app.py

import gradio as gr
from huggingface_hub import InferenceClient
from datasets import load_dataset


dataset = load_dataset("pdf2dataset/d6a9686436c50afc3bcf0276ae378fe8")

"""
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 process_context():
    # if input_text.lower() == "sair":
    #     return "Encerrando o chat. Até mais!"
    
    # if input_text.lower() == "mostrar dados":
    #     # Exemplo simples: Mostra alguns registros do dataset
    #     sample_data = dataset['train'][:5]
    #     return f"Aqui estão alguns dados do dataset:\n {sample_data}"

    # Gera uma resposta a partir da LLM, passando o dataset como contexto
    # Você pode passar partes do dataset e a pergunta do usuário para a LLM responder
    context = f"O dataset contém dados que podem ser descritos assim: {dataset['train'][:2]}"
    
    # Cria a prompt de pergunta para a LLM
    # prompt = f"{context}\n\nPergunta: {input_text}\nResposta:"
    
    # # Envia a prompt para o modelo LLM e obtém a resposta
    # response = client.text_generation(prompt, max_new_tokens=200)
    
    return context


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]})
    context = process_context()

    prompt = f"{context}\n\nPergunta: {message}\nResposta:"
    messages.append({"role": "user", "content": prompt})

    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. your name is MatchGpt", 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()