Ubuntu14.04下安裝caffe環境

本教程安裝caffe的前提是在已安裝好cuda8.0和cuddn5.0的基礎上安裝caffe，安裝cuda8.0和安裝cudnn5.0的教程見本部落格其他博文~~安裝caffe環境如下~~~~

1.安裝依賴的庫

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.安裝其餘依賴

sudo apt-get install libgflags-dev libgoogle-glog-dev liblmdb-dev

3.安裝git，拉取源碼

apt-get install git

git clone https://github.com/BVLC/caffe.git

4.再轉到caffe的python目錄，安裝scipy

cd caffe/python

sudo apt-get install python-numpy python-scipy python-matplotlib ipython ipython-notebook python-pandas python-sympy python-nose

5.安裝python的pip和easy_install，友善安裝軟體包

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.安裝MKL或者Atlas，我裝的Atlas

安裝指令

sudo apt-get install libatlas-base-dev 

7.安裝python依賴(路徑根據自己的目錄可能要調一下)

cd caffe/python

執行 su root 

轉到root使用者執行

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

8.編輯caffe所需的Makefile檔案

cd caffe 

cp Makefile.config.example Makefile.config 

vim Makefile.config 

取消對# USE_CUDNN := 1的注釋  #為了啟用cudnn加速

# cuDNN acceleration switch (uncomment to build with cuDNN).

# USE_CUDNN := 1

9.編譯caffe

make -j4 

測試一下編譯結果 

make test 

make runtest

10.編譯pycaffe

make pycaffe -j4