目录一、下载cuda和cudnn二、配置conda三、下载pytorch一、下载cuda和cudnn1、打开命令提示符输入 nvidia-smi 查看版本2、cudaCUDA Toolkit Archive | NVIDIA Developer根据自己的需求选择对应的版本根据给出命令依次执行最后输入 nvcc -V 出现版本号则安装成功。添加环境变量vi ~/.bashrc#复制到文件末尾记得替换自己安装路径可以通过运行whereis cuda查看cuda路径 export PATH/usr/local/cuda-12.6/bin${PATH::${PATH}} export LD_LIBRARY_PATH/usr/local/cuda-12.6/lib64${LD_LIBRARY_PATH::${LD_LIBRARY_PATH}}重新加载环境配置文件source ~/.bashrc3、cudnncuDNN Archive | NVIDIA Developer我是wsl下的ubuntu所以直接将文件复制到/home/user/下然后进行解压tar -xvf cudnn-linux-x86_64-8.9.7.29_cuda12-archive.tar.xz然后执行下述命令记得修改成自己的路径sudo cp cudnn-linux-x86_64-8.9.7.29_cuda12-archive/include/* /usr/local/cuda/include sudo cp cudnn-linux-x86_64-8.9.7.29_cuda12-archive/lib/libcudnn* /usr/local/cuda/lib64 sudo chmod ar /usr/local/cuda/include/cudnn.h /usr/local/cuda/lib64/libcudnn*查看cudnn的版本cat /usr/local/cuda/include/cudnn_version.h | grep CUDNN_MAJOR -A 2出现下图内容则安装成功二、配置condawget https://repo.anaconda.com/miniconda/Miniconda3-latest-Linux-x86_64.sh bash Miniconda3-latest-Linux-x86_64.sh -b -p $HOME/miniconda3 source ~/miniconda3/bin/activate conda init bash验证conda --version # ≥24.x conda create -n py310 python3.10 #若出现选择则全部选accept(a)输入a即可若验证成功根据提示激活虚拟环境conda activate py310三、下载pytorch在虚拟环境py310中执行以下命令​pip install --no-cache-dir torch2.6.0 torchvision0.21.0 torchaudio2.6.0 --index-url https://download.pytorch.org/whl/cu126验证python -c import torch, sys, platform;\ print(PyTorch, torch.__version__);\ print(CUDA , torch.version.cuda);\ print(GPU , torch.cuda.get_device_name(0) if torch.cuda.is_available() else None)参考https://zhuanlan.zhihu.com/p/691711768https://www.cnblogs.com/HorizonTree/articles/18806799