下載caffe并保存到一個目錄下(推薦放到 /home 目錄)安裝依賴項：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

sudo?apt-get?install?libatlas-base-dev

sudo?apt-get?install?python-dev

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

sudo?apt-get?install?python-lmdb

sudo?apt-get?install?python-pip

5.下載Caffe：

sudo apt-get install git

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

并修改配置文件

cd caffe

備份文件cp?Makefile.config.example?Makefile.configgedit Makefile.config

修改需編譯的選項(針對ubuntu15.04以及之后的版本，之前的版本不用修改，用于解決無法找到hdf5.h等文件)，還有#?Whatever?else?you?find?you?need?goes?here.

INCLUDE_DIRS?:=?$(Python_INCLUDE)?/usr/local/include

LIBRARY_DIRS?:=?$(PYTHON_LIB)?/usr/local/lib?/usr/lib

修改為

#?Whatever?else?you?find?you?need?goes?here.

INCLUDE_DIRS?:=?$(PYTHON_INCLUDE)?/usr/local/include?/usr/include/hdf5/serial?/usr/local/cuda-7.5/targets/x86_64-linux/

LIBRARY_DIRS?:=?$(PYTHON_LIB)?/usr/local/lib?/usr/lib/?/usr/lib/x86_64-linux-gnu/hdf5/serial?/usr/local/cuda-7.5/targets/x86_64-linux/

LIBRARIES?+=?glog?gflags?protobuf?leveldb?snappy?????????lmdb?boost_system?hdf5_hl?hdf5?m?????????opencv_core?opencv_highgui?opencv_imgproc?opencv_imgcodecs

編譯caffe：

make?all?-j4

make?test?-j4

make?runtest

make?all時有cublas_v2.h:?No?such?file?or?directory的錯誤

原以為是caffe編譯時沒有正確編譯CPU版本，或者caffe/util/device_alternate.hpp里未能正確判斷，但查看源文件發現此處正常，原來是我的項目中，在編譯時需要指定CPU_ONLY選項，即在我的項目Makefile里添加下面一句即可

CXXFLAGS?+=?-DCPU_ONLY=1

如果只時使用CPU，打開注釋CPU_ONLY=1即可

去掉CPU—only選項前的注釋,使之生效,若需Python layer,則還需去掉WITH_PYTHON_LAYER ：=1前的注釋

如需編譯MATLAB接口，則需修改MATLAB_DIR,改成當前系統下的MATLAB路徑

編譯python接口sudo?pip?install?-r?python/requirements.txt

make?pycaffe

編譯matlab接口make?matcaffe

gedit?~/.bashrc

最后面添加：export PYTHONPATH=/home/francis/caffe/python

rm *~ 可考慮刪除因修改而自動生成的備份文件

以上可成功配置好cpu_only

安裝GPU版本

應先裝好CUDA，條件允許可以可加入cudnn,用于加速卷積運算

CUDA安裝步驟待后續完善

注釋掉CPU-only，其他與前面相同