Manjaro Linux 安装GPU版本的tensorflow cuda cudnn教程
https://tensorflow.google.cn/install/source#linux 中文社区
Tensorflow CUDA及CUDNN版本对应关系表查询
https://tensorflow.google.cn/install/source#linux
安装此版本
下载Anaconda
sudo wget https://repo.anaconda.com/archive/Anaconda3-2019.03-Linux-x86_64.sh
执行安装
sh Anaconda3-2019.03-Linux-x86_64.sh
环境变量
cd /home/yangyang/anaconda3/
[yangyang@yangyangmanjaro anaconda3]$ echo 'export PATH ="/home/yangyang/anaconda3/bin:$PATH"' >> ~/.bashrc
[yangyang@yangyangmanjaro anaconda3]$ source ~/.bashrc
安装keras 我本来是测试conda 的 结果这个下载半天 还吧tensorflow 下载下来了 违背了初衷（在安装tensorflow之前不要安装）
conda install keras
开始安装tensorflow GPU版本
http://www.tensorfly.cn/tfdoc/get_started/os_setup.html
Linux 安装
安装 Bazel
首先依照 教程 安装 Bazel 的依赖. 然后使用下列命令下载和编译 Bazel 的源码:
https://github.com/bazelbuild/bazel
下载地址
https://docs.bazel.build/versions/master/install.html
查看系统架构和发行版本
我用的 Manjaro x86_64
uname -m && cat /etc/*releasex86_64
Manjaro Linux
DISTRIB_ID=ManjaroLinux
DISTRIB_RELEASE=18.0.4
DISTRIB_CODENAME=Illyria
DISTRIB_DESCRIPTION="Manjaro Linux"
Manjaro Linux
NAME="Manjaro Linux"
ID=manjaro
ID_LIKE=arch
PRETTY_NAME="Manjaro Linux"
ANSI_COLOR="1;32"
HOME_URL="https://www.manjaro.org/"
SUPPORT_URL="https://www.manjaro.org/"
BUG_REPORT_URL="https://bugs.manjaro.org/"
conda info
active environment : base
active env location : /home/yangyang/anaconda3
shell level : 1
user config file : /home/yangyang/.condarc
populated config files : /home/yangyang/.condarc
conda version : 4.6.11
conda-build version : 3.17.8
python version : 3.7.3.final.0
base environment : /home/yangyang/anaconda3 (writable)
channel URLs : https://repo.anaconda.com/pkgs/main/linux-64
https://repo.anaconda.com/pkgs/main/noarch
https://repo.anaconda.com/pkgs/free/linux-64
https://repo.anaconda.com/pkgs/free/noarch
https://repo.anaconda.com/pkgs/r/linux-64
https://repo.anaconda.com/pkgs/r/noarch
package cache : /home/yangyang/anaconda3/pkgs
/home/yangyang/.conda/pkgs
envs directories : /home/yangyang/anaconda3/envs
/home/yangyang/.conda/envs
platform : linux-64
user-agent : conda/4.6.11 requests/2.21.0 CPython/3.7.3 Linux/4.19.56-1-MANJARO manjaro/18.0.4 glibc/2.29
UID:GID : 1000:1000
netrc file : None
offline mode : False
结果 linux 4.19.56-1-manjaro python 3.7 环境
查看电脑的gcc版本
gcc -v
使用内建 specs。
COLLECT_GCC=gcc
COLLECT_LTO_WRAPPER=/usr/lib/gcc/x86_64-pc-linux-gnu/9.1.0/lto-wrapper
目标：x86_64-pc-linux-gnu
配置为：/build/gcc/src/gcc/configure --prefix=/usr --libdir=/usr/lib --libexecdir=/usr/lib --mandir=/usr/share/man --infodir=/usr/share/info --with-bugurl=https://bugs.archlinux.org/ --enable-languages=c,c++,ada,fortran,go,lto,objc,obj-c++ --enable-shared --enable-threads=posix --with-system-zlib --with-isl --enable-__cxa_atexit --disable-libunwind-exceptions --enable-clocale=gnu --disable-libstdcxx-pch --disable-libssp --enable-gnu-unique-object --enable-linker-build-id --enable-lto --enable-plugin --enable-install-libiberty --with-linker-hash-style=gnu --enable-gnu-indirect-function --enable-multilib --disable-werror --enable-checking=release --enable-default-pie --enable-default-ssp --enable-cet=auto
