Ubuntu16.04+Anaconda3虚拟环境下配置caffe+opencv3.4.2+python3.5

我会带着你远行 2021-11-16 11:20 259阅读 0赞

本机Geforce gtx1060 驱动版本384.130， 已安装cuda-9.0，cudnn7.0.5，Anaconda3

常用命令：

find /|grep filename #寻找某个文件
    ls -l  # 列出该文件夹下的所有文件信息
    sudo gedit file
    sudo chmod 777 filename
    sudo cp filename1 filename2 
    sudo ln -s filename1 filename2 # 把filename1建立软链接为filename2
    ldconfig -v  # 在默认搜寻目录(/lib和/usr/lib)以及动态库配置文件/etc/ld.so.conf内所列的目录下,搜索出可共享的动态链接库(格式如,lib*.so*),
    进而创建出动态装入程序(ld.so), 所以往/lib和/usr/lib里面加东西，是不用修改/etc/ld.so.conf的，但是添加之后之后要运行一下 ldconfig
    
    sudo gedit ~/.bashrc  # 添加环境变量需要，每个用户都有一个.bashrc文件，在用户目录下.  
    # /etc/bashrc或/etc/bash.bashrc:为每一个运行bash shell的用户执行此文件
    source ~/.bashrc  #重新执行刚修改的初始化文件，使之立即生效，而不必注销并重新登录

**第一步 安装caffe环境**

按照[https://www.jianshu.com/p/8d309dcc8b99][https_www.jianshu.com_p_8d309dcc8b99]的步骤一步一步走

安装caffe for python的虚拟环境--->安装基础依赖库--->克隆Caffe代码--->修改Makefile.config、Makefile

\--->编译pycaffe--->拷贝编译后的libcaffe.so至caffe的虚拟环境。

*我的最终版Makefile.config内容如下：*

***注意：** 在没有安装opencv之前，我没有去取消23和108行的注释，是为了一步一步来，防止不知道问题出在caffe还是opencv。*

*对于该文件的修改，我还结合了这篇[https://www.cnblogs.com/whb-20160329/p/10256766.html][https_www.cnblogs.com_whb-20160329_p_10256766.html]，哪里该注释掉还是取消注释，自己最好搞清楚，会有所收获。*

