GPU无法使用[🐛BUG] 用一句话描述您的问题。

[ahukmr]

描述这个 bug 在运行recbole1.2.0时，配置文件里use_gpu=true,gpuid =0都配置了，但是运行时一直显示没有使用gpu是为什么

如何复现 复现这个 bug 的步骤：

1. 您引入的额外 yaml 文件
2. 您的代码
3. 您的运行脚本

预期 对您的预期作清晰简明的描述。

屏幕截图

链接 添加能够复现 bug 的代码链接，如 Colab 或者其他在线 Jupyter 平台。（可选）

实验环境（请补全下列信息）：

• 操作系统: [如 Linux, macOS 或 Windows]
  • RecBole 版本 [如1.2.0]
• Python 版本 [如 3.8]

[BoXiaohe]

Thanks for your attention to RecBole! As for your problem, you can try advice below. 1. CUDA and cuDNN Compatibility: Verify that your CUDA and cuDNN versions are compatible with RecBole. Different versions of deep learning libraries may require specific versions of CUDA and cuDNN to work properly. 2. Environment Variables: Double-check the setting of the CUDA_VISIBLE_DEVICES environment variable. Setting gpuid=0 in the configuration file should generally be sufficient, but there might be other configurations or environmental factors affecting GPU usage. 3. GPU Availability: Ensure that your GPU is available and not being used by other processes. You can use tools like nvidia-smi to check GPU status and usage. If another process is occupying the GPU, RecBole won't be able to use it. Hope this could help you!