Linux aarch64 wheels for tensorflow-addons

[ameyp]

It would be great if the package on pip included a manylinux wheel with the aarch64 platform. This would enable using this package on e.g. AWS Graviton instances.

Relevant information

• Are you willing to contribute it (yes/no): yes no
• Are you willing to maintain it going forward? (yes/no): no
• Is there a relevant academic paper? (if so, where): no
• Does the relavent academic paper exceed 50 citations? (yes/no): N/A
• Is there already an implementation in another framework? (if so, where): No new implementation would presumably be needed, just a change to the Github workflow, and maybe to the build process.
• Was it part of tf.contrib? (if so, where): N/A

Which API type would this fall under (layer, metric, optimizer, etc.): N/A

Who will benefit with this feature?: Users deploying to arm64 linux machines

Any other info.

[ameyp]

Good idea? Terrible idea?

[ameyp]

Ping.

[bhack]

We need to wait the next SIG meeting to understand what we want to do with new arch wheels.

[ameyp]

Has the date for the next SIG meeting been decided?

[ameyp]

Ping.

[bhack]

The last SIG meeting was skipped few days ago. /cc @seanpmorgan

[ameyp]

When's the next meeting then, and will this issue be discussed during that meeting?

[seanpmorgan]

When's the next meeting then, and will this issue be discussed during that meeting?

Next meeting is tomorrow at 11am PT. I've added this to the agenda. Apologies for the delay.
https://docs.google.com/document/d/1kxg5xIHWLY7EMdOJCdSGgaPu27a9YKpupUz2VTXqTJg/edit#

[ameyp]

What was the outcome of the meeting?

[ameyp]

@seanpmorgan ?

[gfalcone]

Hello ! any news on this ? what was the output of the last SIG meeting ?

[seanpmorgan]

Want to apologize for the lack of responses here. As you can see from our recent announcement TFA will be ending new contributions and will transition to a minimal maintenance and release mode for the next year:
#2807

If you still have interest in contributing this, I'm okay with this being the last feature that we would support including ( as we discussed at our last SIG Meeting).

[ameyp]

We moved off the direct dependency that was introducing an indirect dependency on this package. Sorry to hear about the package entering minimal maintenance mode.

[brainchip-india]

Hey @seanpmorgan, hope you're doing great.
any update on this I couldn't able to install tensorflow-addons on aarch64.