1 ## Refer to http://caffe.berkeleyvision.org/installation.html
      2 # Contributions simplifying and improving our build system are welcome!
      3 
      4 # cuDNN acceleration switch (uncomment to build with cuDNN).
      5   USE_CUDNN := 1
      6 
      7 # CPU-only switch (uncomment to build without GPU support).
      8 # CPU_ONLY := 1
      9 
     10 # uncomment to disable IO dependencies and corresponding data layers
     11 # USE_OPENCV := 0
     12 # USE_LEVELDB := 0
     13 # USE_LMDB := 0
     14 # This code is taken from https://github.com/sh1r0/caffe-android-lib
     15 # USE_HDF5 := 0
     16 
     17 # uncomment to allow MDB_NOLOCK when reading LMDB files (only if necessary)
     18 #    You should not set this flag if you will be reading LMDBs with any
     19 #    possibility of simultaneous read and write
     20 # ALLOW_LMDB_NOLOCK := 1
     21 
     22 # Uncomment if you're using OpenCV 3
     23   OPENCV_VERSION := 3
     24 
     25 # To customize your choice of compiler, uncomment and set the following.
     26 # N.B. the default for Linux is g++ and the default for OSX is clang++
     27 # CUSTOM_CXX := g++
     28 
     29 # CUDA directory contains bin/ and lib/ directories that we need.
     30 CUDA_DIR := /usr/local/cuda-9.0
     31 # On Ubuntu 14.04, if cuda tools are installed via
     32 # "sudo apt-get install nvidia-cuda-toolkit" then use this instead:
     33 # CUDA_DIR := /usr
     34 
     35 # CUDA architecture setting: going with all of them.
     36 # For CUDA < 6.0, comment the *_50 through *_61 lines for compatibility.
     37 # For CUDA < 8.0, comment the *_60 and *_61 lines for compatibility.
     38 # For CUDA >= 9.0, comment the *_20 and *_21 lines for compatibility.
     39 CUDA_ARCH := -gencode arch=compute_30,code=sm_30 \
     40         -gencode arch=compute_35,code=sm_35 \
     41         -gencode arch=compute_50,code=sm_50 \
     42         -gencode arch=compute_52,code=sm_52 \
     43         -gencode arch=compute_60,code=sm_60 \
     44         -gencode arch=compute_61,code=sm_61 \
     45         -gencode arch=compute_61,code=compute_61
     46 
     47 # BLAS choice:
     48 # atlas for ATLAS (default)
     49 # mkl for MKL
     50 # open for OpenBlas
     51 BLAS := atlas
     52 # Custom (MKL/ATLAS/OpenBLAS) include and lib directories.
     53 # Leave commented to accept the defaults for your choice of BLAS
     54 # (which should work)!
     55 # BLAS_INCLUDE := /path/to/your/blas
     56 # BLAS_LIB := /path/to/your/blas
     57 
     58 # Homebrew puts openblas in a directory that is not on the standard search path
     59 # BLAS_INCLUDE := $(shell brew --prefix openblas)/include
     60 # BLAS_LIB := $(shell brew --prefix openblas)/lib
     61 
     62 # This is required only if you will compile the matlab interface.
     63 # MATLAB directory should contain the mex binary in /bin.
     64 # MATLAB_DIR := /usr/local
     65 # MATLAB_DIR := /Applications/MATLAB_R2012b.app
     66 
     67 # NOTE: this is required only if you will compile the python interface.
     68 # We need to be able to find Python.h and numpy/arrayobject.h.
     69 # PYTHON_INCLUDE := /usr/include/python2.7 \
     70         /usr/lib/python2.7/dist-packages/numpy/core/include
     71 # Anaconda Python distribution is quite popular. Include path:
     72 # Verify anaconda location, sometimes it's in root.
     73   ANACONDA_HOME := $(HOME)/anaconda3/envs/caffe
     74   PYTHON_INCLUDE := $(ANACONDA_HOME)/include \
     75           $(ANACONDA_HOME)/include/python3.5m \
     76           $(ANACONDA_HOME)/lib/python3.5/site-packages/numpy/core/include
     77 
     78 # Uncomment to use Python 3 (default is Python 2)
     79 # PYTHON_LIBRARIES := boost_python3 python3.5m
     80 # PYTHON_INCLUDE := /usr/include/python3.5m \
     81 #                 /usr/lib/python3.5/dist-packages/numpy/core/include
     82 
     83 # We need to be able to find libpythonX.X.so or .dylib.
     84 # PYTHON_LIB := /usr/lib
     85   PYTHON_LIB := $(ANACONDA_HOME)/lib
     86 
     87 # Homebrew installs numpy in a non standard path (keg only)
     88 # PYTHON_INCLUDE += $(dir $(shell python -c 'import numpy.core; print(numpy.core.__file__)'))/include
     89 # PYTHON_LIB += $(shell brew --prefix numpy)/lib
     90 
     91 # Uncomment to support layers written in Python (will link against Python libs)
     92   WITH_PYTHON_LAYER := 1
     93 
     94 # Whatever else you find you need goes here.
     95 INCLUDE_DIRS := $(PYTHON_INCLUDE) /usr/local/include /usr/include/hdf5/serial/ /usr/local/cuda-9.0/include/
     96 LIBRARY_DIRS := $(PYTHON_LIB) /usr/local/lib /usr/lib /usr/local/cuda-9.0/lib64/
     97 
     98 # If Homebrew is installed at a non standard location (for example your home directory) and you use it for general dependencies
     99 # INCLUDE_DIRS += $(shell brew --prefix)/include
    100 # LIBRARY_DIRS += $(shell brew --prefix)/lib
    101 
    102 # NCCL acceleration switch (uncomment to build with NCCL)
    103 # https://github.com/NVIDIA/nccl (last tested version: v1.2.3-1+cuda8.0)
    104 # USE_NCCL := 1
    105 
    106 # Uncomment to use `pkg-config` to specify OpenCV library paths.
    107 # (Usually not necessary -- OpenCV libraries are normally installed in one of the above $LIBRARY_DIRS.)
    108   USE_PKG_CONFIG := 1
    109 
    110 # N.B. both build and distribute dirs are cleared on `make clean`
    111 BUILD_DIR := build
    112 DISTRIBUTE_DIR := distribute
    113 
    114 # Uncomment for debugging. Does not work on OSX due to https://github.com/BVLC/caffe/issues/171
    115 # DEBUG := 1
    116 
    117 # The ID of the GPU that 'make runtest' will use to run unit tests.
    118 TEST_GPUID := 0
    119 
    120 # enable pretty build (comment to see full commands)
    121 Q ?= @