线程模型：posix
gcc 版本 9.1.0 (GCC)
查询完之后 目前情况是 Gcc 9.1 linux 4.19 python 3.7
《《《《《《《《《《《《《《《《《《《《《《《《《《《《《《《《《《《《《《《《《《《《《《《《《《《《《《《
编译版本的安装
一。 安装bazel 版本0.24.1版本
//但是我准备安装0.27.1最新版本//还是下次吧 这次还是0.24.1
https://github.com/bazelbuild/bazel/releases
https://github.com/bazelbuild/bazel/releases/tag/0.24.1
下载
wget https://github.com/bazelbuild/bazel/releases/download/0.24.1/bazel-0.24.1-installer-linux-x86_64.sh镜像下载
https://mirror.bazel.build/openjdk/index.html
wget https://mirror.bazel.build/openjdk/azul-zulu8.38.0.13-ca-jdk8.0.212/zulu8.38.0.13-ca-jdk8.0.212-linux_x64.tar.gz
解压
tar zxvf zulu8.38.0.13-ca-jdk8.0.212-linux_x64.tar.gz
安装
安装nvidia驱动
https://www.nvidia.com/Download/index.aspx?lang=cn
<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<
<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<,
安装已经编译好的版本 Tensorflow
安装cuda cuDNN 下载地址
https://developer.nvidia.com/cuda-downloads
搜索cuda
conda search cuda
No match found for: cuda. Search: *cuda*
# Name Version Build Channel
cudatoolkit 7.5 0 pkgs/free
cudatoolkit 7.5 2 pkgs/free
cudatoolkit 8.0 1 pkgs/free
cudatoolkit 8.0 3 pkgs/free
cudatoolkit 9.0 h13b8566_0 pkgs/main
cudatoolkit 9.2 0 pkgs/main
cudatoolkit 10.0.130 0 pkgs/main
cudatoolkit 10.1.168 0 pkgs/main
numbapro_cudalib 0.1 0 pkgs/free
numbapro_cudalib 0.2 0 pkgs/free conda search cuDNN
conda search cuDNN
WARNING: The conda.compat module is deprecated and will be removed in a future release.
Loading channels: done
# Name Version Build Channel
cudnn 5.1 0 pkgs/free
cudnn 5.1.10 cuda7.5_0 pkgs/free
cudnn 5.1.10 cuda8.0_0 pkgs/free
cudnn 6.0 0 pkgs/free
cudnn 6.0.21 cuda7.5_0 pkgs/free
cudnn 6.0.21 cuda8.0_0 pkgs/free
cudnn 7.0.5 cuda8.0_0 pkgs/main
cudnn 7.1.2 cuda9.0_0 pkgs/main
cudnn 7.1.3 cuda8.0_0 pkgs/main
cudnn 7.2.1 cuda9.2_0 pkgs/main
cudnn 7.3.1 cuda10.0_0 pkgs/main
cudnn 7.3.1 cuda9.0_0 pkgs/main
cudnn 7.3.1 cuda9.2_0 pkgs/main
cudnn 7.6.0 cuda10.0_0 pkgs/main
cudnn 7.6.0 cuda10.1_0 pkgs/main
cudnn 7.6.0 cuda9.0_0 pkgs/main
cudnn 7.6.0 cuda9.2_0 pkgs/main 安装non-free驱动
查询显卡驱动
inxi -G
Graphics: Device-1: Intel HD Graphics 630 driver: i915 v: kernel
Device-2: NVIDIA GK208B [GeForce GT 710] driver: nouveau v: kernel
Display: x11 server: X.org 1.20.5 driver: nouveau
resolution:
OpenGL: renderer: Mesa DRI Intel HD Graphics 630 (Kaby Lake GT2)
v: 4.5 Mesa 19.1.1
https://blog.csdn.net/qq_39828850/article/details/87919188
参考博客
查询tensorflow 版本
conda search tensorflow
WARNING: The conda.compat module is deprecated and will be removed in a future release.
