本人编译如下版本caffe后，记录下需要注意（修改）Makefile.config的选项
系统：ubuntu 14.04
CUDA：CUDA7.5
Python：Python2.7
仅支持CPU
cat Makefile.config
## Refer to http://caffe.berkeleyvision.org/installation.html
# Contributions simplifying and improving our build system are welcome!
# cuDNN acceleration switch (uncomment to build with cuDNN).
# 是否使用cudnn，默认不使用
# USE_CUDNN := 1
# CPU-only switch (uncomment to build without GPU support).
# 是否仅支持cup，注释后会编译gpu部分，因人而异，我自己学习支持cpu就好
CPU_ONLY := 1
# uncomment to disable IO dependencies and corresponding data layers
# 选择支持库，可以先按默认都不选，包括下面对opencv，leveldb，lmdb库的选项都可以注释
# USE_OPENCV := 0
# USE_LEVELDB := 0
# USE_LMDB := 0
# uncomment to allow MDB_NOLOCK when reading LMDB files (only if necessary)
# You should not set this flag if you will be reading LMDBs with any
# possibility of simultaneous read and write
# ALLOW_LMDB_NOLOCK := 1
# Uncomment if you're using OpenCV 3
# OPENCV_VERSION := 3
# To customize your choice of compiler, uncomment and set the following.
# N.B. the default for Linux is g++ and the default for OSX is clang++
# 可选g++版本，如果编译报错因g++库问题，可以尝试换个g++版本，比如曾经报错然后试过g++-5.x，g++-4.x等
CUSTOM_CXX := g++
# CUDA directory contains bin/ and lib/ directories that we need.
# cuda安装命令，其实使用NVADIA gpu才会用到cuda,但是可以使用sudo apt-get install nvidia-cuda-toolkit
# 安装后，which cuda，添加路径编译
CUDA_DIR := /usr/local/cuda
# On Ubuntu 14.04, if cuda tools are installed via
# "sudo apt-get install nvidia-cuda-toolkit" then use this instead:
# CUDA_DIR := /usr
# CUDA architecture setting: going with all of them.
# For CUDA < 6.0, comment the *_50 through *_61 lines for compatibility.
# For CUDA < 8.0, comment the *_60 and *_61 lines for compatibility.
# For CUDA >= 9.0, comment the *_20 and *_21 lines for compatibility.
# 这个根据cuda版本不同选择注释，看上面注释，我是cuda7.5，所以注释*_60等
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
# BLAS choice:
# atlas for ATLAS (default)
# mkl for MKL
# open for OpenBlas
# BLAS库，安装atlas就使用这个吧
BLAS := atlas
# Custom (MKL/ATLAS/OpenBLAS) include and lib directories.
# Leave commented to accept the defaults for your choice of BLAS
# (which should work)!
# BLAS_INCLUDE := /path/to/your/blas
# BLAS_LIB := /path/to/your/blas
# Homebrew puts openblas in a directory that is not on the standard search path
# BLAS_INCLUDE := $(shell brew --prefix openblas)/include
# BLAS_LIB := $(shell brew --prefix openblas)/lib
# This is required only if you will compile the matlab interface.
# MATLAB directory should contain the mex binary in /bin.
# MATLAB_DIR := /usr/local
# MATLAB_DIR := /Applications/MATLAB_R2012b.app
# NOTE: this is required only if you will compile the python interface.
# We need to be able to find Python.h and numpy/arrayobject.h.
# 这个是python库路径，可通过ubuntu的python库在 /usr/lib/x86-xxxx(忘了全拼，自己按tab补全吧)/python2.7/里面
PYTHON_LIBRARIES := boost_python-py27 python2.7
# Python头文件路径，根据实际情况填写
PYTHON_INCLUDE := /usr/include/python2.7 \
/usr/lib/python2.7/dist-packages/numpy/core/include
