ubuntu16.04 install cuda8.0+cudnn6.0+tensorflow_gpu==1.4.0

最新推荐文章于 2023-08-05 17:53:23 发布

原创最新推荐文章于 2023-08-05 17:53:23 发布 · 405 阅读

0 ·

CC 4.0 BY-SA版权

安装教程专栏收录该内容

6 篇文章

订阅专栏

本文详细介绍了如何在Linux环境下为TensorFlow 1.4版本配置CUDA 8.0和cuDNN 6.0，包括依赖安装、CUDA下载与安装、cuDNN下载与配置，以及环境变量设置等关键步骤。

部署运行你感兴趣的模型镜像

Different versions of Tensorflow require correspondence with CUDA and CUDNN versions

https://blog.youkuaiyun.com/omodao1/article/details/83241074

#install dependence
sudo apt-get install openjdk-8-jdk git python-dev python3-dev python-numpy python3-numpy python-six python3-six build-essential python-pip python3-pip python-virtualenv swig python-wheel python3-wheel libcurl3-dev libcupti-dev



#download cuda8.0
wget https://developer.nvidia.com/compute/cuda/8.0/Prod2/local_installers/cuda_8.0.61_375.26_linux-run
#rename
sudo mv cuda_8.0.61_375.26_linux-run cuda_8.0.61_375.26_linux.run
#install
sudo sh cuda_8.0.61_375.26_linux.run --override --silent --toolkit



#download cudnn6.0
#unzip file
tar -xzvf cudnn-8.0-linux-x64-v6.0.tgz
sudo cp cuda/include/cudnn.h /usr/local/cuda/include
sudo cp cuda/lib64/libcudnn* /usr/local/cuda/lib64
sudo chmod a+r /usr/local/cuda/include/cudnn.h /usr/local/cuda/lib64/libcudnn*

#Configure environment variables
sudo gedit ~/.bashrc
#write them to the end of the file
export LD_LIBRARY_PATH="$LD_LIBRARY_PATH:/usr/local/cuda/lib64:/usr/local/cuda/extras/CUPTI/lib64"
export CUDA_HOME=/usr/local/cuda
sudo source ~/.bashrc



#install tensorflow_gpu==1.4.0
sudo pip install tensorflow-gpu==1.4.0

您可能感兴趣的与本文相关的镜像