app.py

import gradio as gr
from peft import PeftModel, PeftTokenizer
from transformers import TextStreamer

# Load model directly
from transformers import AutoModel, AutoTokenizer

"""
For more information on `huggingface_hub` Inference API support, please check the docs: https://huggingface.co/docs/huggingface_hub/v0.22.2/en/guides/inference
Info of how to use a model after training on hf https://huggingface.co/docs/trl/main/en/use_model
"""

model_name_or_path = "unsloth/Llama-3.2-3B-Instruct"
adapter_name = "samlama111/lora_model"

model = AutoModel.from_pretrained(model_name_or_path)
tokenizer = AutoTokenizer.from_pretrained(model_name_or_path)

model = PeftModel.from_pretrained(model, adapter_name)
tokenizer = PeftTokenizer.from_pretrained(tokenizer, adapter_name)


def respond(
    message,
    history: list[tuple[str, str]],
    system_message,
    max_tokens,
    temperature,
    top_p,
):
    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})

    response = ""

    inputs = tokenizer.apply_chat_template(
        messages, tokenize=True, add_generation_prompt=True, return_tensors="pt"
    )

    text_streamer = TextStreamer(tokenizer)
    # TODO: Doesn't stream ATM
    for message in model.generate(
        input_ids=inputs, streamer=text_streamer, max_new_tokens=1024, use_cache=True
    ):
        # Decode the tensor to a string
        decoded_message = tokenizer.decode(message, skip_special_tokens=True)

        # Manually getting the response
        response = decoded_message.split("assistant")[
            -1
        ].strip()  # Extract only the assistant's response
        print(response)

        yield response


"""
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()