-
Notifications
You must be signed in to change notification settings - Fork 150
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
Add support to release linux aarch64 wheels #114
Comments
hi @odidev , our wheels releasing build environment is based on pypa image qq, are you using mlmd with tfx? if so, consider to open a bug on tensorflow/tfx, currently all wheels images are shared between tfx projects. |
I'd appreciate if it would be possible to widen the platform support to ppc64le as well. However the same requirement also counts for that arch, as it is needs |
@hughmiao, as per issue comment tensorflow/tfx#3829 (comment), ml-metadata has dropped tf dependency but here https://github.com/google/ml-metadata/blob/master/ml_metadata/tools/docker_build/Dockerfile.manylinux2010#L18 it looks like build tool requires tf image. Please correct me if I am wrong or let me know if you are using a different wheel builder. |
@BrianSong, can you please suggest on this comment: #114 (comment) |
Hi @odidev, Yes, MLMD OSS has dropped its tf dependency. The image you pointed out above is just a wheel image that we share between tfx projects. Do you encounter any problem when using it? |
Hi @BrianSong, |
Currently this is under our radar but not in a high priority. We can easily update the image to manylinux2014, but as we are testing the build packages natively(without container), so we need put some effort to figure out a way to test aarch64 wheels in this setup(by running tests inside the container image.) I wonder does Building manually from source work for you? |
This is working fine but the availability of the wheel will save a lot of time of Linux AArch64 users. |
Problem
On aarch64, pip install ml-metadata builds the wheels from source code and then install it. It requires user to have development environment installed on his system. also, it take some time to build the wheels than downloading and extracting the wheels from pypi.
Resolution
On aarch64, pip install ml-metadata should download the wheels from pypi
@hughmiao, Please let me know your interest on releasing aarch64 wheels. I can help in this.
The text was updated successfully, but these errors were encountered: