Please help me！I has an error when i do ca_cuda.cu

C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v8.0\bin\nvcc.exe -c E:\pytorch\shiyan\awesome-semantic-segmentation-pytorch\core\nn\csrc\cuda\ca_cuda.cu -o build\temp.win-amd64-3.6\Release\pytorch\shiyan\awesome-semantic-segmentation-pytorch\core\nn\csrc\cuda\ca_cuda.obj -IE:\pytorch\shiyan\awesome-semantic-segmentation-pytorch\core\nn\csrc -ID:\ProgramData\Anaconda2\envs\py3\lib\site-packages\torch\include -ID:\ProgramData\Anaconda2\envs\py3\lib\site-packages\torch\include\torch\csrc\api\include -ID:\ProgramData\Anaconda2\envs\py3\lib\site-packages\torch\include\TH -ID:\ProgramData\Anaconda2\envs\py3\lib\site-packages\torch\include\THC "-IC:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v8.0\include" -ID:\ProgramData\Anaconda2\envs\py3\include -ID:\ProgramData\Anaconda2\envs\py3\include "-ID:\Program Files\VS2015\VC\INCLUDE" "-ID:\Program Files\VS2015\VC\ATLMFC\INCLUDE" "-IC:\Program Files (x86)\Windows Kits\10\include\10.0.10240.0\ucrt" "-IC:\Program Files (x86)\Windows Kits\NETFXSDK\4.6.1\include\um" "-IC:\Program Files (x86)\Windows Kits\8.1\include\shared" "-IC:\Program Files (x86)\Windows Kits\8.1\include\um" "-IC:\Program Files (x86)\Windows Kits\8.1\include\winrt" -Xcompiler /EHsc -Xcompiler /wd4819 -Xcompiler /MD -DCUDA_NO_HALF_OPERATORS -DCUDA_NO_HALF_CONVERSIONS -DCUDA_NO_HALF2_OPERATORS__ --expt-relaxed-constexpr -DTORCH_API_INCLUDE_EXTENSION_H -DTORCH_EXTENSION_NAME=_C -D_GLIBCXX_USE_CXX11_ABI=0 nvcc warning : The 'compute_20', 'sm_20', and 'sm_21' architectures are deprecated, and may be removed in a future release (Use -Wno-deprecated-gpu-targets to suppress warning). ca_cuda.cu D:/ProgramData/Anaconda2/envs/py3/lib/site-packages/torch/include\c10/util/flat_hash_map.h(1334): error C2059: 语法错误:“” D:/ProgramData/Anaconda2/envs/py3/lib/site-packages/torch/include\c10/core/TensorTypeIdRegistration.h(49): note: 参见对正在编译的类模板实例化“ska::flat_hash_map<c10::TensorTypeId,std::basic_string<char,std::char_traits,std::allocator>,std::hash,std::equal_to,std::allocator<std::pair<K,V>>>”的引用 with [ K=c10::TensorTypeId, V=std::basic_string<char,std::char_traits,std::allocator> ] D:/ProgramData/Anaconda2/envs/py3/lib/site-packages/torch/include\c10/util/flat_hash_map.h(1350): error C2059: 语法错误:“” D:/ProgramData/Anaconda2/envs/py3/lib/site-packages/torch/include\c10/util/flat_hash_map.h(292): error C2516: “EntryAlloc”: 是非法基类 D:/ProgramData/Anaconda2/envs/py3/lib/site-packages/torch/include\c10/util/flat_hash_map.h(1350): note: 参见“EntryAlloc”的声明 D:/ProgramData/Anaconda2/envs/py3/lib/site-packages/torch/include\c10/util/flat_hash_map.h(1350): note: 参见对正在编译的类模板实例化“ska::detailv3::sherwood_v3_table<std::pair<K,V>,K,H,ska::detailv3::KeyOrValueHasher<K,std::pair<K,V>,H>,E,ska::detailv3::KeyOrValueEquality<K,std::pair<K,V>,E>,A,int>”的引用 with [ K=c10::TensorTypeId, V=std::basic_string<char,std::char_traits,std::allocator>, H=std::hash, E=std::equal_to, A=std::allocator<std::pair<c10::TensorTypeId,std::basic_string<char,std::char_traits,std::allocator>>> ] D:/Program Files/VS2015/VC/INCLUDE\xmemory0(556): error C2825: '_Alloc': 当后面跟“::”时必须为类或命名空间 D:/ProgramData/Anaconda2/envs/py3/lib/site-packages/torch/include\c10/util/flat_hash_map.h(296): note: 参见对正在编译的类模板实例化“std::allocator_traits”的引用 with [ EntryAlloc=int ] D:/Program Files/VS2015/VC/INCLUDE\xmemory0(556): error C2510: “_Alloc”:“::”的左边必须是类/结构/联合 D:/Program Files/VS2015/VC/INCLUDE\xmemory0(556): error C3646: “value_type”: 未知重写说明符 D:/Program Files/VS2015/VC/INCLUDE\xmemory0(556): error C4430: 缺少类型说明符 - 假定为 int。注意: C++ 不支持默认 int D:/Program Files/VS2015/VC/INCLUDE\xmemory0(314): error C2825: '_Ty': 当后面跟“::”时必须为类或命名空间 D:/Program Files/VS2015/VC/INCLUDE\xmemory0(558): note: 参见对正在编译的类模板实例化“std::_Get_pointer_type<_Alloc>”的引用 with [ _Alloc=int ] D:/Program Files/VS2015/VC/INCLUDE\xmemory0(314): error C2510: “_Ty”:“::”的左边必须是类/结构/联合 D:/Program Files/VS2015/VC/INCLUDE\xmemory0(314): error C2146: 语法错误: 缺少“>”(在标识符“value_type”的前面) D:/Program Files/VS2015/VC/INCLUDE\xmemory0(314): error C2672: “std::_Get_pointer_type<_Alloc>::_Fn”: 未找到匹配的重载函数 with [ _Alloc=int ] D:/Program Files/VS2015/VC/INCLUDE\xmemory0(314): error C2893: 未能使函数模板“std::_Identity std::_Get_pointer_type<_Alloc>::_Fn(std::_Wrap_int)”专用化 with [ _Alloc=int ] D:/Program Files/VS2015/VC/INCLUDE\xmemory0(314): note: 用下列模板参数: D:/Program Files/VS2015/VC/INCLUDE\xmemory0(314): note: “_Uty=int” D:/Program Files/VS2015/VC/INCLUDE\xmemory0(314): error C2770: “std::_Identity<_Uty::pointer> std::_Get_pointer_type<_Alloc>::_Fn(int)”的显式模板参数无效 with [ _Alloc=int ] D:/Program Files/VS2015/VC/INCLUDE\xmemory0(314): note: 参见“std::_Get_pointer_type<_Alloc>::_Fn”的声明 with [ _Alloc=int ] D:/Program Files/VS2015/VC/INCLUDE\xmemory0(320): error C2825: '_Ty': 当后面跟“::”时必须为类或命名空间 D:/Program Files/VS2015/VC/INCLUDE\xmemory0(560): note: 参见对正在编译的类模板实例化“std::_Get_const_pointer_type<_Alloc>”的引用 with [ _Alloc=int ] D:/Program Files/VS2015/VC/INCLUDE\xmemory0(320): error C2510: “_Ty”:“::”的左边必须是类/结构/联合 D:/Program Files/VS2015/VC/INCLUDE\xmemory0(320): error C4430: 缺少类型说明符 - 假定为 int。注意: C++ 不支持默认 int D:/Program Files/VS2015/VC/INCLUDE\xmemory0(320): error C2146: 语法错误: 缺少“>”(在标识符“value_type”的前面) D:/Program Files/VS2015/VC/INCLUDE\xmemory0(320): error C2672: “std::_Get_const_pointer_type<_Alloc>::_Fn”: 未找到匹配的重载函数 with [ _Alloc=int ] D:/Program Files/VS2015/VC/INCLUDE\xmemory0(320): error C2770: “std::_Identity<pointer_traits<_Get_pointer_type<_Alloc>::_Decltype::type>::rebind> std::_Get_const_pointer_type<_Alloc>::_Fn(std::_Wrap_int)”的显式模板参数无效 with [ _Alloc=int ] D:/Program Files/VS2015/VC/INCLUDE\xmemory0(320): note: 参见“std::_Get_const_pointer_type<_Alloc>::_Fn”的声明 with [ _Alloc=int ] D:/Program Files/VS2015/VC/INCLUDE\xmemory0(320): error C2770: “std::_Identity<_Uty::const_pointer> std::_Get_const_pointer_type<_Alloc>::_Fn(int)”的显式模板参数无效 with [ _Alloc=int ] D:/Program Files/VS2015/VC/INCLUDE\xmemory0(320): note: 参见“std::_Get_const_pointer_type<_Alloc>::_Fn”的声明 with [ _Alloc=int ] D:/Program Files/VS2015/VC/INCLUDE\xmemory0(327): error C2672: “std::_Get_void_pointer_type<_Alloc>::_Fn”: 未找到匹配的重载函数 with [ _Alloc=int ] D:/Program Files/VS2015/VC/INCLUDE\xmemory0(562): note: 参见对正在编译的类模板实例化“std::_Get_void_pointer_type<_Alloc>”的引用 with [ _Alloc=int ] D:/Program Files/VS2015/VC/INCLUDE\xmemory0(327): error C2770: “std::_Identity<pointer_traits<_Get_pointer_type<_Alloc>::_Decltype::type>::rebind> std::_Get_void_pointer_type<_Alloc>::_Fn(std::_Wrap_int)”的显式模板参数无效 with [ _Alloc=int ] D:/Program Files/VS2015/VC/INCLUDE\xmemory0(327): note: 参见“std::_Get_void_pointer_type<_Alloc>::_Fn”的声明 with [ _Alloc=int ] D:/Program Files/VS2015/VC/INCLUDE\xmemory0(327): error C2770: “std::_Identity<_Uty::void_pointer> std::_Get_void_pointer_type<_Alloc>::_Fn(int)”的显式模板参数无效 with [ _Alloc=int ] D:/Program Files/VS2015/VC/INCLUDE\xmemory0(327): note: 参见“std::_Get_void_pointer_type<_Alloc>::_Fn”的声明 with [ _Alloc=int ] D:/Program Files/VS2015/VC/INCLUDE\xmemory0(334): error C2672: “std::_Get_const_void_pointer_type<_Alloc>::_Fn”: 未找到匹配的重载函数 with [ _Alloc=int ] D:/Program Files/VS2015/VC/INCLUDE\xmemory0(564): note: 参见对正在编译的类模板实例化“std::_Get_const_void_pointer_type<_Alloc>”的引用 with [ _Alloc=int ] D:/Program Files/VS2015/VC/INCLUDE\xmemory0(334): error C2770: “std::_Identity<pointer_traits<_Get_pointer_type<_Alloc>::_Decltype::type>::rebind> std::_Get_const_void_pointer_type<_Alloc>::_Fn(std::_Wrap_int)”的显式模板参数无效 with [ _Alloc=int ] D:/Program Files/VS2015/VC/INCLUDE\xmemory0(334): note: 参见“std::_Get_const_void_pointer_type<_Alloc>::_Fn”的声明 with [ _Alloc=int ] D:/Program Files/VS2015/VC/INCLUDE\xmemory0(334): error C2770: “std::_Identity<_Uty::const_void_pointer> std::_Get_const_void_pointer_type<_Alloc>::_Fn(int)”的显式模板参数无效 with [ _Alloc=int ] D:/Program Files/VS2015/VC/INCLUDE\xmemory0(334): note: 参见“std::_Get_const_void_pointer_type<_Alloc>::_Fn”的声明 with [ _Alloc=int ] D:/Program Files/VS2015/VC/INCLUDE\xmemory0(341): error C2672: “std::_Get_difference_type<_Ty>::_Fn”: 未找到匹配的重载函数 with [ _Ty=int ] D:/Program Files/VS2015/VC/INCLUDE\xmemory0(348): note: 参见对正在编译的类模板实例化“std::_Get_difference_type<_Ty>”的引用 with [ _Ty=int ] D:/Program Files/VS2015/VC/INCLUDE\xmemory0(567): note: 参见对正在编译的类模板实例化“std::_Get_size_type<_Alloc>”的引用 with [ _Alloc=int ] D:/Program Files/VS2015/VC/INCLUDE\xmemory0(341): error C2770: “std::_Identity<_Get_ptr_difference_type<_Get_pointer_type<_Alloc>::_Decltype::type>::type> std::_Get_difference_type<_Ty>::_Fn(std::_Wrap_int)”的显式模板参数无效 with [ _Alloc=int, _Ty=int ] D:/Program Files/VS2015/VC/INCLUDE\xmemory0(341): note: 参见“std::_Get_difference_type<_Ty>::_Fn”的声明 with [ _Ty=int ] D:/Program