1. 在无独显的情况下，安装ubuntu14

 

2.插上gtx1080, 开机进入BIOS，设置显卡模式为(iGPU)

 

3.开机，显示器接入集显的插口

 

4.安装driver

 可以直接安装cuda.deb的安装包，cuda-repo-ubuntu1404-7-5-local_7.5-18_amd64.deb(不可行)

应该安装cuda8

CUDA Toolkit 8 RC Now Available 

New in CUDA 8

Pascal Architecture Support 

· Out of box performance improvements on Tesla P100, supports GeForce GTX 1080

· Simplify programming using Unified memory on Pascal including support for large datasets, concurrent data access and atomics* 

· Optimize Unified Memory performance using new data migration APIs* 

· Faster Deep Learning using optimized cuBLAS routines for native FP16 computation

 

 

sudo dpkg -i cuda-repo-ubuntu1404-8-0-rc_8.0.27-1_amd64.deb

sudo apt-get ipdate

sudo apt-get install cuda

 

运行nvidia-smi 出现“nvidia-3xx drivers”找不到

sudo add-apt-repository ppa:graphics-drivers/ppa

sudo apt-get update

sudo apt-get install nvidia-367

sudo apt-get install mesa-common-dev

sudo apt-get install freeglut3-dev

 

sduo vi /etc/profile 添加

export PATH=/usr/local/cuda/bin:$PATH

export LD_LIBRARY_PATH=/usr/local/cuda/lib64:$LD_LIBRARY_PATH

 

5.重启，显示器接到独显

 

6. cudnn

解压cudnn-8.0-linux-x64-v5.1.tgz

cd lib64/

sudo cp lib* /usr/local/cuda/lib64/

cd ../include/

sudo cp cudnn.h /usr/local/cuda/include/

cd /usr/local/cuda/lib64/

sudo rm -r libcudnn.so libcudnn.so.5

sudo ln -sf libcudnn.so.5 libcudnn.so

sudo ln -sf libcudnn.so.5.1.5 libcudnn.so.5

sudo ldconfig

 

 7.caffe安装

http://caffe.berkeleyvision.org/installation.html