PSRoiAlign: SymInt support + meta-implem#8058
Conversation
test/optests_failures_dict.json
Outdated
| "TestPSRoIAlign.test_autograd_registration__test_backward[True-mps-0]": { | ||
| "comment": "NotImplementedError: autograd_registration_check: NYI devices other than CPU/CUDA, got {'mps'}", | ||
| "status": "xfail" |
There was a problem hiding this comment.
CC @zou3519
I just opened pytorch/pytorch#111797 which I believe could be a fix for the problem I'm facing here:
This test (and a bunch of others)
Lines 186 to 189 in 3fb88b3
is parametrized over cpu, CUDA and MPS, and fails on MPS.
I'm trying to xfail the MPS parametrization as above (lines 10-12), but optests complains with:
E RuntimeError: In failures dict, got test name 'TestPSRoIAlign.test_autograd_registration__test_backward[True-mps-0]'. We parsed this as running test 'test_autograd_registration' on 'test_backward[True-mps-0]', but test_backward[True-mps-0] does not exist on the TestCase 'TestPSRoIAlign]. Maybe you need to change the test name?
The problem is, if I replace TestPSRoIAlign.test_autograd_registration__test_backward[True-mps-0] with TestPSRoIAlign.test_autograd_registration__test_backward in line 10, then I'm getting and "unexpected success":
E torch.testing._internal.optests.generate_tests.OpCheckError: generate_opcheck_tests: Unexpected success for operator torchvision::ps_roi_align on test TestPSRoIAlign.test_autograd_registration__test_backward. This may mean that you have fixed this test failure. Please rerun the test with PYTORCH_OPCHECK_ACCEPT=1 to automatically update the test runner or manually remove the expected failure in the failure dict at /home/nicolashug/dev/vision/test/optests_failures_dict.jsonFor more details, see https://docs.google.com/document/d/1Pj5HRZvdOq3xpFpbEjUZp2hBovhy7Wnxw14m6lF2154/edit
There was a problem hiding this comment.
@NicolasHug can the tests be run under unittest, or are they pytest only?
|
The MPS checks are going to fail until pytorch/pytorch#112251 is resovled on torch core's side. I don't know how long it will take, so I'm going to remove the opchecks from this PR (and from the other ones) so that we can still have symint / torch.compile support on these ops (albeit temporarily untested) |
Reviewed By: vmoens Differential Revision: D50789103 fbshipit-source-id: 5da312242bdd9f01b9ee4d45f791ee564add9f68
Towards #8055
cc @pmeier