
If you don't know the answer, just say that "I don't know", don't try to make up an answer.
----------------
{summaries}

You MUST answer in Korean and in Markdown format:"""

messages = [
    SystemMessagePromptTemplate.from_template(system_template),
    HumanMessagePromptTemplate.from_template("{question}")
]

prompt = ChatPromptTemplate.from_messages(messages)

from langchain.chat_models import ChatOpenAI
from langchain.chains import RetrievalQAWithSourcesChain

chain_type_kwargs = {"prompt": prompt}

llm = ChatOpenAI(model_name="gpt-4", temperature=0)  # Modify model_name if you have access to GPT-4

chain = RetrievalQAWithSourcesChain.from_chain_type(
    llm=llm,
    chain_type="stuff",
    retriever = retriever,
    return_source_documents=True,
    chain_type_kwargs=chain_type_kwargs
)

query = "행복한 인생이란?"
result = chain(query)


for doc in result['source_documents']:
    print('내용 : ' + doc.page_content[0:100].replace('\n', ' '))
    print('파일 : ' + doc.metadata['source'])
    print('페이지 : ' + str(doc.metadata['page']))


def respond(message, chat_history):  # 채팅봇의 응답을 처리하는 함수를 정의합니다.

    result = chain(message)

    bot_message = result['answer']

    for i, doc in enumerate(result['source_documents']):
        bot_message += '[' + str(i+1) + '] ' + doc.metadata['source'] + '(' + str(doc.metadata['page']) + ') '

    chat_history.append((message, bot_message))  # 채팅 기록에 사용자의 메시지와 봇의 응답을 추가합니다.

    return "", chat_history  # 수정된 채팅 기록을 반환합니다.

with gr.Blocks(theme='gstaff/sketch') as demo:  # gr.Blocks()를 사용하여 인터페이스를 생성합니다.
    gr.Markdown("# 침착하세요. \n 어려운 민원에 대응하는 방법을 알려주는 챗봇입니다.")
    chatbot = gr.Chatbot(label="채팅창")  # '채팅창'이라는 레이블을 가진 채팅봇 컴포넌트를 생성합니다.
    msg = gr.Textbox(label="입력")  # '입력'이라는 레이블을 가진 텍스트박스를 생성합니다.
    clear = gr.Button("초기화")  # '초기화'라는 레이블을 가진 버튼을 생성합니다.

    msg.submit(respond, [msg, chatbot], [msg, chatbot])  # 텍스트박스에 메시지를 입력하고 제출하면 respond 함수가 호출되도록 합니다.
    clear.click(lambda: None, None, chatbot, queue=False)  # '초기화' 버튼을 클릭하면 채팅 기록을 초기화합니다.
demo.launch(debug=True)  # 인터페이스를 실행합니다. 실행하면 사용자는 '입력' 텍스트박스에 메시지를 작성하고 제출할 수 있으며, '초기화' 버튼을 통해 채팅 기록을 초기화 할 수 있습니다.