search

torch-points3d / torch-points-kernels

Pytorch kernels for spatial operations on point clouds
MIT License
96 stars 25 forks source link

Unable to build wheel for torch-points-kernels #69

Closed chaitjo closed 2 years ago

chaitjo commented 3 years ago

Hello @nicolas-chaulet, thank you for the library. I'm attempting to install the kernels (as well as torch-points3d) across multiple servers and am running into installation issues for the kernels. I am basically unable to build the wheel and it leads to an extremely long error trace.

I was wondering if you could assist me here or if my system configurations are just incompatible with the kernels?

System:

Before installation, I ensure that CUDA is configured properly and nvcc works.

$ export PATH=/usr/local/cuda/bin:$PATH
$ export CPATH=/usr/local/cuda/include:$CPATH
$ export LD_LIBRARY_PATH=/usr/local/cuda/lib64:$LD_LIBRARY_PATH

$ python -c "import torch; print(torch.version.cuda)"
11.0

$ nvcc --version
nvcc: NVIDIA (R) Cuda compiler driver
Copyright (c) 2005-2020 NVIDIA Corporation
Built on Wed_Jul_22_19:09:09_PDT_2020
Cuda compilation tools, release 11.0, V11.0.221
Build cuda_11.0_bu.TC445_37.28845127_0

When installing torch-points-kernels, I get the following error trace: (The error seems to be nvcc fatal : Unsupported gpu architecture compute_86)

$ pip3 install torch-points-kernels
Collecting torch-points-kernels
  Downloading torch-points-kernels-0.7.0.tar.gz (44 kB)
     |████████████████████████████████| 44 kB 7.3 MB/s 
Requirement already satisfied: torch>=1.1.0 in /home/user/miniconda3/envs/pointcloud/lib/python3.7/site-packages (from torch-points-kernels) (1.7.1)
Requirement already satisfied: numba in /home/user/miniconda3/envs/pointcloud/lib/python3.7/site-packages (from torch-points-kernels) (0.49.1)
Requirement already satisfied: numpy<1.20 in /home/user/miniconda3/envs/pointcloud/lib/python3.7/site-packages (from torch-points-kernels) (1.19.4)
Requirement already satisfied: scikit-learn in /home/user/miniconda3/envs/pointcloud/lib/python3.7/site-packages (from torch-points-kernels) (0.23.2)
Requirement already satisfied: typing_extensions in /home/user/miniconda3/envs/pointcloud/lib/python3.7/site-packages (from torch>=1.1.0->torch-points-kernels) (3.7.4.3)
Requirement already satisfied: llvmlite<=0.33.0.dev0,>=0.31.0.dev0 in /home/user/miniconda3/envs/pointcloud/lib/python3.7/site-packages (from numba->torch-points-kernels) (0.32.1)
Requirement already satisfied: setuptools in /home/user/miniconda3/envs/pointcloud/lib/python3.7/site-packages (from numba->torch-points-kernels) (52.0.0.post20210125)
Requirement already satisfied: scipy>=0.19.1 in /home/user/miniconda3/envs/pointcloud/lib/python3.7/site-packages (from scikit-learn->torch-points-kernels) (1.5.4)
Requirement already satisfied: threadpoolctl>=2.0.0 in /home/user/miniconda3/envs/pointcloud/lib/python3.7/site-packages (from scikit-learn->torch-points-kernels) (2.1.0)
Requirement already satisfied: joblib>=0.11 in /home/user/miniconda3/envs/pointcloud/lib/python3.7/site-packages (from scikit-learn->torch-points-kernels) (0.17.0)
Building wheels for collected packages: torch-points-kernels
  Building wheel for torch-points-kernels (setup.py) ... error
  ERROR: Command errored out with exit status 1:
   command: /home/user/miniconda3/envs/pointcloud/bin/python -u -c 'import sys, setuptools, tokenize; sys.argv[0] = '"'"'/tmp/pip-install-3twio4ga/torch-points-kernels_4d1bdf9ec06f4aa8984512eb395e074d/setup.py'"'"'; __file__='"'"'/tmp/pip-install-3twio4ga/torch-points-kernels_4d1bdf9ec06f4aa8984512eb395e074d/setup.py'"'"';f=getattr(tokenize, '"'"'open'"'"', open)(__file__);code=f.read().replace('"'"'\r\n'"'"', '"'"'\n'"'"');f.close();exec(compile(code, __file__, '"'"'exec'"'"'))' bdist_wheel -d /tmp/pip-wheel-mk73r6oo
       cwd: /tmp/pip-install-3twio4ga/torch-points-kernels_4d1bdf9ec06f4aa8984512eb395e074d/
  Complete output (126 lines):
  running bdist_wheel
  running build
  running build_py
  creating build
  creating build/lib.linux-x86_64-3.7
  creating build/lib.linux-x86_64-3.7/torch_points_kernels
  copying torch_points_kernels/knn.py -> build/lib.linux-x86_64-3.7/torch_points_kernels
  copying torch_points_kernels/cluster.py -> build/lib.linux-x86_64-3.7/torch_points_kernels
  copying torch_points_kernels/chamfer_dist.py -> build/lib.linux-x86_64-3.7/torch_points_kernels
  copying torch_points_kernels/gridding.py -> build/lib.linux-x86_64-3.7/torch_points_kernels
  copying torch_points_kernels/cubic_feature_sampling.py -> build/lib.linux-x86_64-3.7/torch_points_kernels
  copying torch_points_kernels/__init__.py -> build/lib.linux-x86_64-3.7/torch_points_kernels
  copying torch_points_kernels/torchpoints.py -> build/lib.linux-x86_64-3.7/torch_points_kernels
  copying torch_points_kernels/metrics.py -> build/lib.linux-x86_64-3.7/torch_points_kernels
  running build_ext
  building 'torch_points_kernels.points_cuda' extension
  creating build/temp.linux-x86_64-3.7
  creating build/temp.linux-x86_64-3.7/cuda
  creating build/temp.linux-x86_64-3.7/cuda/src
  gcc -pthread -B /home/user/miniconda3/envs/pointcloud/compiler_compat -Wl,--sysroot=/ -Wsign-compare -DNDEBUG -g -fwrapv -O3 -Wall -Wstrict-prototypes -fPIC -Icuda/include -I/home/user/miniconda3/envs/pointcloud/lib/python3.7/site-packages/torch/include -I/home/user/miniconda3/envs/pointcloud/lib/python3.7/site-packages/torch/include/torch/csrc/api/include -I/home/user/miniconda3/envs/pointcloud/lib/python3.7/site-packages/torch/include/TH -I/home/user/miniconda3/envs/pointcloud/lib/python3.7/site-packages/torch/include/THC -I/usr/local/cuda/include -I/home/user/miniconda3/envs/pointcloud/include/python3.7m -c cuda/src/interpolate.cpp -o build/temp.linux-x86_64-3.7/cuda/src/interpolate.o -O3 -DVERSION_GE_1_3 -DTORCH_API_INCLUDE_EXTENSION_H -DTORCH_API_INCLUDE_EXTENSION_H -DPYBIND11_COMPILER_TYPE="_gcc" -DPYBIND11_COMPILER_TYPE="_gcc" -DPYBIND11_STDLIB="_libstdcpp" -DPYBIND11_STDLIB="_libstdcpp" -DPYBIND11_BUILD_ABI="_cxxabi1011" -DPYBIND11_BUILD_ABI="_cxxabi1011" -DTORCH_EXTENSION_NAME=points_cuda -DTORCH_EXTENSION_NAME=points_cuda -D_GLIBCXX_USE_CXX11_ABI=0 -D_GLIBCXX_USE_CXX11_ABI=0 -std=c++14
  cc1plus: warning: command line option ‘-Wstrict-prototypes’ is valid for C/ObjC but not for C++
  In file included from /home/user/miniconda3/envs/pointcloud/lib/python3.7/site-packages/torch/include/ATen/Parallel.h:149:0,
                   from /home/user/miniconda3/envs/pointcloud/lib/python3.7/site-packages/torch/include/torch/csrc/api/include/torch/utils.h:3,
                   from /home/user/miniconda3/envs/pointcloud/lib/python3.7/site-packages/torch/include/torch/csrc/api/include/torch/nn/cloneable.h:5,
                   from /home/user/miniconda3/envs/pointcloud/lib/python3.7/site-packages/torch/include/torch/csrc/api/include/torch/nn.h:3,
                   from /home/user/miniconda3/envs/pointcloud/lib/python3.7/site-packages/torch/include/torch/csrc/api/include/torch/all.h:12,
                   from /home/user/miniconda3/envs/pointcloud/lib/python3.7/site-packages/torch/include/torch/extension.h:4,
                   from cuda/include/interpolate.h:3,
                   from cuda/src/interpolate.cpp:1:
  /home/user/miniconda3/envs/pointcloud/lib/python3.7/site-packages/torch/include/ATen/ParallelOpenMP.h:84:0: warning: ignoring #pragma omp parallel [-Wunknown-pragmas]
   #pragma omp parallel for if ((end - begin) >= grain_size)

