Xavier_NX_cuda10.0_cudnn7.5_opencv3.4.7_caffe环境配置

安装各种依赖包

• sudo apt-get update
• sudo apt-get install build-essential cmake git pkg-config
• sudo apt-get install libprotobuf-dev libleveldb-dev libsnappy-dev libhdf5-serial-dev protobuf-compiler
• sudo apt-get install libatlas-base-dev
• sudo apt-get install libboost-all-dev
• sudo apt-get install libgflags-dev libgoogle-glog-dev liblmdb-dev
• sudo apt-get install python-pip
• sudo apt-get install libopencv-dev
• sudo apt-get install python-dev
• sudo apt-get install python-numpy python-scipy
• sudo apt-get install libopenblas-dev liblapack-dev libatlas-base-dev

安装caffe

• git clone https://github.com/BVLC/caffe.git      //从github上git caffe

修改Makefile.conf配置文件

• cd caffe
• sudo cp Makefile.config.example Makefile.config  
• sudo gedit Makefile.config
• 若使用cudnn，则将# USE_CUDNN := 1 修改成： USE_CUDNN := 1
• 若要使用python来编写layer，则需要将# WITH_PYTHON_LAYER := 1 修改为 WITH_PYTHON_LAYER := 1
• 若使用的opencv版本是3的，则将# OPENCV_VERSION := 3 修改为： OPENCV_VERSION := 3

如下图所示：

• 重要的一项将# Whatever else you find you need goes here.下面的
• INCLUDE_DIRS := $(PYTHON_INCLUDE) /usr/local/include
• LIBRARY_DIRS := $(PYTHON_LIB) /usr/local/lib /usr/lib

修改为：

• INCLUDE_DIRS := $(PYTHON_INCLUDE) /usr/local/include /usr/include/hdf5/serial
• LIBRARY_DIRS := $(PYTHON_LIB) /usr/local/lib /usr/lib /usr/lib/aarch64-linux-gnu /usr/lib/aarch64-linux-gnu/hdf5/serial

将

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

中的

        -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 \

删除，如果不删除，对caffe进行编译时会报错。报错如下图：

修改Makefile文件

将Makefile文件中的

NVCCFLAGS += -ccbin=$(CXX) -Xcompiler -fPIC $(COMMON_FLAGS)

替换为
NVCCFLAGS +=-D_FORCE_INLINES-ccbin=$(CXX) -Xcompiler-fPIC $(COMMON_FLAGS)将LIBRARIES+= glog gflags protobuf boost_system boost_filesystem m hdf5_hl hdf5
修改为：
LIBRARIES += glog gflags protobuf boost_system boost_filesystem boost_regex m hdf5_hl hdf5

修改host_config.h文件

• sudo gedit /usr/local/cuda/include/crt/host_config.h
• 将#error -- unsupported GNU version! gcc versions later than 7 are not supported!

改为

//#error -- unsupported GNU version! gcc versions later than 7 are not supported!

设置环境变量

• sudo gedit ~/.bashrc
• 在最后加入以下内容：

export PYTHONPATH=/home/nvidia/caffe/python:$PYTHONPATH

export PATH=$PATH:/usr/local/cuda-10.0/bin

export PATH=/usr/local/cuda-10.0/bin:$PATH

进入caffe目录进行编译

• cd caffe
• make all
• make test
• make runtest
• make pycaffe
• 最后，终端输入python，然后再输入import caffe，如果无问题则安装成功。