app.py

import io
import os
import torch
import streamlit as st
from PyPDF2 import PdfReader
from langchain.text_splitter import CharacterTextSplitter
from langchain.chains.question_answering import load_qa_chain
from langchain_huggingface import HuggingFaceEmbeddings, HuggingFacePipeline
from langchain_community.vectorstores import FAISS
from transformers import AutoModelForCausalLM, AutoTokenizer, pipeline


# Global variables
knowledge_base = None
qa_chain = None

# PDF 파일 로드 및 텍스트 추출
def load_pdf(pdf_file):
    pdf_reader = PdfReader(pdf_file)
    text = "".join(page.extract_text() for page in pdf_reader.pages)
    return text

# 텍스트를 청크로 분할
def split_text(text):
    text_splitter = CharacterTextSplitter(
        separator="\n", 
        chunk_size=1000, 
        chunk_overlap=200, 
        length_function=len
    )
    return text_splitter.split_text(text)

# FAISS 벡터 저장소 생성
def create_knowledge_base(chunks):
    model_name = "sentence-transformers/all-mpnet-base-v2"  # 임베딩 모델을 명시
    embeddings = HuggingFaceEmbeddings(model_name=model_name)
    return FAISS.from_texts(chunks, embeddings)

# Hugging Face 모델 로드
def load_model():
    model_name = "google/gemma-2-2b"  # Hugging Face 모델 ID
    access_token = os.getenv("HF_TOKEN")
    try:
        device = 0 if torch.cuda.is_available() else -1
        tokenizer = AutoTokenizer.from_pretrained(model_name, token=access_token, clean_up_tokenization_spaces=False)
        model = AutoModelForCausalLM.from_pretrained(model_name, token=access_token)
        device = 0 if torch.cuda.is_available() else -1
        return pipeline("text-generation", model=model, tokenizer=tokenizer, max_new_tokens=150, temperature=0.1)
    except Exception as e:
        print(f"Error loading model: {e}")
        return None

# QA 체인 설정
def setup_qa_chain():
    global qa_chain
    try:
        pipe = load_model()  
    except Exception as e:
        print(f"Error loading model: {e}")
        return
    llm = HuggingFacePipeline(pipeline=pipe)
    qa_chain = load_qa_chain(llm, chain_type="map_rerank")
    
# 메인 페이지 UI
def main_page():
    st.title("Welcome to GemmaPaperQA")
    st.subheader("Upload Your Paper")

    paper = st.file_uploader("Upload Here!", type="pdf", label_visibility="hidden")
    if paper:
        st.write(f"Upload complete! File name: {paper.name}")
        # 파일 크기 확인
        file_size = paper.size  # 파일 크기를 파일 포인터 이동 없이 확인
        if file_size > 10 * 1024 * 1024:  # 10MB 제한
            st.error("File is too large! Please upload a file smaller than 10MB.")
            return

        # 중간 확인 절차 - PDF 내용 미리보기
        with st.spinner('Processing PDF...'):
            try:
                paper.seek(0)  # 파일 읽기 포인터를 처음으로 되돌림
                contents = paper.read()
                pdf_file = io.BytesIO(contents)
                text = load_pdf(pdf_file)

                # 텍스트가 추출되지 않을 경우 에러 처리
                if len(text.strip()) == 0:
                    st.error("The PDF appears to have no extractable text. Please check the file and try again.")
                    return

                st.text_area("Preview of extracted text", text[:1000], height=200)
                st.write(f"Total characters extracted: {len(text)}")
                global knowledge_base
                if st.button("Proceed with this file"):
                    chunks = split_text(text)
                    knowledge_base = create_knowledge_base(chunks)
                    
                    if knowledge_base is None:
                        st.error("Failed to create knowledge base.")
                        return
                    
                    setup_qa_chain()

                    st.session_state.paper_name = paper.name[:-4]
                    st.session_state.page = "chat"
                    st.success("PDF successfully processed! You can now ask questions.")

            except Exception as e:
                st.error(f"Failed to process the PDF: {str(e)}")


# 채팅 페이지 UI
def chat_page():
    st.title(f"Ask anything about {st.session_state.paper_name}")

    if "messages" not in st.session_state:
        st.session_state.messages = []

    for message in st.session_state.messages:
        with st.chat_message(message["role"]):
            st.markdown(message["content"])

    if prompt := st.chat_input("Chat here!"):
        st.session_state.messages.append({"role": "user", "content": prompt})

        with st.chat_message("user"):
            st.markdown(prompt)

        response = get_response_from_model(prompt)

        with st.chat_message("assistant"):
            st.markdown(response)

        st.session_state.messages.append({"role": "assistant", "content": response})

    if st.button("Go back to main page"):
        st.session_state.page = "main"

# 모델 응답 처리
def get_response_from_model(prompt):
    try:
        global knowledge_base, qa_chain
        if not knowledge_base:
            return "No PDF has been uploaded yet."
        if not qa_chain:
            return "QA chain is not initialized."

        docs = knowledge_base.similarity_search(prompt)
        response = qa_chain.run(input_documents=docs, question=prompt)

        if "Helpful Answer:" in response:
            response = response.split("Helpful Answer:")[1].strip()

        return response
    except Exception as e:
        return f"Error: {str(e)}"

# 페이지 설정
if "page" not in st.session_state:
    st.session_state.page = "main"

if "paper_name" not in st.session_state:
    st.session_state.paper_name = ""

# 페이지 렌더링
if st.session_state.page == "main":
    main_page()
elif st.session_state.page == "chat":
    chat_page()