Ubuntu16.04+NVIDIA驱动 + CUDA9.0 + cuDNN7.0.5 + tensorflow-gpu1.10.0 + keras安装

内容导读

互联网集市收集整理的这篇技术教程文章主要介绍了Ubuntu16.04+NVIDIA驱动 + CUDA9.0 + cuDNN7.0.5 + tensorflow-gpu1.10.0 + keras安装，小编现在分享给大家，供广大互联网技能从业者学习和参考。文章包含4443字，纯文字阅读大概需要7分钟。

内容图文

1. 配置安装环境

sudo apt-get update
sudo apt-get install python-dev python-pip python-nose gcc g++ git gfortran
sudo apt-get install libprotobuf-dev libleveldb-dev libsnappy-dev libopencv-dev libhdf5-serial-dev protobuf-compiler
sudo apt-get install --no-install-recommends libboost-all-dev
sudo apt-get install libopenblas-dev liblapack-dev libatlas-base-dev
sudo apt-get install libgflags-dev libgoogle-glog-dev liblmdb-dev
sudo apt-get install git cmake build-essential

2. 安装NVIDIA驱动

• Step 1. Identify your NVIDIA VGA card.The below commands will allow you to identify your Nvidia card model:

$  lshw -numeric -C display
or
$ lspci -vnn | grep VGA

• Step 2. Download the Official Nvidia Driver. Using your web browser navigate to the official Nvidia website and download an appropriate driver for your Nvidia graphic card. Save the file into your home directory. Example:

$ ls
NVIDIA-Linux-x86_64-410.73.bin

• Step 3. Install Prerequisites. The following prerequisites are required to compile and install Nvidia driver:

$ sudo dpkg --add-architecture i386
$ sudo apt update
$ sudo apt install build-essential libc6:i386

• Step 4. Next step is to disable the default nouveau Nvidia driver.

$ sudo bash -c "echo blacklist nouveau > /etc/modprobe.d/blacklist-nvidia-nouveau.conf"
$ sudo bash -c "echo options nouveau modeset=0 >> /etc/modprobe.d/blacklist-nvidia-nouveau.conf"

       Confirm the content of the new modprobe config file:

$ cat /etc/modprobe.d/blacklist-nvidia-nouveau.conf
blacklist nouveau
options nouveau modeset=0

• Step 5. Make sure you reboot your system before you proceed to the next step.

• Step 6. stop the current display server. 

$ sudo telinit 3

        Hit CTRL+ALT+F1 and login with your username and password to open a new TTY1 session.

• Step 7. start installation of Nvidia driver execute

$ bash NVIDIA-Linux-x86_64-410.73.bin

      You now need to Accept License and follow the wizard to completed the installation. You man also be asked questions like:

    The distribution-provided pre-install script failed!
    Are you sure you want to continue? -> CONTINUE INSTALLATION
    Would you like to run the nvidia-xconfig utility? -> YES 

• Step 8. Reboot your system and configure NVIDIA X Server Settings. After reboot you should be able to start NVIDIA X Server Settings app from the Activities menu.

3. 安装CUDA和CuDNN SDK

3.1 安装CUDA所需库

sudo apt-get install freeglut3-dev build-essential libx11-dev libxmu-dev libxi-dev libgl1-mesa-glx libglu1-mesa libglu1-mesa-dev

3.2 下载CUDA9.0 和CuDNN SDK 7.0.5

3.3 安装CUDA

• Ctrl-Alt+F1 进入命令行界面

cd Downloads

sudo service lightdm stop #禁用X服务
sudo /etc/init.d/lightdm stop #一样的命令

sudo sh cuda-8.0.44_linux.run --no-opengl-libs

       注：第一条问是否安装nvidia驱动，选择no，后面一直输入yes就行

• 安装完成后

sudo /etc/init.d/lightdm start #启用服务
reboot

• 重启之后进入修改环境配置

sudo gedit ~/.bashrc

export PATH=/usr/local/cuda-8.0/bin:$PATH
export LD_LIBRARY_PATH=/usr/local/cuda/lib64:$LD_LIBRARY_PATH

source ~/.bashrc

• 测试安装是否成功：

cd /usr/local/cuda-8.0/samples/1_Utilities/deviceQuery
sudo make
./deviceQuery

        显示gpu信息，说明安装成功。

3.4 安装CuDNN

cd Downloads/cuda
sudo cp -a include/cudnn.h /usr/local/cuda/include/ # 复制头文件
cd ./lib64
sudo cp -a lib* /usr/local/cuda/lib64/ #复制动态链接库
cd /usr/local/cuda/lib64/
sudo rm -rf libcudnn.so libcudnn.so.5 #删除原有动态文件
sudo ln -s libcudnn.so.5.1.5 libcudnn.so.5 #生成软衔接
sudo ln -s libcudnn.so.5 libcudnn.so #生成软链接

4. 安装其他资源

4.1 安装NCCL 2.3.5

sudo dpkg -i nccl-repo-2.3.5.deb
sudo apt update
sudo apt install libnccl2 libnccl-dev

4.2 用virtualenv安装tensorflow1.10.0

• 安装virtualenv环境

sudo apt-get install python3
sudo apt-get install python3-pip python3-dev python-virtualenv

• 安装完pip3、virtualenv后，可进入virtualenv安装tensorflow-gpu，

mkdir -p ~/tensorflow/venv
virtualenv -p python3 ~/tensorflow/venv #创建virtualenv环境
cd ~/tensorflow/venv
source bin/activate #环境用的时候需要激活
pip3 install --index-url http://pypi.douban.com/simple --trusted-host pypi.douban.com --upgrade tensorflow-gpu #采用国内源下载tensorflow-gpu

      激活方法：source ~/tensorflow/venv/bin/activate

4.3 安装keras

pip install keras

注：不能使用sudo，因为环境是vitualenv，用sudo会安装在系统环境下。

参考资料：

1. ubuntu16.04 + NVIDIA驱动 + CUDA + cuDNN + tensorflow-gpu + keras安装

2. 深度学习应用系列（一）| 在Ubuntu 18.04安装tensorflow 1.10 GPU版本

3. How to install the NVIDIA drivers on Ubuntu 18.04 Bionic Beaver Linux

内容总结

以上是互联网集市为您收集整理的Ubuntu16.04+NVIDIA驱动 + CUDA9.0 + cuDNN7.0.5 + tensorflow-gpu1.10.0 + keras安装全部内容，希望文章能够帮你解决Ubuntu16.04+NVIDIA驱动 + CUDA9.0 + cuDNN7.0.5 + tensorflow-gpu1.10.0 + keras安装所遇到的程序开发问题。 如果觉得互联网集市技术教程内容还不错，欢迎将互联网集市网站推荐给程序员好友。

内容备注

版权声明：本文内容由互联网用户自发贡献，该文观点与技术仅代表作者本人。本站仅提供信息存储空间服务，不拥有所有权，不承担相关法律责任。如发现本站有涉嫌侵权/违法违规的内容， 请发送邮件至 gblab@vip.qq.com 举报，一经查实，本站将立刻删除。

内容手机端

---