【书生浦语实战】MindSearch 部署到HuggingFace Space

结果速览

欢迎来玩：https://huggingface.co/spaces/LLyn/mindsearch_exercise

配置开发环境

使用github codespace

第一次使用github的codespace～本质上跟在intern studio一样，但是页面是vscode效果（intern studio是linux cli窗口），浏览器会自动在新的页面打开一个web版的vscode，比起用ssh访问intern studio的优势是部署服务后不需要设置本地ssh和远程再链接啦

步骤：左侧导航栏选择 Codespace、选择Blank 空白模版， Use this template进入IDE

安装环境

mkdir -p /workspaces/mindsearch
cd /workspaces/mindsearch
git clone https://github.com/InternLM/MindSearch.git
cd MindSearch && git checkout b832275 && cd ..

# 创建环境
conda create -n mindsearch python=3.10 -y
# 激活环境
conda activate mindsearch
# 安装依赖
pip install -r /workspaces/mindsearch/MindSearch/requirements.txt

除了requirements.txt里面的依赖包，还要安装class_registry、（不然运行lagent会报错、我之前是直接去修改了源码导入，但现在部署到hf了只能装包解决了）

pip install class_registry

获取硅基流动 API Key

因为要使用硅基流动的 API Key，所以接下来便是注册并获取 API Key 了。

首先，我们打开 https://account.siliconflow.cn/login 来注册硅基流动的账号（如果注册过，则直接登录即可）。

在完成注册后，打开 https://cloud.siliconflow.cn/account/ak 来准备 API Key。首先创建新 API 密钥，然后点击密钥进行复制，以备后续使用。

调试前后端

再github space先运行一下代码看下效果

export SILICON_API_KEY=第二步中复制的密钥
conda activate mindsearch
cd /workspaces/mindsearch/MindSearch
python -m mindsearch.app --lang cn --model_format internlm_silicon --search_engine DuckDuckGoSearch

conda activate mindsearch
cd /workspaces/mindsearch/MindSearch
python frontend/mindsearch_gradio.py

效果如下：

部署到 HuggingFace Space

配置sillionflow的key

再hugging face创建好的space中 setting里面找到如下模块，配置好SILICON_API_KEY

（看图上两条key就知道小白踩过雷…）

记得secret的变量名称要叫SILICON_API_KEY，不要改成别的（除非mindset.agent.models里面定义好了其他环境变量key的名称！，下图是SILICON_API_KEY引用来源）

编写app.py

# 创建新目录
mkdir -p /workspaces/mindsearch/mindsearch_deploy
# 准备复制文件
cd /workspaces/mindsearch
cp -r /workspaces/mindsearch/MindSearch/mindsearch /workspaces/mindsearch/mindsearch_deploy
cp /workspaces/mindsearch/MindSearch/requirements.txt /workspaces/mindsearch/mindsearch_deploy
# 创建 app.py 作为程序入口
touch /workspaces/mindsearch/mindsearch_deploy/app.py

本人稍微调整了一下gradio页面布局，把输入框放到结果框前面了

import json
import os

import gradio as gr
import requests
from lagent.schema import AgentStatusCode

os.system("python -m mindsearch.app --lang cn --model_format internlm_silicon &")

PLANNER_HISTORY = []
SEARCHER_HISTORY = []


def rst_mem(history_planner: list, history_searcher: list):
    '''
    Reset the chatbot memory.
    '''
    history_planner = []
    history_searcher = []
    if PLANNER_HISTORY:
        PLANNER_HISTORY.clear()
    return history_planner, history_searcher


def format_response(gr_history, agent_return):
    if agent_return['state'] in [
            AgentStatusCode.STREAM_ING, AgentStatusCode.ANSWER_ING
    ]:
        gr_history[-1][1] = agent_return['response']
    elif agent_return['state'] == AgentStatusCode.PLUGIN_START:
        thought = gr_history[-1][1].split('```')[0]
        if agent_return['response'].startswith('```'):
            gr_history[-1][1] = thought + '\n' + agent_return['response']
    elif agent_return['state'] == AgentStatusCode.PLUGIN_END:
        thought = gr_history[-1][1].split('```')[0]
        if isinstance(agent_return['response'], dict):
            gr_history[-1][
                1] = thought + '\n' + f'```json\n{json.dumps(agent_return["response"], ensure_ascii=False, indent=4)}\n```'  # noqa: E501
    elif agent_return['state'] == AgentStatusCode.PLUGIN_RETURN:
        assert agent_return['inner_steps'][-1]['role'] == 'environment'
        item = agent_return['inner_steps'][-1]
        gr_history.append([
            None,
            f"```json\n{json.dumps(item['content'], ensure_ascii=False, indent=4)}\n```"
        ])
        gr_history.append([None, ''])
    return


def predict(history_planner, history_searcher):

    def streaming(raw_response):
        for chunk in raw_response.iter_lines(chunk_size=8192,
                                             decode_unicode=False,
                                             delimiter=b'\n'):
            if chunk:
                decoded = chunk.decode('utf-8')
                if decoded == '\r':
                    continue
                if decoded[:6] == 'data: ':
                    decoded = decoded[6:]
                elif decoded.startswith(': ping - '):
                    continue
                response = json.loads(decoded)
                yield (response['response'], response['current_node'])

    global PLANNER_HISTORY
    PLANNER_HISTORY.append(dict(role='user', content=history_planner[-1][0]))
    new_search_turn = True

    url = 'http://localhost:8002/solve'
    headers = {'Content-Type': 'application/json'}
    data = {'inputs': PLANNER_HISTORY}
    raw_response = requests.post(url,
                                 headers=headers,
                                 data=json.dumps(data),
                                 timeout=20,
                                 stream=True)

