Yong Tang

Binaries: ``` $ sha256sum *.whl b7f606f7be35c5d303ddd60f34e8efbb787ea2735078c0d48b70034ed2c2d808 tensorflow_io-0.28.0-cp310-cp310-macosx_10_14_x86_64.whl 3e0bb7e2678e8b726f2511068348f352cc13e7e309145c1c55f3479a2276c937 tensorflow_io-0.28.0-cp310-cp310-manylinux_2_12_x86_64.manylinux2010_x86_64.whl 3a8b2fbe85c4965748170ad65dcc99d1494dc09f23cb20ef15514240337f9f42 tensorflow_io-0.28.0-cp310-cp310-win_amd64.whl fa67977ed4472c41dd505a1c6169ceb8e84fc9535f538e1e4ac5125bcdea941c tensorflow_io-0.28.0-cp311-cp311-macosx_10_14_x86_64.whl 8ad743479d4058d49710df95e6aaf03032462beeded49c1387d5e2ba365ef9ad tensorflow_io-0.28.0-cp311-cp311-manylinux_2_12_x86_64.manylinux2010_x86_64.whl 195d2324019dbf0e6519ef30cf190f517fba5bd17eb9d0fa988da67dddb4753c tensorflow_io-0.28.0-cp311-cp311-win_amd64.whl 240d35f3ab656fbc23761646fc0e03158d8fc9d30239ccdeaccff174b572fb51 tensorflow_io-0.28.0-cp37-cp37m-macosx_10_14_x86_64.whl 9d006d62aa2696191722e4e6bff50c6175bbda066391f03f2399913840b42de6 tensorflow_io-0.28.0-cp37-cp37m-manylinux_2_12_x86_64.manylinux2010_x86_64.whl 5574f4a9924d5cca8a1ba8267dd5de0b40d0994096f5f1233a9062bb673605d8 tensorflow_io-0.28.0-cp37-cp37m-win_amd64.whl 1842f1a70c175959cd8cfeb686042b198374969943a91cfec0d7820987849c96 tensorflow_io-0.28.0-cp38-cp38-macosx_10_14_x86_64.whl f53e9b8cd44f206e90dd5e349b70f6c009d2a683b787cde6f039a66e36962aaa tensorflow_io-0.28.0-cp38-cp38-manylinux_2_12_x86_64.manylinux2010_x86_64.whl 0dce55dca0c48a28a0c0bd65114424e6cf07369d6f84da95c40a802d370ad563 tensorflow_io-0.28.0-cp38-cp38-win_amd64.whl ce8b2105b06118d5a384f592b6467ba917cae9384997013c47e43aaf36ca3a0c...

Release: https://pypi.org/project/tensorflow-io-gcs-filesystem/0.28.0/ https://pypi.org/project/tensorflow-io/0.28.0/ https://github.com/tensorflow/io/releases/tag/v0.28.0

aarch64: ``` $ sha256sum *.whl 23b6ec4ee269c6f900cbd583dc480800095981d46df92f34defda7647322381c tensorflow_io-0.28.0-cp310-cp310-manylinux_2_17_aarch64.manylinux2014_aarch64.whl 97e23e804df19390bd60c51bef1e312f3a51c2e610322fbf0bdf7a1331c70238 tensorflow_io-0.28.0-cp37-cp37m-manylinux_2_17_aarch64.manylinux2014_aarch64.whl 0df8219a6eee315dde1c7ea8acbef5e82538055b6f2fe148da929e906d46f49b tensorflow_io-0.28.0-cp38-cp38-manylinux_2_17_aarch64.manylinux2014_aarch64.whl d9e12fba20196cee6d4129995f63bee4211fe4bc16e99bfb9c1a9bc5299cce38 tensorflow_io-0.28.0-cp39-cp39-manylinux_2_17_aarch64.manylinux2014_aarch64.whl 9cf11d36d5aac6365f9f6c4bc71eb23346a2f5dc139c91af1e2ea06cfd7f69a7 tensorflow_io_gcs_filesystem-0.28.0-cp310-cp310-manylinux_2_17_aarch64.manylinux2014_aarch64.whl b8f4b1b83ee28906881221c0a40822cbf30ebfcc3e96430f91f396309e6b452c tensorflow_io_gcs_filesystem-0.28.0-cp37-cp37m-manylinux_2_17_aarch64.manylinux2014_aarch64.whl 3ebb4724594bb897f74c637c256cb9509b25fc6b83860a2a6e2a6d42fd1a1a2a tensorflow_io_gcs_filesystem-0.28.0-cp38-cp38-manylinux_2_17_aarch64.manylinux2014_aarch64.whl 47944fba38f46c9137b7bc27fb293953249d437f225bf46b96bac6a177bdfe04 tensorflow_io_gcs_filesystem-0.28.0-cp39-cp39-manylinux_2_17_aarch64.manylinux2014_aarch64.whl ```

@johnthagen my fault - was out sick for several weeks and didn't see this in notification. The wheels have been uploaded now.

Let me reopen this one for the following comment: > Do we plan to use GPU operation like > https://docs.nvidia.com/cuda/npp/group__nv12torgb.html#details?

@bhack We are still in exploration mode.. but will circle back soon.

I think this will also help reduce the overall cost of compute resources on GitHub actions, as many open source projects can minimize the GitHub actions minutes they use for...

I believe this issue has been fixed in the latest nightly build of `tf-nightly==2.12.0.dev20230108`. Will close this issue, but feel free to re-open if this issue persist.

I think one way to address the issue without break the compatibility is to have the default value of `--health-start-period` as the `--interval` (currently `--health-start-period=0` by default). I will take...

Created a PR #33918 with the following changes: 1. The first healthcheck is performed on startup (at time 0) 2. The first healthcheck is only valid when start-period < 0...