-
Notifications
You must be signed in to change notification settings - Fork 1.9k
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
Update converter TF version to 2.12.0 #7615
Conversation
This is failing presubmits because tf 2.12.0 requires python>=3.8.0, and the docker container changes in this PR have not been published yet. |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Thanks Matt, interestingly the python nightly test passed with this change, since it depends on the python3_wheel target.
/~https://github.com/tensorflow/tfjs/blob/master/tfjs-converter/python/BUILD.bazel#L68
I believe this part needs to reflect the same change you made to requirements.txt
Good catch, @pyu10055. I've changed the dependencies listed in the BUILD target to match requirements.txt. Converter python nightly tests - Note these are using the new docker with Python 3.9. |
Update tfjs-converter python dependencies to use tensorflow>=2.12.0. This is required for protobuf versions above 3.20.x. Otherwise, it throws
TypeError: Descriptors cannot not be created directly.
See this stackoverflow for more details.Orbax requires at least python 3.8.0. Update the CI docker from buster to bullseye, which uses python 3.9.
Additionally, make the converter python nightly tests cloudbuild file use the release docker when running tests.
To see the logs from the Cloud Build CI, please join either our discussion or announcement mailing list.