utils.py

from langchain.embeddings import HuggingFaceEmbeddings


def setup_embedding_model(model_name: str):
    """
    Setup embedding model for RAG chatbot.
    Returns:
        Embedding_model: HuggingFaceEmbeddings object.
    """
    # Load embedding model
    embedding_model = HuggingFaceEmbeddings(model_name=model_name)
    return embedding_model


def load_text_data(file_path: str):
    """
    Load text data from a file.
    Args:
        file_path (str): Path to the file.
    Returns:
        data (list): List of strings representing the text data.
    """
    try:
        with open(file_path, "r") as file:
            data = file.read()
        return data
    except FileNotFoundError:
        raise FileNotFoundError(f"The file at {file_path} was not found.")
    except Exception as e:
        raise RuntimeError(f"An error occurred while reading the file: {e}")


def semantic_search(prompt, retriever):
    """
    Perform semantic search using the retriever.

    :param prompt: str, The user query or prompt.
    :param retriever: retriever object, Initialized retriever for the database.
    :return: List of relevant documents.
    """
    # Query the retriever with the user prompt
    results = retriever.get_relevant_documents(prompt)
    final = ""
    # Display the results
    for idx, doc in enumerate(results, 1):
        final += doc.page_content + "\n\n"

    return final


def resume_chat_completion(client, model, user_question, relevant_excerpts):
    """
    Generate a response to the user's question using the pre-trained model.
    Args:
        client (Groq): Initialized Groq client.
        model (str): The model to use for the chat completion.
        user_question (str): The user's question.
        relevant_excerpts (str): The relevant excerpts from the resume.
    Returns:
        response (str): The generated response to the user's question.
    """

    # Define the system prompt
    system_prompt = """
    You are an intelligent assistant designed to answer queries about Kacem Mathlouthi's professional background and experiences based on his resume. 

    Guidelines for generating responses:
    - Only use information directly found in the provided resume excerpts.
    - If the information is incomplete or ambiguous in the excerpts, inform the user that you lack sufficient data to answer.
    - If a user asks a general or unrelated question (e.g., about something that isn't part of the resume), you should politely indicate that you can only respond related to Kacem's resume.

    Please ensure that your answers are factual and reflect only the information available in the resume. Do not provide opinions or speculate beyond what is provided in the document.
    """

    # Generate a response to the user's question using the pre-trained model
    chat_completion = client.chat.completions.create(
        messages=[
            {"role": "system", "content": system_prompt},
            {
                "role": "user",
                "content": "User Question: "
                + user_question
                + "\n Relevant Kacem's Resume/CV Exerpt(s): \n"
                + relevant_excerpts,
            },
        ],
        model=model,
    )

    # Extract the response from the chat completion
    response = chat_completion.choices[0].message.content

    return response