window+CUDA+CUDNN 下载和安装

1.安装cuda

1.1 官网下载CUDA Toolkit Archive | NVIDIA DeveloperPrevious releases of the CUDA Toolkit, GPU Computing SDK, documentation and developer drivers can be found using the links below. Please select the release you want from the list below, and be sure to check www.nvidia.com/drivers for more recent production drivers appropriate for your hardware configuration.https://developer.nvidia.com/cuda-toolkit-archive下载版本11.4，选择windows+x86_64+10/11+exe(local)

1.2 自定义安装，只勾选cuda

1.3 验证

查看环境变量是否存在

CUDA_PATH

CUDA_PATH_V11.4

NVCUDASAMPLES_ROOT

NVCUDASAMPLES11_4_ROOT

查看版本号

nvcc -V

2.安装cuDNN

2.1 官网下载

CUDA 深度神经网络库 (cuDNN) | NVIDIA 开发者cuDNN 能够以高度优化的方式实现标准例程如前向和反向卷积、池化层、归一化和激活层。https://developer.nvidia.cn/cudnn

2.2 下载压缩包cudnn-11.4-windows-x64-v8.2.2.26

2.3 cuDNN拷贝到cuda

    将cuDNN中的bin、include、lib文件夹拷贝到“C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v11.4”下。（CUDA 的安装目录中，有和 cuDNN 解压缩后的同名文件夹，直接复制即可。cuDNN 其实就是 CUDA 的一个补丁，专为深度学习运算进行优化的）

2.4 环境变量（默认安装cuda路径）

C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v11.4\bin

C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v11.4\include

C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v11.4\lib

C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v11.4\libnvvp

2.5 验证

    win+R启动cmd

cd “C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v11.4\extras\demo_suite”

bandwidthTest.exe

deviceQuery.exe