app.py

import os
from langchain.document_loaders import PyPDFLoader
from langchain.text_splitter import RecursiveCharacterTextSplitter
from langchain.embeddings import OpenAIEmbeddings
from langchain.vectorstores import FAISS
from langchain.llms import OpenAI
from langchain.chains import ConversationalRetrievalChain
import pickle
import gradio as gr
import time




def upload_file(file, key):

    # Set the Enviroment variable
    os.environ["OPENAI_API_KEY"] = key

    # load document
    loader = PyPDFLoader(file.name)
    documents = loader.load()

    # split the documents into chunks
    text_splitter = RecursiveCharacterTextSplitter(chunk_size=1000, chunk_overlap=200)
    texts = text_splitter.split_documents(documents)

    # OPENAI embegddings
    embeddings = OpenAIEmbeddings()

    # create the vectorestore to use as the index
    db = FAISS.from_documents(documents, embeddings)

    with open("vectorstore.pkl", "wb") as f:
        pickle.dump(db, f)

    return file.name


with gr.Blocks() as demo:

    openai_key = gr.Textbox(label="OPENAI API KEY")
    file_output = gr.File(label="Please select a pdf file wait for the document to be displayed here")
    upload_button = gr.UploadButton("Click to upload a pdf document", file_types=["pdf"], file_count="single")
    upload_button.upload(upload_file, inputs = [upload_button, openai_key], outputs= file_output)
    
    
    chatbot = gr.Chatbot(label="Chat")
    msg = gr.Textbox(label="Enter your query")
    clear = gr.Button("Clear")

    def user(user_message, history):
        return "", history + [[user_message, None]]

    def bot(history):
        user_message = history[-1][0]
        with open("vectorstore.pkl", "rb") as f:
            vectorstore = pickle.load(f)
        llm = OpenAI(temperature=0)
        qa = ConversationalRetrievalChain.from_llm(llm, vectorstore.as_retriever(), return_source_documents=True)
        hist = []
        if history[-1][1] != None:
            hist = history
            
        result = qa({"question": user_message, "chat_history": hist})
        history[-1][1] = result['answer']
        return history
    
    msg.submit(user, [msg, chatbot], [msg, chatbot], queue=False).then(
        bot, chatbot, chatbot
    )
    clear.click(lambda: None, None, chatbot, queue=False)


demo.launch()