Linux（redhat）CUDA,CUDNN,tensorflow-gpu离线安装

最新推荐文章于 2025-10-22 15:09:25 发布

原创

最新推荐文章于 2025-10-22 15:09:25 发布 · 1.4k 阅读

0 ·

CC 4.0 BY-SA版权

本文详细介绍了在Linux Redhat系统中如何离线安装GPU驱动、CUDA 9.0、CUDNN以及TensorFlow-GPU 1.12.0。首先，通过rpm包安装GPU驱动并禁用 nouveau。接着，手动安装CUDA，更新环境变量并验证安装。然后，解压并复制CUDNN文件到相应目录。最后，使用pip安装TensorFlow-GPU，并列举了TensorFlow的依赖库。此外，还提供了Jupyter Notebook的配置和启动方法。

+++++++++++++++++++++++++++++++++++++++++++++++++++++++++++
安装GPU驱动
nvidia-smi
lsmod | grep nouveau
sudo rpm -ivh nvidia-diag-driver-local-repo-rhel7-XXXXXX.x86_64.rpm
++++++++++++++++++++++++++++++++++++++++++++++++++++++++++
安装CUDA,CUDNN
sudo rm /tmp/.X0-lock
sudo sh cuda_9.0.176_384.81_linux.run

sudo vi ~/.bashrc
export PATH=/usr/local/cuda-9.0/bin${PATH:+:${PATH}}
export LD_LIBRARY_PATH=/usr/local/cuda-9.0/lib64${LD_LIBRARY_PATH:+:${LD_LIBRARY_PATH}}

source ~/.bashrc
sudo ldconfig
nvcc --version
ls ../NVIDIA_CUDA-9.0_Samples/

sudo tar -xvf cudnn-9.0-linux-x64-v7.3.1.20.tgz
sudo cp cuda/include/cudnn.h /usr/local/cuda/include/
sudo cp -a cuda/lib64/libcudnn* /usr/local/cuda/lib64/
sudo chmod a+r /usr/local/cuda/include/cudnn.h