app.py

import streamlit as st
from dotenv import load_dotenv
from PyPDF2 import PdfReader
from langchain.text_splitter import CharacterTextSplitter,RecursiveCharacterTextSplitter
from langchain.llms import CTransformers  # For loading transformer models.
from langchain.embeddings import OpenAIEmbeddings, HuggingFaceInstructEmbeddings
from langchain.vectorstores import FAISS, Chroma
from langchain.embeddings import HuggingFaceEmbeddings  # General embeddings from HuggingFace models.
from langchain.chat_models import ChatOpenAI
from langchain.memory import ConversationBufferMemory
from langchain.chains import ConversationalRetrievalChain
from htmlTemplates import css, bot_template, user_template
from langchain.llms import HuggingFaceHub

def get_pdf_text(pdf_docs):
    text = ''
    pdf_file_ = open(pdf_docs,'rb')
    # text = "example hojin"
    pdf_reader = PdfReader(pdf_file_)
    for page in pdf_reader.pages:
        text += page.extract_text()

    return text


def get_text_chunks(text):
    print('text = ',text)
    text_splitter = RecursiveCharacterTextSplitter(
        chunk_size = 300,
        chunk_overlap = 20,
        length_function= len
    )
    # text_splitter = CharacterTextSplitter(
    #     separator="\n",
    #     chunk_size=1000,
    #     chunk_overlap=200,
    #     length_function=len
    # )
    chunks = text_splitter.split_text(text)
    print('chunks = ', chunks)
    return chunks


def get_vectorstore(text_chunks):
    # Load the desired embeddings model.
    embeddings = HuggingFaceEmbeddings(model_name='sentence-transformers/all-MiniLM-L12-v2',
                                       model_kwargs={'device': 'cpu'})
    # embeddings = OpenAIEmbeddings()
    # embeddings = HuggingFaceInstructEmbeddings(model_name="hkunlp/instructor-xl")
    # vectorstore = FAISS.from_texts(texts=text_chunks, embedding=embeddings)
    vectorstore = Chroma.from_texts(texts=text_chunks, embedding=embeddings)

    return vectorstore


def get_conversation_chain(vectorstore):
    # llm = ChatOpenAI()
    # llm = HuggingFaceHub(repo_id="google/flan-t5-xxl", model_kwargs={"temperature":0.5, "max_length":512})
    config = {'max_new_tokens': 2048}
    llm = CTransformers(model="llama-2-7b-chat.ggmlv3.q2_K.bin", model_type="llama", config=config)
    memory = ConversationBufferMemory(
        memory_key='chat_history', return_messages=True)
    conversation_chain = ConversationalRetrievalChain.from_llm(
        llm=llm,
        retriever=vectorstore.as_retriever(),
        memory=memory
    )
    return conversation_chain


def handle_userinput(user_question):
    response = st.session_state.conversation({'question': user_question})
    st.session_state.chat_history = response['chat_history']

    for i, message in enumerate(st.session_state.chat_history):
        if i % 2 == 0:
            st.write(user_template.replace(
                "{{MSG}}", message.content), unsafe_allow_html=True)
        else:
            st.write(bot_template.replace(
                "{{MSG}}", message.content), unsafe_allow_html=True)

def get_text_file(docs):
    text = f.read()
    return text

def get_csv_file(docs):
    import pandas as pd
    text = ''

    data = pd.read_csv(docs)

    for index, row in data.iterrows():
        item_name = row[0]
        row_text = item_name
        for col_name in data.columns[1:]:
            row_text += '{} is {} '.format(col_name, row[col_name])
        text += row_text + '\n'

    return text

def get_json_file(docs):
    import json
    text = ''
    with open(docs, 'r') as f:
        json_data = json.load(f)

    for f_key, f_value in json_data.items():
        for s_value in f_value:
            text += str(f_key) + str(s_value)
        text += '\n'
    #print(text)
    return text

def get_hwp_file(docs):
    pass

def get_docs_file(docs):
    pass


def main():
    load_dotenv()
    st.set_page_config(page_title="Chat with multiple PDFs",
                       page_icon=":books:")
    st.write(css, unsafe_allow_html=True)

    if "conversation" not in st.session_state:
        st.session_state.conversation = None
    if "chat_history" not in st.session_state:
        st.session_state.chat_history = None

    st.header("Chat with multiple PDFs :books:")
    user_question = st.text_input("Ask a question about your documents:")
    if user_question:
        handle_userinput(user_question)

    with st.sidebar:
        st.subheader("Your documents")
        docs = st.file_uploader(
            "Upload your PDFs here and click on 'Process'", accept_multiple_files=True)
        if st.button("Process"):
            with st.spinner("Processing"):
                # get pdf text
                raw_text = ""
                
                for file in docs:
                    print('file - type : ', file.type)
                    if file.type == 'text/plain':
                        #file is .txt
                        raw_text += get_text_file(file)
                    elif file.type in ['application/octet-stream', 'application/pdf']:
                        #file is .pdf
                        raw_text += get_pdf_text(file)
                    elif file.type == 'text/csv':
                        #file is .csv
                        raw_text += get_csv_file(file)
                    elif file.type == 'application/json':
                        # file is .json
                        raw_text += get_json_file(file)
                    elif file.type == 'application/x-hwp':
                        # file is .hwp
                        raw_text += get_hwp_file(file)
                    elif file.type == 'application/vnd.openxmlformats-officedocument.wordprocessingml.document':
                        # file is .docs
                        raw_text += get_docs_file(file)


                # get the text chunks
                text_chunks = get_text_chunks(raw_text)

                # create vector store
                vectorstore = get_vectorstore(text_chunks)

                # create conversation chain
                st.session_state.conversation = get_conversation_chain(
                    vectorstore)


if __name__ == '__main__':
    main()