首先，大家如果想省事情，请直接换成ubuntu16.04，网上很多配置教程很好用。
不然，有很多坑要填！
gcc和g++版本用5.0以上版本
python用原生的python，最好用python2
gflags，下载源码安装
opencv下载源码安装，参考我的另一篇博客：https://www.imooc.com/article/284786
/usr/local/lib/libopencv_imgcodecs.so：对‘TIFFClose@LIBTIFF_4.0’ 未定义的引用 /usr/local/lib/libopencv_imgcodecs.so：对‘TIFFRGBAImageOK@LIBTIFF_4.0’ 未定义的引用 /usr/local/lib/libopencv_imgcodecs.so：对‘TIFFReadRGBAStrip@LIBTIFF_4.0’ 未定义的引用 collect2: error: ld returned 1 exit statusmake: *** [.build_release/tools/convert_imageset.bin] 错误 1 解决：在cmake编译opencv时候加参数编译-DBUILD_TIFF=ON
5. boost源码安装：使用1.58版本的，安装前需要先删除和卸载相关的东西库
$ make [ 1%] Built target proto [ 80%] Built target caffe [ 82%] Linking CXX executable upgrade_solver_proto_text ../lib/libcaffe.so.1.0.0-rc3: undefined reference to `boost::re_detail_106300::cpp_regex_traits_implementation<char>::transform(char const*, char const*) const' ../lib/libcaffe.so.1.0.0-rc3: undefined reference to `boost::re_detail_106300::cpp_regex_traits_implementation<char>::transform_primary(char const*, char const*) const' collect2: error: ld returned 1 exit status make[2]: *** [tools/CMakeFiles/upgrade_solver_proto_text.dir/build.make:132: tools/upgrade_solver_proto_text] Error 1 make[1]: *** [CMakeFiles/Makefile2:442: tools/CMakeFiles/upgrade_solver_proto_text.dir/all] Error 2 make: *** [Makefile:128: all] Error 2
最后，建议大家还是用ubuntu16.04。配置环境是真的累！
随时随地看视频
