Torch: test on 2.0 and latest versions + explicitly load with weights_only=True
#2488
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
To merge before #2440 (needed to test it). Two things in this PR:
we are now running the torch-test CI with
torch~=2.0
andtorch=="latest"
. This is to ensure a sense of backward-compatibility and still testing on latest updates.2.0
was chosen arbitrarily. It has been released ~15 months ago. It doesn't meanhuggingface_hub
does not supporttorch 1.0
(we would have received reports^^), just that we are not guaranteeing it.Pytorch tests are currently failing in CI with this
FutureWarning
:This is actually a good opportunity to add
weights_only=True
explicitly in the Pytorch hub mixin when loading from an unsafe pickle. Note that the default file type is still.safetensors
when saving / loading weights.Expectations: a green CI and relaxed users!