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
remove componentwise comparison of complex values in TestCase.assertEqual #63572
Conversation
…qual [ghstack-poisoned]
🔗 Helpful links
💊 CI failures summary and remediationsAs of commit 2b0074a (more details on the Dr. CI page):
🕵️ 3 new failures recognized by patternsThe following CI failures do not appear to be due to upstream breakages: pytorch_pure_torch_linux_xenial_py3_6_gcc5_4_build (1/3)Step: "(Optional) Merge target branch" (full log | diagnosis details | 🔁 rerun)
|
…qual ghstack-source-id: cb9f2ff52c2f62fc3fd72290370667f4854b1367 Pull Request resolved: #63572
…ase.assertEqual" Addresses #61906. Issue will be fixed later in the stack when `torch.testing.assert_close` got the same treatment. [ghstack-poisoned]
…qual ghstack-source-id: 476d75b23a46fad7a8baeb0e8395a79824b774f5 Pull Request resolved: #63572
There is one issue that materializes for example in >>> t = torch.tensor(0, dtype=torch.complex32)
>>> torch.allclose(t, t)
RuntimeError: "eq_cpu" not implemented for 'ComplexHalf'
>>> t = t.cuda()
>>> torch.allclose(t, t)
RuntimeError: "eq_cuda" not implemented for 'ComplexHalf' Previously one |
…ase.assertEqual" Addresses #61906. Issue will be fixed later in the stack when `torch.testing.assert_close` got the same treatment. [ghstack-poisoned]
…qual ghstack-source-id: edde9eb62299b0cf4e8fdeb22ab47c06f8dd89b2 Pull Request resolved: #63572
Some (one?) of the test failures seem real:
A fix for this could be to implement allclose for complexhalf or to upcast complexhalf tensors in assertEqual/compareTensors/assert_close. |
cc @zasdfgbnm |
It is not >>> t = torch.tensor(0, dtype=torch.complex32)
>>> t == t
RuntimeError: "eq_cpu" not implemented for 'ComplexHalf'
>>> t = t.cuda()
>>> t == t
RuntimeError: "eq_cuda" not implemented for 'ComplexHalf' Feels weird that this fundamental method is not there. |
There's bare minimum support for As for the failing tests for this PR, it's safe to disable any test for |
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.
Cool
@mruberry has imported this pull request. If you are a Facebook employee, you can view this diff on Phabricator. |
Stack from ghstack:
Addresses #61906. Issue will be fixed later in the stack when
torch.testing.assert_close
got the same treatment.cc @ezyang @gchanan
Differential Revision: D30633527