  gcc -pthread -B /home/user/miniconda3/envs/pointcloud/compiler_compat -Wl,--sysroot=/ -Wsign-compare -DNDEBUG -g -fwrapv -O3 -Wall -Wstrict-prototypes -fPIC -Icuda/include -I/home/user/miniconda3/envs/pointcloud/lib/python3.7/site-packages/torch/include -I/home/user/miniconda3/envs/pointcloud/lib/python3.7/site-packages/torch/include/torch/csrc/api/include -I/home/user/miniconda3/envs/pointcloud/lib/python3.7/site-packages/torch/include/TH -I/home/user/miniconda3/envs/pointcloud/lib/python3.7/site-packages/torch/include/THC -I/usr/local/cuda/include -I/home/user/miniconda3/envs/pointcloud/include/python3.7m -c cuda/src/cubic_feature_sampling.cpp -o build/temp.linux-x86_64-3.7/cuda/src/cubic_feature_sampling.o -O3 -DVERSION_GE_1_3 -DTORCH_API_INCLUDE_EXTENSION_H -DTORCH_API_INCLUDE_EXTENSION_H -DPYBIND11_COMPILER_TYPE="_gcc" -DPYBIND11_COMPILER_TYPE="_gcc" -DPYBIND11_STDLIB="_libstdcpp" -DPYBIND11_STDLIB="_libstdcpp" -DPYBIND11_BUILD_ABI="_cxxabi1011" -DPYBIND11_BUILD_ABI="_cxxabi1011" -DTORCH_EXTENSION_NAME=points_cuda -DTORCH_EXTENSION_NAME=points_cuda -D_GLIBCXX_USE_CXX11_ABI=0 -D_GLIBCXX_USE_CXX11_ABI=0 -std=c++14
  cc1plus: warning: command line option ‘-Wstrict-prototypes’ is valid for C/ObjC but not for C++
  In file included from /home/user/miniconda3/envs/pointcloud/lib/python3.7/site-packages/torch/include/ATen/Parallel.h:149:0,
                   from /home/user/miniconda3/envs/pointcloud/lib/python3.7/site-packages/torch/include/torch/csrc/api/include/torch/utils.h:3,
                   from /home/user/miniconda3/envs/pointcloud/lib/python3.7/site-packages/torch/include/torch/csrc/api/include/torch/nn/cloneable.h:5,
                   from /home/user/miniconda3/envs/pointcloud/lib/python3.7/site-packages/torch/include/torch/csrc/api/include/torch/nn.h:3,
                   from /home/user/miniconda3/envs/pointcloud/lib/python3.7/site-packages/torch/include/torch/csrc/api/include/torch/all.h:12,
                   from /home/user/miniconda3/envs/pointcloud/lib/python3.7/site-packages/torch/include/torch/extension.h:4,
                   from cuda/include/cubic_feature_sampling.h:4,
                   from cuda/src/cubic_feature_sampling.cpp:1:
  /home/user/miniconda3/envs/pointcloud/lib/python3.7/site-packages/torch/include/ATen/ParallelOpenMP.h:84:0: warning: ignoring #pragma omp parallel [-Wunknown-pragmas]
   #pragma omp parallel for if ((end - begin) >= grain_size)

  gcc -pthread -B /home/user/miniconda3/envs/pointcloud/compiler_compat -Wl,--sysroot=/ -Wsign-compare -DNDEBUG -g -fwrapv -O3 -Wall -Wstrict-prototypes -fPIC -Icuda/include -I/home/user/miniconda3/envs/pointcloud/lib/python3.7/site-packages/torch/include -I/home/user/miniconda3/envs/pointcloud/lib/python3.7/site-packages/torch/include/torch/csrc/api/include -I/home/user/miniconda3/envs/pointcloud/lib/python3.7/site-packages/torch/include/TH -I/home/user/miniconda3/envs/pointcloud/lib/python3.7/site-packages/torch/include/THC -I/usr/local/cuda/include -I/home/user/miniconda3/envs/pointcloud/include/python3.7m -c cuda/src/ball_query.cpp -o build/temp.linux-x86_64-3.7/cuda/src/ball_query.o -O3 -DVERSION_GE_1_3 -DTORCH_API_INCLUDE_EXTENSION_H -DTORCH_API_INCLUDE_EXTENSION_H -DPYBIND11_COMPILER_TYPE="_gcc" -DPYBIND11_COMPILER_TYPE="_gcc" -DPYBIND11_STDLIB="_libstdcpp" -DPYBIND11_STDLIB="_libstdcpp" -DPYBIND11_BUILD_ABI="_cxxabi1011" -DPYBIND11_BUILD_ABI="_cxxabi1011" -DTORCH_EXTENSION_NAME=points_cuda -DTORCH_EXTENSION_NAME=points_cuda -D_GLIBCXX_USE_CXX11_ABI=0 -D_GLIBCXX_USE_CXX11_ABI=0 -std=c++14
  cc1plus: warning: command line option ‘-Wstrict-prototypes’ is valid for C/ObjC but not for C++
  In file included from /home/user/miniconda3/envs/pointcloud/lib/python3.7/site-packages/torch/include/ATen/Parallel.h:149:0,
                   from /home/user/miniconda3/envs/pointcloud/lib/python3.7/site-packages/torch/include/torch/csrc/api/include/torch/utils.h:3,
                   from /home/user/miniconda3/envs/pointcloud/lib/python3.7/site-packages/torch/include/torch/csrc/api/include/torch/nn/cloneable.h:5,
                   from /home/user/miniconda3/envs/pointcloud/lib/python3.7/site-packages/torch/include/torch/csrc/api/include/torch/nn.h:3,
                   from /home/user/miniconda3/envs/pointcloud/lib/python3.7/site-packages/torch/include/torch/csrc/api/include/torch/all.h:12,
                   from /home/user/miniconda3/envs/pointcloud/lib/python3.7/site-packages/torch/include/torch/extension.h:4,
                   from cuda/include/ball_query.h:2,
                   from cuda/src/ball_query.cpp:1:
  /home/user/miniconda3/envs/pointcloud/lib/python3.7/site-packages/torch/include/ATen/ParallelOpenMP.h:84:0: warning: ignoring #pragma omp parallel [-Wunknown-pragmas]
   #pragma omp parallel for if ((end - begin) >= grain_size)

