app.py

import gradio as gr
import pandas as pd
from langchain.embeddings import HuggingFaceEmbeddings
from langchain.vectorstores import faiss
from langchain_community.llms import HuggingFaceHub
from langchain.chains import ConversationalRetrievalChain, LLMChain
from langchain_community.document_loaders import PyPDFLoader
from langchain.text_splitter import RecursiveCharacterTextSplitter
from langchain import vectorstores
from langchain.prompts import PromptTemplate
from langchain.memory import ConversationBufferMemory
import os
import re

api_token = os.environ.get('HUGGINGFACEHUB_API_TOKEN')

memory = ConversationBufferMemory(
    memory_key="chat_history",
    return_messages=True
)

model = HuggingFaceHub(
    huggingfacehub_api_token=api_token,
    repo_id="mistralai/Mistral-7B-Instruct-v0.2",
    task="conversational",
    model_kwargs={"temperature": 0.8, "max_length": 1000},
)
template = """Use the following pieces of context to answer the question at the end. If you don't know the answer, just say that you don't know, don't try to make up an answer. Use three sentences maximum. Keep the answer as concise as possible. Always say "thanks for asking!" at the end of the answer. 
{context}
Question: {question}
Helpful Answer:"""
QA_CHAIN_PROMPT = PromptTemplate.from_template(template)

def load_db(file, k):
    # load documents
    loader = PyPDFLoader(file)
    documents = loader.load()
    # split documents
    text_splitter = RecursiveCharacterTextSplitter(chunk_size=1000, chunk_overlap=150)
    docs = text_splitter.split_documents(documents)
    # define embedding
    embeddings = HuggingFaceEmbeddings()
    # create vector database from data
    db = vectorstores.FAISS.from_documents(docs, embeddings)
    # define retriever
    retriever = db.as_retriever(search_type="similarity", search_kwargs={"k": k})
    # create a chatbot chain. Memory is managed externally.
    question_generator_chain = LLMChain(llm=model, prompt=QA_CHAIN_PROMPT)

    qa = ConversationalRetrievalChain.from_llm(
        llm=model,
        chain_type="stuff",
        retriever=retriever,
        return_source_documents=True,
        return_generated_question=True,
        memory=memory,
    )

    return qa

def chat(input_text, pdf_file):
    qa = load_db(pdf_file, 3)
    print("MEMORY")
    print(memory)
    if not memory.history:
        # If no previous conversation, start with a greeting
        response = qa.invoke({"question": "Hi, how can I help you today?", "chat_history": []})
        memory.update(response["chat_history"])

    response = qa.invoke({"question": input_text, "chat_history": memory.history})
    
    # Extracting the helpful answer from the response
    match = re.search(r'Helpful Answer:(.*)', response['answer'])
    if match:
        helpful_answer = match.group(1).strip()
    else:
        helpful_answer = "No helpful answer found."

    # Update the chat history
    memory.update([(input_text, helpful_answer)])

    return helpful_answer


iface = gr.Interface(fn=chat, inputs=["text", "file"], outputs="text")
iface.launch(share=True)