autogen+ollma+wsl2本地代码编辑样例

1、wsl2环境

sudo apt install python3-full
sudo apt install python3.10-venv

2、新建目录和环境

mkdir autogen-example
cd autogen-example
python3 -m venv .venv

3、激活环境

source .venv/bin/activate

4、拉取模型

ollama pull codellama

5、安装autogen

pip install pyautogen

6、编写代码

from autogen import AssistantAgent, UserProxyAgent

config_list = [
  {
    "model": "codellama",
    "base_url": "http://localhost:11434/v1",
    "api_key": "ollama",
  }
]

assistant = AssistantAgent("assistant", llm_config={"config_list": config_list})

user_proxy = UserProxyAgent("user_proxy", code_execution_config={"work_dir": "coding", "use_docker": False})
user_proxy.initiate_chat(assistant, message="Plot a chart of NVDA and TESLA stock price change YTD.")

7、运行例子

python example.py

好，那我帮你装上

pip install pandas

又报错

那就再帮你装

 pip install matplotlib

哈哈，它倒是好，使唤着 我去搞文件

行嘛