Add tan_cuda for complex dtypes by anjali411 · Pull Request #38400 · pytorch/pytorch

anjali411 · 2020-05-13T15:35:19Z

Stack from ghstack:

Add tan_cuda for complex dtypes #38400 Add tan_cuda for complex dtypes

Differential Revision: D21572209

[ghstack-poisoned]

dr-ci · 2020-05-13T15:52:23Z

💊 CI failures summary and remediations

As of commit d912829 (more details on the Dr. CI page):

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:

pytorch_linux_backward_compatibility_check_test (1/1)

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

May 15 13:03:35 The PR is introducing backward incompatible changes to the operator library. Please contact PyTorch team to confirm whether this change is wanted or not.

May 15 13:03:35 processing existing schema:  aten::std.dim(Tensor self, int[1] dim, bool unbiased=True, bool keepdim=False) -> (Tensor) 
May 15 13:03:35 processing existing schema:  aten::std.out(Tensor self, int[1] dim, bool unbiased=True, bool keepdim=False, *, Tensor(a!) out) -> (Tensor(a!)) 
May 15 13:03:35 processing existing schema:  aten::std.names_dim(Tensor self, str[1] dim, bool unbiased=True, bool keepdim=False) -> (Tensor) 
May 15 13:03:35 processing existing schema:  aten::std.names_out(Tensor self, str[1] dim, bool unbiased=True, bool keepdim=False, *, Tensor(a!) out) -> (Tensor(a!)) 
May 15 13:03:35 processing existing schema:  aten::std_mean(Tensor self, bool unbiased=True) -> (Tensor, Tensor) 
May 15 13:03:35 processing existing schema:  aten::std_mean.dim(Tensor self, int[1] dim, bool unbiased=True, bool keepdim=False) -> (Tensor, Tensor) 
May 15 13:03:35 processing existing schema:  aten::std_mean.names_dim(Tensor self, str[1] dim, bool unbiased=True, bool keepdim=False) -> (Tensor, Tensor) 
May 15 13:03:35 processing existing schema:  aten::t(Tensor(a) self) -> (Tensor(a)) 
May 15 13:03:35 processing existing schema:  aten::t_(Tensor(a!) self) -> (Tensor(a!)) 
May 15 13:03:35 processing existing schema:  aten::tan_(Tensor(a!) self) -> (Tensor(a!)) 
May 15 13:03:35 The PR is introducing backward incompatible changes to the operator library. Please contact PyTorch team to confirm whether this change is wanted or not.  
May 15 13:03:35  
May 15 13:03:35 Broken ops: [ 
May 15 13:03:35 	profiler::_call_end_callbacks_on_jit_fut(Tensor x, Future(t) y) -> (Future(t)) 
May 15 13:03:35 	aten::_sparse_softmax_backward_data(Tensor grad_output, Tensor output, int dim, Tensor self) -> (Tensor) 
May 15 13:03:35 	aten::_sparse_softmax(Tensor self, int dim, bool half_to_float) -> (Tensor) 
May 15 13:03:35 	aten::_sparse_softmax.int(Tensor self, int dim, int? dtype=None) -> (Tensor) 
May 15 13:03:35 	aten::_sparse_softmax.Dimname(Tensor self, str dim, *, int? dtype=None) -> (Tensor) 
May 15 13:03:35 	aten::_sparse_log_softmax_backward_data(Tensor grad_output, Tensor output, int dim, Tensor self) -> (Tensor) 
May 15 13:03:35 	aten::_sparse_log_softmax(Tensor self, int dim, bool half_to_float) -> (Tensor) 
May 15 13:03:35 	aten::_sparse_log_softmax.int(Tensor self, int dim, int? dtype=None) -> (Tensor)

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

Follow this link to opt-out of these comments for your Pull Requests.

Please report bugs/suggestions on the GitHub issue tracker.

See how this bot performed.

This comment has been revised 41 times.

albanD · 2020-05-13T15:54:27Z

    ('rsqrt', '', lambda t, d: _small_3d(t, d) + 1, lambda t, d: [], 1e-2, 1e-5, 1e-4, _float_types_no_half),
    ('sinh', '', lambda t, d: _small_3d(t, d).clamp(-1, 1), lambda t, d: [], 1e-3, 1e-5, 1e-5, _float_types),
    ('tan', '', lambda t, d: _small_3d(t, d).clamp(-1, 1), lambda t, d: [], 1e-3, 1e-5, 1e-5, _float_types),
+    ('tan', 'complex', lambda t, d: _small_3d(t, d), lambda t, d: [], 1e-3, 1e-5, 1e-5, _complex_types)


I am surprised that _complex_types did not exist yet here. How do we test the other methods that already support complex?

Unfortunately there are no existing tests for these ops for complex dtypes. we should add it soon though ...

albanD · 2020-05-13T15:54:58Z

        test_case.assertEqual(self_variable.size(), self_variable.grad.size())

-separate_complex_tests = ['log', 'log10', 'log1p', 'log2', 'reciprocal']
+separate_complex_tests = ['log', 'log10', 'log1p', 'log2', 'reciprocal', 'tan']


Why is this one in the reparate list? It should work for real numbers as well no?

the test for floating points uses clamp in the input and clamp does not work for complex

[ghstack-poisoned]

ghstack-source-id: f2bada5 Pull Request resolved: #38400

albanD

LGTM

[ghstack-poisoned]

ghstack-source-id: 9f3a926 Pull Request resolved: #38400

Differential Revision: [D21556867](https://our.internmc.facebook.com/intern/diff/D21556867) [ghstack-poisoned]

ghstack-source-id: 6d1d74a Pull Request resolved: #38400

* #38399 Added autograd tests, disabled jit autograd tests for complex and added a separate list for tests for complex dtype only [ghstack-poisoned]

ghstack-source-id: 647e636 Pull Request resolved: #38400

Differential Revision: [D21572209](https://our.internmc.facebook.com/intern/diff/D21572209) [ghstack-poisoned]

ghstack-source-id: dce14f0 Pull Request resolved: #38400

facebook-github-bot · 2020-05-15T16:14:06Z

@anjali411 merged this pull request in 242af6c.

facebook-github-bot · 2020-05-15T16:14:16Z

@anjali411 merged this pull request in 242af6c.

