本教程安装caffe的前提是在已安装好cuda8.0和cuddn5.0的基础上安装caffe，安装cuda8.0和安装cudnn5.0的教程见本博客其他博文~~安装caffe环境如下~~~~
1.安装依赖的库
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 2.安装其余依赖 sudo apt-get install libgflags-dev libgoogle-glog-dev liblmdb-dev 3.安装git，拉取源码 apt-get install git git clone https://github.com/BVLC/caffe.git 4.再转到caffe的python目录，安装scipy cd caffe/python sudo apt-get install python-numpy python-scipy python-matplotlib ipython ipython-notebook python-pandas python-sympy python-nose 5.安装python的pip和easy_install，方便安装软件包 wget --no-check-certificate https://bootstrap.pypa.io/ez_setup.py python ez_setup.py --insecure wget https://bootstrap.pypa.io/get-pip.py python get-pip.py 6.安装MKL或者Atlas，我装的Atlas 安装命令 sudo apt-get install libatlas-base-dev 7.安装python依赖(路径根据自己的目录可能要调一下) cd caffe/python执行 su root
转到root用户执行
for req in $(cat requirements.txt); do pip install $req; done
8.编辑caffe所需的Makefile文件
cd caffe cp Makefile.config.example Makefile.config vim Makefile.config 取消对# USE_CUDNN := 1的注释 #为了启用cudnn加速
# cuDNN acceleration switch (uncomment to build with cuDNN).
# USE_CUDNN := 19.编译caffe
make -j4 测试一下编译结果 make test make runtest 10.编译pycaffe make pycaffe -j4