Files/VS2015/VC/INCLUDE\xmemory0(341): error C2770: “std::_Identity<_Uty::difference_type> std::_Get_difference_type<_Ty>::_Fn(int)”的显式模板参数无效 with [ _Ty=int ] D:/Program Files/VS2015/VC/INCLUDE\xmemory0(341): note: 参见“std::_Get_difference_type<_Ty>::_Fn”的声明 with [ _Ty=int ] D:/Program Files/VS2015/VC/INCLUDE\xmemory0(348): error C2672: “std::_Get_size_type<_Alloc>::_Fn”: 未找到匹配的重载函数 with [ _Alloc=int ] D:/Program Files/VS2015/VC/INCLUDE\xmemory0(348): error C2770: “std::_Identity<make_unsigned<_Get_difference_type<_Ty>::_Decltype::type>::type> std::_Get_size_type<_Ty>::_Fn(std::_Wrap_int)”的显式模板参数无效 with [ _Ty=int ] D:/Program Files/VS2015/VC/INCLUDE\xmemory0(348): note: 参见“std::_Get_size_type<_Alloc>::_Fn”的声明 with [ _Alloc=int ] D:/Program Files/VS2015/VC/INCLUDE\xmemory0(348): error C2770: “std::_Identity<_Uty::size_type> std::_Get_size_type<_Alloc>::_Fn(int)”的显式模板参数无效 with [ _Alloc=int ] D:/Program Files/VS2015/VC/INCLUDE\xmemory0(348): note: 参见“std::_Get_size_type<_Alloc>::_Fn”的声明 with [ _Alloc=int ] D:/Program Files/VS2015/VC/INCLUDE\xmemory0(586): warning C4494: “std::allocator_traits::allocate”: 因为函数的返回类型不是指针或引用，将忽略 declspec(allocator) with [ EntryAlloc=int ] D:/Program Files/VS2015/VC/INCLUDE\xmemory0(592): warning C4494: “std::allocator_traits::allocate”: 因为函数的返回类型不是指针或引用，将忽略 __declspec(allocator) with [ EntryAlloc=int ] D:/ProgramData/Anaconda2/envs/py3/lib/site-packages/torch/include\c10/util/flat_hash_map.h(1350): error C2602: “ska::detailv3::sherwood_v3_table<std::pair<K,V>,K,H,ska::detailv3::KeyOrValueHasher<K,std::pair<K,V>,H>,E,ska::detailv3::KeyOrValueEquality<K,std::pair<K,V>,E>,A,int>::{ctor}”不是“ska::flat_hash_map<c10::TensorTypeId,std::basic_string<char,std::char_traits,std::allocator>,std::hash,std::equal_to,std::allocator<std::pair<K,V>>>”基类的成员 with [ K=c10::TensorTypeId, V=std::basic_string<char,std::char_traits,std::allocator>, H=std::hash, E=std::equal_to, A=std::allocator<std::pair<c10::TensorTypeId,std::basic_string<char,std::char_traits,std::allocator>>> ] and [ K=c10::TensorTypeId, V=std::basic_string<char,std::char_traits,std::allocator> ] D:/ProgramData/Anaconda2/envs/py3/lib/site-packages/torch/include\c10/core/TensorTypeIdRegistration.h(49): note: 参见“ska::flat_hash_map<c10::TensorTypeId,std::basic_string<char,std::char_traits,std::allocator>,std::hash,std::equal_to,std::allocator<std::pair<K,V>>>”的声明 with [ K=c10::TensorTypeId, V=std::basic_string<char,std::char_traits,std::allocator> ] D:/ProgramData/Anaconda2/envs/py3/lib/site-packages/torch/include\c10/util/flat_hash_map.h(1378): error C2059: 语法错误:“” D:/ProgramData/Anaconda2/envs/py3/lib/site-packages/torch/include\c10/util/flat_hash_map.h(1378): error C2602: “ska::detailv3::sherwood_v3_table<std::pair<K,V>,K,H,ska::detailv3::KeyOrValueHasher<K,std::pair<K,V>,H>,E,ska::detailv3::KeyOrValueEquality<K,std::pair<K,V>,E>,A,int>::emplace”不是“ska::flat_hash_map<c10::TensorTypeId,std::basic_string<char,std::char_traits,std::allocator>,std::hash,std::equal_to,std::allocator<std::pair<K,V>>>”基类的成员 