好好学习的郝
是能够稳定复现的吗？注意到了两个问题： 1. 报错中出现了 D:\MetaGPT-0.6.4\workspace\snake_game error: [WinError 32] 另一个程序正在使用此文件，进程无法访问。: 'D:\MetaGPT-0.6.4\workspace\20240121104955' 。这个需要重启电脑后重新跑一下试试 2. 使用的是 MetaGPT-0.6.4 版本，这个版本里是不支持 ZHIPUAI_API_MODEL 参数的，可以使用最新版的main分支尝试下（同时需要更新依赖）。
从报错看：Incorrect API key provided: sk-SNBWJzaub. You can find your API key at https://platform.openai.com/account/api-keys. 说明你的key已经提供了，但是是错误的，可以检查下你的环境变量 `env | grep -i key`
You can refer to [unittest.yaml](https://github.com/geekan/MetaGPT/blob/main/.github/workflows/unittest.yaml) and [run_install_deps.sh](https://github.com/geekan/MetaGPT/blob/main/tests/scripts/run_install_deps.sh).
OK, I will add it to the docs soon. 1. Install unittest dependencies ```bash bash run_install_deps.sh ``` 2. Prepare config file `key.yaml` ```yaml OPENAI_API_KEY: "sk-xxx" OPENAI_BASE_URL: "https://api.openai.com/v1" OPENAI_API_MODEL: "gpt-3.5-turbo-16k" MAX_TOKENS:...
json is from LLM, so it is random. We have to handle it to make it more standardized.
运行代码前，本地配置全局代理。例如： ```bash export http_proxy=http://127.0.0.1:7890 export https_proxy=http://127.0.0.1:7890 ```
试试重启电脑，然后重新跑这个代码。
好建议，添加这行代码后日志更清晰，提交个PR？😁
It seem because of `# sqlite3 is included in Python's standard library, but versioning is for consistency`. It can not parse the comments.