  gcc -pthread -B /home/user/miniconda3/envs/pointcloud/compiler_compat -Wl,--sysroot=/ -Wsign-compare -DNDEBUG -g -fwrapv -O3 -Wall -Wstrict-prototypes -fPIC -Icuda/include -I/home/user/miniconda3/envs/pointcloud/lib/python3.7/site-packages/torch/include -I/home/user/miniconda3/envs/pointcloud/lib/python3.7/site-packages/torch/include/torch/csrc/api/include -I/home/user/miniconda3/envs/pointcloud/lib/python3.7/site-packages/torch/include/TH -I/home/user/miniconda3/envs/pointcloud/lib/python3.7/site-packages/torch/include/THC -I/usr/local/cuda/include -I/home/user/miniconda3/envs/pointcloud/include/python3.7m -c cuda/src/bindings.cpp -o build/temp.linux-x86_64-3.7/cuda/src/bindings.o -O3 -DVERSION_GE_1_3 -DTORCH_API_INCLUDE_EXTENSION_H -DTORCH_API_INCLUDE_EXTENSION_H -DPYBIND11_COMPILER_TYPE="_gcc" -DPYBIND11_COMPILER_TYPE="_gcc" -DPYBIND11_STDLIB="_libstdcpp" -DPYBIND11_STDLIB="_libstdcpp" -DPYBIND11_BUILD_ABI="_cxxabi1011" -DPYBIND11_BUILD_ABI="_cxxabi1011" -DTORCH_EXTENSION_NAME=points_cuda -DTORCH_EXTENSION_NAME=points_cuda -D_GLIBCXX_USE_CXX11_ABI=0 -D_GLIBCXX_USE_CXX11_ABI=0 -std=c++14
  cc1plus: warning: command line option ‘-Wstrict-prototypes’ is valid for C/ObjC but not for C++
  In file included from /home/user/miniconda3/envs/pointcloud/lib/python3.7/site-packages/torch/include/ATen/Parallel.h:149:0,
                   from /home/user/miniconda3/envs/pointcloud/lib/python3.7/site-packages/torch/include/torch/csrc/api/include/torch/utils.h:3,
                   from /home/user/miniconda3/envs/pointcloud/lib/python3.7/site-packages/torch/include/torch/csrc/api/include/torch/nn/cloneable.h:5,
                   from /home/user/miniconda3/envs/pointcloud/lib/python3.7/site-packages/torch/include/torch/csrc/api/include/torch/nn.h:3,
                   from /home/user/miniconda3/envs/pointcloud/lib/python3.7/site-packages/torch/include/torch/csrc/api/include/torch/all.h:12,
                   from /home/user/miniconda3/envs/pointcloud/lib/python3.7/site-packages/torch/include/torch/extension.h:4,
                   from cuda/include/ball_query.h:2,
                   from cuda/src/bindings.cpp:1:
  /home/user/miniconda3/envs/pointcloud/lib/python3.7/site-packages/torch/include/ATen/ParallelOpenMP.h:84:0: warning: ignoring #pragma omp parallel [-Wunknown-pragmas]
   #pragma omp parallel for if ((end - begin) >= grain_size)

  gcc -pthread -B /home/user/miniconda3/envs/pointcloud/compiler_compat -Wl,--sysroot=/ -Wsign-compare -DNDEBUG -g -fwrapv -O3 -Wall -Wstrict-prototypes -fPIC -Icuda/include -I/home/user/miniconda3/envs/pointcloud/lib/python3.7/site-packages/torch/include -I/home/user/miniconda3/envs/pointcloud/lib/python3.7/site-packages/torch/include/torch/csrc/api/include -I/home/user/miniconda3/envs/pointcloud/lib/python3.7/site-packages/torch/include/TH -I/home/user/miniconda3/envs/pointcloud/lib/python3.7/site-packages/torch/include/THC -I/usr/local/cuda/include -I/home/user/miniconda3/envs/pointcloud/include/python3.7m -c cuda/src/sampling.cpp -o build/temp.linux-x86_64-3.7/cuda/src/sampling.o -O3 -DVERSION_GE_1_3 -DTORCH_API_INCLUDE_EXTENSION_H -DTORCH_API_INCLUDE_EXTENSION_H -DPYBIND11_COMPILER_TYPE="_gcc" -DPYBIND11_COMPILER_TYPE="_gcc" -DPYBIND11_STDLIB="_libstdcpp" -DPYBIND11_STDLIB="_libstdcpp" -DPYBIND11_BUILD_ABI="_cxxabi1011" -DPYBIND11_BUILD_ABI="_cxxabi1011" -DTORCH_EXTENSION_NAME=points_cuda -DTORCH_EXTENSION_NAME=points_cuda -D_GLIBCXX_USE_CXX11_ABI=0 -D_GLIBCXX_USE_CXX11_ABI=0 -std=c++14
  cc1plus: warning: command line option ‘-Wstrict-prototypes’ is valid for C/ObjC but not for C++
  In file included from /home/user/miniconda3/envs/pointcloud/lib/python3.7/site-packages/torch/include/ATen/Parallel.h:149:0,
                   from /home/user/miniconda3/envs/pointcloud/lib/python3.7/site-packages/torch/include/torch/csrc/api/include/torch/utils.h:3,
                   from /home/user/miniconda3/envs/pointcloud/lib/python3.7/site-packages/torch/include/torch/csrc/api/include/torch/nn/cloneable.h:5,
                   from /home/user/miniconda3/envs/pointcloud/lib/python3.7/site-packages/torch/include/torch/csrc/api/include/torch/nn.h:3,
                   from /home/user/miniconda3/envs/pointcloud/lib/python3.7/site-packages/torch/include/torch/csrc/api/include/torch/all.h:12,
                   from /home/user/miniconda3/envs/pointcloud/lib/python3.7/site-packages/torch/include/torch/extension.h:4,
                   from cuda/include/sampling.h:2,
                   from cuda/src/sampling.cpp:1:
  /home/user/miniconda3/envs/pointcloud/lib/python3.7/site-packages/torch/include/ATen/ParallelOpenMP.h:84:0: warning: ignoring #pragma omp parallel [-Wunknown-pragmas]
   #pragma omp parallel for if ((end - begin) >= grain_size)

  gcc -pthread -B /home/user/miniconda3/envs/pointcloud/compiler_compat -Wl,--sysroot=/ -Wsign-compare -DNDEBUG -g -fwrapv -O3 -Wall -Wstrict-prototypes -fPIC -Icuda/include -I/home/user/miniconda3/envs/pointcloud/lib/python3.7/site-packages/torch/include -I/home/user/miniconda3/envs/pointcloud/lib/python3.7/site-packages/torch/include/torch/csrc/api/include -I/home/user/miniconda3/envs/pointcloud/lib/python3.7/site-packages/torch/include/TH -I/home/user/miniconda3/envs/pointcloud/lib/python3.7/site-packages/torch/include/THC -I/usr/local/cuda/include -I/home/user/miniconda3/envs/pointcloud/include/python3.7m -c cuda/src/chamfer_dist.cpp -o build/temp.linux-x86_64-3.7/cuda/src/chamfer_dist.o -O3 -DVERSION_GE_1_3 -DTORCH_API_INCLUDE_EXTENSION_H -DTORCH_API_INCLUDE_EXTENSION_H -DPYBIND11_COMPILER_TYPE="_gcc" -DPYBIND11_COMPILER_TYPE="_gcc" -DPYBIND11_STDLIB="_libstdcpp" -DPYBIND11_STDLIB="_libstdcpp" -DPYBIND11_BUILD_ABI="_cxxabi1011" -DPYBIND11_BUILD_ABI="_cxxabi1011" -DTORCH_EXTENSION_NAME=points_cuda -DTORCH_EXTENSION_NAME=points_cuda -D_GLIBCXX_USE_CXX11_ABI=0 -D_GLIBCXX_USE_CXX11_ABI=0 -std=c++14
  cc1plus: warning: command line option ‘-Wstrict-prototypes’ is valid for C/ObjC but not for C++
  In file included from /home/user/miniconda3/envs/pointcloud/lib/python3.7/site-packages/torch/include/ATen/Parallel.h:149:0,
                   from /home/user/miniconda3/envs/pointcloud/lib/python3.7/site-packages/torch/include/torch/csrc/api/include/torch/utils.h:3,
                   from /home/user/miniconda3/envs/pointcloud/lib/python3.7/site-packages/torch/include/torch/csrc/api/include/torch/nn/cloneable.h:5,
                   from /home/user/miniconda3/envs/pointcloud/lib/python3.7/site-packages/torch/include/torch/csrc/api/include/torch/nn.h:3,
                   from /home/user/miniconda3/envs/pointcloud/lib/python3.7/site-packages/torch/include/torch/csrc/api/include/torch/all.h:12,
                   from /home/user/miniconda3/envs/pointcloud/lib/python3.7/site-packages/torch/include/torch/extension.h:4,
                   from cuda/include/chamfer_dist.h:1,
                   from cuda/src/chamfer_dist.cpp:1:
  /home/user/miniconda3/envs/pointcloud/lib/python3.7/site-packages/torch/include/ATen/ParallelOpenMP.h:84:0: warning: ignoring #pragma omp parallel [-Wunknown-pragmas]
   #pragma omp parallel for if ((end - begin) >= grain_size)