    for resp in streaming(raw_response):
        agent_return, node_name = resp
        if node_name:
            if node_name in ['root', 'response']:
                continue
            agent_return = agent_return['nodes'][node_name]['detail']
            if new_search_turn:
                history_searcher.append([agent_return['content'], ''])
                new_search_turn = False
            format_response(history_searcher, agent_return)
            if agent_return['state'] == AgentStatusCode.END:
                new_search_turn = True
            yield history_planner, history_searcher
        else:
            new_search_turn = True
            format_response(history_planner, agent_return)
            if agent_return['state'] == AgentStatusCode.END:
                PLANNER_HISTORY = agent_return['inner_steps']
            yield history_planner, history_searcher
    return history_planner, history_searcher


with gr.Blocks() as demo:
    gr.HTML("""<h1 align="center">MindSearch Gradio Demo</h1>""")
    gr.HTML("""<p style="text-align: center; font-family: Arial, sans-serif;">MindSearch is an open-source AI Search Engine Framework with Perplexity.ai Pro performance. You can deploy your own Perplexity.ai-style search engine using either closed-source LLMs (GPT, Claude) or open-source LLMs (InternLM2.5-7b-chat).</p>""")
    gr.HTML("""
    <div style="text-align: center; font-size: 16px;">
        <a href="https://github.com/InternLM/MindSearch" style="margin-right: 15px; text-decoration: none; color: #4A90E2;">🔗 GitHub</a>
        <a href="https://arxiv.org/abs/2407.20183" style="margin-right: 15px; text-decoration: none; color: #4A90E2;">📄 Arxiv</a>
        <a href="https://huggingface.co/papers/2407.20183" style="margin-right: 15px; text-decoration: none; color: #4A90E2;">📚 Hugging Face Papers</a>
        <a href="https://huggingface.co/spaces/internlm/MindSearch" style="text-decoration: none; color: #4A90E2;">🤗 Hugging Face Demo</a>
    </div>
    """)
    with gr.Row():
        with gr.Column(scale=10):
            with gr.Row():
                user_input = gr.Textbox(show_label=False,
                                        placeholder='介绍一下Mindseach',
                                        lines=5,
                                        container=False)
            with gr.Row():
                with gr.Column(scale=2):
                    submitBtn = gr.Button('Submit')
                with gr.Column(scale=1, min_width=20):
                    emptyBtn = gr.Button('Clear History')

            with gr.Row():
                with gr.Column():
                    planner = gr.Chatbot(label='planner',
                                         height=700,
                                         show_label=True,
                                         show_copy_button=True,
                                         bubble_full_width=False,
                                         render_markdown=True)
                with gr.Column():
                    searcher = gr.Chatbot(label='searcher',
                                          height=700,
                                          show_label=True,
                                          show_copy_button=True,
                                          bubble_full_width=False,
                                          render_markdown=True)

    def user(query, history):
        return '', history + [[query, '']]

    submitBtn.click(user, [user_input, planner], [user_input, planner],
                    queue=False).then(predict, [planner, searcher],
                                      [planner, searcher])
    emptyBtn.click(rst_mem, [planner, searcher], [planner, searcher],
                   queue=False)

demo.queue()
demo.launch(server_name='0.0.0.0',
            server_port=7860,
            inbrowser=True,
            share=True)

提交到 HF space

首先创建一个有写权限的token

记得要选permissions=write！

然后拷贝代码：

教程里面给的git remote set-url稍微有点错，应该是 git remote set-url ；remote-name是远程仓的名称、或者用origin（教程给的“space”会报错找不到指定的仓！）

cd /workspaces/codespaces-blank
git clone https://huggingface.co/spaces/<你的名字>/<仓库名称>
# 把token挂到仓库上，让自己有写权限
git remote set-url origin https://<你的名字>:<上面创建的token>@huggingface.co/spaces/<你的名字>/<仓库名称>

## my case
git clone  https://huggingface.co/spaces/LLyn/mindsearch_exercise
git remote set-url origin https://LLyn:hf_token@huggingface.co/spaces/LLyn/mindsearch_exercise

然后把代码搬运到建好的仓里面（这里教程又错了，最后一期写的有点潦草啊喂…难道老师故意的？？？cp要加上-r不然文件夹拷贝不过来）

cd /workspaces/codespaces-blank/mindsearch_exercise
cp -r /workspaces/mindsearch/mindsearch_deploy/* .

注意mindsearch文件夹其实是Mindsearch项目中的一个子文件夹，文件夹完整内容如下：

(base) @lynnelian ➜ /workspaces/codespaces-blank/mindsearch_exercise (main) $ tree
.
├── README.md
├── app.py
├── mindsearch
│   ├── __pycache__
│   │   ├── app.cpython-310.pyc
│   │   └── app.cpython-312.pyc
│   ├── agent
│   │   ├── __init__.py
│   │   ├── __pycache__
│   │   │   ├── __init__.cpython-310.pyc
│   │   │   ├── mindsearch_agent.cpython-310.pyc
│   │   │   ├── mindsearch_prompt.cpython-310.pyc
│   │   │   └── models.cpython-310.pyc
│   │   ├── mindsearch_agent.py
│   │   ├── mindsearch_prompt.py
│   │   └── models.py
│   ├── app.py
│   └── terminal.py
└── requirements.txt

4 directories, 15 files

然后git提交

git add .
git commit -m "update"
git push

Hugging Face Space效果验证

进入https://huggingface.co/spaces/LLyn/mindsearch_exercise
输入测试query，成功！