annotate a few torch.nn.modules.* modules #45772

guilhermeleobas · 2020-10-02T20:59:50Z

torch/nn/modules/batchnorm.py

albanD · 2020-10-06T13:57:05Z

@rgommers you can ping me when this is ready for merge.

guilhermeleobas · 2020-10-07T15:11:41Z

Current failures doesn't seem to be related to any changes introduced in this PR. I will rebase with master

Traceback (most recent call last):
  File "/opt/conda/lib/python3.6/site-packages/torch/testing/_internal/common_utils.py", line 827, in wrapper
    method(*args, **kwargs)
  File "/opt/conda/lib/python3.6/site-packages/torch/testing/_internal/common_device_type.py", line 273, in instantiated_test
    result = test_fn(self, *args)
  File "/opt/conda/lib/python3.6/site-packages/torch/testing/_internal/common_device_type.py", line 508, in dep_fn
    return fn(slf, device, *args, **kwargs)
  File "/opt/conda/lib/python3.6/site-packages/torch/testing/_internal/common_device_type.py", line 508, in dep_fn
    return fn(slf, device, *args, **kwargs)
  File "test_torch.py", line 20843, in test_svd_square
    self._test_svd_helper((10, 10), True, False, device, dtype)
  File "test_torch.py", line 20828, in _test_svd_helper
    device_result = torch.svd(device_tensor, some=some)
RuntimeError: "svd_cuda" not implemented for 'ComplexDouble'

dr-ci · 2020-10-16T03:08:28Z

💊 CI failures summary and remediations

As of commit b207772 (more details on the Dr. CI page):

3/3 failures introduced in this PR

🕵️ 3 new failures recognized by patterns

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

pytorch_linux_xenial_py3_clang5_asan_build (1/3)

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

Oct 30 18:44:39 sccache: error: couldn't connect to server

Oct 30 18:44:39 +++ eval 'extract_trap_cmd ' 
Oct 30 18:44:39 ++++ extract_trap_cmd 
Oct 30 18:44:39 ++++ printf '%s\n' '' 
Oct 30 18:44:39 +++ printf '%s\n' cleanup 
Oct 30 18:44:39 ++ trap -- ' 
Oct 30 18:44:39 cleanup' EXIT 
Oct 30 18:44:39 ++ [[ pytorch-linux-xenial-py3-clang5-asan-build != *pytorch-win-* ]] 
Oct 30 18:44:39 ++ which sccache 
Oct 30 18:44:39 ++ sccache --stop-server 
Oct 30 18:44:39 Stopping sccache server... 
Oct 30 18:44:39 sccache: error: couldn't connect to server 
Oct 30 18:44:39 sccache: caused by: Connection refused (os error 111) 
Oct 30 18:44:39 ++ true 
Oct 30 18:44:39 ++ rm /var/lib/jenkins/sccache_error.log 
Oct 30 18:44:39 rm: cannot remove '/var/lib/jenkins/sccache_error.log': No such file or directory 
Oct 30 18:44:39 ++ true 
Oct 30 18:44:39 ++ [[ pytorch-linux-xenial-py3-clang5-asan-build == *rocm* ]] 
Oct 30 18:44:39 ++ SCCACHE_ERROR_LOG=/var/lib/jenkins/sccache_error.log 
Oct 30 18:44:39 ++ SCCACHE_IDLE_TIMEOUT=1200 
Oct 30 18:44:39 ++ RUST_LOG=sccache::server=error 
Oct 30 18:44:39 ++ sccache --start-server

pytorch_xla_linux_bionic_py3_6_clang9_test (2/3)

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

Oct 30 20:04:41 FAIL [0.790s]: test_muldiv_scalar_xla_bfloat16 (__main__.TestTorchDeviceTypeXLA)

Oct 30 20:03:40   test_where_scalar_valid_combination_xla_float16 (__main__.TestTorchDeviceTypeXLA) ... skip (0.002s) 
Oct 30 20:03:43   test_where_scalar_valid_combination_xla_float32 (__main__.TestTorchDeviceTypeXLA) ... ok (2.848s) 
Oct 30 20:04:37   test_where_scalar_valid_combination_xla_float64 (__main__.TestTorchDeviceTypeXLA) ... ok (54.645s) 
Oct 30 20:04:38   test_where_scalar_valid_combination_xla_int16 (__main__.TestTorchDeviceTypeXLA) ... ok (0.518s) 
Oct 30 20:04:38   test_where_scalar_valid_combination_xla_int32 (__main__.TestTorchDeviceTypeXLA) ... ok (0.468s) 
Oct 30 20:04:40   test_where_scalar_valid_combination_xla_int64 (__main__.TestTorchDeviceTypeXLA) ... ok (1.470s) 
Oct 30 20:04:40   test_where_scalar_valid_combination_xla_int8 (__main__.TestTorchDeviceTypeXLA) ... ok (0.507s) 
Oct 30 20:04:41   test_where_scalar_valid_combination_xla_uint8 (__main__.TestTorchDeviceTypeXLA) ... ok (0.467s) 
Oct 30 20:04:41  
Oct 30 20:04:41 ====================================================================== 
Oct 30 20:04:41 FAIL [0.790s]: test_muldiv_scalar_xla_bfloat16 (__main__.TestTorchDeviceTypeXLA) 
Oct 30 20:04:41 ---------------------------------------------------------------------- 
Oct 30 20:04:41 Traceback (most recent call last): 
Oct 30 20:04:41   File "/opt/conda/lib/python3.6/site-packages/torch/testing/_internal/common_device_type.py", line 274, in instantiated_test 
Oct 30 20:04:41     result = test_fn(self, *args) 
Oct 30 20:04:41   File "/var/lib/jenkins/workspace/xla/test/../../test/test_torch.py", line 19475, in test_muldiv_scalar 
Oct 30 20:04:41     self.assertEqual(s / x, y / x) 
Oct 30 20:04:41   File "/var/lib/jenkins/workspace/xla/test/pytorch_test_base.py", line 551, in assertEqual 
Oct 30 20:04:41     return DeviceTypeTestBase.assertEqual(self, x, y, *args, **kwargs) 
Oct 30 20:04:41   File "/opt/conda/lib/python3.6/site-packages/torch/testing/_internal/common_utils.py", line 1145, in assertEqual 
Oct 30 20:04:41     self.assertTrue(result, msg=msg)