  gcc -pthread -B /home/user/miniconda3/envs/pointcloud/compiler_compat -Wl,--sysroot=/ -Wsign-compare -DNDEBUG -g -fwrapv -O3 -Wall -Wstrict-prototypes -fPIC -Icuda/include -I/home/user/miniconda3/envs/pointcloud/lib/python3.7/site-packages/torch/include -I/home/user/miniconda3/envs/pointcloud/lib/python3.7/site-packages/torch/include/torch/csrc/api/include -I/home/user/miniconda3/envs/pointcloud/lib/python3.7/site-packages/torch/include/TH -I/home/user/miniconda3/envs/pointcloud/lib/python3.7/site-packages/torch/include/THC -I/usr/local/cuda/include -I/home/user/miniconda3/envs/pointcloud/include/python3.7m -c cuda/src/metrics.cpp -o build/temp.linux-x86_64-3.7/cuda/src/metrics.o -O3 -DVERSION_GE_1_3 -DTORCH_API_INCLUDE_EXTENSION_H -DTORCH_API_INCLUDE_EXTENSION_H -DPYBIND11_COMPILER_TYPE="_gcc" -DPYBIND11_COMPILER_TYPE="_gcc" -DPYBIND11_STDLIB="_libstdcpp" -DPYBIND11_STDLIB="_libstdcpp" -DPYBIND11_BUILD_ABI="_cxxabi1011" -DPYBIND11_BUILD_ABI="_cxxabi1011" -DTORCH_EXTENSION_NAME=points_cuda -DTORCH_EXTENSION_NAME=points_cuda -D_GLIBCXX_USE_CXX11_ABI=0 -D_GLIBCXX_USE_CXX11_ABI=0 -std=c++14
  cc1plus: warning: command line option ‘-Wstrict-prototypes’ is valid for C/ObjC but not for C++
  In file included from /home/user/miniconda3/envs/pointcloud/lib/python3.7/site-packages/torch/include/ATen/Parallel.h:149:0,
                   from /home/user/miniconda3/envs/pointcloud/lib/python3.7/site-packages/torch/include/torch/csrc/api/include/torch/utils.h:3,
                   from /home/user/miniconda3/envs/pointcloud/lib/python3.7/site-packages/torch/include/torch/csrc/api/include/torch/nn/cloneable.h:5,
                   from /home/user/miniconda3/envs/pointcloud/lib/python3.7/site-packages/torch/include/torch/csrc/api/include/torch/nn.h:3,
                   from /home/user/miniconda3/envs/pointcloud/lib/python3.7/site-packages/torch/include/torch/csrc/api/include/torch/all.h:12,
                   from /home/user/miniconda3/envs/pointcloud/lib/python3.7/site-packages/torch/include/torch/extension.h:4,
                   from cuda/include/metrics.h:2,
                   from cuda/src/metrics.cpp:1:
  /home/user/miniconda3/envs/pointcloud/lib/python3.7/site-packages/torch/include/ATen/ParallelOpenMP.h:84:0: warning: ignoring #pragma omp parallel [-Wunknown-pragmas]
   #pragma omp parallel for if ((end - begin) >= grain_size)

  gcc -pthread -B /home/user/miniconda3/envs/pointcloud/compiler_compat -Wl,--sysroot=/ -Wsign-compare -DNDEBUG -g -fwrapv -O3 -Wall -Wstrict-prototypes -fPIC -Icuda/include -I/home/user/miniconda3/envs/pointcloud/lib/python3.7/site-packages/torch/include -I/home/user/miniconda3/envs/pointcloud/lib/python3.7/site-packages/torch/include/torch/csrc/api/include -I/home/user/miniconda3/envs/pointcloud/lib/python3.7/site-packages/torch/include/TH -I/home/user/miniconda3/envs/pointcloud/lib/python3.7/site-packages/torch/include/THC -I/usr/local/cuda/include -I/home/user/miniconda3/envs/pointcloud/include/python3.7m -c cuda/src/gridding.cpp -o build/temp.linux-x86_64-3.7/cuda/src/gridding.o -O3 -DVERSION_GE_1_3 -DTORCH_API_INCLUDE_EXTENSION_H -DTORCH_API_INCLUDE_EXTENSION_H -DPYBIND11_COMPILER_TYPE="_gcc" -DPYBIND11_COMPILER_TYPE="_gcc" -DPYBIND11_STDLIB="_libstdcpp" -DPYBIND11_STDLIB="_libstdcpp" -DPYBIND11_BUILD_ABI="_cxxabi1011" -DPYBIND11_BUILD_ABI="_cxxabi1011" -DTORCH_EXTENSION_NAME=points_cuda -DTORCH_EXTENSION_NAME=points_cuda -D_GLIBCXX_USE_CXX11_ABI=0 -D_GLIBCXX_USE_CXX11_ABI=0 -std=c++14
  cc1plus: warning: command line option ‘-Wstrict-prototypes’ is valid for C/ObjC but not for C++
  In file included from /home/user/miniconda3/envs/pointcloud/lib/python3.7/site-packages/torch/include/ATen/Parallel.h:149:0,
                   from /home/user/miniconda3/envs/pointcloud/lib/python3.7/site-packages/torch/include/torch/csrc/api/include/torch/utils.h:3,
                   from /home/user/miniconda3/envs/pointcloud/lib/python3.7/site-packages/torch/include/torch/csrc/api/include/torch/nn/cloneable.h:5,
                   from /home/user/miniconda3/envs/pointcloud/lib/python3.7/site-packages/torch/include/torch/csrc/api/include/torch/nn.h:3,
                   from /home/user/miniconda3/envs/pointcloud/lib/python3.7/site-packages/torch/include/torch/csrc/api/include/torch/all.h:12,
                   from /home/user/miniconda3/envs/pointcloud/lib/python3.7/site-packages/torch/include/torch/extension.h:4,
                   from cuda/include/gridding.h:4,
                   from cuda/src/gridding.cpp:1:
  /home/user/miniconda3/envs/pointcloud/lib/python3.7/site-packages/torch/include/ATen/ParallelOpenMP.h:84:0: warning: ignoring #pragma omp parallel [-Wunknown-pragmas]
   #pragma omp parallel for if ((end - begin) >= grain_size)

