app.py

import gradio as gr
from langchain_community.document_loaders import PyPDFLoader
from langchain_text_splitters import RecursiveCharacterTextSplitter
from langchain_chroma import Chroma
from langchain_huggingface.embeddings import HuggingFaceEmbeddings
from langchain_huggingface import HuggingFaceEndpoint, ChatHuggingFace
from langchain_core.prompts import ChatPromptTemplate
from langchain.chains.combine_documents import create_stuff_documents_chain
from langchain.chains import create_retrieval_chain


def initialise_chatbot(pdf, llm, progress=gr.Progress()):
    progress(0, desc="Reading PDF")

    loader = PyPDFLoader(pdf.name)
    pages = loader.load()
    text_splitter = RecursiveCharacterTextSplitter(chunk_size=1000, chunk_overlap=200)
    splits = text_splitter.split_documents(pages)

    progress(0.25, desc="Initialising Vectorstore")

    vectorstore = Chroma.from_documents(splits, embedding=HuggingFaceEmbeddings())

    progress(0.85, desc="Initialising LLM")

    llm = HuggingFaceEndpoint(
        repo_id=llm,
        task="text-generation",
        max_new_tokens=512,
        top_k=4,
        temperature=0.05,
    )

    chat = ChatHuggingFace(llm=llm, verbose=True)

    retriever = vectorstore.as_retriever(search_kwargs={"k": 8})

    system_prompt = (
        "You are an assistant for question-answering tasks. "
        "Use the following pieces of retrieved context to answer "
        "the question. If you don't know the answer, say that you "
        "don't know. Use two sentences maximum and keep the "
        "answer concise and to the point."
        "\n\n"
        "{context}"
    )

    prompt = ChatPromptTemplate.from_messages(
        [
            ("system", system_prompt),
            ("human", "{input}"),
        ]
    )

    question_answer_chain = create_stuff_documents_chain(chat, prompt)
    rag_chain = create_retrieval_chain(retriever, question_answer_chain)

    return rag_chain, "Complete!"


def send(message, rag_chain, chat_history):
    response = rag_chain.invoke({"input": message})
    chat_history.append((message, response["answer"]))

    return "", chat_history


with gr.Blocks() as demo:
    vectorstore = gr.State()
    rag_chain = gr.State()

    gr.Markdown("<H1>Talk to Documents</H1>")
    gr.Markdown("<H3>Upload and ask questions about your PDF files</H3>")
    gr.Markdown(
        "<H6>Note: This project uses LangChain to perform RAG (Retrieval Augmented Generation) on PDF files, allowing users to ask any questions related to their contents. When a PDF file is uploaded, it is embedded and stored in an in-memory Chroma vectorstore, which the chatbot uses as a source of knowledge when aswering user questions.</H6>"
    )

    with gr.Row():
        with gr.Column(scale=1):
            input_pdf = gr.File(label="1. Upload PDF")
            language_model = gr.Radio(
                label="2. Choose LLM",
                choices=[
                    "microsoft/Phi-3-mini-4k-instruct",
                    "mistralai/Mistral-7B-Instruct-v0.2",
                    "HuggingFaceH4/zephyr-7b-beta",
                    "mistralai/Mixtral-8x7B-Instruct-v0.1",
                ],
            )
            initialise_chatbot_btn = gr.Button(
                value="3. Initialise Chatbot", variant="primary"
            )
            chatbot_initialisation_progress = gr.Textbox(
                value="Not Started", label="Initialization Progress"
            )

        with gr.Column(scale=4):
            chatbot = gr.Chatbot(scale=1)
            message = gr.Textbox(label="4. Ask questions about your PDF")

    initialise_chatbot_btn.click(
        fn=initialise_chatbot,
        inputs=[input_pdf, language_model],
        outputs=[rag_chain, chatbot_initialisation_progress],
    )
    message.submit(
        fn=send, inputs=[message, rag_chain, chatbot], outputs=[message, chatbot]
    )

demo.launch()