"Building wheel for torch-sparse (setup.py) ... error" when installing torch_sparse via pip3

[Jerry-jwz]

😵 Describe the installation problem

When I was installing torch-sparse through the command:
pip install torch-sparse -f https://pytorch-geometric.com/whl/torch-1.10.0+cu113.html,
I was totally confused by the following error:

Looking in links: https://pytorch-geometric.com/whl/torch-1.10.0+cu113.html
Collecting torch-sparse
  Using cached torch_sparse-0.6.14.tar.gz (51 kB)
  Preparing metadata (setup.py) ... done
Requirement already satisfied: scipy in ./anaconda3/envs/pytorch/lib/python3.8/site-packages (from torch-sparse) (1.8.1)
Requirement already satisfied: numpy<1.25.0,>=1.17.3 in ./anaconda3/envs/pytorch/lib/python3.8/site-packages (from scipy->torch-sparse) (1.23.0)
Building wheels for collected packages: torch-sparse
  Building wheel for torch-sparse (setup.py) ... error
  error: subprocess-exited-with-error
  
  × python setup.py bdist_wheel did not run successfully.
  │ exit code: 1
  ╰─> [55 lines of output]
      running bdist_wheel
      running build
      running build_py
      creating build
      creating build/lib.linux-x86_64-3.8
      creating build/lib.linux-x86_64-3.8/torch_sparse
      copying torch_sparse/__init__.py -> build/lib.linux-x86_64-3.8/torch_sparse
      copying torch_sparse/add.py -> build/lib.linux-x86_64-3.8/torch_sparse
      copying torch_sparse/bandwidth.py -> build/lib.linux-x86_64-3.8/torch_sparse
      copying torch_sparse/cat.py -> build/lib.linux-x86_64-3.8/torch_sparse
      copying torch_sparse/coalesce.py -> build/lib.linux-x86_64-3.8/torch_sparse
      copying torch_sparse/convert.py -> build/lib.linux-x86_64-3.8/torch_sparse
      copying torch_sparse/diag.py -> build/lib.linux-x86_64-3.8/torch_sparse
      copying torch_sparse/eye.py -> build/lib.linux-x86_64-3.8/torch_sparse
      copying torch_sparse/index_select.py -> build/lib.linux-x86_64-3.8/torch_sparse
      copying torch_sparse/masked_select.py -> build/lib.linux-x86_64-3.8/torch_sparse
      copying torch_sparse/matmul.py -> build/lib.linux-x86_64-3.8/torch_sparse
      copying torch_sparse/metis.py -> build/lib.linux-x86_64-3.8/torch_sparse
      copying torch_sparse/mul.py -> build/lib.linux-x86_64-3.8/torch_sparse
      copying torch_sparse/narrow.py -> build/lib.linux-x86_64-3.8/torch_sparse
      copying torch_sparse/padding.py -> build/lib.linux-x86_64-3.8/torch_sparse
      copying torch_sparse/permute.py -> build/lib.linux-x86_64-3.8/torch_sparse
      copying torch_sparse/reduce.py -> build/lib.linux-x86_64-3.8/torch_sparse
      copying torch_sparse/rw.py -> build/lib.linux-x86_64-3.8/torch_sparse
      copying torch_sparse/saint.py -> build/lib.linux-x86_64-3.8/torch_sparse
      copying torch_sparse/sample.py -> build/lib.linux-x86_64-3.8/torch_sparse
      copying torch_sparse/select.py -> build/lib.linux-x86_64-3.8/torch_sparse
      copying torch_sparse/spadd.py -> build/lib.linux-x86_64-3.8/torch_sparse
      copying torch_sparse/spmm.py -> build/lib.linux-x86_64-3.8/torch_sparse
      copying torch_sparse/spspmm.py -> build/lib.linux-x86_64-3.8/torch_sparse
      copying torch_sparse/storage.py -> build/lib.linux-x86_64-3.8/torch_sparse
