Jamba: update integration tests #32250
Merged
Conversation
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
|
cc @ydshieh |
gante
commented
Jul 26, 2024
Comment on lines
+653
to
+655
| # This variable is used to determine which CUDA device are we using for our runners (A10 or T4) | ||
| # Depending on the hardware we get different logits / generations | ||
| cuda_compute_capability_major_version = None |
There was a problem hiding this comment.
This cuda_compute_capability_major_version pattern is copied from other models like e.g. gemma
|
The docs for this PR live here. All of your documentation changes will be reflected on that endpoint. The docs are available until 30 days after the last update. |
|
(thank you for trigger the tests on the runner 🙏 ) |
amyeroberts
approved these changes
Aug 7, 2024
Co-authored-by: amyeroberts <[email protected]>
Co-authored-by: amyeroberts <[email protected]>
ydshieh
reviewed
Aug 22, 2024
| torch.testing.assert_close(logits[0, -1, :40].cpu(), EXPECTED_LOGITS_NO_GRAD_0, rtol=1e-3, atol=1e-3) | ||
| torch.testing.assert_close(logits[1, -1, :40].cpu(), EXPECTED_LOGITS_NO_GRAD_1, rtol=1e-3, atol=1e-3) | ||
| # TODO: there are significant differences in the logits across major cuda versions, which shouldn't exist | ||
| if self.cuda_compute_capability_major_version == 8: |
There was a problem hiding this comment.
maybe better to use
self.skipTest(reason="Skipping for T4 runners because ...")
There was a problem hiding this comment.
oops, merged before seeing this comment!
You have a good point, in fact we should split the test in two to test (/skip) the logits separately
zucchini-nlp
pushed a commit
to zucchini-nlp/transformers
that referenced
this pull request
Aug 30, 2024
* try test updates * a few more changes * a few more changes * a few more changes * [run slow] jamba * skip logits checks on older gpus * [run slow] jamba * oops * [run slow] jamba * Update tests/models/jamba/test_modeling_jamba.py Co-authored-by: amyeroberts <[email protected]> * Update tests/models/jamba/test_modeling_jamba.py Co-authored-by: amyeroberts <[email protected]> --------- Co-authored-by: amyeroberts <[email protected]>
zucchini-nlp
pushed a commit
to zucchini-nlp/transformers
that referenced
this pull request
Aug 30, 2024
* try test updates * a few more changes * a few more changes * a few more changes * [run slow] jamba * skip logits checks on older gpus * [run slow] jamba * oops * [run slow] jamba * Update tests/models/jamba/test_modeling_jamba.py Co-authored-by: amyeroberts <[email protected]> * Update tests/models/jamba/test_modeling_jamba.py Co-authored-by: amyeroberts <[email protected]> --------- Co-authored-by: amyeroberts <[email protected]>
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
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.
What does this PR do?
🟢 Fixes
generate-related integration tests for jamba 🟢I've checked them against:
Detective work 🕵️
ai21labs/Jamba-tiny-random, the generation text quality doesn't matter.