[TensorFlow] Ubuntu 16.04 配置 tensorflow gpu 环境

参考内容
http://blog.csdn.net/u012759136/article/details/53355781
http://blog.csdn.net/zafir_410/article/details/73188228
https://devtalk.nvidia.com/default/topic/996474/linux/unable-to-load-the-nvidia-drm-kernel-module-fedora-25-kernel-4-9-11/
https://devtalk.nvidia.com/default/topic/926967/unable-to-load-kernel-module-for-364-12/

配置环境

• 操作系统: Ubuntu 16.04
• 显卡: GTX 950M
• NVIDIA 驱动: NVIDIA-Linux-x86_64-384.90.run
• Cuda 8.0.44
• cudnn v5.1

流程

安装 NVIDIA 显卡驱动

去官网 驱动程序 | GeForce 下载对应驱动。我下载时最新的是 384.90 版本，下载后文件为 NVIDIA-Linux-x86_64-384.90.run。

卸载原有驱动

sudo apt-get remove –purge nvidia*

禁用 nouveau

1. 添加黑名单

sudo vim /etc/modprobe.d/blacklist-nouveau.conf

输入以下内容并保存

blacklist nouveau
options nouveau modeset=0

2. 执行

sudo update-initramfs -u

3. 重启电脑后检查

lsmod | grep nouveau

若无输出则可以

开始安装

1. 进入本地控制台

Ctrl + Alt + F1

2. 关闭图形界面

sudo service lightdm stop

3. runfile 安装

sudo chmod a+x NVIDIA-Linux-x86_64-384.90.run
sudo ./NVIDIA-Linux-x86_64-384.90.run –no-x-check –no-nouveau-check –no-opengl-files

参数含义

• –no-opengl-files 只安装驱动文件，不安装 OpenGL 文件。这个参数最重要
• –no-x-check 安装驱动时不检查 X 服务
• –no-nouveau-check 安装驱动时不检查 nouveau

按安装提示一步一步就可以，基本上都选 Accept。

5. 启动图形界面

sudo service lightdm start

这时候可以进终端检查是否有 nvidia-smi，执行后应该会显示类似这样的结果，说明安装成功。

问题及解决方法

1. ERROR: Unable to load the ‘nvidia-drm’ kernel module.
检查 BIOS 的 Secure Boot 是否关闭，如果没有关闭，把他关掉即可，

安装 Cuda

官网下载 CUDA Toolkit Archive，选择 CUDA Toolkit 8.0 GA1 (Sept 2016)，下载 runfile 文件 cuda_8.0.44_linux.run。

runfile 安装

sudo chmod a+x cuda_8.0.44_linux.run
sudo ./cuda_8.0.44_linux.run

依次按如下选择

Description

This package includes over 100+ CUDA examples that demonstrate
various CUDA programming principles, and efficient CUDA
implementation of algorithms in specific application domains.
The NVIDIA CUDA Samples License Agreement is available in
Do you accept the previously read EULA?
accept/decline/quit: accept

Install NVIDIA Accelerated Graphics Driver for Linux-x86_64 367.48?
(y)es/(n)o/(q)uit: n

Install the CUDA 8.0 Toolkit?
(y)es/(n)o/(q)uit: y

Enter Toolkit Location
 [ default is /usr/local/cuda-8.0 ]:

Do you want to install a symbolic link at /usr/local/cuda?
(y)es/(n)o/(q)uit: y

Install the CUDA 8.0 Samples?
(y)es/(n)o/(q)uit: y

Enter CUDA Samples Location
 [ default is /home/c302 ]:

Installing the CUDA Toolkit in /usr/local/cuda-8.0 ...
Installing the CUDA Samples in /home/c302 ...
Copying samples to /home/c302/NVIDIA_CUDA-8.0_Samples now...
Finished copying samples.

添加环境变量

打开 ~/.bashrc 添加

export PATH=/usr/local/cuda-8.0/bin:$PATH
export LD_LIBRARY_PATH="$LD_LIBRARY_PATH:/usr/local/cuda/lib64:/usr/local/cuda/extras/CUPTI/lib64"
export CUDA_HOME=/usr/local/cuda

安装 cuDNN

官网下载 NVIDIA cuDNN，要注册账号还有填一个很简单的问卷。

解压并放到相应位置

tar xvzf cudnn-8.0-linux-x64-v5.1-ga.tgz
sudo cp -P cuda/include/cudnn.h /usr/local/cuda/include
sudo cp -P cuda/lib64/libcudnn* /usr/local/cuda/lib64
sudo chmod a+r /usr/local/cuda/include/cudnn.h /usr/local/cuda/lib64/libcudnn*

安装 tensorflow-gpu

这个就很简单了，按照官网教程用 ｀pip｀ 安装，选择 GPU 版本。

# Ubuntu/Linux 64-bit, CPU only, Python 2.7
$ export TF_BINARY_URL=https://storage.googleapis.com/tensorflow/linux/cpu/tensorflow-0.12.0rc1-cp27-none-linux_x86_64.whl

# Ubuntu/Linux 64-bit, GPU enabled, Python 2.7
# Requires CUDA toolkit 8.0 and CuDNN v5. For other versions, see "Installing from sources" below.
$ export TF_BINARY_URL=https://storage.googleapis.com/tensorflow/linux/gpu/tensorflow_gpu-0.12.0rc1-cp27-none-linux_x86_64.whl

# Mac OS X, CPU only, Python 2.7:
$ export TF_BINARY_URL=https://storage.googleapis.com/tensorflow/mac/cpu/tensorflow-0.12.0rc1-py2-none-any.whl

# Mac OS X, GPU enabled, Python 2.7:
$ export TF_BINARY_URL=https://storage.googleapis.com/tensorflow/mac/gpu/tensorflow_gpu-0.12.0rc1-py2-none-any.whl

# Ubuntu/Linux 64-bit, CPU only, Python 3.4
$ export TF_BINARY_URL=https://storage.googleapis.com/tensorflow/linux/cpu/tensorflow-0.12.0rc1-cp34-cp34m-linux_x86_64.whl

# Ubuntu/Linux 64-bit, GPU enabled, Python 3.4
# Requires CUDA toolkit 8.0 and CuDNN v5. For other versions, see "Installing from sources" below.
$ export TF_BINARY_URL=https://storage.googleapis.com/tensorflow/linux/gpu/tensorflow_gpu-0.12.0rc1-cp34-cp34m-linux_x86_64.whl

# Ubuntu/Linux 64-bit, CPU only, Python 3.5
$ export TF_BINARY_URL=https://storage.googleapis.com/tensorflow/linux/cpu/tensorflow-0.12.0rc1-cp35-cp35m-linux_x86_64.whl

# Ubuntu/Linux 64-bit, GPU enabled, Python 3.5
# Requires CUDA toolkit 8.0 and CuDNN v5. For other versions, see "Installing from sources" below.
$ export TF_BINARY_URL=https://storage.googleapis.com/tensorflow/linux/gpu/tensorflow_gpu-0.12.0rc1-cp35-cp35m-linux_x86_64.whl

# Mac OS X, CPU only, Python 3.4 or 3.5:
$ export TF_BINARY_URL=https://storage.googleapis.com/tensorflow/mac/cpu/tensorflow-0.12.0rc1-py3-none-any.whl

# Mac OS X, GPU enabled, Python 3.4 or 3.5:
$ export TF_BINARY_URL=https://storage.googleapis.com/tensorflow/mac/gpu/tensorflow_gpu-0.12.0rc1-py3-none-any.whl

安装

# Python 2
$ sudo pip install --upgrade $TF_BINARY_URL

# Python 3
$ sudo pip3 install --upgrade $TF_BINARY_URL

至此，就全部安装好了，进入 python　导入包后应该会显示