  /usr/local/cuda/bin/nvcc -Icuda/include -I/home/user/miniconda3/envs/pointcloud/lib/python3.7/site-packages/torch/include -I/home/user/miniconda3/envs/pointcloud/lib/python3.7/site-packages/torch/include/torch/csrc/api/include -I/home/user/miniconda3/envs/pointcloud/lib/python3.7/site-packages/torch/include/TH -I/home/user/miniconda3/envs/pointcloud/lib/python3.7/site-packages/torch/include/THC -I/usr/local/cuda/include -I/home/user/miniconda3/envs/pointcloud/include/python3.7m -c cuda/src/chamfer_dist_gpu.cu -o build/temp.linux-x86_64-3.7/cuda/src/chamfer_dist_gpu.o -D__CUDA_NO_HALF_OPERATORS__ -D__CUDA_NO_HALF_CONVERSIONS__ -D__CUDA_NO_HALF2_OPERATORS__ --expt-relaxed-constexpr --compiler-options '-fPIC' -O3 -DVERSION_GE_1_3 -DTORCH_API_INCLUDE_EXTENSION_H -DTORCH_API_INCLUDE_EXTENSION_H -DPYBIND11_COMPILER_TYPE="_gcc" -DPYBIND11_COMPILER_TYPE="_gcc" -DPYBIND11_STDLIB="_libstdcpp" -DPYBIND11_STDLIB="_libstdcpp" -DPYBIND11_BUILD_ABI="_cxxabi1011" -DPYBIND11_BUILD_ABI="_cxxabi1011" -DTORCH_EXTENSION_NAME=points_cuda -DTORCH_EXTENSION_NAME=points_cuda -D_GLIBCXX_USE_CXX11_ABI=0 -D_GLIBCXX_USE_CXX11_ABI=0 -gencode=arch=compute_86,code=sm_86 -std=c++14
  nvcc fatal   : Unsupported gpu architecture 'compute_86'
  error: command '/usr/local/cuda/bin/nvcc' failed with exit status 1
  ----------------------------------------
  ERROR: Failed building wheel for torch-points-kernels
  Running setup.py clean for torch-points-kernels
Failed to build torch-points-kernels
Installing collected packages: torch-points-kernels
    Running setup.py install for torch-points-kernels ... error
    ERROR: Command errored out with exit status 1:
     command: /home/user/miniconda3/envs/pointcloud/bin/python -u -c 'import sys, setuptools, tokenize; sys.argv[0] = '"'"'/tmp/pip-install-3twio4ga/torch-points-kernels_4d1bdf9ec06f4aa8984512eb395e074d/setup.py'"'"'; __file__='"'"'/tmp/pip-install-3twio4ga/torch-points-kernels_4d1bdf9ec06f4aa8984512eb395e074d/setup.py'"'"';f=getattr(tokenize, '"'"'open'"'"', open)(__file__);code=f.read().replace('"'"'\r\n'"'"', '"'"'\n'"'"');f.close();exec(compile(code, __file__, '"'"'exec'"'"'))' install --record /tmp/pip-record-clyu4km7/install-record.txt --single-version-externally-managed --compile --install-headers /home/user/miniconda3/envs/pointcloud/include/python3.7m/torch-points-kernels
         cwd: /tmp/pip-install-3twio4ga/torch-points-kernels_4d1bdf9ec06f4aa8984512eb395e074d/
    Complete output (126 lines):
    running install
    running build
    running build_py
    creating build
    creating build/lib.linux-x86_64-3.7
    creating build/lib.linux-x86_64-3.7/torch_points_kernels
    copying torch_points_kernels/knn.py -> build/lib.linux-x86_64-3.7/torch_points_kernels
    copying torch_points_kernels/cluster.py -> build/lib.linux-x86_64-3.7/torch_points_kernels
    copying torch_points_kernels/chamfer_dist.py -> build/lib.linux-x86_64-3.7/torch_points_kernels
    copying torch_points_kernels/gridding.py -> build/lib.linux-x86_64-3.7/torch_points_kernels
    copying torch_points_kernels/cubic_feature_sampling.py -> build/lib.linux-x86_64-3.7/torch_points_kernels
    copying torch_points_kernels/__init__.py -> build/lib.linux-x86_64-3.7/torch_points_kernels
    copying torch_points_kernels/torchpoints.py -> build/lib.linux-x86_64-3.7/torch_points_kernels
    copying torch_points_kernels/metrics.py -> build/lib.linux-x86_64-3.7/torch_points_kernels
    running build_ext
    building 'torch_points_kernels.points_cuda' extension
    creating build/temp.linux-x86_64-3.7
    creating build/temp.linux-x86_64-3.7/cuda
    creating build/temp.linux-x86_64-3.7/cuda/src
    gcc -pthread -B /home/user/miniconda3/envs/pointcloud/compiler_compat -Wl,--sysroot=/ -Wsign-compare -DNDEBUG -g -fwrapv -O3 -Wall -Wstrict-prototypes -fPIC -Icuda/include -I/home/user/miniconda3/envs/pointcloud/lib/python3.7/site-packages/torch/include -I/home/user/miniconda3/envs/pointcloud/lib/python3.7/site-packages/torch/include/torch/csrc/api/include -I/home/user/miniconda3/envs/pointcloud/lib/python3.7/site-packages/torch/include/TH -I/home/user/miniconda3/envs/pointcloud/lib/python3.7/site-packages/torch/include/THC -I/usr/local/cuda/include -I/home/user/miniconda3/envs/pointcloud/include/python3.7m -c cuda/src/interpolate.cpp -o build/temp.linux-x86_64-3.7/cuda/src/interpolate.o -O3 -DVERSION_GE_1_3 -DTORCH_API_INCLUDE_EXTENSION_H -DTORCH_API_INCLUDE_EXTENSION_H -DPYBIND11_COMPILER_TYPE="_gcc" -DPYBIND11_COMPILER_TYPE="_gcc" -DPYBIND11_STDLIB="_libstdcpp" -DPYBIND11_STDLIB="_libstdcpp" -DPYBIND11_BUILD_ABI="_cxxabi1011" -DPYBIND11_BUILD_ABI="_cxxabi1011" -DTORCH_EXTENSION_NAME=points_cuda -DTORCH_EXTENSION_NAME=points_cuda -D_GLIBCXX_USE_CXX11_ABI=0 -D_GLIBCXX_USE_CXX11_ABI=0 -std=c++14
    cc1plus: warning: command line option ‘-Wstrict-prototypes’ is valid for C/ObjC but not for C++
    In file included from /home/user/miniconda3/envs/pointcloud/lib/python3.7/site-packages/torch/include/ATen/Parallel.h:149:0,
                     from /home/user/miniconda3/envs/pointcloud/lib/python3.7/site-packages/torch/include/torch/csrc/api/include/torch/utils.h:3,
                     from /home/user/miniconda3/envs/pointcloud/lib/python3.7/site-packages/torch/include/torch/csrc/api/include/torch/nn/cloneable.h:5,
                     from /home/user/miniconda3/envs/pointcloud/lib/python3.7/site-packages/torch/include/torch/csrc/api/include/torch/nn.h:3,
                     from /home/user/miniconda3/envs/pointcloud/lib/python3.7/site-packages/torch/include/torch/csrc/api/include/torch/all.h:12,
                     from /home/user/miniconda3/envs/pointcloud/lib/python3.7/site-packages/torch/include/torch/extension.h:4,
                     from cuda/include/interpolate.h:3,
                     from cuda/src/interpolate.cpp:1:
    /home/user/miniconda3/envs/pointcloud/lib/python3.7/site-packages/torch/include/ATen/ParallelOpenMP.h:84:0: warning: ignoring #pragma omp parallel [-Wunknown-pragmas]
     #pragma omp parallel for if ((end - begin) >= grain_size)

