-
Notifications
You must be signed in to change notification settings - Fork 30.9k
TF: XLA Logits Warpers #16899
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
TF: XLA Logits Warpers #16899
Conversation
|
The documentation is not available anymore as the PR was closed or merged. |
6f8b045 to
3ececfd
Compare
|
@patrickvonplaten Sorry, I know you've already reviewed this, but I'm going to re-request your review. I realized the tests were much easier to understand (and with fewer lines) if they were parametrized, instead of having two tests (one for XLA, another for non-XLA) with shared code 😅 |
|
|
||
| def _get_repetition_penalty_inputs(self): | ||
| @parameterized.expand([(False,), (True,)]) | ||
| def test_repetition_penalty_dist_process(self, use_xla): |
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.
changed this one from a two-test format (added in the previous PR) to the parametrized format
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.
Looks good to me!
(Also you don't actually have to add my suggestions, I'm just being a jerk)
Co-authored-by: Matt <Rocketknight1@users.noreply.github.com>
Co-authored-by: Matt <Rocketknight1@users.noreply.github.com>
What does this PR do?
This PR enables XLA on the logits warpers... which actually needed no changes. In essence, it adds XLA tests to ensure we don't regress.