Error during installation

[remarkSD]

I am using python 3.6, Pytorch 1.0.

I am getting the following error when installing:

pip install torch-cluster
Collecting torch-cluster
  Using cached https://files.pythonhosted.org/packages/33/b7/05b9ce9afc76f5709efe04d6344fbed09ea217f916f94e63f2fe9659eb62/torch_cluster-1.2.2.tar.gz
Requirement already satisfied: scipy in /home/airscan-razer04/miniconda3/envs/pytorch/lib/python3.6/site-packages (from torch-cluster) (1.1.0)
Building wheels for collected packages: torch-cluster
  Running setup.py bdist_wheel for torch-cluster ... error
  Complete output from command /home/airscan-razer04/miniconda3/envs/pytorch/bin/python -u -c "import setuptools, tokenize;__file__='/tmp/pip-install-_id9r1qr/torch-cluster/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-ij35ca3y --python-tag cp36:
  running bdist_wheel
  running build
  running build_py
  creating build
  creating build/lib.linux-x86_64-3.6
  creating build/lib.linux-x86_64-3.6/torch_cluster
  copying torch_cluster/__init__.py -> build/lib.linux-x86_64-3.6/torch_cluster
  copying torch_cluster/nearest.py -> build/lib.linux-x86_64-3.6/torch_cluster
  copying torch_cluster/fps.py -> build/lib.linux-x86_64-3.6/torch_cluster
  copying torch_cluster/knn.py -> build/lib.linux-x86_64-3.6/torch_cluster
  copying torch_cluster/graclus.py -> build/lib.linux-x86_64-3.6/torch_cluster
  copying torch_cluster/grid.py -> build/lib.linux-x86_64-3.6/torch_cluster
  copying torch_cluster/radius.py -> build/lib.linux-x86_64-3.6/torch_cluster
  creating build/lib.linux-x86_64-3.6/test
  copying test/test_radius.py -> build/lib.linux-x86_64-3.6/test
  copying test/__init__.py -> build/lib.linux-x86_64-3.6/test
  copying test/test_fps.py -> build/lib.linux-x86_64-3.6/test
  copying test/test_knn.py -> build/lib.linux-x86_64-3.6/test
  copying test/test_nearest.py -> build/lib.linux-x86_64-3.6/test
  copying test/test_graclus.py -> build/lib.linux-x86_64-3.6/test
  copying test/utils.py -> build/lib.linux-x86_64-3.6/test
  copying test/test_grid.py -> build/lib.linux-x86_64-3.6/test
  creating build/lib.linux-x86_64-3.6/torch_cluster/_ext
  copying torch_cluster/_ext/__init__.py -> build/lib.linux-x86_64-3.6/torch_cluster/_ext
  creating build/lib.linux-x86_64-3.6/torch_cluster/_ext/ffi
  copying torch_cluster/_ext/ffi/__init__.py -> build/lib.linux-x86_64-3.6/torch_cluster/_ext/ffi
  running build_ext
  building 'graclus_cpu' extension
  creating build/temp.linux-x86_64-3.6
  creating build/temp.linux-x86_64-3.6/cpu
  gcc -pthread -B /home/airscan-razer04/miniconda3/envs/pytorch/compiler_compat -Wl,--sysroot=/ -Wsign-compare -DNDEBUG -g -fwrapv -O3 -Wall -Wstrict-prototypes -fPIC -I/home/airscan-razer04/miniconda3/envs/pytorch/lib/python3.6/site-packages/torch/lib/include -I/home/airscan-razer04/miniconda3/envs/pytorch/lib/python3.6/site-packages/torch/lib/include/TH -I/home/airscan-razer04/miniconda3/envs/pytorch/lib/python3.6/site-packages/torch/lib/include/THC -I/home/airscan-razer04/miniconda3/envs/pytorch/include/python3.6m -c cpu/graclus.cpp -o build/temp.linux-x86_64-3.6/cpu/graclus.o -DTORCH_EXTENSION_NAME=graclus_cpu -D_GLIBCXX_USE_CXX11_ABI=0 -std=c++11
  cc1plus: warning: command line option ‘-Wstrict-prototypes’ is valid for C/ObjC but not for C++
  cpu/graclus.cpp:1:29: fatal error: torch/extension.h: No such file or directory
  compilation terminated.
  error: command 'gcc' failed with exit status 1
  