    gcc -pthread -B /home/user/miniconda3/envs/pointcloud/compiler_compat -Wl,--sysroot=/ -Wsign-compare -DNDEBUG -g -fwrapv -O3 -Wall -Wstrict-prototypes -fPIC -Icuda/include -I/home/user/miniconda3/envs/pointcloud/lib/python3.7/site-packages/torch/include -I/home/user/miniconda3/envs/pointcloud/lib/python3.7/site-packages/torch/include/torch/csrc/api/include -I/home/user/miniconda3/envs/pointcloud/lib/python3.7/site-packages/torch/include/TH -I/home/user/miniconda3/envs/pointcloud/lib/python3.7/site-packages/torch/include/THC -I/usr/local/cuda/include -I/home/user/miniconda3/envs/pointcloud/include/python3.7m -c cuda/src/cubic_feature_sampling.cpp -o build/temp.linux-x86_64-3.7/cuda/src/cubic_feature_sampling.o -O3 -DVERSION_GE_1_3 -DTORCH_API_INCLUDE_EXTENSION_H -DTORCH_API_INCLUDE_EXTENSION_H -DPYBIND11_COMPILER_TYPE="_gcc" -DPYBIND11_COMPILER_TYPE="_gcc" -DPYBIND11_STDLIB="_libstdcpp" -DPYBIND11_STDLIB="_libstdcpp" -DPYBIND11_BUILD_ABI="_cxxabi1011" -DPYBIND11_BUILD_ABI="_cxxabi1011" -DTORCH_EXTENSION_NAME=points_cuda -DTORCH_EXTENSION_NAME=points_cuda -D_GLIBCXX_USE_CXX11_ABI=0 -D_GLIBCXX_USE_CXX11_ABI=0 -std=c++14
    cc1plus: warning: command line option ‘-Wstrict-prototypes’ is valid for C/ObjC but not for C++
    In file included from /home/user/miniconda3/envs/pointcloud/lib/python3.7/site-packages/torch/include/ATen/Parallel.h:149:0,
                     from /home/user/miniconda3/envs/pointcloud/lib/python3.7/site-packages/torch/include/torch/csrc/api/include/torch/utils.h:3,
                     from /home/user/miniconda3/envs/pointcloud/lib/python3.7/site-packages/torch/include/torch/csrc/api/include/torch/nn/cloneable.h:5,
                     from /home/user/miniconda3/envs/pointcloud/lib/python3.7/site-packages/torch/include/torch/csrc/api/include/torch/nn.h:3,
                     from /home/user/miniconda3/envs/pointcloud/lib/python3.7/site-packages/torch/include/torch/csrc/api/include/torch/all.h:12,
                     from /home/user/miniconda3/envs/pointcloud/lib/python3.7/site-packages/torch/include/torch/extension.h:4,
                     from cuda/include/cubic_feature_sampling.h:4,
                     from cuda/src/cubic_feature_sampling.cpp:1:
    /home/user/miniconda3/envs/pointcloud/lib/python3.7/site-packages/torch/include/ATen/ParallelOpenMP.h:84:0: warning: ignoring #pragma omp parallel [-Wunknown-pragmas]
     #pragma omp parallel for if ((end - begin) >= grain_size)

    gcc -pthread -B /home/user/miniconda3/envs/pointcloud/compiler_compat -Wl,--sysroot=/ -Wsign-compare -DNDEBUG -g -fwrapv -O3 -Wall -Wstrict-prototypes -fPIC -Icuda/include -I/home/user/miniconda3/envs/pointcloud/lib/python3.7/site-packages/torch/include -I/home/user/miniconda3/envs/pointcloud/lib/python3.7/site-packages/torch/include/torch/csrc/api/include -I/home/user/miniconda3/envs/pointcloud/lib/python3.7/site-packages/torch/include/TH -I/home/user/miniconda3/envs/pointcloud/lib/python3.7/site-packages/torch/include/THC -I/usr/local/cuda/include -I/home/user/miniconda3/envs/pointcloud/include/python3.7m -c cuda/src/ball_query.cpp -o build/temp.linux-x86_64-3.7/cuda/src/ball_query.o -O3 -DVERSION_GE_1_3 -DTORCH_API_INCLUDE_EXTENSION_H -DTORCH_API_INCLUDE_EXTENSION_H -DPYBIND11_COMPILER_TYPE="_gcc" -DPYBIND11_COMPILER_TYPE="_gcc" -DPYBIND11_STDLIB="_libstdcpp" -DPYBIND11_STDLIB="_libstdcpp" -DPYBIND11_BUILD_ABI="_cxxabi1011" -DPYBIND11_BUILD_ABI="_cxxabi1011" -DTORCH_EXTENSION_NAME=points_cuda -DTORCH_EXTENSION_NAME=points_cuda -D_GLIBCXX_USE_CXX11_ABI=0 -D_GLIBCXX_USE_CXX11_ABI=0 -std=c++14
    cc1plus: warning: command line option ‘-Wstrict-prototypes’ is valid for C/ObjC but not for C++
    In file included from /home/user/miniconda3/envs/pointcloud/lib/python3.7/site-packages/torch/include/ATen/Parallel.h:149:0,
                     from /home/user/miniconda3/envs/pointcloud/lib/python3.7/site-packages/torch/include/torch/csrc/api/include/torch/utils.h:3,
                     from /home/user/miniconda3/envs/pointcloud/lib/python3.7/site-packages/torch/include/torch/csrc/api/include/torch/nn/cloneable.h:5,
                     from /home/user/miniconda3/envs/pointcloud/lib/python3.7/site-packages/torch/include/torch/csrc/api/include/torch/nn.h:3,
                     from /home/user/miniconda3/envs/pointcloud/lib/python3.7/site-packages/torch/include/torch/csrc/api/include/torch/all.h:12,
                     from /home/user/miniconda3/envs/pointcloud/lib/python3.7/site-packages/torch/include/torch/extension.h:4,
                     from cuda/include/ball_query.h:2,
                     from cuda/src/ball_query.cpp:1:
    /home/user/miniconda3/envs/pointcloud/lib/python3.7/site-packages/torch/include/ATen/ParallelOpenMP.h:84:0: warning: ignoring #pragma omp parallel [-Wunknown-pragmas]
     #pragma omp parallel for if ((end - begin) >= grain_size)

    gcc -pthread -B /home/user/miniconda3/envs/pointcloud/compiler_compat -Wl,--sysroot=/ -Wsign-compare -DNDEBUG -g -fwrapv -O3 -Wall -Wstrict-prototypes -fPIC -Icuda/include -I/home/user/miniconda3/envs/pointcloud/lib/python3.7/site-packages/torch/include -I/home/user/miniconda3/envs/pointcloud/lib/python3.7/site-packages/torch/include/torch/csrc/api/include -I/home/user/miniconda3/envs/pointcloud/lib/python3.7/site-packages/torch/include/TH -I/home/user/miniconda3/envs/pointcloud/lib/python3.7/site-packages/torch/include/THC -I/usr/local/cuda/include -I/home/user/miniconda3/envs/pointcloud/include/python3.7m -c cuda/src/bindings.cpp -o build/temp.linux-x86_64-3.7/cuda/src/bindings.o -O3 -DVERSION_GE_1_3 -DTORCH_API_INCLUDE_EXTENSION_H -DTORCH_API_INCLUDE_EXTENSION_H -DPYBIND11_COMPILER_TYPE="_gcc" -DPYBIND11_COMPILER_TYPE="_gcc" -DPYBIND11_STDLIB="_libstdcpp" -DPYBIND11_STDLIB="_libstdcpp" -DPYBIND11_BUILD_ABI="_cxxabi1011" -DPYBIND11_BUILD_ABI="_cxxabi1011" -DTORCH_EXTENSION_NAME=points_cuda -DTORCH_EXTENSION_NAME=points_cuda -D_GLIBCXX_USE_CXX11_ABI=0 -D_GLIBCXX_USE_CXX11_ABI=0 -std=c++14
    cc1plus: warning: command line option ‘-Wstrict-prototypes’ is valid for C/ObjC but not for C++
    In file included from /home/user/miniconda3/envs/pointcloud/lib/python3.7/site-packages/torch/include/ATen/Parallel.h:149:0,
                     from /home/user/miniconda3/envs/pointcloud/lib/python3.7/site-packages/torch/include/torch/csrc/api/include/torch/utils.h:3,
                     from /home/user/miniconda3/envs/pointcloud/lib/python3.7/site-packages/torch/include/torch/csrc/api/include/torch/nn/cloneable.h:5,
                     from /home/user/miniconda3/envs/pointcloud/lib/python3.7/site-packages/torch/include/torch/csrc/api/include/torch/nn.h:3,
                     from /home/user/miniconda3/envs/pointcloud/lib/python3.7/site-packages/torch/include/torch/csrc/api/include/torch/all.h:12,
                     from /home/user/miniconda3/envs/pointcloud/lib/python3.7/site-packages/torch/include/torch/extension.h:4,
                     from cuda/include/ball_query.h:2,
                     from cuda/src/bindings.cpp:1:
    /home/user/miniconda3/envs/pointcloud/lib/python3.7/site-packages/torch/include/ATen/ParallelOpenMP.h:84:0: warning: ignoring #pragma omp parallel [-Wunknown-pragmas]
     #pragma omp parallel for if ((end - begin) >= grain_size)