# Anaconda Python distribution is quite popular. Include path:
# Verify anaconda location, sometimes it's in root.
# anaconda是一个封装的环境，里面有编译caffe所需的库，如果通过anaconda安装caffe时候打开配置，并
# 注释上面python2.7的配置，修改成anaconda的环境路径，这里我不需要
# ANACONDA_HOME := $(HOME)/anaconda
# PYTHON_INCLUDE := $(ANACONDA_HOME)/include \
# $(ANACONDA_HOME)/include/python2.7 \
# $(ANACONDA_HOME)/lib/python2.7/site-packages/numpy/core/include
# Uncomment to use Python 3 (default is Python 2)
# python3的路径配置，我是2，所以不需要，根据实际情况配置
# PYTHON_LIBRARIES := boost_python3 python3.5m
# PYTHON_INCLUDE := /usr/include/python3.5m \
# /usr/lib/python3.5/dist-packages/numpy/core/include
# We need to be able to find libpythonX.X.so or .dylib.
# 按默认的，如果使用anaconda，则换成下面的
PYTHON_LIB := /usr/lib
# PYTHON_LIB := $(ANACONDA_HOME)/lib
# Homebrew installs numpy in a non standard path (keg only)
# PYTHON_INCLUDE += $(dir $(shell python -c 'import numpy.core; print(numpy.core.__file__)'))/include
# PYTHON_LIB += $(shell brew --prefix numpy)/lib
# Uncomment to support layers written in Python (will link against Python libs)
# 如果想要caffe支持python接口，则这个打开，我会编译make && make pycaffe，所以需要，如果不编译，则
# 使用C++接口
WITH_PYTHON_LAYER := 1
# Whatever else you find you need goes here.
# 库的配置，这里添加hdf的库，我环境会编译报错：can not find hdf5.h
INCLUDE_DIRS := $(PYTHON_INCLUDE) /usr/local/include /usr/include/hdf5/serial/
LIBRARY_DIRS := $(PYTHON_LIB) /usr/local/lib /usr/lib
# If Homebrew is installed at a non standard location (for example your home directory) and you use it for general dependencies
# 库的配置，默认
# INCLUDE_DIRS += $(shell brew --prefix)/include
# LIBRARY_DIRS += $(shell brew --prefix)/lib
# NCCL acceleration switch (uncomment to build with NCCL)
# https://github.com/NVIDIA/nccl (last tested version: v1.2.3-1+cuda8.0)
# gpu相关，跟我没关系
# USE_NCCL := 1
# Uncomment to use `pkg-config` to specify OpenCV library paths.
# (Usually not necessary -- OpenCV libraries are normally installed in one of the above $LIBRARY_DIRS.)
# opencv支持，暂时不需要
# USE_PKG_CONFIG := 1
# N.B. both build and distribute dirs are cleared on `make clean`
BUILD_DIR := build
DISTRIBUTE_DIR := distribute
# Uncomment for debugging. Does not work on OSX due to https://github.com/BVLC/caffe/issues/171
# DEBUG := 1
# The ID of the GPU that 'make runtest' will use to run unit tests.
TEST_GPUID := 0
# enable pretty build (comment to see full commands)
Q ?= @另外补充intel分支的Makefile.config增加项说明：
# 这两项是intel对使用mkl2017和mkldnn的优化，选择成功后（可以看下/src/caffe/net.cpp中部分代码），
# caffe.Net解析网络时会对Scale层进行融合，认为是Drop层，计算时忽略。
# USE_MKL2017_AS_DEFAULT_ENGINE := 1
# or put this at the top your train_val.protoxt or solver.prototxt file:
# engine: "MKL2017"
# or use this option with caffe tool:
# -engine "MKL2017"
USE_MKLDNN_AS_DEFAULT_ENGINE := 1
# Put this at the top your train_val.protoxt or solver.prototxt file:
# engine: "MKLDNN"
# or use this option with caffe tool:
# -engine "MKLDNN"
# caffe.Net解析网络时会对Bn层进行融合，认为是Drop层，计算时忽略，下面也是相关算法优化
# Use remove batch norm optimization to boost inference
DISABLE_BN_FOLDING := 0
# Use Conv + Relu fusion to boost inference
DISABLE_CONV_RELU_FUSION:= 0
# Use Bn + ReLU fusion to boost inference
DISABLE_BN_RELU_FUSION := 0
# Use Conv + Eltwise + Relu layer fusion to boost inference.
DISABLE_CONV_SUM_FUSION := 0
# Use sparse to boost inference.
DISABLE_SPARSE := 0