training error about second.pytorch

[traveller59]

This looks like your protobuf version is old, use pip install -U protobuf to install a new version.

[zwyzwy]

I am installing the sparseconvnet and meet the problems below, I have already update the google package.

running build_ext
/home/users/wenyong.zheng/.local/lib/python3.6/site-packages/torch/utils/cpp_extension.py:118: UserWarning:

                           !! WARNING !!

!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
Your compiler (g++ 4.8) may be ABI-incompatible with PyTorch!
Please use a compiler that is ABI-compatible with GCC 4.9 and above.
See https://gcc.gnu.org/onlinedocs/libstdc++/manual/abi.html.

See https://gist.github.com/goldsborough/d466f43e8ffc948ff92de7486c5216d6
for instructions on how to install GCC 4.9 or higher.
!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!

                          !! WARNING !!

warnings.warn(ABI_INCOMPATIBILITY_WARNING.format(compiler))
building 'sparseconvnet.SCN' extension
creating build/temp.linux-x86_64-3.6
creating build/temp.linux-x86_64-3.6/sparseconvnet
creating build/temp.linux-x86_64-3.6/sparseconvnet/SCN
/usr/local/cuda/bin/nvcc -I/home/users/wenyong.zheng/.local/include -I/mnt/data-3/data/wenyong.zheng/vxlnet/second.pytorch/SparseConvNet/sparseconvnet/SCN/ -I/home/users/wenyong.zheng/.local/lib/python3.6/site-packages/torch/lib/include -I/home/users/wenyong.zheng/.local/lib/python3.6/site-packages/torch/lib/include/TH -I/home/users/wenyong.zheng/.local/lib/python3.6/site-packages/torch/lib/include/THC -I/usr/local/cuda/include -I/usr/include/python3.6m -c sparseconvnet/SCN/cuda.cu -o build/temp.linux-x86_64-3.6/sparseconvnet/SCN/cuda.o -std=c++11 -Xcompiler -fopenmp -DTORCH_EXTENSION_NAME=SCN -D_GLIBCXX_USE_CXX11_ABI=0 --compiler-options '-fPIC'
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).
In file included from sparseconvnet/SCN/cuda.cu:8:0:
/mnt/data-3/data/wenyong.zheng/vxlnet/second.pytorch/SparseConvNet/sparseconvnet/SCN/Metadata/Metadata.h:15:33: fatal error: google/dense_hash_map: No such file or directory
#include <google/dense_hash_map>
^
compilation terminated.
error: command '/usr/local/cuda/bin/nvcc' failed with exit status 1

[traveller59]

In my experience, the pip version of pytorch 0.4.1 use gcc 4.9.2 and conda 0.4.1 version use gcc 5. you need to use corresponding gcc version to compile SparseConvNet.

[zwyzwy]

when I update the gcc version to 4.9, errors occurred as below:

gcc -pthread -Wno-unused-result -Wsign-compare -DDYNAMIC_ANNOTATIONS_ENABLED=1 -DNDEBUG -O2 -g -pipe -Wall -Wp,-D_FORTIFY_SOURCE=2 -fexceptions -fstack-protector-strong --param=ssp-buffer-size=4 -grecord-gcc-switches -m64 -mtune=generic -D_GNU_SOURCE -fPIC -fwrapv -fPIC -I/home/users/wenyong.zheng/.local/include -I/home/users/wenyong.zheng/.local/lib/python3.6/site-packages/torch/lib/include/torch/csrc -I/mnt/data-3/data/wenyong.zheng/vxlnet/second.pytorch/SparseConvNet/sparseconvnet/SCN/ -I/home/users/wenyong.zheng/.local/lib/python3.6/site-packages/torch/lib/include -I/home/users/wenyong.zheng/.local/lib/python3.6/site-packages/torch/lib/include/TH -I/home/users/wenyong.zheng/.local/lib/python3.6/site-packages/torch/lib/include/THC -I/usr/local/cuda/include -I/usr/include/python3.6m -c sparseconvnet/SCN/sparseconvnet_cuda.cpp -o build/temp.linux-x86_64-3.6/sparseconvnet/SCN/sparseconvnet_cuda.o -std=c++11 -fopenmp -DTORCH_EXTENSION_NAME=SCN -D_GLIBCXX_USE_CXX11_ABI=0
sparseconvnet/SCN/sparseconvnet_cuda.cpp:14:29: fatal error: torch/extension.h: No such file or directory
#include <torch/extension.h>
^
compilation terminated.
error: command 'gcc' failed with exit status 1

it can't find extension.h on my computer, though I locate such file, nothing returns.

