app.py

import gradio as gr
from transformers import GPT2LMHeadModel, GPT2Tokenizer
import torch
import time

# Load model and tokenizer from Hugging Face Hub
model_name = "Electricarchmage/cookbookgpt"
model = GPT2LMHeadModel.from_pretrained(model_name)
tokenizer = GPT2Tokenizer.from_pretrained(model_name)

# Set the pad_token to eos_token and padding_side to 'left'
tokenizer.pad_token = tokenizer.eos_token
tokenizer.padding_side = 'left'

# Define the respond function with logging for debugging
def respond(
    message,
    history: list[tuple[str, str]],
    system_message,
    max_tokens,
    temperature,
    top_p,
):
    # Preparing the messages for context (the history and the new message)
    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]})
    
    messages.append({"role": "user", "content": message})

    # Tokenize the input
    inputs = tokenizer([msg["content"] for msg in messages], return_tensors="pt", padding=True, truncation=True)
    attention_mask = inputs.get('attention_mask', torch.ones_like(inputs['input_ids']))

    start_time = time.time()  # Start the timer

    # Generate output tokens
    try:
        output = model.generate(
            inputs["input_ids"],
            attention_mask=attention_mask,
            max_length=max_tokens + len(inputs["input_ids"][0]),
            temperature=temperature,
            top_p=top_p,
            num_return_sequences=1,
            do_sample=True,
            no_repeat_ngram_size=2,
        )
    except Exception as e:
        return f"Error during generation: {str(e)}"

    generation_time = time.time() - start_time  # Time taken for generation

    # Decode the output tokens into text
    response = tokenizer.decode(output[0], skip_special_tokens=True)

    # Extract only the assistant's reply
    assistant_reply = response.split("Assistant:")[-1].strip()

    # Add generation time in the response for debugging
    return f"Response: {assistant_reply}\nGeneration time: {generation_time:.2f} seconds"

# Define the Gradio interface
demo = gr.ChatInterface(
    respond,
    additional_inputs=[
        gr.Textbox(value="You are a friendly Chatbot.", 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)",
        ),
    ],
)

# Launch the app
if __name__ == "__main__":
    demo.launch()