| Git Clone URL: | https://aur.archlinux.org/python-apex-git.git (read-only, click to copy) |
|---|---|
| Package Base: | python-apex-git |
| Description: | A PyTorch Extension: Tools for easy mixed precision and distributed training in Pytorch |
| Upstream URL: | https://github.com/NVIDIA/apex |
| Keywords: | pytorch |
| Licenses: | BSD |
| Conflicts: | python-apex |
| Provides: | python-apex |
| Submitter: | leomao |
| Maintainer: | petronny |
| Last Packager: | petronny |
| Votes: | 0 |
| Popularity: | 0.000000 |
| First Submitted: | 2018-12-14 06:07 (UTC) |
| Last Updated: | 2024-08-02 07:00 (UTC) |
gcc11 dependency can be dropped as suggested by the building message.
Also, the source in PKGBUILD needs to set the tag; otherwise, the main branch will likely fail the build.
Mine PKGBUILD file, in case anyone needs one,
https://gist.github.com/chengscott/d279f3f898f0c09921b598cf38a9f889
CC=/opt/cuda/bin/gcc CXX=/opt/cuda/bin/g++
is also needed for python setup.py install
otherwise, the build shows lto build failed (quite confusing error message) This is due to mismatched g++ version: /usr/bin/g++: g++ (GCC) 12.2.0 /opt/cuda/bin/g++: g++ (GCC) 11.3.0
Prebuilt binaries of this package can be found in the arch4edu repository.
@petronny Yes, it is clear that you need qt5-base because the error message showed that it needed libQt5Test.so.5. But this error was triggered by import torch.
I have verified that in a clean chroot environment, import torch is not working after just installing python-pytorch-cuda. So it seems that the dependencies of some packages in the official repo are not properly set...
I think the reason is qt5-base is not added to makedepends.
I can build the package with qt5-base in makedepends.
@petronny I think there is something wrong with your python-pytorch-cuda because the error message indicated that import torch failed...
Could you try python -c 'import torch; print(torch.__version__)' to verify that you can import it without the error?
I will add python-pip into the makedepends later.
Getting
==> Starting build()...
==> Building Python 3
Traceback (most recent call last):
File "setup.py", line 1, in <module>
import torch
File "/usr/lib/python3.7/site-packages/torch/__init__.py", line 81, in <module>
from torch._C import *
ImportError: libQt5Test.so.5: cannot open shared object file: No such file or directory
==> ERROR: A failure occurred in build().
now.
There might be something missing in makedepends.
There is one more thing to fix to pass the build.
==> Building Python 3
Traceback (most recent call last):
File "setup.py", line 5, in <module>
from pip._internal import main as pipmain
ModuleNotFoundError: No module named 'pip'
==> ERROR: A failure occurred in build().
Aborting...
Please add python-pip to makedepends.
@petronny you're right. Fixed.
No CUDA runtime is found, using CUDA_HOME='/opt/cuda'
Traceback (most recent call last):
File "setup.py", line 64, in <module>
check_cuda_torch_binary_vs_bare_metal(torch.utils.cpp_extension.CUDA_HOME)
File "setup.py", line 43, in check_cuda_torch_binary_vs_bare_metal
torch_binary_major = torch.version.cuda.split(".")[0]
AttributeError: 'NoneType' object has no attribute 'split'
Please change python-pytorch to python-pytorch-cuda.
Pinned Comments
petronny commented on 2022-08-31 06:43 (UTC)
Prebuilt binaries of this package can be found in the arch4edu repository.