[BE] Test CrossEntropyLoss for torch.half

[malfet]

To test it on MPS and CUDA devices
Also, move some float64 skip-tests for MPS to xfail, same as CPU tests for torch.half

[pytorch-bot]

🔗 Helpful Links

🧪 See artifacts and rendered test results at hud.pytorch.org/pr/116681

• 📄 Preview Python docs built from this PR
• 📄 Preview C++ docs built from this PR
• ❓ Need help or want to give feedback on the CI? Visit the bot commands wiki or our office hours

Note: Links to docs will display an error until the docs builds have been completed.

✅ No Failures

As of commit 46cf278 with merge base 3fe437b ():
💚 Looks good so far! There are no failures yet. 💚

This comment was automatically generated by Dr. CI and updates every 15 minutes.

[malfet]

@pytorchbot drci

[mikaylagawarecki]

Thanks for fixing the skip --> expected failure, will submit a follow up PR to fix some other skips in the file to expected failure

[mikaylagawarecki]

For this case, this isn't that relevant since CPU does not have a float16 implementation for CrossEntropyLoss, but I wanted to note that test_cpu_gpu_parity is not tested for MPS as the test is decorated with @onlyCUDA,

pytorch/test/test_modules.py

Lines 512 to 517 in 2ae55e9

	@onlyCUDA
	@with_tf32_off # Turn off TF32 to compute at full precision https://github.com/pytorch/pytorch/issues/86798
	@toleranceOverride({torch.float32: tol(5e-2, 0),
	torch.float64: tol(4e-4, 0)})
	@modules(module_db)
	def test_cpu_gpu_parity(self, device, dtype, module_info, training):

In general, do we want to enable this test for MPS as well? (I can do so in a followup if we should)

[malfet]

Sure, let me submit a followup PR (as I guess it will require few more suppressions on MPS side)

[malfet]

@pytorchbot merge

[pytorchmergebot]

Merge started

Your change will be merged once all checks pass (ETA 0-4 Hours).

Learn more about merging in the wiki.

Questions? Feedback? Please reach out to the PyTorch DevX Team

Advanced Debugging

Check the merge workflow status
here

[pytorchmergebot]

Merge failed

Reason: 2 mandatory check(s) failed. The first few are:

• pull / linux-focal-cuda12.1-py3.10-gcc9 / test (default, 1, 5, linux.4xlarge.nvidia.gpu)
• pull / linux-focal-cuda12.1-py3.10-gcc9-sm86 / test (default, 1, 5, linux.g5.4xlarge.nvidia.gpu)

Dig deeper by viewing the failures on hud

Details for Dev Infra team

Raised by workflow job

Failing merge rule: Core Maintainers

[malfet]

@pytorchbot merge

[pytorchmergebot]

Merge started

Your change will be merged once all checks pass (ETA 0-4 Hours).

Learn more about merging in the wiki.

Questions? Feedback? Please reach out to the PyTorch DevX Team

Advanced Debugging

Check the merge workflow status
here