Loading channels: done
# Name Version Build Channel
tensorflow 0.10.0rc0 np111py27_0 pkgs/free
tensorflow 0.10.0rc0 np111py34_0 pkgs/free
tensorflow 0.10.0rc0 np111py35_0 pkgs/free
tensorflow 1.0.1 np112py27_0 pkgs/free
tensorflow 1.0.1 np112py35_0 pkgs/free
tensorflow 1.0.1 np112py36_0 pkgs/free
tensorflow 1.1.0 np111py27_0 pkgs/free
tensorflow 1.1.0 np111py35_0 pkgs/free
tensorflow 1.1.0 np111py36_0 pkgs/free
tensorflow 1.1.0 np112py27_0 pkgs/free
tensorflow 1.1.0 np112py35_0 pkgs/free
tensorflow 1.1.0 np112py36_0 pkgs/free
tensorflow 1.2.1 py27_0 pkgs/free
tensorflow 1.2.1 py35_0 pkgs/free
tensorflow 1.2.1 py36_0 pkgs/free
tensorflow 1.3.0 0 pkgs/free
tensorflow 1.4.1 0 pkgs/main
tensorflow 1.5.0 0 pkgs/main
tensorflow 1.6.0 0 pkgs/main
tensorflow 1.7.0 0 pkgs/main
tensorflow 1.8.0 h01c6a4e_0 pkgs/main
tensorflow 1.8.0 h16da8f2_0 pkgs/main
tensorflow 1.8.0 h2742514_0 pkgs/main
tensorflow 1.8.0 h469b60b_0 pkgs/main
tensorflow 1.8.0 h57681fa_0 pkgs/main
tensorflow 1.8.0 h5c3c37f_0 pkgs/main
tensorflow 1.8.0 h645107b_0 pkgs/main
tensorflow 1.8.0 h7b2774c_0 pkgs/main
tensorflow 1.8.0 hb11d968_0 pkgs/main
tensorflow 1.8.0 hb1b1514_0 pkgs/main
tensorflow 1.8.0 hb381393_0 pkgs/main
tensorflow 1.8.0 hc2d9325_0 pkgs/main
tensorflow 1.9.0 eigen_py27hf386fcc_1 pkgs/main
tensorflow 1.9.0 eigen_py35h8c89287_1 pkgs/main
tensorflow 1.9.0 eigen_py36h8c89287_0 pkgs/main
tensorflow 1.9.0 eigen_py36hbec2359_0 pkgs/main
tensorflow 1.9.0 eigen_py36hbec2359_1 pkgs/main
tensorflow 1.9.0 eigen_py36hf386fcc_0 pkgs/main
tensorflow 1.9.0 gpu_py27h233f449_1 pkgs/main
tensorflow 1.9.0 gpu_py27h395d940_1 pkgs/main
tensorflow 1.9.0 gpu_py27hd3a791e_1 pkgs/main
tensorflow 1.9.0 gpu_py35h42d5ad8_1 pkgs/main
tensorflow 1.9.0 gpu_py35h60c0932_1 pkgs/main
tensorflow 1.9.0 gpu_py35hb39db67_1 pkgs/main
tensorflow 1.9.0 gpu_py36h02c5d5e_1 pkgs/main
tensorflow 1.9.0 gpu_py36h220e158_1 pkgs/main
tensorflow 1.9.0 gpu_py36h313df88_1 pkgs/main
tensorflow 1.9.0 mkl_py27h0cb61a4_1 pkgs/main
tensorflow 1.9.0 mkl_py35h5be851a_1 pkgs/main
tensorflow 1.9.0 mkl_py36h0cb61a4_0 pkgs/main
tensorflow 1.9.0 mkl_py36h5be851a_0 pkgs/main
tensorflow 1.9.0 mkl_py36h6d6ce78_0 pkgs/main
tensorflow 1.9.0 mkl_py36h6d6ce78_1 pkgs/main
tensorflow 1.10.0 eigen_py27ha0ab958_0 pkgs/main
tensorflow 1.10.0 eigen_py35h5ed898b_0 pkgs/main
tensorflow 1.10.0 eigen_py36hb995bb4_0 pkgs/main
tensorflow 1.10.0 gpu_py27h67ad7fe_0 pkgs/main
tensorflow 1.10.0 gpu_py27h6f941b3_0 pkgs/main
tensorflow 1.10.0 gpu_py27h9580370_0 pkgs/main
tensorflow 1.10.0 gpu_py35h566a776_0 pkgs/main
tensorflow 1.10.0 gpu_py35ha6119f3_0 pkgs/main
tensorflow 1.10.0 gpu_py35hd9c640d_0 pkgs/main
tensorflow 1.10.0 gpu_py36h8dbd23f_0 pkgs/main
tensorflow 1.10.0 gpu_py36h97a2126_0 pkgs/main
tensorflow 1.10.0 gpu_py36hcebf108_0 pkgs/main
tensorflow 1.10.0 mkl_py27h857755f_0 pkgs/main