    gcc -pthread -B /home/user/miniconda3/envs/pointcloud/compiler_compat -Wl,--sysroot=/ -Wsign-compare -DNDEBUG -g -fwrapv -O3 -Wall -Wstrict-prototypes -fPIC -Icuda/include -I/home/user/miniconda3/envs/pointcloud/lib/python3.7/site-packages/torch/include -I/home/user/miniconda3/envs/pointcloud/lib/python3.7/site-packages/torch/include/torch/csrc/api/include -I/home/user/miniconda3/envs/pointcloud/lib/python3.7/site-packages/torch/include/TH -I/home/user/miniconda3/envs/pointcloud/lib/python3.7/site-packages/torch/include/THC -I/usr/local/cuda/include -I/home/user/miniconda3/envs/pointcloud/include/python3.7m -c cuda/src/sampling.cpp -o build/temp.linux-x86_64-3.7/cuda/src/sampling.o -O3 -DVERSION_GE_1_3 -DTORCH_API_INCLUDE_EXTENSION_H -DTORCH_API_INCLUDE_EXTENSION_H -DPYBIND11_COMPILER_TYPE="_gcc" -DPYBIND11_COMPILER_TYPE="_gcc" -DPYBIND11_STDLIB="_libstdcpp" -DPYBIND11_STDLIB="_libstdcpp" -DPYBIND11_BUILD_ABI="_cxxabi1011" -DPYBIND11_BUILD_ABI="_cxxabi1011" -DTORCH_EXTENSION_NAME=points_cuda -DTORCH_EXTENSION_NAME=points_cuda -D_GLIBCXX_USE_CXX11_ABI=0 -D_GLIBCXX_USE_CXX11_ABI=0 -std=c++14
    cc1plus: warning: command line option ‘-Wstrict-prototypes’ is valid for C/ObjC but not for C++
    In file included from /home/user/miniconda3/envs/pointcloud/lib/python3.7/site-packages/torch/include/ATen/Parallel.h:149:0,
                     from /home/user/miniconda3/envs/pointcloud/lib/python3.7/site-packages/torch/include/torch/csrc/api/include/torch/utils.h:3,
                     from /home/user/miniconda3/envs/pointcloud/lib/python3.7/site-packages/torch/include/torch/csrc/api/include/torch/nn/cloneable.h:5,
                     from /home/user/miniconda3/envs/pointcloud/lib/python3.7/site-packages/torch/include/torch/csrc/api/include/torch/nn.h:3,
                     from /home/user/miniconda3/envs/pointcloud/lib/python3.7/site-packages/torch/include/torch/csrc/api/include/torch/all.h:12,
                     from /home/user/miniconda3/envs/pointcloud/lib/python3.7/site-packages/torch/include/torch/extension.h:4,
                     from cuda/include/sampling.h:2,
                     from cuda/src/sampling.cpp:1:
    /home/user/miniconda3/envs/pointcloud/lib/python3.7/site-packages/torch/include/ATen/ParallelOpenMP.h:84:0: warning: ignoring #pragma omp parallel [-Wunknown-pragmas]
     #pragma omp parallel for if ((end - begin) >= grain_size)

    gcc -pthread -B /home/user/miniconda3/envs/pointcloud/compiler_compat -Wl,--sysroot=/ -Wsign-compare -DNDEBUG -g -fwrapv -O3 -Wall -Wstrict-prototypes -fPIC -Icuda/include -I/home/user/miniconda3/envs/pointcloud/lib/python3.7/site-packages/torch/include -I/home/user/miniconda3/envs/pointcloud/lib/python3.7/site-packages/torch/include/torch/csrc/api/include -I/home/user/miniconda3/envs/pointcloud/lib/python3.7/site-packages/torch/include/TH -I/home/user/miniconda3/envs/pointcloud/lib/python3.7/site-packages/torch/include/THC -I/usr/local/cuda/include -I/home/user/miniconda3/envs/pointcloud/include/python3.7m -c cuda/src/chamfer_dist.cpp -o build/temp.linux-x86_64-3.7/cuda/src/chamfer_dist.o -O3 -DVERSION_GE_1_3 -DTORCH_API_INCLUDE_EXTENSION_H -DTORCH_API_INCLUDE_EXTENSION_H -DPYBIND11_COMPILER_TYPE="_gcc" -DPYBIND11_COMPILER_TYPE="_gcc" -DPYBIND11_STDLIB="_libstdcpp" -DPYBIND11_STDLIB="_libstdcpp" -DPYBIND11_BUILD_ABI="_cxxabi1011" -DPYBIND11_BUILD_ABI="_cxxabi1011" -DTORCH_EXTENSION_NAME=points_cuda -DTORCH_EXTENSION_NAME=points_cuda -D_GLIBCXX_USE_CXX11_ABI=0 -D_GLIBCXX_USE_CXX11_ABI=0 -std=c++14
    cc1plus: warning: command line option ‘-Wstrict-prototypes’ is valid for C/ObjC but not for C++
    In file included from /home/user/miniconda3/envs/pointcloud/lib/python3.7/site-packages/torch/include/ATen/Parallel.h:149:0,
                     from /home/user/miniconda3/envs/pointcloud/lib/python3.7/site-packages/torch/include/torch/csrc/api/include/torch/utils.h:3,
                     from /home/user/miniconda3/envs/pointcloud/lib/python3.7/site-packages/torch/include/torch/csrc/api/include/torch/nn/cloneable.h:5,
                     from /home/user/miniconda3/envs/pointcloud/lib/python3.7/site-packages/torch/include/torch/csrc/api/include/torch/nn.h:3,
                     from /home/user/miniconda3/envs/pointcloud/lib/python3.7/site-packages/torch/include/torch/csrc/api/include/torch/all.h:12,
                     from /home/user/miniconda3/envs/pointcloud/lib/python3.7/site-packages/torch/include/torch/extension.h:4,
                     from cuda/include/chamfer_dist.h:1,
                     from cuda/src/chamfer_dist.cpp:1:
    /home/user/miniconda3/envs/pointcloud/lib/python3.7/site-packages/torch/include/ATen/ParallelOpenMP.h:84:0: warning: ignoring #pragma omp parallel [-Wunknown-pragmas]
     #pragma omp parallel for if ((end - begin) >= grain_size)

    gcc -pthread -B /home/user/miniconda3/envs/pointcloud/compiler_compat -Wl,--sysroot=/ -Wsign-compare -DNDEBUG -g -fwrapv -O3 -Wall -Wstrict-prototypes -fPIC -Icuda/include -I/home/user/miniconda3/envs/pointcloud/lib/python3.7/site-packages/torch/include -I/home/user/miniconda3/envs/pointcloud/lib/python3.7/site-packages/torch/include/torch/csrc/api/include -I/home/user/miniconda3/envs/pointcloud/lib/python3.7/site-packages/torch/include/TH -I/home/user/miniconda3/envs/pointcloud/lib/python3.7/site-packages/torch/include/THC -I/usr/local/cuda/include -I/home/user/miniconda3/envs/pointcloud/include/python3.7m -c cuda/src/metrics.cpp -o build/temp.linux-x86_64-3.7/cuda/src/metrics.o -O3 -DVERSION_GE_1_3 -DTORCH_API_INCLUDE_EXTENSION_H -DTORCH_API_INCLUDE_EXTENSION_H -DPYBIND11_COMPILER_TYPE="_gcc" -DPYBIND11_COMPILER_TYPE="_gcc" -DPYBIND11_STDLIB="_libstdcpp" -DPYBIND11_STDLIB="_libstdcpp" -DPYBIND11_BUILD_ABI="_cxxabi1011" -DPYBIND11_BUILD_ABI="_cxxabi1011" -DTORCH_EXTENSION_NAME=points_cuda -DTORCH_EXTENSION_NAME=points_cuda -D_GLIBCXX_USE_CXX11_ABI=0 -D_GLIBCXX_USE_CXX11_ABI=0 -std=c++14
    cc1plus: warning: command line option ‘-Wstrict-prototypes’ is valid for C/ObjC but not for C++
    In file included from /home/user/miniconda3/envs/pointcloud/lib/python3.7/site-packages/torch/include/ATen/Parallel.h:149:0,
                     from /home/user/miniconda3/envs/pointcloud/lib/python3.7/site-packages/torch/include/torch/csrc/api/include/torch/utils.h:3,
                     from /home/user/miniconda3/envs/pointcloud/lib/python3.7/site-packages/torch/include/torch/csrc/api/include/torch/nn/cloneable.h:5,
                     from /home/user/miniconda3/envs/pointcloud/lib/python3.7/site-packages/torch/include/torch/csrc/api/include/torch/nn.h:3,
                     from /home/user/miniconda3/envs/pointcloud/lib/python3.7/site-packages/torch/include/torch/csrc/api/include/torch/all.h:12,
                     from /home/user/miniconda3/envs/pointcloud/lib/python3.7/site-packages/torch/include/torch/extension.h:4,
                     from cuda/include/metrics.h:2,
                     from cuda/src/metrics.cpp:1:
    /home/user/miniconda3/envs/pointcloud/lib/python3.7/site-packages/torch/include/ATen/ParallelOpenMP.h:84:0: warning: ignoring #pragma omp parallel [-Wunknown-pragmas]
     #pragma omp parallel for if ((end - begin) >= grain_size)

