web_multiturn.py

# Copyright (c) Alibaba Cloud.
#
# This source code is licensed under the license found in the
# LICENSE file in the root directory of this source tree.

"""A simple web interactive chat demo based on gradio."""

from argparse import ArgumentParser
from threading import Thread

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

DEFAULT_CKPT_PATH = '../Qwen1.5-7B-Chat'


def _get_args():
    parser = ArgumentParser()
    parser.add_argument("-m", "--base_model", type=str, default=DEFAULT_CKPT_PATH,
                        help="Checkpoint name or path, default to %(default)r")
    parser.add_argument("--cpu_only", action="store_true", help="Run demo with CPU only")

    parser.add_argument("--server_port", type=int, default=8000,
                        help="Demo server port.")
    parser.add_argument("--server_name", type=str, default="0.0.0.0",
                        help="Demo server name.")

    args = parser.parse_args()
    return args


def _load_model_tokenizer(args):
    tokenizer = AutoTokenizer.from_pretrained(
        args.base_model, resume_download=True, trust_remote_code=True,
    )

    if args.cpu_only:
        device_map = "cpu"
    else:
        device_map = "auto"

    model = AutoModelForCausalLM.from_pretrained(
        args.base_model,
        torch_dtype="auto",
        device_map=device_map,
        resume_download=True,
        trust_remote_code=True,
    ).eval()
    model.generation_config.max_new_tokens = 2048   # For chat.

    return model, tokenizer


def _chat_stream(model, tokenizer, query, system, history):
    sys = 'You are a helpful assistant.'
    if len(system) > 0:
        sys = system
    
    conversation = [
        {'role': 'system', 'content': sys },
    ]
    for query_h, response_h in history:
        conversation.append({'role': 'user', 'content': query_h})
        conversation.append({'role': 'assistant', 'content': response_h})
    conversation.append({'role': 'user', 'content': query})
    #print(conversation)
    inputs = tokenizer.apply_chat_template(
        conversation,
        add_generation_prompt=True,
        return_tensors='pt',
    )
    inputs = inputs.to(model.device)
    streamer = TextIteratorStreamer(tokenizer=tokenizer, skip_prompt=True, timeout=60.0, skip_special_tokens=True)
    generation_kwargs = dict(
        input_ids=inputs,
        streamer=streamer,
    )
    thread = Thread(target=model.generate, kwargs=generation_kwargs)
    thread.start()

    for new_text in streamer:
        yield new_text


def _gc():
    import gc
    gc.collect()
    if torch.cuda.is_available():
        torch.cuda.empty_cache()


def _launch_demo(args, model, tokenizer):

    def predict(_query, _system, _chatbot, _task_history):
        print(f"System: {_system}")
        print(f"User: {_query}")
        _chatbot.append((_query, ""))
        full_response = ""
        response = ""
        for new_text in _chat_stream(model, tokenizer, _query, _system, history=_task_history):
            response += new_text
            _chatbot[-1] = (_query, response)

            yield _chatbot
            full_response = response

        print(f"History: {_task_history}")
        _task_history.append((_query, full_response))
        print(f"Chatbot: {full_response}")

    def regenerate(_system, _chatbot, _task_history):
        if not _task_history:
            yield _chatbot
            return
        item = _task_history.pop(-1)
        _chatbot.pop(-1)
        yield from predict(item[0], _system, _chatbot, _task_history)

    def reset_user_input():
        return gr.update(value="")

    def reset_state(_chatbot, _task_history):
        _task_history.clear()
        _chatbot.clear()
        _gc()
        return _chatbot

    with gr.Blocks() as demo:
        gr.Markdown("""<center><font size=8>LLMUtils Chatbot</center>""")

        chatbot = gr.Chatbot(elem_classes="control-height", height=500)
        system = gr.Textbox(lines=2, label='System')
        query = gr.Textbox(lines=1, label='Input')
        task_history = gr.State([])

        with gr.Row():
            empty_btn = gr.Button("🧹 Clear History (清除历史)")
            submit_btn = gr.Button("🚀 Submit (发送)")
            regen_btn = gr.Button("🤔️ Regenerate (重试)")

        submit_btn.click(predict, [query, system, chatbot, task_history], [chatbot], show_progress=True)
        submit_btn.click(reset_user_input, [], [query])
        empty_btn.click(reset_state, [chatbot, task_history], outputs=[chatbot], show_progress=True)
        regen_btn.click(regenerate, [system, chatbot, task_history], [chatbot], show_progress=True)

    demo.queue().launch(
        server_port=args.server_port,
        server_name=args.server_name,
    )


def main():
    args = _get_args()

    model, tokenizer = _load_model_tokenizer(args)

    _launch_demo(args, model, tokenizer)


if __name__ == '__main__':
    main()