编译caffe，要在你git下载的caffe目录下编译

make -j8 \# make clean是清除之前编译产生的内容

这里你可能会遇到一些问题，仔细检擦自己的Makefile.config先（要搞明白这个文件），再去百度搜索问题。

**第二步 安装opencv3.4.2**

同时参考[https://blog.csdn.net/dl\_chenbo/article/details/78687461/][https_blog.csdn.net_dl_chenbo_article_details_78687461]和[https://www.jianshu.com/p/c42fc8702f2b][https_www.jianshu.com_p_c42fc8702f2b]，下面只说与他们的不同之处：

1. 本博客所述整个过程只要是安装东西，我都是在激活虚拟环境下安装的，因为本身用虚拟环境就是为了不影响别的用户。

2. 我下载的是3.4.2版本，解压之后如下

![1595953-20190709142910876-733228789.png][]

3. CMake 配置如下：（根据自己的目录修改几个小地方）

cmake -D CMAKE_BUILD_TYPE=Release -D INSTALL_PYTHON_EXAMPLES=ON -D INSTALL_C_EXAMPLES=OFF -D BUILD_EXAMPLES=ON -D OPENCV_EXTRA_MODULES_PATH=/home/fly/opencv/opencv_contrib-3.4.2/modules -D CMAKE_LIBRARY_PATH=/usr/local/cuda-9.0/lib64/stubs -D WITH_CUDA=ON -DBUILD_TIFF=ON -DBUILD_opencv_java=OFF -DWITH_OPENGL=ON -DWITH_OPENCL=ON -DWITH_IPP=ON -DWITH_TBB=ON -DWITH_EIGEN=ON -DWITH_V4L=ON -DWITH_VTK=OFF -DBUILD_TESTS=OFF -DBUILD_opencv_dnn=ON -D CUDA_NVCC_FLAGS="-D_FORCE_INLINES --expt-relaxed-constexpr" -DCMAKE_C_COMPILER=/usr/bin/cc -DCMAKE_CXX_COMPILER=/usr/bin/c++ -DBUILD_LIBPROTOBUF_FROM_SOURCES=ON -DBUILD_PERF_TESTS=OFF -DCMAKE_INSTALL_PREFIX=$(python -c "import sys; print(sys.prefix)") -DPYTHON3_EXECUTABLE=$(which python) -DPYTHON_INCLUDE_DIR=$(python -c "from distutils.sysconfig import get_python_inc; print(get_python_inc())") -DPYTHON_PACKAGES_PATH=$(python -c "from distutils.sysconfig import get_python_lib; print(get_python_lib())") -DPYTHON3_NUMPY_INCLUDE_DIRS=$(python -c "import numpy; import os; print(os.path.join(numpy.__path__[0], 'core', 'include'))") -DPYTHON_LIBRARY=/home/fly/anaconda3/envs/caffe/lib/libpython3.5m.so ..

