linux服务器端 + cuda10 + anaconda3 + py35 + opencv3 之 caffe安装

linux服务器端 + cuda10 + anaconda3 + py35 + opencv3 之 caffe安装

本文算是caffe踩坑记录，折腾了两天，最终解决了，欣喜之余赶紧记录一波。

前提条件：

• 已安装cuda , 但cudnn没有安装（因为没权限）
  • 更正，使用conda install cudnn即可安装，编译caffe的时候打开cudnn加速也是可以的
• 无法使用sudo权限
• 没有图形界面，是服务器端

0. 如何查看版本号

1. 查看cuda版本 cat /usr/local/cuda/version.txt
2. 查看cudnn版本 cat /usr/local/cuda/include/cudnn.h | grep CUDNN_MAJOR -A 2
3. 查看python版本 python -V
4. 查看opencv版本

进入python
import cv2
cv2.__version__

1. 安装anaconda

上官网 https://www.anaconda.com/distribution/, 选择对应的anaconda版本，这里我选择的是linux + python 3.7 版本，使用wget进行下载：

wget https://repo.anaconda.com/archive/Anaconda3-2019.03-Linux-x86_64.sh

然后进行安装 sh Anaconda3-2019.03-Linux-x86_64.sh，注意记得修改安装路径，一般不建议装到/里，因为linux系统挂载方式决定了/相当于是win的C盘，最好是放在/data*/里面。

完成后修改用户环境变量vim ~/.bashrc，在文件末尾添加anaconda3的安装路径，例如：

# add anaconda env
export PATH=/data*/***/tool/anaconda3/bin:$PATH

接着更新环境变量source ~/.bashrc

2. 安装编译环境

1. 新建虚拟环境
   如果不想多个项目依赖冲突，还是建立个虚拟环境把各个项目给隔开吧

conda create -n caffe_py35 python=3.5

这里-n代表命名，python=3.5规定python的版本号

• conda env list 输出所有虚拟环境名
• conda activate XXX 激活XXX环境
• conda deactivate 关闭已激活的XXX环境
• conda remove -n XXX --all 删除XXX环境

因此此处键入conda activate caffe_py35，激活python3.5环境，这是终端应该会显示(caffe_py35)xxx@xxx: xxx$，查看python的版本号也可以发现，是py35版。

2. 安装opencv

conda install opencv

这里py35对应的是opencv3，所以也不需要刻意规定opencv的版本，可通过查看opencv版本的方法判断是否成功安装。

3. 编译caffe

0. 进入caffe_py35虚拟环境

1. 找个好地方把caffe源码git下来，不知道是下在release版，还是直接git好，这里我就直接git了

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

2. 进入caffe目录，配置Makefile.config

巨坑预警

这里把自带的config复制过来

cp Makefile.config.example Makefile.config

在它的基础上进行修改。这时候你得仔细阅读其中的每一项，找到有关python和anaconda的行，进行修改操作，修改总结一下，大致有6处：

# Uncomment if you're using OpenCV 3
OPENCV_VERSION := 3
# ---------------------
# 因为是py35，因此这里需要修改
# ---------------------

# 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.
CUDA_ARCH := -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
#-gencode arch=compute_20,code=sm_20 \
#-gencode arch=compute_20,code=sm_21 \
# ---------------------
# 因为是cuda10，所以把 *_20 and *_21 lines 删去
# ---------------------

# 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.
# 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_HOME := ../anaconda3/envs/caffe_py35
PYTHON_INCLUDE := $(ANACONDA_HOME)/include \
		$(ANACONDA_HOME)/include/python3.5m \
		$(ANACONDA_HOME)/lib/python3.5/site-packages/numpy/core/include
# ---------------------
# 这里是anaconda的安装位置，因为是虚拟环境，一定要注意ANACONDA_HOME的位置
# ---------------------

# Uncomment to use Python 3 (default is Python 2)
# PYTHON_LIBRARIES := boost_python3 python3.5m
PYTHON_LIBRARIES := boost_python-py35
# ---------------------
# 需要libboost，去/usr/lib/x86_64-linux-gnu/找一找libboost_python-py35.so.1.58.0是否匹配
# ---------------------

# Whatever else you find you need goes here.
INCLUDE_DIRS := $(PYTHON_INCLUDE) /usr/local/include /usr/include/hdf5/serial 
LIBRARY_DIRS := $(PYTHON_LIB) /usr/local/lib /usr/lib /usr/lib/x86_64-linux-gnu/hdf5/serial
LINKFLAGS := -Wl,-rpath,$(PYTHON_LIB)
# ---------------------
# LINKFLAGS是 * [.build_release/tools/compute_image_mean.bin] Error 1 的坑，需要重新定义lib的位置
# ---------------------