tensorflow 1.10.0 mkl_py35heddcb22_0 pkgs/main
tensorflow 1.10.0 mkl_py36hdb377fd_0 pkgs/main
tensorflow 1.11.0 eigen_py27h06aee4b_0 pkgs/main
tensorflow 1.11.0 eigen_py36he3f7ef1_0 pkgs/main
tensorflow 1.11.0 gpu_py27h99ab47f_0 pkgs/main
tensorflow 1.11.0 gpu_py27hd8bfc1a_0 pkgs/main
tensorflow 1.11.0 gpu_py36h4459f94_0 pkgs/main
tensorflow 1.11.0 gpu_py36h9c9050a_0 pkgs/main
tensorflow 1.11.0 mkl_py27h25e0b76_0 pkgs/main
tensorflow 1.11.0 mkl_py36ha6f0bda_0 pkgs/main
tensorflow 1.12.0 eigen_py27hfe19c55_0 pkgs/main
tensorflow 1.12.0 eigen_py36hbd5f568_0 pkgs/main
tensorflow 1.12.0 gpu_py27h2a0f108_0 pkgs/main
tensorflow 1.12.0 gpu_py27h956c076_0 pkgs/main
tensorflow 1.12.0 gpu_py36he68c306_0 pkgs/main
tensorflow 1.12.0 gpu_py36he74679b_0 pkgs/main
tensorflow 1.12.0 mkl_py27hc55d17a_0 pkgs/main
tensorflow 1.12.0 mkl_py36h69b6ba0_0 pkgs/main
tensorflow 1.13.1 eigen_py27h5e92bea_0 pkgs/main
tensorflow 1.13.1 eigen_py36hc59b85e_0 pkgs/main
tensorflow 1.13.1 eigen_py37h7cb7401_0 pkgs/main
tensorflow 1.13.1 gpu_py27hc5faae7_0 pkgs/main
tensorflow 1.13.1 gpu_py27hcb41dfa_0 pkgs/main
tensorflow 1.13.1 gpu_py27hd3b962e_0 pkgs/main
tensorflow 1.13.1 gpu_py36h26cf82e_0 pkgs/main
tensorflow 1.13.1 gpu_py36h3991807_0 pkgs/main
tensorflow 1.13.1 gpu_py36h9b25d83_0 pkgs/main
tensorflow 1.13.1 gpu_py37h49933a4_0 pkgs/main
tensorflow 1.13.1 gpu_py37hc158e3b_0 pkgs/main
tensorflow 1.13.1 gpu_py37hd37c573_0 pkgs/main
tensorflow 1.13.1 mkl_py27h74ee40f_0 pkgs/main
tensorflow 1.13.1 mkl_py36h27d456a_0 pkgs/main
tensorflow 1.13.1 mkl_py37h54b294f_0 pkgs/main
tensorflow 1.14.0 eigen_py27h99c1539_0 pkgs/main
tensorflow 1.14.0 eigen_py36hb2cf719_0 pkgs/main
tensorflow 1.14.0 eigen_py37h195cb1b_0 pkgs/main
tensorflow 1.14.0 mkl_py27h957988d_0 pkgs/main
tensorflow 1.14.0 mkl_py36h2526735_0 pkgs/main
tensorflow 1.14.0 mkl_py37h45c423b_0 pkgs/main
安装cuda 10.0.130
conda install cudatoolkit==10.0.130安装cuDNN7.6.0
conda install cudnn==7.6.0
安装tensorflow 1.31..1 gpu-py37hd27c573_0 tensorflow 1.13.1 gpu_py37hd37c573_0
conda install tensorflow-gpu==1.13.1做copy 代码只是示例
解压并拷贝 CUDNN 文件到 Cuda Toolkit 7.0 安装路径下. 假设 Cuda Toolkit 7.0 安装 在 /usr/local/cuda, 执行以下命令:
tar xvzf cudnn-6.5-linux-x64-v2.tgz
sudo cp cudnn-6.5-linux-x64-v2/cudnn.h /usr/local/cuda/include
sudo cp cudnn-6.5-linux-x64-v2/libcudnn* /usr/local/cuda/lib64
测试
验证失败 因为驱动没搞定
<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<
<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<
conda 安装tensorflow CPU版本
conda create -n tensorflow_cpu
激活 conda activate tensorflow_cpu 推出 conda deactivate tensorflow_cpu
conda install tensorflow==1.14.0
验证 OK
anaconda-navigator 启动界面检查一下环境
<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<
安装pyCharm
PyCharm does not have write access to /usr/share/pycharm. Please run it by a privileged user to update
修改写入权限就可以了
sudo chown -R $USER:$USER /usr/share/pycharm