[traveller59]

I don't know how to solve this error... I think you can consider to switch to Anaconda environment (don't need root)
Is torch/extension.h in the -I/home/users/wenyong.zheng/.local/lib/python3.6/site-packages/torch/lib/include gcc argument?

[zwyzwy]

after I fix such error, it seems another error occurred:

In file included from sparseconvnet/SCN/sparseconvnet_cpu.cpp:14:0:
sparseconvnet/SCN/Metadata/Metadata.cpp: In member function 'std::vectorat::Tensor Metadata::compareSparseHelper(Metadata&, at::Tensor)':
sparseconvnet/SCN/Metadata/Metadata.cpp:591:66: error: no matching function for call to 'at::TypeExtendedInterface::tensor()'
at::Tensor cL_ = torch::CPU(at::kLong).tensor({(long)cL.size()});
^
In file included from /home/users/wenyong.zheng/.local/lib/python3.6/site-packages/torch/lib/include/ATen/Context.h:6:0,
from /home/users/wenyong.zheng/.local/lib/python3.6/site-packages/torch/lib/include/ATen/ATen.h:6,
from /home/users/wenyong.zheng/.local/lib/python3.6/site-packages/torch/lib/include/torch/csrc/utils/pybind.h:5,
from /home/users/wenyong.zheng/.local/lib/python3.6/site-packages/torch/lib/include/torch/extension.h:5,
from sparseconvnet/SCN/sparseconvnet_cpu.cpp:12:
/home/users/wenyong.zheng/.local/lib/python3.6/site-packages/torch/lib/include/ATen/TypeExtendedInterface.h:254:18: note: candidate: virtual at::Tensor at::TypeExtendedInterface::tensor(at::Storage, int64_t, at::IntList, at::IntList) const
virtual Tensor tensor(Storage storage, int64_t storageOffset, IntList size, IntList stride) const = 0;
^
/home/users/wenyong.zheng/.local/lib/python3.6/site-packages/torch/lib/include/ATen/TypeExtendedInterface.h:254:18: note: candidate expects 4 arguments, 1 provided
/home/users/wenyong.zheng/.local/lib/python3.6/site-packages/torch/lib/include/ATen/TypeExtendedInterface.h:255:18: note: candidate: virtual at::Tensor at::TypeExtendedInterface::tensor(at::IntList, at::IntList) const
virtual Tensor tensor(IntList size, IntList stride) const = 0;
^
/home/users/wenyong.zheng/.local/lib/python3.6/site-packages/torch/lib/include/ATen/TypeExtendedInterface.h:255:18: note: candidate expects 2 arguments, 1 provided
In file included from sparseconvnet/SCN/sparseconvnet_cpu.cpp:14:0:
sparseconvnet/SCN/Metadata/Metadata.cpp:593:66: error: no matching function for call to 'at::TypeExtendedInterface::tensor()'
at::Tensor cR_ = torch::CPU(at::kLong).tensor({(long)cR.size()});
^
In file included from /home/users/wenyong.zheng/.local/lib/python3.6/site-packages/torch/lib/include/ATen/Context.h:6:0,
from /home/users/wenyong.zheng/.local/lib/python3.6/site-packages/torch/lib/include/ATen/ATen.h:6,
from /home/users/wenyong.zheng/.local/lib/python3.6/site-packages/torch/lib/include/torch/csrc/utils/pybind.h:5,
from /home/users/wenyong.zheng/.local/lib/python3.6/site-packages/torch/lib/include/torch/extension.h:5,
from sparseconvnet/SCN/sparseconvnet_cpu.cpp:12:
/home/users/wenyong.zheng/.local/lib/python3.6/site-packages/torch/lib/include/ATen/TypeExtendedInterface.h:254:18: note: candidate: virtual at::Tensor at::TypeExtendedInterface::tensor(at::Storage, int64_t, at::IntList, at::IntList) const
virtual Tensor tensor(Storage storage, int64_t storageOffset, IntList size, IntList stride) const = 0;
^
/home/users/wenyong.zheng/.local/lib/python3.6/site-packages/torch/lib/include/ATen/TypeExtendedInterface.h:254:18: note: candidate expects 4 arguments, 1 provided
/home/users/wenyong.zheng/.local/lib/python3.6/site-packages/torch/lib/include/ATen/TypeExtendedInterface.h:255:18: note: candidate: virtual at::Tensor at::TypeExtendedInterface::tensor(at::IntList, at::IntList) const
virtual Tensor tensor(IntList size, IntList stride) const = 0;
^
/home/users/wenyong.zheng/.local/lib/python3.6/site-packages/torch/lib/include/ATen/TypeExtendedInterface.h:255:18: note: candidate expects 2 arguments, 1 provided