with [ K=c10::TensorTypeId, V=std::basic_string<char,std::char_traits,std::allocator>, H=std::hash, E=std::equal_to, A=std::allocator<std::pair<c10::TensorTypeId,std::basic_string<char,std::char_traits,std::allocator>>> ] and [ K=c10::TensorTypeId, V=std::basic_string<char,std::char_traits,std::allocator> ] D:/ProgramData/Anaconda2/envs/py3/lib/site-packages/torch/include\c10/core/TensorTypeIdRegistration.h(49): note: 参见“ska::flat_hash_map<c10::TensorTypeId,std::basic_string<char,std::char_traits,std::allocator>,std::hash,std::equal_to,std::allocator<std::pair<K,V>>>”的声明 with [ K=c10::TensorTypeId, V=std::basic_string<char,std::char_traits,std::allocator> ] D:/ProgramData/Anaconda2/envs/py3/lib/site-packages/torch/include\c10/util/flat_hash_map.h(1379): error C2059: 语法错误:“” D:/ProgramData/Anaconda2/envs/py3/lib/site-packages/torch/include\c10/util/flat_hash_map.h(1383): error C2059: 语法错误:“” D:/ProgramData/Anaconda2/envs/py3/lib/site-packages/torch/include\c10/util/flat_hash_map.h(1391): error C2059: 语法错误:“” D:/ProgramData/Anaconda2/envs/py3/lib/site-packages/torch/include\c10/util/flat_hash_map.h(1399): error C2059: 语法错误:“” D:/ProgramData/Anaconda2/envs/py3/lib/site-packages/torch/include\c10/util/flat_hash_map.h(1400): error C2059: 语法错误:“” D:/ProgramData/Anaconda2/envs/py3/lib/site-packages/torch/include\c10/util/flat_hash_map.h(1404): error C2059: 语法错误:“” D:/ProgramData/Anaconda2/envs/py3/lib/site-packages/torch/include\c10/util/flat_hash_map.h(1405): error C2059: 语法错误:“” D:/ProgramData/Anaconda2/envs/py3/lib/site-packages/torch/include\THC/THCBlas.h(10): warning C4190: “THCudaBlas_Hdot”有指定的 C 链接，但返回了与 C 不兼容的 UDT“c10::Half” D:/ProgramData/Anaconda2/envs/py3/lib/site-packages/torch/include\c10/util/Half.h(324): note: 参见“c10::Half”的声明 D:/ProgramData/Anaconda2/envs/py3/lib/site-packages/torch/include\TH/generic/THStorage.h(45): warning C4190: “THHalfStorage_get”有指定的 C 链接，但返回了与 C 不兼容的 UDT“c10::Half” D:/ProgramData/Anaconda2/envs/py3/lib/site-packages/torch/include\c10/util/Half.h(324): note: 参见“c10::Half”的声明 D:/ProgramData/Anaconda2/envs/py3/lib/site-packages/torch/include\TH/generic/THStorage.h(45): warning C4190: “THBFloat16Storage_get”有指定的 C 链接，但返回了与 C 不兼容的 UDT“c10::BFloat16” D:/ProgramData/Anaconda2/envs/py3/lib/site-packages/torch/include\c10/util/BFloat16.h(48): note: 参见“c10::BFloat16”的声明 D:/ProgramData/Anaconda2/envs/py3/lib/site-packages/torch/include\THC/generic/THCStorage.h(26): warning C4190: “THCudaHalfStorage_get”有指定的 C 链接，但返回了与 C 不兼容的 UDT“c10::Half” D:/ProgramData/Anaconda2/envs/py3/lib/site-packages/torch/include\c10/util/Half.h(324): note: 参见“c10::Half”的声明 D:/ProgramData/Anaconda2/envs/py3/lib/site-packages/torch/include\THC/generic/THCStorage.h(26): warning C4190: “THCudaBFloat16Storage_get”有指定的 C 链接，但返回了与 