import gradio as gr
from transformers import AutoTokenizer, AutoModelForCausalLM
import torch

model_name = "PostNetworkAcademy/gpt2-robotics-PostNetworkAcademy"

# Load tokenizer & model
tokenizer = AutoTokenizer.from_pretrained(model_name)
model = AutoModelForCausalLM.from_pretrained(model_name)

# Ensure tokens exist
if tokenizer.pad_token is None:
    tokenizer.pad_token = tokenizer.eos_token or "<|endoftext|>"

# Force model to CPU (avoids device mismatch)
device = torch.device("cuda" if torch.cuda.is_available() else "cpu")
model = model.to(device)

def generate_text(prompt, max_length=100, temperature=0.7, top_p=0.9):
    inputs = tokenizer(prompt, return_tensors="pt").to(device)
    outputs = model.generate(
        **inputs,
        max_length=max_length,
        temperature=temperature,
        top_p=top_p,
        do_sample=True,
        pad_token_id=tokenizer.pad_token_id,
        eos_token_id=tokenizer.eos_token_id
    )
    return tokenizer.decode(outputs[0], skip_special_tokens=True)

# Gradio UI
demo = gr.Interface(
    fn=generate_text,
    inputs=[
        gr.Textbox(lines=4, placeholder="Enter your robotics prompt...", label="Prompt"),
        gr.Slider(20, 500, value=100, step=10, label="Max Length"),
        gr.Slider(0.1, 1.5, value=0.7, step=0.1, label="Temperature"),
        gr.Slider(0.1, 1.0, value=0.9, step=0.05, label="Top-p"),
    ],
    outputs=gr.Textbox(label="Generated Response"),
    title="GPT-2 Robotics - PostNetworkAcademy",
    description="Fine-tuned GPT-2 model for robotics."
)

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