In file included from sparseconvnet/SCN/sparseconvnet_cpu.cpp:14:0:
sparseconvnet/SCN/Metadata/Metadata.cpp:595:64: error: no matching function for call to 'at::TypeExtendedInterface::tensor()'
at::Tensor L_ = torch::CPU(at::kLong).tensor({(long)L.size()});
^
In file included from /home/users/wenyong.zheng/.local/lib/python3.6/site-packages/torch/lib/include/ATen/Context.h:6:0,
from /home/users/wenyong.zheng/.local/lib/python3.6/site-packages/torch/lib/include/ATen/ATen.h:6,
from /home/users/wenyong.zheng/.local/lib/python3.6/site-packages/torch/lib/include/torch/csrc/utils/pybind.h:5,
from /home/users/wenyong.zheng/.local/lib/python3.6/site-packages/torch/lib/include/torch/extension.h:5,
from sparseconvnet/SCN/sparseconvnet_cpu.cpp:12:
/home/users/wenyong.zheng/.local/lib/python3.6/site-packages/torch/lib/include/ATen/TypeExtendedInterface.h:254:18: note: candidate: virtual at::Tensor at::TypeExtendedInterface::tensor(at::Storage, int64_t, at::IntList, at::IntList) const
virtual Tensor tensor(Storage storage, int64_t storageOffset, IntList size, IntList stride) const = 0;
^
/home/users/wenyong.zheng/.local/lib/python3.6/site-packages/torch/lib/include/ATen/TypeExtendedInterface.h:254:18: note: candidate expects 4 arguments, 1 provided
/home/users/wenyong.zheng/.local/lib/python3.6/site-packages/torch/lib/include/ATen/TypeExtendedInterface.h:255:18: note: candidate: virtual at::Tensor at::TypeExtendedInterface::tensor(at::IntList, at::IntList) const
virtual Tensor tensor(IntList size, IntList stride) const = 0;
^
/home/users/wenyong.zheng/.local/lib/python3.6/site-packages/torch/lib/include/ATen/TypeExtendedInterface.h:255:18: note: candidate expects 2 arguments, 1 provided
In file included from sparseconvnet/SCN/sparseconvnet_cpu.cpp:14:0:
sparseconvnet/SCN/Metadata/Metadata.cpp:597:64: error: no matching function for call to 'at::TypeExtendedInterface::tensor()'
at::Tensor R_ = torch::CPU(at::kLong).tensor({(long)R.size()});
^
In file included from /home/users/wenyong.zheng/.local/lib/python3.6/site-packages/torch/lib/include/ATen/Context.h:6:0,
from /home/users/wenyong.zheng/.local/lib/python3.6/site-packages/torch/lib/include/ATen/ATen.h:6,
from /home/users/wenyong.zheng/.local/lib/python3.6/site-packages/torch/lib/include/torch/csrc/utils/pybind.h:5,
from /home/users/wenyong.zheng/.local/lib/python3.6/site-packages/torch/lib/include/torch/extension.h:5,
from sparseconvnet/SCN/sparseconvnet_cpu.cpp:12:
/home/users/wenyong.zheng/.local/lib/python3.6/site-packages/torch/lib/include/ATen/TypeExtendedInterface.h:254:18: note: candidate: virtual at::Tensor at::TypeExtendedInterface::tensor(at::Storage, int64_t, at::IntList, at::IntList) const
virtual Tensor tensor(Storage storage, int64_t storageOffset, IntList size, IntList stride) const = 0;
^
/home/users/wenyong.zheng/.local/lib/python3.6/site-packages/torch/lib/include/ATen/TypeExtendedInterface.h:254:18: note: candidate expects 4 arguments, 1 provided
/home/users/wenyong.zheng/.local/lib/python3.6/site-packages/torch/lib/include/ATen/TypeExtendedInterface.h:255:18: note: candidate: virtual at::Tensor at::TypeExtendedInterface::tensor(at::IntList, at::IntList) const
virtual Tensor tensor(IntList size, IntList stride) const = 0;
^
/home/users/wenyong.zheng/.local/lib/python3.6/site-packages/torch/lib/include/ATen/TypeExtendedInterface.h:255:18: note: candidate expects 2 arguments, 1 provided
error: command 'gcc' failed with exit status 1

[lucasjinreal]

@zwyzwy I also got this error, it's caused by SparseNet no longer support pytorch 0.41-, it upgrade into pytorch 1.0, you should probably using pytorch 1.0 to support SparseNet, I don't know whether newest version of pytorch compatiable with this repo though