app.py

import gradio as gr
from huggingface_hub import InferenceClient
from datasets import load_dataset  # Import datasets library

# Load the PleIAs/common_corpus dataset
common_corpus = load_dataset("PleIAs/common_corpus")

# Function to retrieve an example from the dataset
def get_example_from_corpus(dataset, index):
    if "train" in dataset:
        example = dataset["train"][index]
        return example
    else:
        raise ValueError("Dataset does not have a 'train' split.")

# Initialize inference client
client = InferenceClient("unsloth/Llama-3.2-1B-Instruct")

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

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

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

    # Get response from model
    response = client.chat_completion(
        messages,
        max_tokens=max_tokens,
        temperature=temperature,
        top_p=top_p,
    ).choices[0].message.content

    return response

# Example usage of the dataset
example_data = get_example_from_corpus(common_corpus, index=0)
print("Example from PleIAs/common_corpus:", example_data)

# Gradio ChatInterface
demo = gr.ChatInterface(
    respond,
    additional_inputs=[
        gr.Textbox(value="You are a friendly Chatbot. Your name is Juninho.", 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()