Ubuntu GPU版 caffe 安装

环境准备

基础环境

sudo apt-get update
sudo apt-get upgrade
sudo apt-get install -y build-essential cmake git pkg-config nano
sudo apt-get install -y libprotobuf-dev libleveldb-dev libsnappy-dev libhdf5-serial-dev protobuf-compiler libatlas-base-dev libboost-all-dev libgflags-dev libgoogle-glog-dev liblmdb-dev

# (Python general)

sudo apt-get install -y python-pip

# (Python 2.7 development files)
sudo apt-get install -y python-dev
sudo apt-get install -y python-numpy python-scipy

# (or, Python 3.5 development files)
sudo apt-get install -y python3-dev
sudo apt-get install -y python3-numpy python3-scipy

安装显卡驱动（英伟达）

安装方法见其他博客，ubuntu安装显卡驱动

安装cuda/cudnn

安装链接见我的博客文章，ubuntu安装cuda

下载caffe工程

官方工程地址：https://github.com/BVLC/caffe
百度云链接: https://pan.baidu.com/s/1k8SK_cKCXXiRBh5AYIRsQw
提取码: yu4q

编译安装

配置文件修改

解压工程，cd到工程目录执行下面命令：

cp Makefile.config.example Makefile.config

修改配置文件

nano Makefile.config

配置cuda相关

打开# USE_CUDNN := 1
配置cuda路径 CUDA_DIR := /usr/local/cuda
注释掉

-gencode arch=compute_20,code=sm_20 \
-gencode arch=compute_20,code=sm_21 \

CUDA_ARCH := #-gencode arch=compute_20,code=sm_20 \
		#-gencode arch=compute_20,code=sm_21 \
		-gencode arch=compute_30,code=sm_30 \
		-gencode arch=compute_35,code=sm_35 \
		-gencode arch=compute_50,code=sm_50 \
		-gencode arch=compute_52,code=sm_52 \
		-gencode arch=compute_60,code=sm_60 \
		-gencode arch=compute_61,code=sm_61 \
		-gencode arch=compute_61,code=compute_61

配置python

• Python2.7
  保持打开

PYTHON_INCLUDE := /usr/include/python2.7 \
		/usr/lib/python2.7/dist-packages/numpy/core/include

注释下面的Python3.5（默认情况就是已经注释好的）

# PYTHON_LIBRARIES := boost_python3 python3.5m
# PYTHON_INCLUDE := /usr/include/python3.5m \
#                 /usr/lib/python3.5/dist-packages/numpy/core/include

打开下面开关

WITH_PYTHON_LAYER := 1

• Python3.5
  注释下面的python2.7部分

#PYTHON_INCLUDE := /usr/include/python2.7 \
#		/usr/lib/python2.7/dist-packages/numpy/core/include

打开下面的Python3.5

PYTHON_LIBRARIES := boost_python3 python3.5m
PYTHON_INCLUDE := /usr/include/python3.5m \
                /usr/lib/python3.5/dist-packages/numpy/core/include

打开下面开关

WITH_PYTHON_LAYER := 1

• 其他Python版本（python3.x）
  需要找到该版本的python3.xm文件，改写python3.5配置，如下：

PYTHON_LIBRARIES := boost_python3 python3.xm
PYTHON_INCLUDE := /usr/include/python3.xm \
                /usr/lib/python3.x/dist-packages/numpy/core/include

其他配置
修改下面配置

# Whatever else you find you need goes here.
INCLUDE_DIRS := $(PYTHON_INCLUDE) /usr/local/include
LIBRARY_DIRS := $(PYTHON_LIB) /usr/local/lib /usr/lib

修改添加后如下：

INCLUDE_DIRS := $(PYTHON_INCLUDE) /usr/local/include /usr/include/hdf5/serial
LIBRARY_DIRS := $(PYTHON_LIB) /usr/local/lib /usr/lib /usr/lib/x86_64-linux-gnu /usr/lib/x86_64-linux-gnu/hdf5/serial

安装Python依赖

cd python

安装命令（for python2.7）

for req in $(cat requirements.txt); do pip install $req; done

安装命令（for python3）

for req in $(cat requirements.txt); do pip3 install $req; done

如需镜像加速：

for req in $(cat requirements.txt); do pip3 install $req -i https://pypi.douban.com/simple; done

一切顺利的话，进入下一步

编译caffe

依次执行下面命令：

make all
make test
make runtest
make pycaffe

加速编译：
每个命令后面加上-jn ，n为cpu核数，如果有4个空闲cpu，则命令可以为：

make all -j4
make test -j4
make runtest -j4
make pycaffe -j4

pycaffe环境配置

nano ~/.bashrc

最后面添加如下：

export PYTHONPATH=/path/to/caffe-master/python:$PYTHONPATH

问题解决