retrieval.py

"""
LLM chain retrieval
"""

import json
import gradio as gr

from langchain.chains.conversational_retrieval.base import ConversationalRetrievalChain
from langchain.memory import ConversationBufferMemory
from langchain_huggingface import HuggingFaceEndpoint
from langchain_core.prompts import PromptTemplate


# Initialize langchain LLM chain
def initialize_llmchain(
    llm_model,
    huggingfacehub_api_token,
    temperature,
    max_tokens,
    top_k,
    vector_db,
    progress=gr.Progress(),
):
    """Initialize Langchain LLM chain"""

    progress(0.1, desc="Initializing HF tokenizer...")
    progress(0.5, desc="Initializing HF Hub...")

    llm = HuggingFaceEndpoint(
        repo_id=llm_model,
        task="text-generation",
        provider="hf-inference",
        temperature=temperature,
        max_new_tokens=max_tokens,
        top_k=top_k,
        huggingfacehub_api_token=huggingfacehub_api_token,
    )

    progress(0.75, desc="Defining buffer memory...")
    memory = ConversationBufferMemory(
        memory_key="chat_history",
        output_key="answer",
        return_messages=True,
    )
    retriever = vector_db.as_retriever(search_type="similarity", search_kwargs={'k': top_k})

    progress(0.8, desc="Defining retrieval chain...")
    with open('prompt_template.json', 'r') as file:
        system_prompt = json.load(file)
    prompt_template = system_prompt["prompt"]
    rag_prompt = PromptTemplate(
        template=prompt_template, input_variables=["context", "question"]
    )

    qa_chain = ConversationalRetrievalChain.from_llm(
        llm,
        retriever=retriever,
        chain_type="stuff",
        memory=memory,
        combine_docs_chain_kwargs={"prompt": rag_prompt},
        return_source_documents=True,
        verbose=False,
    )

    progress(0.9, desc="Done!")
    return qa_chain


# Format chat history
def format_chat_history(message, chat_history):
    """Format chat history for LLM"""
    formatted_chat_history = []
    for user_message, bot_message in chat_history:
        formatted_chat_history.append(f"User: {user_message}")
        formatted_chat_history.append(f"Assistant: {bot_message}")
    return formatted_chat_history


# Invoke QA chain with history
def invoke_qa_chain(qa_chain, message, history):
    """Invoke question-answering chain"""
    formatted_chat_history = format_chat_history(message, history)

    response = qa_chain.invoke({
        "question": message,
        "chat_history": formatted_chat_history,
    })

    response_sources = response["source_documents"]
    response_answer = response["answer"]

    # Clean up if "Helpful Answer:" is included
    if "Helpful Answer:" in response_answer:
        response_answer = response_answer.split("Helpful Answer:")[-1].strip()

    new_history = history + [(message, response_answer)]
    return qa_chain, new_history, response_sources