app.py

import gradio as gr
from huggingface_hub import InferenceClient
from transformers import TextStreamer
# from peft import AutoPeftModelForCausalLM
# from transformers import AutoTokenizer
from unsloth import FastLanguageModel
from unsloth.chat_templates import get_chat_template


"""
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
"""

model_name_or_path = "samlama111/lora_model"

# client = InferenceClient(model_name_or_path)
model, tokenizer = FastLanguageModel.from_pretrained(
    model_name = model_name_or_path,
    max_seq_length = 8192,
    load_in_4bit = True,
    # token = "hf_...", # No need since our model is public
)   

tokenizer = get_chat_template(
    tokenizer,
    chat_template = "llama-3.1",
    mapping = {"role" : "from", "content" : "value", "user" : "human", "assistant" : "gpt"}, # ShareGPT style
)
FastLanguageModel.for_inference(model) # Enable native 2x faster inference

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()