pytorch_linux_xenial_py3_clang7_onnx_build (3/3)

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

Oct 30 18:44:41 sccache: error: couldn't connect to server

Oct 30 18:44:41 +++ eval 'extract_trap_cmd ' 
Oct 30 18:44:41 ++++ extract_trap_cmd 
Oct 30 18:44:41 ++++ printf '%s\n' '' 
Oct 30 18:44:41 +++ printf '%s\n' cleanup 
Oct 30 18:44:41 ++ trap -- ' 
Oct 30 18:44:41 cleanup' EXIT 
Oct 30 18:44:41 ++ [[ pytorch-linux-xenial-py3-clang7-onnx-build != *pytorch-win-* ]] 
Oct 30 18:44:41 ++ which sccache 
Oct 30 18:44:41 ++ sccache --stop-server 
Oct 30 18:44:41 Stopping sccache server... 
Oct 30 18:44:41 sccache: error: couldn't connect to server 
Oct 30 18:44:41 sccache: caused by: Connection refused (os error 111) 
Oct 30 18:44:41 ++ true 
Oct 30 18:44:41 ++ rm /var/lib/jenkins/sccache_error.log 
Oct 30 18:44:41 rm: cannot remove '/var/lib/jenkins/sccache_error.log': No such file or directory 
Oct 30 18:44:41 ++ true 
Oct 30 18:44:41 ++ [[ pytorch-linux-xenial-py3-clang7-onnx-build == *rocm* ]] 
Oct 30 18:44:41 ++ SCCACHE_ERROR_LOG=/var/lib/jenkins/sccache_error.log 
Oct 30 18:44:41 ++ SCCACHE_IDLE_TIMEOUT=1200 
Oct 30 18:44:41 ++ RUST_LOG=sccache::server=error 
Oct 30 18:44:41 ++ sccache --start-server

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 or post in the (internal) Dr. CI Users group.

See how this bot performed.

This comment has been revised 15 times.

guilhermeleobas · 2020-10-16T19:22:12Z

CI failure is not related to any changes introduced in this PR.

Traceback (most recent call last):
  File "C:\Users\circleci\project\build\win_tmp\build\torch\testing\_internal\common_quantized.py", line 122, in test_fn
    qfunction(*args, **kwargs)
  File "C:\Users\circleci\project\test\quantization\test_quantize_jit.py", line 3033, in test_linear
    "quantized::linear_dynamic", tracing=tracing, dynamic=True)
  File "C:\Users\circleci\project\build\win_tmp\build\torch\testing\_internal\common_quantization.py", line 499, in checkGraphModeOp
    models[d] = quantize_dynamic_jit(model, qconfig_dict, debug=d)
  File "C:\Users\circleci\project\build\win_tmp\build\torch\quantization\quantize_jit.py", line 207, in quantize_dynamic_jit
    return _quantize_jit(model, qconfig_dict, inplace=inplace, debug=debug, quant_type=QuantType.DYNAMIC)
  File "C:\Users\circleci\project\build\win_tmp\build\torch\quantization\quantize_jit.py", line 105, in _quantize_jit
    model = convert_dynamic_jit(model, True, debug)
  File "C:\Users\circleci\project\build\win_tmp\build\torch\quantization\quantize_jit.py", line 98, in convert_dynamic_jit
    return _convert_jit(model, inplace, debug, quant_type=QuantType.DYNAMIC, preserved_attrs=preserved_attrs)
  File "C:\Users\circleci\project\build\win_tmp\build\torch\quantization\quantize_jit.py", line 78, in _convert_jit
    model_c = torch._C._jit_pass_insert_quant_dequant(model_c, 'forward', inplace, debug, quant_type)
RuntimeError: 0 INTERNAL ASSERT FAILED at "..\\torch\\csrc\\jit\\ir\\alias_analysis.cpp":461, please report a bug to PyTorch. We don't have an op for aten::quantize_per_channel but it isn't a special case.  Argument types: Tensor, float, int, int,

rgommers

Thanks @guilhermeleobas, overall this looks good. One comment that needs a closer look / different solution.

torch/nn/modules/linear.py

codecov · 2020-10-30T04:23:21Z

Codecov Report

Merging #45772 into master will decrease coverage by 0.00%.
The diff coverage is 85.00%.