C 不兼容的 UDT“c10::BFloat16” D:/ProgramData/Anaconda2/envs/py3/lib/site-packages/torch/include\c10/util/BFloat16.h(48): note: 参见“c10::BFloat16”的声明 D:/ProgramData/Anaconda2/envs/py3/lib/site-packages/torch/include\c10/util/flat_hash_map.h(1350): error C2602: “ska::detailv3::sherwood_v3_table<std::pair<K,V>,K,H,ska::detailv3::KeyOrValueHasher<K,std::pair<K,V>,H>,E,ska::detailv3::KeyOrValueEquality<K,std::pair<K,V>,E>,A,int>::{ctor}”不是“ska::flat_hash_map<c10::IValue,c10::IValue,c10::detail::DictKeyHash,c10::detail::DictKeyEqualTo,std::allocator<std::pair<K,V>>>”基类的成员 with [ K=c10::IValue, V=c10::IValue, H=c10::detail::DictKeyHash, E=c10::detail::DictKeyEqualTo, A=std::allocator<std::pair<c10::IValue,c10::IValue>> ] and [ K=c10::IValue, V=c10::IValue ] D:/ProgramData/Anaconda2/envs/py3/lib/site-packages/torch/include\ATen/core/Dict.h(40): note: 参见“ska::flat_hash_map<c10::IValue,c10::IValue,c10::detail::DictKeyHash,c10::detail::DictKeyEqualTo,std::allocator<std::pair<K,V>>>”的声明 with [ K=c10::IValue, V=c10::IValue ] D:/ProgramData/Anaconda2/envs/py3/lib/site-packages/torch/include\c10/util/flat_hash_map.h(1378): error C2602: “ska::detailv3::sherwood_v3_table<std::pair<K,V>,K,H,ska::detailv3::KeyOrValueHasher<K,std::pair<K,V>,H>,E,ska::detailv3::KeyOrValueEquality<K,std::pair<K,V>,E>,A,int>::emplace”不是“ska::flat_hash_map<c10::IValue,c10::IValue,c10::detail::DictKeyHash,c10::detail::DictKeyEqualTo,std::allocator<std::pair<K,V>>>”基类的成员 with [ K=c10::IValue, V=c10::IValue, H=c10::detail::DictKeyHash, E=c10::detail::DictKeyEqualTo, A=std::allocator<std::pair<c10::IValue,c10::IValue>> ] and [ K=c10::IValue, V=c10::IValue ] D:/ProgramData/Anaconda2/envs/py3/lib/site-packages/torch/include\ATen/core/Dict.h(40): note: 参见“ska::flat_hash_map<c10::IValue,c10::IValue,c10::detail::DictKeyHash,c10::detail::DictKeyEqualTo,std::allocator<std::pair<K,V>>>”的声明 with [ K=c10::IValue, V=c10::IValue ] D:/ProgramData/Anaconda2/envs/py3/lib/site-packages/torch/include\TH/generic/THTensor.h(123): warning C4190: “THHalfTensor_get0d”有指定的 C 链接，但返回了与 C 不兼容的 UDT“c10::Half” D:/ProgramData/Anaconda2/envs/py3/lib/site-packages/torch/include\c10/util/Half.h(324): note: 参见“c10::Half”的声明 D:/ProgramData/Anaconda2/envs/py3/lib/site-packages/torch/include\TH/generic/THTensor.h(124): warning C4190: “THHalfTensor_get1d”有指定的 C 链接，但返回了与 C 不兼容的 UDT“c10::Half” D:/ProgramData/Anaconda2/envs/py3/lib/site-packages/torch/include\c10/util/Half.h(324): note: 参见“c10::Half”的声明 D:/ProgramData/Anaconda2/envs/py3/lib/site-packages/torch/include\TH/generic/THTensor.h(125): warning C4190: “THHalfTensor_get2d”有指定的 C 链接，但返回了与 C 不兼容的 UDT“c10::Half” D:/ProgramData/Anaconda2/envs/py3/lib/site-packages/torch/include\c10/util/Half.h(324): note: 参见“c10::Half”的声明 D:/ProgramData/Anaconda2/envs/py3/lib/site-packages/torch/include\TH/generic/THTensor.h(126): warning C4190: “THHalfTensor_get3d”有指定的 C 链接，但返回了与 C 