五，安装caffe2

1，创建anaconda3下面的虚拟环境：

conda create -n caffe2_py3 python=3.6

source activate caffe2_py

官网安装

ubuntu16.04.5  cuda9.0 cudnn7_7.1.4

sudo apt-get update
sudo apt-get install -y --no-install-recommends \
      build-essential \
      git \
      libgoogle-glog-dev \
      libgtest-dev \
      libiomp-dev \
      libleveldb-dev \
      liblmdb-dev \
      libopencv-dev \
      libopenmpi-dev \
      libsnappy-dev \
      libprotobuf-dev \
      openmpi-bin \
      openmpi-doc \
      protobuf-compiler \
      python-dev \
      python-pip                          
pip install --user \
      future \
      numpy \
      protobuf \
      typing \
      hypothesis

sudo apt-get install -y --no-install-recommends \
      libgflags-dev \
      cmake
git clone https://github.com/pytorch/pytorch.git && cd pytorch
git submodule update --init --recursive
python setup.py install

 

 

 

2,安装python需要的包

conda install -y future
conda install -y gflags
conda install -y leveldb
conda install -y mkl   
conda install -y mkl-include  
conda install -y numpy   
conda install -y opencv   
conda install -y protobuf  
conda install -y six
conda install -y lmdb

conda install -y matplotlib

conda install -y flask
conda install -y graphviz

conda install -y hypothesis
conda install -y jupyter
conda install -y pydot
conda install -y pyyaml
conda install -y requests
conda install -y scikit-image
conda install -y scipy
conda install -y setuptools
conda install -y tornado

conda install -c menpo opencv3 

conda install -y cython  ipython h5py nose pandas protobuf

conda install -y spyder

3,下载caffe2 编译

cd ~

git clone --recursive https://github.com/pytorch/pytorch.git

cd pytorch

git submodule update --init

 mkdir build && cd build

cmake -DCMAKE_PREFIX_PATH=~/anaconda3/envs/caffe2_py3 -DCMAKE_INSTALL_PREFIX=~/anaconda3/envs/caffe2_py3 -DUSE_NATIVE_ARCH=ON  ..

make

caffe2的好处是: Makefile中会根据你的库进行make，省去了makefile.config的设置步骤。

make install

4，测试：

cd  /home/xxxxx/Downloads/caffe/pytorch/build

1）python -c 'from caffe2.python import core' 2>/dev/null && echo "Success" || echo "Failure"

2）执行 python

import caffe2.python.core #导入这个包，看看是否成功，如果没报错，就OK了。