查看CUDA版本方法一在桌面单击鼠标右键->NVIDIA控制面板->系统信息->组件方法二cmd命令行下输入命令：nvcc --version方法三进入 CUDA 的安装目录查看：C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA查看cuDNN版本进入cuDNN安装目录下查看文件 cudnn.h...

from torch.utils.tensorboard import SummaryWriter报错：ImportError: TensorBoard logging requires TensorBoard version 1.15 or above解决方法：在terminal中安装tensorboard1.15.0pip3 install tensorboard==1.15.0

from torch.utils.tensorboard import SummaryWriter报错：ImportError: TensorBoard logging requires TensorBoard version 1.15 or above解决方法：在terminal中安装tensorboard1.15.0pip3 install tensorboard==1.15.0

（默认服务器上安装好了conda环境和jupyter notebook）1. 生成默认配置文件jupyter notebook --generate-config2. 进入Python终端，生成秘钥ipythonIn [1]: from notebook.auth import passwdIn [2]: passwd()Enter password:Verify password:Out[2]: