深度学习入门教程-Ubuntu18.04系统安装cuDNN7和NCCL2

深度学习入门教程-Ubuntu18.04系统安装cuDNN7和NCCL2

说明：

• 介绍如何安装在Ubuntu18.04系统安装cuDNN7和NCCL2

环境：

• 系统：ubuntu 18.04
• 显卡：GTX 1080Ti

安装PPA：

• 找到合适的PPA版本，https://developer.download.nvidia.com/compute/machine-learning/repos/ubuntu1804/x86_64/
• 目前选择nvidia-machine-learning-repo-ubuntu1804_1.0.0-1_amd64.deb
• 下载安装：

wget https://developer.download.nvidia.com/compute/machine-learning/repos/ubuntu1804/x86_64/nvidia-machine-learning-repo-ubuntu1804_1.0.0-1_amd64.deb

sudo dpkg -i nvidia-machine-learning-repo-ubuntu1804_1.0.0-1_amd64.deb

sudo apt update

• 安装cuDNN

sudo apt install -y libcudnn7 libcudnn7-dev libnccl2 libc-ares-dev

• 链接到正常位置

sudo mkdir -p /usr/local/cuda/nccl/lib
sudo ln -s /usr/lib/x86_64-linux-gnu/libnccl.so.2 /usr/local/cuda/nccl/lib/
sudo ln -s /usr/lib/x86_64-linux-gnu/libcudnn.so.7 /usr/local/cuda/lib64/

检测：

• 执行

cat /usr/include/cudnn.h | grep CUDNN_MAJOR -A 2

• 效果如下：

ubuntu@AiDLHost:~$ cat /usr/include/cudnn.h | grep CUDNN_MAJOR -A 2
#define CUDNN_MAJOR 7
#define CUDNN_MINOR 5
#define CUDNN_PATCHLEVEL 1
--
#define CUDNN_VERSION (CUDNN_MAJOR * 1000 + CUDNN_MINOR * 100 + CUDNN_PATCHLEVEL)

#include "driver_types.h"

• 下载测试例子，点击下载

wget http://file.ncnynl.com/ros/2019/libcudnn7-doc_7.5.0.56-1+cuda10.1_amd64.deb
sudo apt install ./libcudnn7-doc_7.5.0.56-1+cuda10.1_amd64.deb
cp -r /usr/src/cudnn_samples_v7/ ~/tools/

• 安装编译：

cd ~/tools/cudnn_samples_v7/mnistCUDNN
make clean && make 

• 测试如下：

ubuntu@AiDLHost:~/tools/cudnn_samples_v7/mnistCUDNN$ ./mnistCUDNN 
cudnnGetVersion() : 7501 , CUDNN_VERSION from cudnn.h : 7501 (7.5.1)
Host compiler version : GCC 7.4.0
There are 1 CUDA capable devices on your machine :
device 0 : sms 28  Capabilities 6.1, SmClock 1632.5 Mhz, MemSize (Mb) 11175, MemClock 5505.0 Mhz, Ecc=0, boardGroupID=0
Using device 0

Testing single precision
Loading image data/one_28x28.pgm
Performing forward propagation ...
Testing cudnnGetConvolutionForwardAlgorithm ...
Fastest algorithm is Algo 1
Testing cudnnFindConvolutionForwardAlgorithm ...
^^^^ CUDNN_STATUS_SUCCESS for Algo 0: 0.014336 time requiring 0 memory
^^^^ CUDNN_STATUS_SUCCESS for Algo 1: 0.026176 time requiring 3464 memory
^^^^ CUDNN_STATUS_SUCCESS for Algo 2: 0.032768 time requiring 57600 memory
^^^^ CUDNN_STATUS_SUCCESS for Algo 5: 0.059424 time requiring 203008 memory
^^^^ CUDNN_STATUS_SUCCESS for Algo 4: 0.062240 time requiring 207360 memory
Resulting weights from Softmax:
0.0000000 0.9999399 0.0000000 0.0000000 0.0000561 0.0000000 0.0000012 0.0000017 0.0000010 0.0000000 
Loading image data/three_28x28.pgm
Performing forward propagation ...
Resulting weights from Softmax:
0.0000000 0.0000000 0.0000000 0.9999288 0.0000000 0.0000711 0.0000000 0.0000000 0.0000000 0.0000000 
Loading image data/five_28x28.pgm
Performing forward propagation ...
Resulting weights from Softmax:
0.0000000 0.0000008 0.0000000 0.0000002 0.0000000 0.9999820 0.0000154 0.0000000 0.0000012 0.0000006 

Result of classification: 1 3 5

Test passed!

Testing half precision (math in single precision)
Loading image data/one_28x28.pgm
Performing forward propagation ...
Testing cudnnGetConvolutionForwardAlgorithm ...
Fastest algorithm is Algo 1
Testing cudnnFindConvolutionForwardAlgorithm ...
^^^^ CUDNN_STATUS_SUCCESS for Algo 0: 0.013312 time requiring 0 memory
^^^^ CUDNN_STATUS_SUCCESS for Algo 1: 0.025600 time requiring 3464 memory
^^^^ CUDNN_STATUS_SUCCESS for Algo 2: 0.032800 time requiring 28800 memory
^^^^ CUDNN_STATUS_SUCCESS for Algo 5: 0.043008 time requiring 203008 memory
^^^^ CUDNN_STATUS_SUCCESS for Algo 4: 0.061280 time requiring 207360 memory
Resulting weights from Softmax:
0.0000001 1.0000000 0.0000001 0.0000000 0.0000563 0.0000001 0.0000012 0.0000017 0.0000010 0.0000001 
Loading image data/three_28x28.pgm
Performing forward propagation ...
Resulting weights from Softmax:
0.0000000 0.0000000 0.0000000 1.0000000 0.0000000 0.0000714 0.0000000 0.0000000 0.0000000 0.0000000 
Loading image data/five_28x28.pgm
Performing forward propagation ...
Resulting weights from Softmax:
0.0000000 0.0000008 0.0000000 0.0000002 0.0000000 1.0000000 0.0000154 0.0000000 0.0000012 0.0000006 

Result of classification: 1 3 5

Test passed!