    gcc -pthread -B /home/user/miniconda3/envs/pointcloud/compiler_compat -Wl,--sysroot=/ -Wsign-compare -DNDEBUG -g -fwrapv -O3 -Wall -Wstrict-prototypes -fPIC -Icuda/include -I/home/user/miniconda3/envs/pointcloud/lib/python3.7/site-packages/torch/include -I/home/user/miniconda3/envs/pointcloud/lib/python3.7/site-packages/torch/include/torch/csrc/api/include -I/home/user/miniconda3/envs/pointcloud/lib/python3.7/site-packages/torch/include/TH -I/home/user/miniconda3/envs/pointcloud/lib/python3.7/site-packages/torch/include/THC -I/usr/local/cuda/include -I/home/user/miniconda3/envs/pointcloud/include/python3.7m -c cuda/src/gridding.cpp -o build/temp.linux-x86_64-3.7/cuda/src/gridding.o -O3 -DVERSION_GE_1_3 -DTORCH_API_INCLUDE_EXTENSION_H -DTORCH_API_INCLUDE_EXTENSION_H -DPYBIND11_COMPILER_TYPE="_gcc" -DPYBIND11_COMPILER_TYPE="_gcc" -DPYBIND11_STDLIB="_libstdcpp" -DPYBIND11_STDLIB="_libstdcpp" -DPYBIND11_BUILD_ABI="_cxxabi1011" -DPYBIND11_BUILD_ABI="_cxxabi1011" -DTORCH_EXTENSION_NAME=points_cuda -DTORCH_EXTENSION_NAME=points_cuda -D_GLIBCXX_USE_CXX11_ABI=0 -D_GLIBCXX_USE_CXX11_ABI=0 -std=c++14
    cc1plus: warning: command line option ‘-Wstrict-prototypes’ is valid for C/ObjC but not for C++
    In file included from /home/user/miniconda3/envs/pointcloud/lib/python3.7/site-packages/torch/include/ATen/Parallel.h:149:0,
                     from /home/user/miniconda3/envs/pointcloud/lib/python3.7/site-packages/torch/include/torch/csrc/api/include/torch/utils.h:3,
                     from /home/user/miniconda3/envs/pointcloud/lib/python3.7/site-packages/torch/include/torch/csrc/api/include/torch/nn/cloneable.h:5,
                     from /home/user/miniconda3/envs/pointcloud/lib/python3.7/site-packages/torch/include/torch/csrc/api/include/torch/nn.h:3,
                     from /home/user/miniconda3/envs/pointcloud/lib/python3.7/site-packages/torch/include/torch/csrc/api/include/torch/all.h:12,
                     from /home/user/miniconda3/envs/pointcloud/lib/python3.7/site-packages/torch/include/torch/extension.h:4,
                     from cuda/include/gridding.h:4,
                     from cuda/src/gridding.cpp:1:
    /home/user/miniconda3/envs/pointcloud/lib/python3.7/site-packages/torch/include/ATen/ParallelOpenMP.h:84:0: warning: ignoring #pragma omp parallel [-Wunknown-pragmas]
     #pragma omp parallel for if ((end - begin) >= grain_size)

    /usr/local/cuda/bin/nvcc -Icuda/include -I/home/user/miniconda3/envs/pointcloud/lib/python3.7/site-packages/torch/include -I/home/user/miniconda3/envs/pointcloud/lib/python3.7/site-packages/torch/include/torch/csrc/api/include -I/home/user/miniconda3/envs/pointcloud/lib/python3.7/site-packages/torch/include/TH -I/home/user/miniconda3/envs/pointcloud/lib/python3.7/site-packages/torch/include/THC -I/usr/local/cuda/include -I/home/user/miniconda3/envs/pointcloud/include/python3.7m -c cuda/src/chamfer_dist_gpu.cu -o build/temp.linux-x86_64-3.7/cuda/src/chamfer_dist_gpu.o -D__CUDA_NO_HALF_OPERATORS__ -D__CUDA_NO_HALF_CONVERSIONS__ -D__CUDA_NO_HALF2_OPERATORS__ --expt-relaxed-constexpr --compiler-options '-fPIC' -O3 -DVERSION_GE_1_3 -DTORCH_API_INCLUDE_EXTENSION_H -DTORCH_API_INCLUDE_EXTENSION_H -DPYBIND11_COMPILER_TYPE="_gcc" -DPYBIND11_COMPILER_TYPE="_gcc" -DPYBIND11_STDLIB="_libstdcpp" -DPYBIND11_STDLIB="_libstdcpp" -DPYBIND11_BUILD_ABI="_cxxabi1011" -DPYBIND11_BUILD_ABI="_cxxabi1011" -DTORCH_EXTENSION_NAME=points_cuda -DTORCH_EXTENSION_NAME=points_cuda -D_GLIBCXX_USE_CXX11_ABI=0 -D_GLIBCXX_USE_CXX11_ABI=0 -gencode=arch=compute_86,code=sm_86 -std=c++14
    nvcc fatal   : Unsupported gpu architecture 'compute_86'
    error: command '/usr/local/cuda/bin/nvcc' failed with exit status 1
    ----------------------------------------
ERROR: Command errored out with exit status 1: /home/user/miniconda3/envs/pointcloud/bin/python -u -c 'import sys, setuptools, tokenize; sys.argv[0] = '"'"'/tmp/pip-install-3twio4ga/torch-points-kernels_4d1bdf9ec06f4aa8984512eb395e074d/setup.py'"'"'; __file__='"'"'/tmp/pip-install-3twio4ga/torch-points-kernels_4d1bdf9ec06f4aa8984512eb395e074d/setup.py'"'"';f=getattr(tokenize, '"'"'open'"'"', open)(__file__);code=f.read().replace('"'"'\r\n'"'"', '"'"'\n'"'"');f.close();exec(compile(code, __file__, '"'"'exec'"'"'))' install --record /tmp/pip-record-clyu4km7/install-record.txt --single-version-externally-managed --compile --install-headers /home/user/miniconda3/envs/pointcloud/include/python3.7m/torch-points-kernels Check the logs for full command output.
nicolas-chaulet commented 3 years ago

Hey! This is the reason image

Not sure what to do with that, it could be a problem in the cuda code that uses features that are not supported by the 3090, I have never tried to install it on the GTX 3090. @CCInc any suggestion ?

CCInc commented 3 years ago

@chaitjo I am pretty sure the 3000 series isn't fully supported on CUDA 11.0. You should try upgrading nvcc and cuda to 11.1.1 or 11.2 which should allow the kernels to compile. You may have to compile pytorch yourself as a result, but maybe it will work as-is. Let me know!

CCInc commented 2 years ago

Closing for now, feel free to reopen if still an issue.