[Experiment] Build Q-A System trained on LangChain.

This code implements a Streamlit application that allows users to upload multiple PDF documents and ask questions about the content of those documents. Here's a breakdown of what each part of the code does:

Imports

import streamlit as st
from dotenv import load_dotenv
from PyPDF2 import PdfReader
from langchain.text_splitter import CharacterTextSplitter
from langchain.embeddings import OpenAIEmbeddings, HuggingFaceInstructEmbeddings
from langchain.vectorstores import FAISS
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

Streamlit: Used for creating the web application.
dotenv: Loads environment variables from a .env file.
PyPDF2: Reads PDF files.
LangChain modules: Various utilities from the LangChain framework for text processing, embedding generation, and conversational AI.
htmlTemplates: Custom HTML templates for styling the chat interface.

Functions

`get_pdf_text(pdf_docs)`

Extracts text from the uploaded PDF documents.

def get_pdf_text(pdf_docs):
    text = ""
    for pdf in pdf_docs:
        pdf_reader = PdfReader(pdf)
        for page in pdf_reader.pages:
            text += page.extract_text()
    return text

`get_text_chunks(text)`

Splits the extracted text into smaller chunks.

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

`get_vectorstore(text_chunks)`

Generates embeddings for the text chunks and creates a vector store using FAISS.

def get_vectorstore(text_chunks):
    embeddings = OpenAIEmbeddings()
    vectorstore = FAISS.from_texts(texts=text_chunks, embedding=embeddings)
    return vectorstore

`get_conversation_chain(vectorstore)`

Creates a conversational retrieval chain using the vector store and a language model.

def get_conversation_chain(vectorstore):
    llm = ChatOpenAI()
    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

`handle_userinput(user_question)`

Handles user input by generating responses using the conversation chain and updating the chat history.

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)

Main Application Logic

`main()`

Sets up the Streamlit application, handles PDF uploads, and processes the PDFs.

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")
        pdf_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 = get_pdf_text(pdf_docs)

                # 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()

Explanation of the Main Logic:

Environment Setup:
- load_dotenv(): Loads environment variables from a .env file.
- st.set_page_config(): Configures the Streamlit app page settings.
- st.write(css, unsafe_allow_html=True): Applies custom CSS styling.
Session State Initialization:
- Initializes session state variables for the conversation chain and chat history if they don't exist.
User Interface:
- Header: Displays the main header of the application.
- User Question Input: Provides a text input for the user to ask questions about the uploaded documents.
- Sidebar: Allows users to upload multiple PDF files and click "Process" to start processing.
Processing Logic:
- When the user clicks "Process", the application:
  - Extracts text from the uploaded PDFs.
  - Splits the text into chunks.
  - Creates a vector store from the text chunks.
  - Initializes a conversational retrieval chain using the vector store.
Handle User Input:
- When the user asks a question, the handle_userinput function generates a response and updates the chat history, displaying it in the user interface.

This application effectively allows users to interact with the content of multiple PDFs through a conversational interface, leveraging powerful language models and vector search capabilities.

Vignana-Jyothi / kp-gen-ai