app.py

import streamlit as st
import ollama
import torch
import os
import logging
import accelerate

from langchain import hub
from langchain_community.document_loaders.web_base import WebBaseLoader
from langchain_community.vectorstores.faiss import FAISS
from langchain.text_splitter import RecursiveCharacterTextSplitter
from typing import List, Tuple, Dict, Any
from langchain_ollama import OllamaEmbeddings



##### Logging

# Logging configuration
logging.basicConfig(
    level=logging.INFO,
    format="%(asctime)s - %(levelname)s - %(message)s",
    datefmt="%Y-%m-%d %H:%M:%S",
)

logger = logging.getLogger(__name__)


prompt = hub.pull("rlm/rag-prompt")

embeddings = OllamaEmbeddings(model='nomic-embed-text', base_url="http://localhost:11434")

@st.cache_resource(show_spinner=True)
def extract_model_names(
    models_info: Dict[str, List[Dict[str, Any]]],
) -> Tuple[str, ...]:
    """
    Extract model names from the provided models information.

    Args:
        models_info (Dict[str, List[Dict[str, Any]]]): Dictionary containing information about available models.

    Returns:
        Tuple[str, ...]: A tuple of model names.
    """
    logger.info("Extracting model names from models_info")
    model_names = tuple(model["name"] for model in models_info["models"])
    logger.info(f"Extracted model names: {model_names}")
    return model_names

def format_docs(docs):
    return "\n\n".join([doc.page_content for doc in docs])

## RAG with URL
def rag_with_url(target_url, model, prompt):

    loader = WebBaseLoader(target_url)
    raw_document = loader.load()
    text_splitter = RecursiveCharacterTextSplitter(chunk_size=1000, chunk_overlap=200, length_function=len)
    splited_document = text_splitter.split_documents(raw_document)
    vector_store = FAISS.from_documents(splited_document, embeddings)
    retriever = vector_store.as_retriever()
    relevant_documents = retriever.get_relevant_documents(prompt)
    final_prompt = prompt + " " + " ".join([doc.page_content for doc in relevant_documents])


    AI_Respose = model.invoke(final_prompt)

    return AI_Respose.content

def main() -> None:

    st.set_page_config(layout="centered")

    st.title("🧠 This is a RAG Chatbot to any URL with LLama-3.2 and Langchain !!!")

    st.write("You can select which local LLM from Ollama to be used !!!")

    from transformers import AutoTokenizer, AutoModelForCausalLM

    token = os.environ["HF_TOKEN"]
    
    tokenizer = AutoTokenizer.from_pretrained("meta-llama/Llama-3.2-3B-Instruct",token=token)
    model = AutoModelForCausalLM.from_pretrained("meta-llama/Llama-3.2-3B-Instruct", token=token)  


    # URL text box for user input
    url_input = st.text_input("Enter a URL to be queried:", "")

    with st.form("llm-form"):
        text = st.text_area("Enter your question below:")
        submit = st.form_submit_button("Submit")

    if "chat_history" not in st.session_state:
        st.session_state['chat_history'] = []

    if submit and text:
        with st.spinner("Generating response..."):
            response = rag_with_url(url_input, model, text)
            st.session_state['chat_history'].append({"user": text, "ollama": response})
            st.write(response)

    st.write("## Chat History")
    for chat in reversed(st.session_state['chat_history']):
        st.write(f"**🧑 User**: {chat['user']}")
        st.write(f"**🧠 Assistant**: {chat['ollama']}")
        st.write("---")

if __name__ == "__main__":
    main()