wow agent day12 metaGPT安装和配置

Jan 26, 2025#MetaGPT #Datawhale组队学习88

AI-generated summary

To set up Metagpt with Python 3.9+, follow these steps: 1. Create a virtual environment: ``` conda create -n metagpt python=3.9 && conda activate metagpt ``` 2. Install Metagpt: ``` pip install metagpt==0.8.0 ``` 3. Initialize the configuration: ``` metagpt --init-config ``` Fill in the `config2.yaml` file located in the config folder with your API settings. For example: ```yaml llm: api_type: 'ollama' base_url: 'http://192.168.0.70:11434/api' model: 'qwen2:7b' repair_llm_output: true ``` Replace `192.168.0.70` with the actual IP of the machine hosting the model. 4. To verify the configuration, use the following code: ```python from metagpt.config2 import Config def print_llm_config(): config = Config.default() llm_config = config.llm if llm_config: print(f"API Type: {llm_config.api_type}") print(f"API Key: {llm_config.api_key}") print(f"Model: {llm_config.model}") else: print("No LLM configured") if __name__ == "__main__": print_llm_config() ```

Reference：Datawhale wow agent day12
适用于 python 3.9+
pip install metagpt==0.8.0

根据Metagpt教程。

步骤#

创建 python 3.9 的虚拟环境：conda create -n metagpt python=3.9 && conda activate metagpt
pip install metagpt
metagpt --init-config 生成 yaml 文件。在 config 的文件夹下的 config2.yaml 里填写 API 配置。
我这里是本地配置 ollama:
llm:
api_type: 'ollama'
base_url: 'http://192.168.0.70:11434/api'
model: 'qwen2:7b'

repair_llm_output: true
代码中 192.168.0.70 就是部署了大模型的电脑的 IP，

请根据实际情况进行替换

检验#

from metagpt.config2 import Config 
def print_llm_config():
    # 加载默认配置
    config = Config.default()

    # 获取LLM配置
    llm_config = config.llm
    # 打印LLM配置的详细信息
    if llm_config:
        print(f"API类型: {llm_config.api_type}")
        print(f"API密钥: {llm_config.api_key}")
        print(f"模型: {llm_config.model}")
    else:
        print("没有配置LLM")

if __name__ == "__main__":
    print_llm_config()