app.py

import gradio as gr
import os
import spaces
from transformers import GemmaTokenizer, AutoModelForCausalLM
from transformers import AutoModelForCausalLM, AutoTokenizer, TextIteratorStreamer
from threading import Thread

# Set an environment variable
HF_TOKEN = os.environ.get("HF_TOKEN", None)


DESCRIPTION = '''
<div>
<h1 style="text-align: center;">HugCode</h1>
<p>CodeLLaMA sfted on [HugCode](https://huggingface.co/datasets/nuojohnchen/hugcode-codesft) data. Made in 2023.9. <a href="https://huggingface.co/FreedomIntelligence/Apollo-7B"><b>Meta Llama3 8b Chat</b></a></p>
</div>
'''

# LICENSE = """
# <p/>
# ---
# Built with Meta Llama 3
# """

PLACEHOLDER = """
<div style="padding: 30px; text-align: center; display: flex; flex-direction: column; align-items: center;">
   <img src="https://ysharma-dummy-chat-app.hf.space/file=/tmp/gradio/8e75e61cc9bab22b7ce3dec85ab0e6db1da5d107/Meta_lockup_positive%20primary_RGB.jpg" style="width: 80%; max-width: 550px; height: auto; opacity: 0.55;  "> 
   <h1 style="font-size: 28px; margin-bottom: 2px; opacity: 0.55;">HugCode</h1>
   <p style="font-size: 18px; margin-bottom: 2px; opacity: 0.65;">Ask me code questions (English/Chinese).</p>
</div>
"""


css = """
h1 {
  text-align: center;
  display: block;
}
#duplicate-button {
  margin: auto;
  color: white;
  background: #1565c0;
  border-radius: 100vh;
}
"""

# Load the tokenizer and model
tokenizer = AutoTokenizer.from_pretrained("nuojohnchen/codellama-7b-sft-v1.3")

model = AutoModelForCausalLM.from_pretrained("nuojohnchen/codellama-7b-sft-v1.3", device_map="auto")  # to("cuda:0") 
terminators = [
    tokenizer.eos_token_id,
    tokenizer.convert_tokens_to_ids("<|eot_id|>")
]

@spaces.GPU(duration=120)
def chat_llama3_8b(message: str, 
              history: list, 
              temperature: float, 
              max_new_tokens: int
             ) -> str:
    """
    Generate a streaming response using the llama3-8b model.
    Args:
        message (str): The input message.
        history (list): The conversation history used by ChatInterface.
        temperature (float): The temperature for generating the response.
        max_new_tokens (int): The maximum number of new tokens to generate.
    Returns:
        str: The generated response.
    """
    # Build conversation as pure array format
    history_messages = []
    for user, assistant in history:
        history_messages.extend(assistant)
    
    # conversation = [
    #     message,          # 当前消息
    #     history_messages  # 历史消息数组
    # ]
    conversation = ""
    for user, assistant in history:
        conversation += f"User: {user}\nAssistant: {assistant}<|endoftext|>\n"
    conversation += f"User: {message}\nAssistant: "
    
    tokenizer.chat_template = "User:{query}\nAssistant:{response}<|endoftext|>"
    input_ids = tokenizer.encode(conversation, return_tensors="pt").to(model.device)
    streamer = TextIteratorStreamer(tokenizer, timeout=10.0, skip_prompt=True, skip_special_tokens=True)

    generate_kwargs = dict(
        input_ids= input_ids,
        streamer=streamer,
        max_new_tokens=max_new_tokens,
        do_sample=True,
        temperature=temperature,
        eos_token_id=terminators,
    )
    # This will enforce greedy generation (do_sample=False) when the temperature is passed 0, avoiding the crash.             
    if temperature == 0:
        generate_kwargs['do_sample'] = False
        
    t = Thread(target=model.generate, kwargs=generate_kwargs)
    t.start()

    outputs = []
    for text in streamer:
        outputs.append(text)
        #print(outputs)
        yield "".join(outputs)
        

# Gradio block
chatbot=gr.Chatbot(height=450, placeholder=PLACEHOLDER, label='Gradio ChatInterface')

with gr.Blocks(fill_height=True, css=css) as demo:
    
    gr.Markdown(DESCRIPTION)
    # gr.DuplicateButton(value="Duplicate Space for private use", elem_id="duplicate-button")
    gr.ChatInterface(
        fn=chat_llama3_8b,
        chatbot=chatbot,
        fill_height=True,
        additional_inputs_accordion=gr.Accordion(label="⚙️ Parameters", open=False, render=False),
        additional_inputs=[
            gr.Slider(minimum=0,
                      maximum=1, 
                      step=0.1,
                      value=0.95, 
                      label="Temperature", 
                      render=False),
            gr.Slider(minimum=128, 
                      maximum=4096,
                      step=1,
                      value=512, 
                      label="Max new tokens", 
                      render=False ),
            ],
        examples=[
            ['Build a REST API that allows users to manage their to-do lists.'],
            ['Implement a machine learning model to predict stock prices based on historical data.'],
            ['Develop a web application that allows users to upload images and apply various filters.']
            ],
        cache_examples=False,
                     )
    
    # gr.Markdown(LICENSE)
    
if __name__ == "__main__":
    demo.launch()