Summary: Pull Request resolved: pytorch#38400 * pytorch#38399 Added autograd tests, disabled jit autograd tests for complex and added a separate list for tests for complex dtype only Test Plan: Imported from OSS Differential Revision: D21572209 Pulled By: anjali411 fbshipit-source-id: 7036029e9f8336139f5d54e0dfff9759f3bf8376

Add tan_cuda for complex dtypes

9cb7331

[ghstack-poisoned]

This was referenced May 13, 2020

Have Device available in torch namespace #38041

Closed

Added autograd tests, disabled jit autograd tests for complex and added a separate list for tests for complex dtype only #38399

Closed

Update on "Add tan_cuda for complex dtypes"

5d3addd

[ghstack-poisoned]

anjali411 requested a review from albanD May 13, 2020 15:42

albanD reviewed May 13, 2020

View reviewed changes

Update on "Add tan_cuda for complex dtypes"

d7a5696

[ghstack-poisoned]

anjali411 added a commit that referenced this pull request May 13, 2020

Add tan_cuda for complex dtypes

5fa40c6

ghstack-source-id: f2bada5 Pull Request resolved: #38400

albanD approved these changes May 13, 2020

View reviewed changes

Update on "Add tan_cuda for complex dtypes"

ac0d159

[ghstack-poisoned]

Update on "Add tan_cuda for complex dtypes"

868ae38

[ghstack-poisoned]

anjali411 added a commit that referenced this pull request May 13, 2020

Add tan_cuda for complex dtypes

009e7a1

ghstack-source-id: 9f3a926 Pull Request resolved: #38400

Update on "Add tan_cuda for complex dtypes"

3b5aabe

Differential Revision: [D21556867](https://our.internmc.facebook.com/intern/diff/D21556867) [ghstack-poisoned]

anjali411 added a commit that referenced this pull request May 14, 2020

Add tan_cuda for complex dtypes

d65e09b

ghstack-source-id: 6d1d74a Pull Request resolved: #38400

Update on "Add tan_cuda for complex dtypes"

3efbdcf

* #38399 Added autograd tests, disabled jit autograd tests for complex and added a separate list for tests for complex dtype only [ghstack-poisoned]

anjali411 added a commit that referenced this pull request May 14, 2020

Add tan_cuda for complex dtypes

ef903c5

ghstack-source-id: 647e636 Pull Request resolved: #38400

krshrimali mentioned this pull request May 14, 2020

Add tanh_cuda support for complex types #38486

Closed

Update on "Add tan_cuda for complex dtypes"

d912829

Differential Revision: [D21572209](https://our.internmc.facebook.com/intern/diff/D21572209) [ghstack-poisoned]

anjali411 added a commit that referenced this pull request May 15, 2020

Add tan_cuda for complex dtypes

9856d36

ghstack-source-id: dce14f0 Pull Request resolved: #38400

krshrimali mentioned this pull request May 15, 2020

Add arcosh, arcsinh and arctanh to unary ops #38388

Closed

2 tasks

facebook-github-bot closed this in 242af6c May 15, 2020

facebook-github-bot added the merged label May 15, 2020

krshrimali added a commit to krshrimali/pytorch that referenced this pull request May 19, 2020

pytorch#38400 is merged now, so finish the TODO

2962788

facebook-github-bot deleted the gh/anjali411/23/head branch May 19, 2020 14:16

mruberry added the Merged label Oct 28, 2020

Provide feedback

Saved searches

Use saved searches to filter your results more quickly

Add tan_cuda for complex dtypes#38400

Add tan_cuda for complex dtypes#38400
anjali411 wants to merge 8 commits intogh/anjali411/23/basefrom
gh/anjali411/23/head

anjali411 commented May 13, 2020 •

edited

Loading

Uh oh!

dr-ci Bot commented May 13, 2020 •

edited

Loading

Uh oh!

albanD May 13, 2020

Uh oh!

anjali411 May 13, 2020

Uh oh!

albanD May 13, 2020

Uh oh!

anjali411 May 13, 2020

Uh oh!

albanD left a comment

Uh oh!

facebook-github-bot commented May 15, 2020

Uh oh!

facebook-github-bot commented May 15, 2020

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

4 participants

Conversation

anjali411 commented May 13, 2020 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Uh oh!

dr-ci Bot commented May 13, 2020 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

💊 CI failures summary and remediations

🕵️ 1 new failure recognized by patterns

pytorch_linux_backward_compatibility_check_test (1/1)

Uh oh!

albanD May 13, 2020

Choose a reason for hiding this comment

Uh oh!

anjali411 May 13, 2020

Choose a reason for hiding this comment

Uh oh!

albanD May 13, 2020

Choose a reason for hiding this comment

Uh oh!

anjali411 May 13, 2020

Choose a reason for hiding this comment

Uh oh!

albanD left a comment

Choose a reason for hiding this comment

Uh oh!

facebook-github-bot commented May 15, 2020

Uh oh!

facebook-github-bot commented May 15, 2020

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

4 participants

anjali411 commented May 13, 2020 •

edited

Loading

dr-ci Bot commented May 13, 2020 •

edited

Loading