app.py

import gradio as gr
import transformers
import torch
import logging
import html
import signal

# Setup logging
logging.basicConfig(level=logging.INFO)

# Constants
MAX_HISTORY_LENGTH = 5  # Adjust as needed

# Graceful shutdown handler
def shutdown_handler(signum, frame):
    logging.info("Shutting down gracefully...")
    exit(0)

signal.signal(signal.SIGINT, shutdown_handler)

def system_message_selector(choice, custom_message):
    if custom_message:
        return custom_message
    elif choice == "Friendly Chatbot":
        return "You are a friendly and helpful chatbot."
    elif choice == "Professional Assistant":
        return "You are a highly knowledgeable and professional assistant."
    elif choice == "Curious Researcher":
        return "You are a curious researcher who loves to explore new ideas."
    else:
        return "You are a helpful assistant."

def sanitize_input(text):
    return html.escape(text)

def validate_parameters(max_tokens, temperature, top_p):
    if not (1 <= max_tokens <= 1024):
        return False, "Error: 'Max new tokens' must be between 1 and 1024."
    if not (0.1 <= temperature <= 4.0):
        return False, "Error: 'Temperature' must be between 0.1 and 4.0."
    if not (0.1 <= top_p <= 1.0):
        return False, "Error: 'Top-p' must be between 0.1 and 1.0."
    return True, ""

# Determine the device
device = torch.device("cuda" if torch.cuda.is_available() else "cpu")

# Load the model and tokenizer
model_name = "gpt2"  # Use GPT-2 model

try:
    tokenizer = transformers.AutoTokenizer.from_pretrained(model_name)
    model = transformers.AutoModelForCausalLM.from_pretrained(model_name)
    model.to(device)
    model.eval()
except Exception as e:
    logging.error(f"Failed to load model {model_name}: {e}")
    exit(1)

def respond(message, history, persona_choice, custom_persona, max_tokens, temperature, top_p):
    is_valid, error_message = validate_parameters(max_tokens, temperature, top_p)
    if not is_valid:
        return error_message

    safe_message = sanitize_input(message)
    safe_history = [(sanitize_input(u), sanitize_input(b)) for u, b in history]
    truncated_history = safe_history[-MAX_HISTORY_LENGTH:]
    system_message = system_message_selector(persona_choice, custom_persona)

    # Build the conversation prompt
    conversation = system_message + "\n\n"
    for user_msg, bot_msg in truncated_history:
        conversation += f"User: {user_msg}\n"
        conversation += f"Assistant: {bot_msg}\n"
    conversation += f"User: {safe_message}\nAssistant:"

    logging.info(f"Received message: {safe_message}")

    try:
        input_ids = tokenizer.encode(conversation, return_tensors="pt").to(device)

        output_ids = model.generate(
            input_ids,
            max_new_tokens=max_tokens,
            temperature=temperature,
            top_p=top_p,
            do_sample=True,
            pad_token_id=tokenizer.eos_token_id,
            eos_token_id=tokenizer.eos_token_id,
        )

        generated_text = tokenizer.decode(
            output_ids[0][input_ids.shape[-1]:], skip_special_tokens=True
        )
        return generated_text.strip()
    except Exception as e:
        logging.error(f"An error occurred: {e}")
        return "I'm sorry, but something went wrong. Please try again."

# Create the UI components
system_message_radio = gr.Radio(
    choices=["Friendly Chatbot", "Professional Assistant", "Curious Researcher"],
    value="Friendly Chatbot",
    label="Choose a Persona",
)

system_message_textbox = gr.Textbox(
    placeholder="Enter custom persona or system message...",
    label="Custom Persona (Optional)",
)

max_tokens_slider = gr.Slider(
    minimum=1, maximum=1024, value=50, step=1, label="Max new tokens"
)

temperature_slider = gr.Slider(
    minimum=0.1, maximum=4.0, value=1.0, step=0.1, label="Temperature"
)

top_p_slider = gr.Slider(
    minimum=0.1, maximum=1.0, value=0.9, step=0.05, label="Top-p (nucleus sampling)"
)

# Create the ChatInterface
demo = gr.ChatInterface(
    fn=respond,
    additional_inputs=[
        system_message_radio,
        system_message_textbox,
        max_tokens_slider,
        temperature_slider,
        top_p_slider,
    ],
    allow_reset_history=True,
    title="Customizable Chatbot Interface",
    description="Choose a persona or enter a custom one, and adjust parameters as needed.",
)

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