dlib开发环境部署

问题1

NOTE: Visual Studio didn't have good enough C++11 support until Visual Studio 2015 update 3 (v19.0.24210.0)

So we aren't enabling things that require full C++11 support (e.g. the deep learning tools).

导致无法使用GPU加速

Disabling CUDA support for dlib. DLIB WILL NOT USE CUDA

解决方案

最好使用Visual Studio 2015 update 3安装包，安装VS2015

否则在编译dlib库的时候，出错:

错误 C2535 “unknown-type dlib::gopt_impl::call_function_and_expand_args<max_unpack>::go(T &&,const dlib::matrix<double,0,1,dlib::default_memory_manager,dlib::row_major_layout> &)”: 已经定义或声明成员函数 (编译源文件dlib-master\dlib\svm\auto.cpp) dlib-master\dlib\global_optimization\find_max_global.h 79

错误 C2535 “unknown-type dlib::gopt_impl::call_function_and_expand_args<max_unpack>::go(T &&,const dlib::matrix<double,0,1,dlib::default_memory_manager,dlib::row_major_layout> &)”: 已经定义或声明成员函数 (编译源文件dlib-master\dlib\filtering\kalman_filter.cpp) dlib-master\dlib\global_optimization\find_max_global.h 77

因为VS2015的最初版本对C++11支持不是很好

解决方案

删除auto.cpp和kalman_filter.cpp

解决方案

更新到VS2017,导致CUDA对VS2017支持编译异常