import gradio as gr
from huggingface_hub import InferenceClient

from transformers import AutoTokenizer, AutoModelForCausalLM
from peft import PeftModel
import torch

# Base model name
base_model_name = "meta-llama/Llama-3.2-3B-Instruct"

# Load the base model and tokenizer
tokenizer = AutoTokenizer.from_pretrained(base_model_name)
base_model = AutoModelForCausalLM.from_pretrained(
    base_model_name,
    device_map="auto",  # Automatically map to GPU if available
    torch_dtype=torch.float16,  # Use float16 for better performance on GPU
)

# Fine-tuned LoRA adapter
lora_model_name = "shanaka95/autotrain-sios2"

# Load the LoRA adapter and merge it with the base model
model = PeftModel.from_pretrained(base_model, lora_model_name)

# Move the model to GPU if available
device = "cuda" if torch.cuda.is_available() else "cpu"
model = model.to(device)

def respond(
    message,
    history: list[tuple[str, str]],
    system_message,
    max_tokens,
    temperature,
    top_p,
):
    inputs = tokenizer(message, return_tensors="pt").to(device)
    
    # Generate response
    outputs = model.generate(
        inputs.input_ids,
        max_length=300,
        temperature=0.2,
        top_p=0.8,
        do_sample=True,
        eos_token_id=tokenizer.eos_token_id,
        max_new_tokens=128,
    )
    
    # Decode the response
    return tokenizer.decode(outputs[0], skip_special_tokens=True)


"""
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.", 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(show_error=True)