  ----------------------------------------
  Failed building wheel for torch-cluster
  Running setup.py clean for torch-cluster
Failed to build torch-cluster
Installing collected packages: torch-cluster
  Running setup.py install for torch-cluster ... error
    Complete output from command /home/airscan-razer04/miniconda3/envs/pytorch/bin/python -u -c "import setuptools, tokenize;__file__='/tmp/pip-install-_id9r1qr/torch-cluster/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-esqv00gn/install-record.txt --single-version-externally-managed --compile:
    running install
    running build
    running build_py
    creating build
    creating build/lib.linux-x86_64-3.6
    creating build/lib.linux-x86_64-3.6/torch_cluster
    copying torch_cluster/__init__.py -> build/lib.linux-x86_64-3.6/torch_cluster
    copying torch_cluster/nearest.py -> build/lib.linux-x86_64-3.6/torch_cluster
    copying torch_cluster/fps.py -> build/lib.linux-x86_64-3.6/torch_cluster
    copying torch_cluster/knn.py -> build/lib.linux-x86_64-3.6/torch_cluster
    copying torch_cluster/graclus.py -> build/lib.linux-x86_64-3.6/torch_cluster
    copying torch_cluster/grid.py -> build/lib.linux-x86_64-3.6/torch_cluster
    copying torch_cluster/radius.py -> build/lib.linux-x86_64-3.6/torch_cluster
    creating build/lib.linux-x86_64-3.6/test
    copying test/test_radius.py -> build/lib.linux-x86_64-3.6/test
    copying test/__init__.py -> build/lib.linux-x86_64-3.6/test
    copying test/test_fps.py -> build/lib.linux-x86_64-3.6/test
    copying test/test_knn.py -> build/lib.linux-x86_64-3.6/test
    copying test/test_nearest.py -> build/lib.linux-x86_64-3.6/test
    copying test/test_graclus.py -> build/lib.linux-x86_64-3.6/test
    copying test/utils.py -> build/lib.linux-x86_64-3.6/test
    copying test/test_grid.py -> build/lib.linux-x86_64-3.6/test
    creating build/lib.linux-x86_64-3.6/torch_cluster/_ext
    copying torch_cluster/_ext/__init__.py -> build/lib.linux-x86_64-3.6/torch_cluster/_ext
    creating build/lib.linux-x86_64-3.6/torch_cluster/_ext/ffi
    copying torch_cluster/_ext/ffi/__init__.py -> build/lib.linux-x86_64-3.6/torch_cluster/_ext/ffi
    running build_ext
    building 'graclus_cpu' extension
    creating build/temp.linux-x86_64-3.6
    creating build/temp.linux-x86_64-3.6/cpu
    gcc -pthread -B /home/airscan-razer04/miniconda3/envs/pytorch/compiler_compat -Wl,--sysroot=/ -Wsign-compare -DNDEBUG -g -fwrapv -O3 -Wall -Wstrict-prototypes -fPIC -I/home/airscan-razer04/miniconda3/envs/pytorch/lib/python3.6/site-packages/torch/lib/include -I/home/airscan-razer04/miniconda3/envs/pytorch/lib/python3.6/site-packages/torch/lib/include/TH -I/home/airscan-razer04/miniconda3/envs/pytorch/lib/python3.6/site-packages/torch/lib/include/THC -I/home/airscan-razer04/miniconda3/envs/pytorch/include/python3.6m -c cpu/graclus.cpp -o build/temp.linux-x86_64-3.6/cpu/graclus.o -DTORCH_EXTENSION_NAME=graclus_cpu -D_GLIBCXX_USE_CXX11_ABI=0 -std=c++11
    cc1plus: warning: command line option ‘-Wstrict-prototypes’ is valid for C/ObjC but not for C++
    cpu/graclus.cpp:1:29: fatal error: torch/extension.h: No such file or directory
    compilation terminated.
    error: command 'gcc' failed with exit status 1
    
    ----------------------------------------
Command "/home/airscan-razer04/miniconda3/envs/pytorch/bin/python -u -c "import setuptools, tokenize;__file__='/tmp/pip-install-_id9r1qr/torch-cluster/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-esqv00gn/install-record.txt --single-version-externally-managed --compile" failed with error code 1 in /tmp/pip-install-_id9r1qr/torch-cluster/

I've already set the cuda/bin and cuda/include in PATH and CPATH.
I'm still trying to figure out if I'm just missing something.
Would appreciate your help on this.

Thanks.

Edit: Apparently, I wasn't using pytorch 1.0. Installation is good after updating pytorch to 1.0.

[rusty1s]

Are you sure you are on PyTorch 1.0.0? It seems that it can not find the torch/extension header, which was introduced in 1.0.0. You need to either upgrade PyTorch or replace the torch/extension.h calls with torch/torch.h.

[remarkSD]

Are you sure you are on PyTorch 1.0.0? It seems that it can not find the torch/extension header, which was introduced in 1.0.0. You need to either upgrade PyTorch or replace the torch/extension.h calls with torch/torch.h.

You're right. Apparently I have been using PyTorch 0.4.1 and torchvision 0.2.1. I didn't notice that my pytorch installation got interrupted so it didn't get updated. I'll reply here again if it works.

[remarkSD]

The pytorch version was the problem. Successfully installed it after updating pytorch to 1.0.
Thanks!

[xuChenSJTU]

However, it seems that this problem still exists with pytorch 1.1