get error when run charRNN with vs2013update5

[kongjiellx]

1>..\..\..\example\charRNN.cpp(101): error C2661: 'mxnet::cpp::transpose' : no overloaded function takes 1 arguments
1>..\..\..\example\charRNN.cpp(149): warning C4267: 'initializing' : conversion from 'size_t' to 'int', possible loss of data
1>..\..\..\example\charRNN.cpp(153): warning C4267: '=' : conversion from 'size_t' to 'int', possible loss of data
1>..\..\..\example\charRNN.cpp(202): warning C4267: 'return' : conversion from 'size_t' to 'int', possible loss of data
1>..\..\..\example\charRNN.cpp(288): error C2664: 'std::tuple<std::unordered_map<wchar_t,mx_float,std::hash<wchar_t>,std::equal_to<_Kty>,std::allocator<std::pair<const _Kty,_Ty>>>,std::vector<wchar_t,std::allocator<wchar_t>>>::tuple(const std::tuple<std::unordered_map<_Kty,_Ty,std::hash<wchar_t>,std::equal_to<_Kty>,std::allocator<std::pair<const _Kty,_Ty>>>,std::vector<wchar_t,std::allocator<wchar_t>>> &)' : cannot convert argument 1 from 'std::unordered_map<wchar_t,mx_float,std::hash<wchar_t>,std::equal_to<_Kty>,std::allocator<std::pair<const _Kty,_Ty>>>' to 'std::allocator_arg_t'
1>          with
1>          [
1>              _Kty=wchar_t
1>  ,            _Ty=mx_float
1>          ]
1>          No user-defined-conversion operator available that can perform this conversion, or the operator cannot be called
1>..\..\..\example\charRNN.cpp(349): warning C4305: 'initializing' : truncation from 'double' to 'mx_float'
1>..\..\..\example\charRNN.cpp(372): warning C4305: 'argument' : truncation from 'double' to 'float'
1>..\..\..\example\charRNN.cpp(376): warning C4305: 'initializing' : truncation from 'double' to 'mx_float'
1>..\..\..\example\charRNN.cpp(377): warning C4305: 'initializing' : truncation from 'double' to 'mx_float'
========== Build: 0 succeeded, 1 failed, 0 up-to-date, 0 skipped ==========

maybe the transpose op should add a arg about axes? how can fixed it?

or maybe vs2015 don't have this problem...? but i can't find a transpose function with only one arg exactly

[kongjiellx]

and I'm new here, I fell confused about the differences about the MXNet.cpp's include file and mxnet source file. I can't find any opreation like conv or FullyConnected in https://github.com/dmlc/mxnet/tree/master/include/mxnet but MXNet.cpp have.

how can i start with mxnet's source code but not MXNet.cpp?

and how can i get the resnet.cpp's train data? i can't find "sf1_train.lst" in http://data.dmlc.ml/mxnet/data/

@hjk41 @zhangchen-qinyinghua @lx75249

[mz24cn]

transpose issue: it is a bug: https://github.com/dmlc/MXNet.cpp/pull/53 and https://github.com/dmlc/MXNet.cpp/pull/54. but, ... even the two issues were resolved, the code still cannot properly run for latest MXNet version after NNVM refactored. I am working on it.

[mz24cn]

I wanna know how you find file charRNN.cpp since it had been reverted by @lx75249. I cannot see it when clone the repository to local.

[kongjiellx]

@mz24cn you can find it here https://github.com/dmlc/MXNet.cpp/tree/06478fe1545bf1c500f38a597efa7ebe6c8c16ee/example