      copying torch_sparse/tensor.py -> build/lib.linux-x86_64-3.8/torch_sparse
      copying torch_sparse/transpose.py -> build/lib.linux-x86_64-3.8/torch_sparse
      copying torch_sparse/utils.py -> build/lib.linux-x86_64-3.8/torch_sparse
      running egg_info
      writing torch_sparse.egg-info/PKG-INFO
      writing dependency_links to torch_sparse.egg-info/dependency_links.txt
      writing requirements to torch_sparse.egg-info/requires.txt
      writing top-level names to torch_sparse.egg-info/top_level.txt
      reading manifest file 'torch_sparse.egg-info/SOURCES.txt'
      reading manifest template 'MANIFEST.in'
      warning: no previously-included files matching '*' found under directory 'test'
      adding license file 'LICENSE'
      writing manifest file 'torch_sparse.egg-info/SOURCES.txt'
      running build_ext
      /home/jiangwenzhao/anaconda3/envs/pytorch/lib/python3.8/site-packages/torch/utils/cpp_extension.py:782: UserWarning: The detected CUDA version (11.6) has a minor version mismatch with the version that was used to compile PyTorch (11.3). Most likely this shouldn't be a problem.
        warnings.warn(CUDA_MISMATCH_WARN.format(cuda_str_version, torch.version.cuda))
      building 'torch_sparse._convert_cpu' extension
      creating build/temp.linux-x86_64-3.8
      creating build/temp.linux-x86_64-3.8/csrc
      creating build/temp.linux-x86_64-3.8/csrc/cpu
      gcc -pthread -B /home/jiangwenzhao/anaconda3/envs/pytorch/compiler_compat -Wl,--sysroot=/ -Wsign-compare -DNDEBUG -g -fwrapv -O3 -Wall -Wstrict-prototypes -fPIC -DWITH_PYTHON -Icsrc -I/home/jiangwenzhao/anaconda3/envs/pytorch/lib/python3.8/site-packages/torch/include -I/home/jiangwenzhao/anaconda3/envs/pytorch/lib/python3.8/site-packages/torch/include/torch/csrc/api/include -I/home/jiangwenzhao/anaconda3/envs/pytorch/lib/python3.8/site-packages/torch/include/TH -I/home/jiangwenzhao/anaconda3/envs/pytorch/lib/python3.8/site-packages/torch/include/THC -I/home/jiangwenzhao/anaconda3/envs/pytorch/include/python3.8 -c csrc/convert.cpp -o build/temp.linux-x86_64-3.8/csrc/convert.o -O2 -Wno-sign-compare -DAT_PARALLEL_OPENMP -fopenmp -DTORCH_API_INCLUDE_EXTENSION_H -DPYBIND11_COMPILER_TYPE=\"_gcc\" -DPYBIND11_STDLIB=\"_libstdcpp\" -DPYBIND11_BUILD_ABI=\"_cxxabi1011\" -DTORCH_EXTENSION_NAME=_convert_cpu -D_GLIBCXX_USE_CXX11_ABI=0 -std=c++14
      gcc: fatal error: cannot execute 'cc1plus': execvp: No such file or directory
      compilation terminated.
      error: command '/home/jiangwenzhao/anaconda3/envs/pytorch/bin/gcc' failed with exit code 1
      [end of output]
  
  note: This error originates from a subprocess, and is likely not a problem with pip.
  ERROR: Failed building wheel for torch-sparse
  Running setup.py clean for torch-sparse
Failed to build torch-sparse

Before this, I had just fixed a warning saying 'Your compiler (g++ 4.8.5) may be ABI-incompatible with PyTorch!
Please use a compiler that is ABI-compatible with GCC 5.0 and above', by 'conda install -c conda-forge gxx=9.4.0 gcc=9.4.0'. However, I am totally stuck with the above error. Hope anyone could help!

If you need any other info, please let me know.

