🔗 Helpful Links

🧪 See artifacts and rendered test results at hud.pytorch.org/pr/159556

• 📄 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.

❌ 38 New Failures

As of commit 2155398 with merge base 178515d ():

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

This PR needs a release notes: label

If your changes are user facing and intended to be a part of release notes, please use a label starting with release notes:.

If not, please add the topic: not user facing label.

To add a label, you can comment to pytorchbot, for example
@pytorchbot label "topic: not user facing"

For more information, see
https://github.com/pytorch/pytorch/wiki/PyTorch-AutoLabel-Bot#why-categorize-for-release-notes-and-how-does-it-work.

Test case?

Thanks for the PR -- I noticed that topk, sort, minimum, maximum all have this behavior. Perhaps it makes sense to extend this PR to those functions as well?

Here are some minimal tests to get one started:

from torch.testing._internal.common_methods_invocations import op_db
from torch.testing._internal.common_device_type import ops

ordered_op_names = {'sort', 'topk', 'lt', 'argmin', 'le', 'ge', 'amax', 'maximum', 'minimum', 'clamp', 'amin', 'gt', 'ceil', 'argmax', 'floor'}
ordered_op_db = tuple(filter(lambda op: op.name in ordered_op_names, op_db))

...
    @ops(ordered_op_db, dtypes=[torch.complex32, torch.complex64, torch.complex128])
    def test_ordered_raises(self, device, dtype, op: OpInfo):
        sample_inputs = op.sample_inputs(device, dtype)

        for sample_input in sample_inputs:
            self.assertRaises(
                NotImplementedError,
                op,
                sample_input.input,
                *sample_input.args,
                **sample_input.kwargs,
            )

I've added the other functions. Sort had a TORCH_CHECK_VALUE instead of a TORCH_CHECK , so to make consistent with the others I've updated that, otherwise it raised a ValueError rather than a RuntimeError,

Not sure what that build error is about?

CI failures are related

It seems some of the operations are implemented for bool but not complex -- we should not error on bool for those.