-
Notifications
You must be signed in to change notification settings - Fork 24.3k
[BE] Parametrize test_min_max
#144249
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
[BE] Parametrize test_min_max
#144249
Conversation
🔗 Helpful Links🧪 See artifacts and rendered test results at hud.pytorch.org/pr/144249
Note: Links to docs will display an error until the docs builds have been completed. ⏳ 10 Pending, 1 Unrelated FailureAs of commit 33eb6eb with merge base f6488d8 ( BROKEN TRUNK - The following job failed but were present on the merge base:👉 Rebase onto the `viable/strict` branch to avoid these failures
This comment was automatically generated by Dr. CI and updates every 15 minutes. |
test_min_max
- `dtype` was not passed as argument to `torch.rand` before - Condition bfloat16 testing on MacOS14+ Pull Request resolved: #144250 Approved by: https://github.com/Skylion007 ghstack dependencies: #144249
Previosly all bitwise operations were aliased to the same type, but this is wrong for shift ops Rather than building an overly complex logic, let's just instantiate using shared `scalarToMetalTypeString` helper function Fixes #144190 Pull Request resolved: #144251 Approved by: https://github.com/Skylion007 ghstack dependencies: #144249, #144250
Previosly all bitwise operations were aliased to the same type, but this is wrong for shift ops Rather than building an overly complex logic, let's just instantiate using shared `scalarToMetalTypeString` helper function Fixes #144190 Pull Request resolved: #144251 Approved by: https://github.com/Skylion007 ghstack dependencies: #144249, #144250 (cherry picked from commit e56768f)
[MPS] Fix bitwise shifts for uint8 (#144251) Previosly all bitwise operations were aliased to the same type, but this is wrong for shift ops Rather than building an overly complex logic, let's just instantiate using shared `scalarToMetalTypeString` helper function Fixes #144190 Pull Request resolved: #144251 Approved by: https://github.com/Skylion007 ghstack dependencies: #144249, #144250 (cherry picked from commit e56768f) Co-authored-by: Nikita Shulga <nikita.shulga@gmail.com>
@pytorchbot cherry-pick --onto release/2.6 -c foobar |
❌ 🤖 pytorchbot command failed:
Try |
@pytorchbto cherry-pick --onto release/2.6 -c fixnewfeature |
@pytorchbot cherry-pick --onto release/2.6 -c fixnewfeature --fixes "test failures on release" |
It's better to have one unit test per dtype rather a combined one Pull Request resolved: #144249 Approved by: https://github.com/Skylion007 (cherry picked from commit 11a0663)
Cherry picking #144249The cherry pick PR is at #144814 and it is linked with issue test failures on release. The following tracker issues are updated: Details for Dev Infra teamRaised by workflow job |
[BE] Parametrize `test_min_max` (#144249) It's better to have one unit test per dtype rather a combined one Pull Request resolved: #144249 Approved by: https://github.com/Skylion007 (cherry picked from commit 11a0663) Co-authored-by: Nikita Shulga <nikita.shulga@gmail.com>
Stack from ghstack (oldest at bottom):
test_min_max_nan_propagation
#144250test_min_max
#144249It's better to have one unit test per dtype rather a combined one