Skip to content

Fix ML presubmits. #38289

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.

Sign up for GitHub

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

Open
claudevdm wants to merge 1 commit into
base: master
Choose a base branch
from
+13 −0
Open

Fix ML presubmits. #38289

Changes from all commits
Commits
Show all changes
1 commit
Select commit
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
13 changes: 13 additions & 0 deletions sdks/python/setup.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -185,6 +185,19 @@ def cythonize(*args, **kwargs):
# tensorflow transitive dep, lower versions not compatible with Python3.10+
'absl-py>=0.12.0',
'tensorflow-hub',
# tokenizers 0.23.0rc0 renamed the PyO3 kwarg of
# processors.RobertaProcessing (and BertProcessing) from `cls` to
# `cls_token` -- the rename was a drive-by inside huggingface/tokenizers
# https://github.com/huggingface/tokenizers/pull/1928.
# transformers' slow CLIP tokenizer still calls
# `processors.RobertaProcessing(sep=..., cls=..., ...)` at
# transformers/models/clip/tokenization_clip.py, so model load fails with
# "RobertaProcessing.__new__() got an unexpected keyword argument 'cls'".
# The ml tox envs run with pip_pre=True (tox.ini:32), so even though no
# 0.23 stable has shipped yet, the rc gets resolved.
# Drop this cap once transformers updates the CLIP call site to
# `cls_token=` or tokenizers reinstates `cls=` as a deprecation alias.
'tokenizers<0.23',
'torch',
'transformers',
]
Expand Down
Loading