【书生3.2】Lagent 自定义你的 Agent 智能体

1 环境准备

直接复用之前的xtuner或者llamaindex环境

# 安装其他依赖包
conda activate llamainde
pip install termcolor==2.4.0
pip install lmdeploy==0.5.2

pip install griffe==0.48 # 重要

2 启动模型

# 这里我使用的是早前下载的internlm2-chat-7b
conda activate llamainde
lmdeploy serve api_server /project/serve/xtuner/Shanghai_AI_Laboratory/internlm2-chat-7b --model-name internlm2-chat-7b

3 部署lagent

另启动一个窗口

# 创建目录以存放代码
mkdir -p /project/serve/camp3
cd /project/serve/camp3
git clone https://github.com/InternLM/lagent.git
cd lagent && git checkout 81e7ace && pip install -e . && cd ..

cd /project/serve/camp3/lagent
conda activate llamaindex
streamlit run examples/internlm2_agent_web_demo.py

4 可视化页面及配置

注意模型ip不要带http

这里用的internlm2-chat-7b， 使用工具的效果不是很直接，它会再询问一下。

5 自定义工具的智能体

(base) [root@localhost ~]# cd /project/serve/camp3/lagent/
(base) [root@localhost lagent]# touch lagent/actions/magicmaker.py
(base) [root@localhost lagent]# vim lagent/actions/magicmaker.py 
(base) [root@localhost lagent]# vim examples/internlm2_agent_web_demo.py 
(base) [root@localhost lagent]# 


#magicmaker.py
import json
import requests

from lagent.actions.base_action import BaseAction, tool_api
from lagent.actions.parser import BaseParser, JsonParser
from lagent.schema import ActionReturn, ActionStatusCode


class MagicMaker(BaseAction):
    styles_option = [
        'dongman',  # 动漫
        'guofeng',  # 国风
        'xieshi',   # 写实
        'youhua',   # 油画
        'manghe',   # 盲盒
    ]
    aspect_ratio_options = [
        '16:9', '4:3', '3:2', '1:1',
        '2:3', '3:4', '9:16'
    ]

    def __init__(self,
                 style='guofeng',
                 aspect_ratio='4:3'):
        super().__init__()
        if style in self.styles_option:
            self.style = style
        else:
            raise ValueError(f'The style must be one of {self.styles_option}')
        
        if aspect_ratio in self.aspect_ratio_options:
            self.aspect_ratio = aspect_ratio
        else:
            raise ValueError(f'The aspect ratio must be one of {aspect_ratio}')
    
    @tool_api
    def generate_image(self, keywords: str) -> dict:
        """Run magicmaker and get the generated image according to the keywords.

        Args:
            keywords (:class:`str`): the keywords to generate image

        Returns:
            :class:`dict`: the generated image
                * image (str): path to the generated image
        """
        try:
            response = requests.post(
                url='https://magicmaker.openxlab.org.cn/gw/edit-anything/api/v1/bff/sd/generate',
                data=json.dumps({
                    "official": True,
                    "prompt": keywords,
                    "style": self.style,
                    "poseT": False,
                    "aspectRatio": self.aspect_ratio
                }),
                headers={'content-type': 'application/json'}
            )
        except Exception as exc:
            return ActionReturn(
                errmsg=f'MagicMaker exception: {exc}',
                state=ActionStatusCode.HTTP_ERROR)
        image_url = response.json()['data']['imgUrl']
        return {'image': image_url}

1. 创建magicmaker.py文件

2. 修改 /root/agent_camp3/lagent/examples/internlm2_agent_web_demo.py 来适配我们的自定义工具。

   1. 在 from lagent.actions import ActionExecutor, ArxivSearch, IPythonInterpreter 的下一行添加 from lagent.actions.magicmaker import MagicMaker
   2. 在第27行添加 MagicMaker()。

3. 重新启动**streamlit run examples/internlm2_agent_web_demo.py

这里与教程不一样的是，图片并没有直接展示出来，要点击才能查看

bug

ModuleNotFoundError: No module named 'griffe.enumerations

pip install griffe==0.48