Skip to content

Conversation

lezcano
Copy link
Collaborator

@lezcano lezcano commented May 27, 2022

As per title

[ghstack-poisoned]
@facebook-github-bot
Copy link
Contributor

facebook-github-bot commented May 27, 2022

🔗 Helpful links

❌ 1 New Failures

As of commit 7ab07c0 (more details on the Dr. CI page):

Expand to see more
  • 1/1 failures introduced in this PR

🕵️ 1 new failure recognized by patterns

The following CI failures do not appear to be due to upstream breakages

See GitHub Actions build pull / linux-focal-py3.7-gcc7 / test (backwards_compat, 1, 1, linux.2xlarge) (1/1)

Step: "Test" (full log | diagnosis details | 🔁 rerun)

2022-05-31T10:03:26.9945532Z The PR is introduc...m to confirm whether this change is wanted or not.
2022-05-31T10:03:26.9933349Z processing existing schema:  text(__torch__.torch.classes.profiling.SourceRef _0) -> str _0
2022-05-31T10:03:26.9934412Z processing existing schema:  count(__torch__.torch.classes.profiling.InstructionStats _0) -> int _0
2022-05-31T10:03:26.9935340Z processing existing schema:  duration_ns(__torch__.torch.classes.profiling.InstructionStats _0) -> int _0
2022-05-31T10:03:26.9936764Z processing existing schema:  source(__torch__.torch.classes.profiling.SourceStats _0) -> __torch__.torch.classes.profiling.SourceRef _0
2022-05-31T10:03:26.9938569Z processing existing schema:  line_map(__torch__.torch.classes.profiling.SourceStats _0) -> Dict(int, __torch__.torch.classes.profiling.InstructionStats) _0
2022-05-31T10:03:26.9939634Z processing existing schema:  __init__(__torch__.torch.classes.profiling._ScriptProfile _0) -> NoneType _0
2022-05-31T10:03:26.9940839Z processing existing schema:  enable(__torch__.torch.classes.profiling._ScriptProfile _0) -> NoneType _0
2022-05-31T10:03:26.9941816Z processing existing schema:  disable(__torch__.torch.classes.profiling._ScriptProfile _0) -> NoneType _0
2022-05-31T10:03:26.9943560Z processing existing schema:  _dump_stats(__torch__.torch.classes.profiling._ScriptProfile _0) -> __torch__.torch.classes.profiling.SourceStats[] _0
2022-05-31T10:03:26.9944968Z processing existing schema:  __init__(__torch__.torch.classes.dist_rpc.WorkerInfo _0, str _1, int _2) -> NoneType _0
2022-05-31T10:03:26.9945532Z The PR is introducing backward incompatible changes to the operator library. Please contact PyTorch team to confirm whether this change is wanted or not. 
2022-05-31T10:03:26.9945563Z 
2022-05-31T10:03:26.9945677Z Broken ops: [
2022-05-31T10:03:26.9945958Z 	aten::special_laguerre_polynomial_l(Tensor x, Tensor n) -> Tensor
2022-05-31T10:03:26.9946193Z 	aten::special_laguerre_polynomial_l.out(Tensor x, Tensor n, *, Tensor(a!) out) -> Tensor(a!)
2022-05-31T10:03:26.9946444Z 	aten::special_laguerre_polynomial_l.n_scalar_out(Tensor x, Scalar n, *, Tensor(a!) out) -> Tensor(a!)
2022-05-31T10:03:26.9946648Z 	aten::special_laguerre_polynomial_l.x_scalar(Scalar x, Tensor n) -> Tensor
2022-05-31T10:03:26.9946895Z 	aten::special_laguerre_polynomial_l.x_scalar_out(Scalar x, Tensor n, *, Tensor(a!) out) -> Tensor(a!)
2022-05-31T10:03:26.9947100Z 	aten::special_laguerre_polynomial_l.n_scalar(Tensor x, Scalar n) -> Tensor
2022-05-31T10:03:26.9947291Z 	aten::special_hermite_polynomial_he(Tensor x, Tensor n) -> Tensor
2022-05-31T10:03:26.9947523Z 	aten::special_hermite_polynomial_he.out(Tensor x, Tensor n, *, Tensor(a!) out) -> Tensor(a!)

This comment was automatically generated by Dr. CI (expand for details).

Please report bugs/suggestions to the (internal) Dr. CI Users group.

Click here to manually regenerate this comment.

@lezcano lezcano added module: testing Issues related to the torch.testing module (not tests) topic: not user facing topic category labels May 27, 2022
@ngimel
Copy link
Collaborator

ngimel commented May 27, 2022

What's up with test failures? They don't seem related, but also trunk doesn't look broken.

Copy link
Collaborator

@ngimel ngimel left a comment

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

This looks good, but what's up with test failures?

kwargs=kwargs))

return inputs
yield SampleInput(t.detach().requires_grad_(requires_grad), args=args, kwargs=kwargs)
Copy link
Collaborator

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

wow previous invocation seems buggy, bringing unneeded history.

Copy link
Collaborator Author

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

I do not understand how the previous implementation worked tbh... I should look further into that.

This little fix needed some extra fixes here and there (hence the test failures). I needed to convert quite a few other functions into generators after this one. Let's see if I managed to not miss any one.

@@ -1381,6 +1378,7 @@ def __init__(
# These are mutually exclusive options
assert not (result_dtype and promotes_int_to_float)
assert not (result_dtype and promotes_int_to_int64)
assert not (result_dtype and complex_to_real)
Copy link
Collaborator

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

but vector_norm accepts dtype arg?

Copy link
Collaborator Author

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

result_dtype is used to specify when a function always returns a given dtype (e.g. eq).

facebook-github-bot pushed a commit that referenced this pull request Jun 2, 2022
Summary:
As per title

Pull Request resolved: #78423

Approved by: https://github.com/ngimel

Test Plan: contbuild & OSS CI, see https://hud.pytorch.org/commit/pytorch/pytorch/9b97d5d625b02ed27a22d9e3c981caa4b1dea5c7

Reviewed By: seemethere

Differential Revision: D36815705

Pulled By: seemethere

fbshipit-source-id: d9e59c329bc63e56719ae39c07c71d8d656b198b
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
cla signed module: testing Issues related to the torch.testing module (not tests) open source topic: not user facing topic category
Projects
None yet
Development

Successfully merging this pull request may close these issues.

4 participants