不兼容的 UDT“c10::Half” D:/ProgramData/Anaconda2/envs/py3/lib/site-packages/torch/include\c10/util/Half.h(324): note: 参见“c10::Half”的声明 D:/ProgramData/Anaconda2/envs/py3/lib/site-packages/torch/include\TH/generic/THTensor.h(127): warning C4190: “THHalfTensor_get4d”有指定的 C 链接，但返回了与 C 不兼容的 UDT“c10::Half” D:/ProgramData/Anaconda2/envs/py3/lib/site-packages/torch/include\c10/util/Half.h(324): note: 参见“c10::Half”的声明 D:/ProgramData/Anaconda2/envs/py3/lib/site-packages/torch/include\TH/generic/THTensor.h(123): warning C4190: “THBFloat16Tensor_get0d”有指定的 C 链接，但返回了与 C 不兼容的 UDT“c10::BFloat16” D:/ProgramData/Anaconda2/envs/py3/lib/site-packages/torch/include\c10/util/BFloat16.h(48): note: 参见“c10::BFloat16”的声明 D:/ProgramData/Anaconda2/envs/py3/lib/site-packages/torch/include\TH/generic/THTensor.h(124): warning C4190: “THBFloat16Tensor_get1d”有指定的 C 链接，但返回了与 C 不兼容的 UDT“c10::BFloat16” D:/ProgramData/Anaconda2/envs/py3/lib/site-packages/torch/include\c10/util/BFloat16.h(48): note: 参见“c10::BFloat16”的声明 D:/ProgramData/Anaconda2/envs/py3/lib/site-packages/torch/include\TH/generic/THTensor.h(125): warning C4190: “THBFloat16Tensor_get2d”有指定的 C 链接，但返回了与 C 不兼容的 UDT“c10::BFloat16” D:/ProgramData/Anaconda2/envs/py3/lib/site-packages/torch/include\c10/util/BFloat16.h(48): note: 参见“c10::BFloat16”的声明 D:/ProgramData/Anaconda2/envs/py3/lib/site-packages/torch/include\TH/generic/THTensor.h(126): warning C4190: “THBFloat16Tensor_get3d”有指定的 C 链接，但返回了与 C 不兼容的 UDT“c10::BFloat16” D:/ProgramData/Anaconda2/envs/py3/lib/site-packages/torch/include\c10/util/BFloat16.h(48): note: 参见“c10::BFloat16”的声明 D:/ProgramData/Anaconda2/envs/py3/lib/site-packages/torch/include\TH/generic/THTensor.h(127): warning C4190: “THBFloat16Tensor_get4d”有指定的 C 链接，但返回了与 C 不兼容的 UDT“c10::BFloat16” D:/ProgramData/Anaconda2/envs/py3/lib/site-packages/torch/include\c10/util/BFloat16.h(48): note: 参见“c10::BFloat16”的声明 D:/ProgramData/Anaconda2/envs/py3/lib/site-packages/torch/include\THC/generic/THCTensor.h(124): warning C4190: “THCudaHalfTensor_get0d”有指定的 C 链接，但返回了与 C 不兼容的 UDT“c10::Half” D:/ProgramData/Anaconda2/envs/py3/lib/site-packages/torch/include\c10/util/Half.h(324): note: 参见“c10::Half”的声明 D:/ProgramData/Anaconda2/envs/py3/lib/site-packages/torch/include\THC/generic/THCTensor.h(125): warning C4190: “THCudaHalfTensor_get1d”有指定的 C 链接，但返回了与 C 不兼容的 UDT“c10::Half” D:/ProgramData/Anaconda2/envs/py3/lib/site-packages/torch/include\c10/util/Half.h(324): note: 参见“c10::Half”的声明 D:/ProgramData/Anaconda2/envs/py3/lib/site-packages/torch/include\THC/generic/THCTensor.h(126): warning C4190: “THCudaHalfTensor_get2d”有指定的 C 链接，但返回了与 C 不兼容的 UDT“c10::Half” D:/ProgramData/Anaconda2/envs/py3/lib/site-packages/torch/include\c10/util/Half.h(324): note: 参见“c10::Half”的声明 D:/ProgramData/Anaconda2/envs/py3/lib/site-packages/torch/include\THC/generic/THCTensor.h(127): warning C4190: “THCudaHalfTensor_get3d”有指定的 C 链接，但返回了与 C 不兼容的 