4. 编译遇到许多文件下载不下来

参考[https://blog.csdn.net/CSDN330/article/details/86747867][https_blog.csdn.net_CSDN330_article_details_86747867]

![1595953-20190709143448383-1632913298.png][]

打开这个文件，ctrl+f 查找failed，自然会发现它在哪些个网址下载哪些包（安装不同的版本的opencv对应不同版本的包）的时候失败，以及这些包应该下载的位置，我利用一个插件（类似于vpn）下载了这些包，并且根据参考链接里的方法将下载地址修改为本地，才成功。

**还要（重要！）**

进入目录：/etc/ld.so.conf.d

创建：opencv.conf \# sudo gedit opencv.conf

添加内容：/home/fly/opencv/opencv-3.4.2/build/lib

执行：ldconfig

最后**make install**别忘了。

补充说明：提供的链接目的是为了更好理解自己的安装。遇到问题，先看链接里面有木有，检查自己的，再百度。

**第三步 再次编译caffe（此时你已经安装了opencv）**

make clean

make -j8

make test -j8

make runtest -j8

make pycaffe -j8

我遇到问题之后就做了下面的事情

进入home/fly/ ， sudo gedit ~/.bashrc，再最后加上

export PYTHONPATH=/home/fly/CAFFE_ROOT/caffe/python:$PYTHONPATH  #将caffe的python路径加入到环境变量中，为了使得import caffe成功
    export PKG_CONFIG_PATH=/home/fly/anaconda3/envs/caffe/lib/pkgconfig:$PKG_CONFIG_PATH  # 将opencv库路径加入环境变量
    
    #查看linux下的opencv安装版本
    pkg-config opencv --modversion

#这里你可以使用pkg-config opencv --libs看你的opencv安装在哪里，也与makefile.config里面的第108行相呼应。

再次参考[https://www.cnblogs.com/whb-20160329/p/10256766.html][https_www.cnblogs.com_whb-20160329_p_10256766.html]和[https://www.jianshu.com/p/8d309dcc8b99][https_www.jianshu.com_p_8d309dcc8b99]和[https://www.jianshu.com/p/c42fc8702f2b][https_www.jianshu.com_p_c42fc8702f2b]后面的部分（根据自己的目录修改），其中这个*pip install scikit-image protobuf ，caffe依赖的库一定要装，不然会import不成功。*

*最终*

*![1595953-20190709145336724-1835471299.png][]*

说明：第一次配置caffe，历时五天，踩了许多坑，参考了很多博客和网站，也收获了不少，希望正在路上安装的不要放弃，搞清楚每一步做了什么，先后顺序。这个博客也是记录一下，有许多不足之处，只能靠各位同仁的努力，会成功的。

转载于:https://www.cnblogs.com/fly-luo/p/11159005.html

[https_www.jianshu.com_p_8d309dcc8b99]: https://www.jianshu.com/p/8d309dcc8b99
[https_www.cnblogs.com_whb-20160329_p_10256766.html]: https://www.cnblogs.com/whb-20160329/p/10256766.html
[https_blog.csdn.net_dl_chenbo_article_details_78687461]: https://blog.csdn.net/dl_chenbo/article/details/78687461/
[https_www.jianshu.com_p_c42fc8702f2b]: https://www.jianshu.com/p/c42fc8702f2b
[1595953-20190709142910876-733228789.png]: /images/20211116/2372661facec42158610092218d9d86a.png
[https_blog.csdn.net_CSDN330_article_details_86747867]: https://blog.csdn.net/CSDN330/article/details/86747867
[1595953-20190709143448383-1632913298.png]: /images/20211116/446c4709bd0c4c468e8bfbba4ee873b2.png
[1595953-20190709145336724-1835471299.png]: /images/20211116/1309fbfa84c0408aab9254e7656a2a80.png