@@            Coverage Diff             @@
##           master   #45772      +/-   ##
==========================================
- Coverage   68.87%   68.87%   -0.01%     
==========================================
  Files         436      436              
  Lines       56368    56376       +8     
==========================================
+ Hits        38823    38828       +5     
- Misses      17545    17548       +3

facebook-github-bot · 2020-10-30T19:03:53Z

Hi @guilhermeleobas!

Thank you for your pull request and welcome to our community. We require contributors to sign our Contributor License Agreement, and we don't seem to have you on file.

In order for us to review and merge your code, please sign at https://code.facebook.com/cla. If you are contributing on behalf of someone else (eg your employer), the individual CLA may not be sufficient and your employer may need to sign the corporate CLA.

If you have received this in error or have any questions, please contact us at cla@fb.com. Thanks!

rgommers

LGTM now, thanks @guilhermeleobas. CI failures are unrelated.

@albanD could you take it from here?

albanD

Thanks @rgommers I can take it yes.

facebook-github-bot

@albanD has imported this pull request. If you are a Facebook employee, you can view this diff on Phabricator.

facebook-github-bot · 2020-11-02T21:17:13Z

@albanD merged this pull request in 9b52654.

guilhermeleobas added the module: typing Related to mypy type annotations label Oct 2, 2020

guilhermeleobas requested a review from apaszke as a code owner October 2, 2020 20:59

guilhermeleobas self-assigned this Oct 2, 2020

guilhermeleobas requested a review from rgommers October 2, 2020 21:00

pytorchbot added the open source label Oct 2, 2020

rgommers removed the request for review from apaszke October 2, 2020 22:08

rgommers reviewed Oct 2, 2020

View reviewed changes

torch/nn/modules/batchnorm.py Outdated Show resolved Hide resolved

albanD added the triaged This issue has been looked at a team member, and triaged and prioritized into an appropriate module label Oct 6, 2020

guilhermeleobas requested a review from rgommers October 16, 2020 19:22

rgommers requested changes Oct 28, 2020

View reviewed changes

torch/nn/modules/linear.py Show resolved Hide resolved

guilhermeleobas and others added 6 commits October 29, 2020 19:41

annotate a few torch.nn.modules.* modules

83f2ac4

fix errors

b14020b

update functional.pyi.in file

d7c46cc

fix jit error

31b34f6

update type annotation

bb403bd

annotate linear.py

2ce2a8f

flake8

b207772

rgommers approved these changes Oct 31, 2020

View reviewed changes

rgommers requested a review from albanD October 31, 2020 12:25

albanD approved these changes Oct 31, 2020

View reviewed changes

facebook-github-bot added the cla signed label Oct 31, 2020

facebook-github-bot reviewed Nov 2, 2020

View reviewed changes

facebook-github-bot closed this in 9b52654 Nov 2, 2020

facebook-github-bot added the Merged label Nov 2, 2020

guilhermeleobas deleted the batchnorm branch November 9, 2020 17:19

Provide feedback

Saved searches

Use saved searches to filter your results more quickly

annotate a few torch.nn.modules.* modules #45772

annotate a few torch.nn.modules.* modules #45772

guilhermeleobas commented Oct 2, 2020

albanD commented Oct 6, 2020

guilhermeleobas commented Oct 7, 2020

dr-ci bot commented Oct 16, 2020 •

edited

guilhermeleobas commented Oct 16, 2020

rgommers left a comment

codecov bot commented Oct 30, 2020

facebook-github-bot commented Oct 30, 2020

rgommers left a comment

albanD left a comment

facebook-github-bot left a comment

facebook-github-bot commented Nov 2, 2020

annotate a few torch.nn.modules.* modules #45772

annotate a few torch.nn.modules.* modules #45772

Conversation

guilhermeleobas commented Oct 2, 2020

albanD commented Oct 6, 2020

guilhermeleobas commented Oct 7, 2020

dr-ci bot commented Oct 16, 2020 • edited

💊 CI failures summary and remediations

🕵️ 3 new failures recognized by patterns

pytorch_linux_xenial_py3_clang5_asan_build (1/3)

pytorch_xla_linux_bionic_py3_6_clang9_test (2/3)

pytorch_linux_xenial_py3_clang7_onnx_build (3/3)

guilhermeleobas commented Oct 16, 2020

rgommers left a comment

Choose a reason for hiding this comment

codecov bot commented Oct 30, 2020

Codecov Report

facebook-github-bot commented Oct 30, 2020

rgommers left a comment

Choose a reason for hiding this comment

albanD left a comment

Choose a reason for hiding this comment

facebook-github-bot left a comment

Choose a reason for hiding this comment

facebook-github-bot commented Nov 2, 2020

dr-ci bot commented Oct 16, 2020 •

edited