Environment

• PyG version:2.04
• PyTorch version:1.10.0
• OS:CentOS Linux release 7.9.2009 (Core)
• Python version:3.8
• CUDA/cuDNN version:11.3
• How you installed PyTorch and PyG (conda, pip, source):pip
• torch_cluster:1.6.0
• torch_scatter:2.0.9
• torch_spline_conv:1.2.1
• gcc & g++: 9.4.0

[downeykking]

I encountered the same question with you and i solved it using this command:
pip install torch-sparse==0.6.13 -f https://pytorch-geometric.com/whl/torch-1.10.0+cu113.html

[Jerry-jwz]

I encountered the same question with you and i solved it using this command: pip install torch-sparse==0.6.13 -f https://pytorch-geometric.com/whl/torch-1.10.0+cu113.html

It works! Thanks! But still don't know why torch-sparse-0.6.14 causes this trouble.

[downeykking]

I encountered the same question with you and i solved it using this command: pip install torch-sparse==0.6.13 -f https://pytorch-geometric.com/whl/torch-1.10.0+cu113.html

It works! Thanks! But still don't know why torch-sparse-0.6.14 causes this trouble.

check 'https://data.pyg.org/whl/index.html'. It only has torch-sparse-0.6.13 and torch-sparse-0.6.14 does not released yet. maybe that is the reason.

[rusty1s]

We only build new versions for the two latest PyTorch releases. That‘s why 0.6.13 is the last version for PyTorch 1.10.

[LingjieBao1998]

wonderful! It help me a lot! Thanks for your goodness

[harshy105]

In my environment, I had pytorch 2.0.1 with cuda 11.8 on windows. And none of the provided comment worked for me. Then I found the comment on PyG installation guide: https://pytorch-geometric.readthedocs.io/en/latest/install/installation.html

pip install torch_scatter torch_sparse torch_cluster torch_spline_conv -f https://data.pyg.org/whl/torch-2.0.0+cu118.html

This worked just fine

[Armanasq]

I have tried to install pip install torch_scatter torch_sparse torch_cluster torch_spline_conv -f https://data.pyg.org/whl/torch-2.0.0+cu118.html

But still I could not load torch_sparse

---

OSError Traceback (most recent call last)
Cell In[2], line 1
----> 1 import torch_sparse

File /media/ssrc/DataBase/cmu/lib/python3.8/site-packages/torch_sparse/init.py:19
17 spec = cuda_spec or cpu_spec
18 if spec is not None:
---> 19 torch.ops.load_library(spec.origin)
20 else: # pragma: no cover
21 raise ImportError(f"Could not find module '{library}_cpu' in "
22 f"{osp.dirname(file)}")

File /media/ssrc/DataBase/cmu/lib/python3.8/site-packages/torch/_ops.py:852, in _Ops.load_library(self, path)
847 path = _utils_internal.resolve_library_path(path)
848 with dl_open_guard():
849 # Import the shared library into the process, thus running its
850 # static (global) initialization code in order to register custom
851 # operators with the JIT.
--> 852 ctypes.CDLL(path)
853 self.loaded_libraries.add(path)

File /usr/lib/python3.8/ctypes/init.py:373, in CDLL.init(self, name, mode, handle, use_errno, use_last_error, winmode)
370 self._FuncPtr = _FuncPtr
372 if handle is None:
--> 373 self._handle = _dlopen(self._name, mode)
374 else:
375 self._handle = handle

OSError: libtorch_cuda_cpp.so: cannot open shared object file: No such file or directory

[rusty1s]

What happens if you run via

pip uninstall torch-sparse
pip install --verbose --no-index torch_sparse -f https://data.pyg.org/whl/torch-2.0.0+cu118.html

[mainguyenanhvu]

I found the solution here.

[zhangnuc]

Sooooooo useful to me! @downeykking
I often encounter the same issue with packages such as torch_spline_conv, torch-cluster, torch-sparse, and torch-spline-conv,
which are all related to Torch and CUDA versions.
I solve torch-sparse error with command "pip install torch-sparse==0.6.12 -f https://pytorch-geometric.com/whl/torch-1.9.1+cu111.html" which can instead your cuda version and torch version inthe link "https://pytorch-geometric.com/whl/torch-1.9.1+cu111.html" , also we can solve torch-sparse error by "pip install torch-sparse==0.X.X -f https://pytorch-geometric.com/whl/torch-1.9.1+cu111.html"