app.py

import gradio as gr
import os
from langchain.document_loaders import WebBaseLoader
from langchain_community.vectorstores import FAISS
from langchain.text_splitter import RecursiveCharacterTextSplitter
from langchain.chains import ConversationalRetrievalChain
from langchain.chains import ConversationChain
from langchain.memory import ConversationBufferMemory
from langchain_community.llms import HuggingFaceEndpoint
from sentence_transformers import SentenceTransformer
from langchain_ollama import ChatOllama
import re
import torch



def preprocessing_text(document: list) -> list:
    document[0].page_content = re.sub(r"\n{2,}", "\n\n", document[0].page_content)
    return document

def loading_the_webpage(url: str) -> list:
    loader = WebBaseLoader(url)
    document = preprocessing_text(loader.load())
    return document


def chunking(document: list) -> list:
    text_splitter = RecursiveCharacterTextSplitter(chunk_size= 1024, 
                                               chunk_overlap= 128, 
                                               separators= ["\n\n", "\n", " ", ""])
    return text_splitter.split_documents(documents= document)



def create_vector_db(chunked_documents):
    embeddings = SentenceTransformer("nomic-ai/nomic-embed-text-v1", trust_remote_code=True)
    vector_db = FAISS.from_documents(chunked_documents, embeddings)
    return vector_db


# Initialize langchain LLM chain
def initialize_llmchain(llm_model, temperature, max_tokens, top_k, vector_db, progress=gr.Progress()):
    llm = ChatOllama(model= "mistral",
                    temperature= temperature,
                    top_k= top_k, 
                    num_predict= max_tokens)
    
    memory = ConversationBufferMemory(
        memory_key="chat_history",
        output_key='answer',
        return_messages=True
    )

    retriever=vector_db.as_retriever()
    qa_chain = ConversationalRetrievalChain.from_llm(
        llm,
        retriever=retriever,
        chain_type="stuff", 
        memory=memory,
        return_source_documents=True,
        verbose=False,
    )
    return qa_chain

def process_url_and_query(url: str, query: str):
    # Load and process the webpage
    documents = loading_the_webpage(url)
    documents = preprocessing_text(documents)
    
    # Chunk the documents
    chunked_documents = chunking(documents)
    
    # Create a vector database from chunked documents
    vector_db = create_vector_db(chunked_documents)
    
    # Initialize the LLM chain
    qa_chain = initialize_llmchain(llm_model="mistral", temperature=0.7, max_tokens=150, top_k=5, vector_db=vector_db)
    
    # Get the answer for the user's query
    answer = qa_chain({"question": query})
    
    return answer['answer']

with gr.Blocks() as demo:
    gr.Markdown("# Webpage Querying App")
    
    url_input = gr.Textbox(label="Enter URL")
    query_input = gr.Textbox(label="Enter your query")
    
    submit_button = gr.Button("Submit")
    
    output_textbox = gr.Textbox(label="Response", interactive=False)

    submit_button.click(process_url_and_query, inputs=[url_input, query_input], outputs=output_textbox)

# Launch the app
demo.launch()































# import gradio as gr
# from huggingface_hub import InferenceClient

# """
# For more information on `huggingface_hub` Inference API support, please check the docs: https://huggingface.co/docs/huggingface_hub/v0.22.2/en/guides/inference
# """
# client = InferenceClient("HuggingFaceH4/zephyr-7b-beta")


# def respond(
#     message,
#     history: list[tuple[str, str]],
#     system_message,
#     max_tokens,
#     temperature,
#     top_p,
# ):
#     messages = [{"role": "system", "content": system_message}]

#     for val in history:
#         if val[0]:
#             messages.append({"role": "user", "content": val[0]})
#         if val[1]:
#             messages.append({"role": "assistant", "content": val[1]})

#     messages.append({"role": "user", "content": message})

#     response = ""

#     for message in client.chat_completion(
#         messages,
#         max_tokens=max_tokens,
#         stream=True,
#         temperature=temperature,
#         top_p=top_p,
#     ):
#         token = message.choices[0].delta.content

#         response += token
#         yield response


# """
# For information on how to customize the ChatInterface, peruse the gradio docs: https://www.gradio.app/docs/chatinterface
# """
# demo = gr.ChatInterface(
#     respond,
#     additional_inputs=[
#         gr.Textbox(value="You are a friendly Chatbot.", label="System message"),
#         gr.Slider(minimum=1, maximum=2048, value=512, step=1, label="Max new tokens"),
#         gr.Slider(minimum=0.1, maximum=4.0, value=0.7, step=0.1, label="Temperature"),
#         gr.Slider(
#             minimum=0.1,
#             maximum=1.0,
#             value=0.95,
#             step=0.05,
#             label="Top-p (nucleus sampling)",
#         ),
#     ],
# )


# if __name__ == "__main__":
#     demo.launch()