UDT“c10::Half” D:/ProgramData/Anaconda2/envs/py3/lib/site-packages/torch/include\c10/util/Half.h(324): note: 参见“c10::Half”的声明 D:/ProgramData/Anaconda2/envs/py3/lib/site-packages/torch/include\THC/generic/THCTensor.h(128): warning C4190: “THCudaHalfTensor_get4d”有指定的 C 链接，但返回了与 C 不兼容的 UDT“c10::Half” D:/ProgramData/Anaconda2/envs/py3/lib/site-packages/torch/include\c10/util/Half.h(324): note: 参见“c10::Half”的声明 D:/ProgramData/Anaconda2/envs/py3/lib/site-packages/torch/include\THC/generic/THCTensor.h(124): warning C4190: “THCudaBFloat16Tensor_get0d”有指定的 C 链接，但返回了与 C 不兼容的 UDT“c10::BFloat16” D:/ProgramData/Anaconda2/envs/py3/lib/site-packages/torch/include\c10/util/BFloat16.h(48): note: 参见“c10::BFloat16”的声明 D:/ProgramData/Anaconda2/envs/py3/lib/site-packages/torch/include\THC/generic/THCTensor.h(125): warning C4190: “THCudaBFloat16Tensor_get1d”有指定的 C 链接，但返回了与 C 不兼容的 UDT“c10::BFloat16” D:/ProgramData/Anaconda2/envs/py3/lib/site-packages/torch/include\c10/util/BFloat16.h(48): note: 参见“c10::BFloat16”的声明 D:/ProgramData/Anaconda2/envs/py3/lib/site-packages/torch/include\THC/generic/THCTensor.h(126): warning C4190: “THCudaBFloat16Tensor_get2d”有指定的 C 链接，但返回了与 C 不兼容的 UDT“c10::BFloat16” D:/ProgramData/Anaconda2/envs/py3/lib/site-packages/torch/include\c10/util/BFloat16.h(48): note: 参见“c10::BFloat16”的声明 D:/ProgramData/Anaconda2/envs/py3/lib/site-packages/torch/include\THC/generic/THCTensor.h(127): warning C4190: “THCudaBFloat16Tensor_get3d”有指定的 C 链接，但返回了与 C 不兼容的 UDT“c10::BFloat16” D:/ProgramData/Anaconda2/envs/py3/lib/site-packages/torch/include\c10/util/BFloat16.h(48): note: 参见“c10::BFloat16”的声明 D:/ProgramData/Anaconda2/envs/py3/lib/site-packages/torch/include\THC/generic/THCTensor.h(128): warning C4190: “THCudaBFloat16Tensor_get4d”有指定的 C 链接，但返回了与 C 不兼容的 UDT“c10::BFloat16” D:/ProgramData/Anaconda2/envs/py3/lib/site-packages/torch/include\c10/util/BFloat16.h(48): note: 参见“c10::BFloat16”的声明 D:/ProgramData/Anaconda2/envs/py3/lib/site-packages/torch/include\THC/generic/THCTensorMathReduce.h(16): warning C4190: “THCudaHalfTensor_minall”有指定的 C 链接，但返回了与 C 不兼容的 UDT“c10::Half” D:/ProgramData/Anaconda2/envs/py3/lib/site-packages/torch/include\c10/util/Half.h(324): note: 参见“c10::Half”的声明 D:/ProgramData/Anaconda2/envs/py3/lib/site-packages/torch/include\THC/generic/THCTensorMathReduce.h(17): warning C4190: “THCudaHalfTensor_maxall”有指定的 C 链接，但返回了与 C 不兼容的 UDT“c10::Half” D:/ProgramData/Anaconda2/envs/py3/lib/site-packages/torch/include\c10/util/Half.h(324): note: 参见“c10::Half”的声明 D:/ProgramData/Anaconda2/envs/py3/lib/site-packages/torch/include\THC/generic/THCTensorMathReduce.h(38): warning C4190: “THCudaHalfTensor_medianall”有指定的 C 链接，但返回了与 C 不兼容的 UDT“c10::Half” D:/ProgramData/Anaconda2/envs/py3/lib/site-packages/torch/include\c10/util/Half.h(324): note: 参见“c10::Half”的声明

Tramac / awesome-semantic-segmentation-pytorch

Please help me！I has an error when i do ca_cuda.cu #101