这里我把我的Makefile.config贴出来供大家参考：

## 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).
# USE_CUDNN := 1
# ---------------------
# 需要用cudnn加速不，我这里没有cudnn，就没有取消注释
# ---------------------

# CPU-only switch (uncomment to build without GPU support).
# CPU_ONLY := 1

# uncomment to disable IO dependencies and corresponding data layers
# USE_OPENCV := 0
# USE_LEVELDB := 0
# USE_LMDB := 0
# This code is taken from https://github.com/sh1r0/caffe-android-lib
# USE_HDF5 := 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
# ---------------------
# 因为是py35，因此这里需要修改
# ---------------------

# 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++
# CUSTOM_CXX := g++

# CUDA directory contains bin/ and lib/ directories that we need.
CUDA_DIR := /usr/local/cuda
# ---------------------
# cuda的安装位置，需要自行注意
# ---------------------

# 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.
CUDA_ARCH := -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
#-gencode arch=compute_20,code=sm_20 \
#-gencode arch=compute_20,code=sm_21 \
# ---------------------
# 因为是cuda10，所以把 *_20 and *_21 lines 删去
# ---------------------

# BLAS choice:
# atlas for ATLAS (default)
# mkl for MKL
# open for OpenBlas
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.
# 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_HOME := ../anaconda3/envs/caffe_py35
PYTHON_INCLUDE := $(ANACONDA_HOME)/include \
		$(ANACONDA_HOME)/include/python3.5m \
		$(ANACONDA_HOME)/lib/python3.5/site-packages/numpy/core/include
# ---------------------
# 这里是anaconda的安装位置，因为是虚拟环境，一定要注意ANACONDA_HOME的位置
# ---------------------

# Uncomment to use Python 3 (default is Python 2)
# PYTHON_LIBRARIES := boost_python3 python3.5m
PYTHON_LIBRARIES := boost_python-py35
# ---------------------
# 需要libboost，去/usr/lib/x86_64-linux-gnu/找一找libboost_python-py35.so.1.58.0是否匹配
# ---------------------

# 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.
# PYTHON_LIB := /usr/lib
PYTHON_LIB := $(ANACONDA_HOME)/lib
# ---------------------
# 这里是anaconda的安装位置
# ---------------------

# 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)
# WITH_PYTHON_LAYER := 1

# Whatever else you find you need goes here.
INCLUDE_DIRS := $(PYTHON_INCLUDE) /usr/local/include /usr/include/hdf5/serial 
LIBRARY_DIRS := $(PYTHON_LIB) /usr/local/lib /usr/lib /usr/lib/x86_64-linux-gnu/hdf5/serial
LINKFLAGS := -Wl,-rpath,$(PYTHON_LIB)
# ---------------------
# LINKFLAGS是 * [.build_release/tools/compute_image_mean.bin] Error 1 的坑，需要重新定义lib的位置
# ---------------------

# If Homebrew is installed at a non standard location (for example your home directory) and you use it for general dependencies
# 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)
# 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.)
# 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 ?= @

3. make

• make clean
• make all
• make test
• make runtest
• make pycaffe

看到一切都是 - RUN OK - ，舒服了

4. 添加caffe到环境变量

vim ~/.bashrc，在末尾加入

export PYTHONPATH=/data*/***/tool/caffe/python

5. import caffe的坑
   1. ImportError: No module named google.protobuf.internal
      到anaconda的bin文件夹下使用pip install protobuf，千万不要直接conda install，不然会没用的。
   2. ImportError: No module named ‘skimage’
      同样地，到anaconda的bin里面使用pip install scikit-image
   3. ImportError: libhdf5.so.101: cannot open shared object file: No such file or directory
      到环境依赖~/.bashrc里面进行修改，把在Makefile.config里面的libhdf5路径复制进去：

export LD_LIBRARY_PATH="/data*/***/tool/anaconda3/lib":$LD_LIBRARY_PATH

记得source ~/.bashrc下。

4. 参考资料与致谢

1. http://yingshu.ink/2017/01/12/Python3-5-Anaconda3-Caffe深度学习框架搭建/
2. https://blog.csdn.net/sinat_35406909/article/details/84198140
3. https://blog.csdn.net/yhaolpz/article/details/71375762
4. https://www.jianshu.com/p/1e405b9fe973
5. https://blog.csdn.net/u011534057/article/details/51659999