Skip to content

[ONNX] Use parameter values in onnx shape inference#49706

Merged
BowenBao merged 9 commits intopytorch:onnx_ms_1from
BowenBao:onnx_shape_inf_param
Jan 14, 2021
Merged

[ONNX] Use parameter values in onnx shape inference#49706
BowenBao merged 9 commits intopytorch:onnx_ms_1from
BowenBao:onnx_shape_inf_param

Conversation

@BowenBao
Copy link
Copy Markdown
Collaborator

@BowenBao BowenBao commented Dec 21, 2020

Adds an additional run of onnx shape inference after constant folding, since initializer may have changed and affected shape inference.

@facebook-github-bot facebook-github-bot added cla signed oncall: jit Add this issue/PR to JIT oncall triage queue labels Dec 21, 2020
@facebook-github-bot
Copy link
Copy Markdown
Contributor

facebook-github-bot commented Dec 21, 2020 •
edited
Loading

💊 CI failures summary and remediations

As of commit 82d3a69 (more details on the Dr. CI page):

❄️ 1 failure tentatively classified as flaky

but reruns have not yet been triggered to confirm:

See CircleCI build pytorch_linux_xenial_cuda10_2_cudnn7_py3_gcc7_test2 (1/1)

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

Jan 14 08:12:39 RuntimeError: CUDA error: an illegal memory access was encountered 
Jan 14 08:12:39                        ~~~~ <--- HERE
Jan 14 08:12:39 RuntimeError: CUDA error: an illegal memory access was encountered
Jan 14 08:12:39 
Jan 14 08:12:39 
Jan 14 08:12:39 ======================================================================
Jan 14 08:12:39 ERROR [0.203s]: test_where_and_typing (__main__.TestTEFuser)
Jan 14 08:12:39 ----------------------------------------------------------------------
Jan 14 08:12:39 Traceback (most recent call last):
Jan 14 08:12:39   File "test_jit_fuser_te.py", line 1142, in test_where_and_typing
Jan 14 08:12:39     x = torch.randn(4, 4, dtype=torch.double, device=device)
Jan 14 08:12:39 RuntimeError: CUDA error: an illegal memory access was encountered
Jan 14 08:12:39 
Jan 14 08:12:39 ======================================================================
Jan 14 08:12:39 ERROR [0.174s]: test_zero_element_tensors_cuda (__main__.TestTEFuser)
Jan 14 08:12:39 ----------------------------------------------------------------------
Jan 14 08:12:39 Traceback (most recent call last):
Jan 14 08:12:39   File "/opt/conda/lib/python3.6/site-packages/torch/testing/_internal/common_utils.py", line 888, in wrapper
Jan 14 08:12:39     method(*args, **kwargs)
Jan 14 08:12:39   File "test_jit_fuser_te.py", line 178, in test_zero_element_tensors_cuda
Jan 14 08:12:39     self._test_zero_element_tensors(device="cuda")
Jan 14 08:12:39   File "test_jit_fuser_te.py", line 174, in _test_zero_element_tensors

ci.pytorch.org: 1 failed

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 to the (internal) Dr. CI Users group.

neginraoof
neginraoof reviewed Dec 22, 2020
View reviewed changes
@BowenBao BowenBao force-pushed the onnx_shape_inf_param branch from 7c66287 to 185d7a9 Compare December 22, 2020 22:24
@BowenBao BowenBao force-pushed the onnx_shape_inf_param branch from 185d7a9 to 0f60072 Compare January 4, 2021 18:31
neginraoof
neginraoof approved these changes Jan 7, 2021
View reviewed changes
Copy link
Copy Markdown
Contributor

@neginraoof neginraoof left a comment

Choose a reason for hiding this comment

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

LGTM. Thanks!

@BowenBao BowenBao force-pushed the onnx_shape_inf_param branch from b9266a0 to 82d3a69 Compare January 14, 2021 05:24
@BowenBao BowenBao merged commit eaad7f2 into pytorch:onnx_ms_1 Jan 14, 2021
BowenBao added a commit that referenced this pull request Jan 21, 2021
@BowenBao
[ONNX] Use parameter values in onnx shape inference (#49706)
9dc2b29 
Adds an additional run of onnx shape inference after constant folding, since initializer may have changed and affected shape inference.

[ghstack-poisoned]
BowenBao added a commit that referenced this pull request Jan 21, 2021
@BowenBao
[ONNX] Use parameter values in onnx shape inference (#49706)
8acf025 
Adds an additional run of onnx shape inference after constant folding, since initializer may have changed and affected shape inference.

[ghstack-poisoned]
This was referenced Jan 21, 2021
Closed
Closed
Closed
Closed
Closed
This was referenced Jan 21, 2021
Closed
Closed
BowenBao added a commit that referenced this pull request Jan 21, 2021
@BowenBao
Update on "[ONNX] Use parameter values in onnx shape inference (#49706)"
ec50ce9 
Adds an additional run of onnx shape inference after constant folding, since initializer may have changed and affected shape inference.

[ghstack-poisoned]
This was referenced Jan 22, 2021
Closed
Closed
BowenBao added a commit that referenced this pull request Jan 22, 2021
@BowenBao
Update on "[ONNX] Use parameter values in onnx shape inference (#49706)"
ff4fcd3 
Adds an additional run of onnx shape inference after constant folding, since initializer may have changed and affected shape inference.

Differential Revision: [D26023935](https://our.internmc.facebook.com/intern/diff/D26023935)

[ghstack-poisoned]
BowenBao added a commit that referenced this pull request Jan 25, 2021
@BowenBao
Update on "[ONNX] Use parameter values in onnx shape inference (#49706)"
16e623f 
Adds an additional run of onnx shape inference after constant folding, since initializer may have changed and affected shape inference.

[ghstack-poisoned]
BowenBao added a commit that referenced this pull request Jan 25, 2021
@BowenBao
Update on "[ONNX] Use parameter values in onnx shape inference (#49706)"
78dcad9 
Adds an additional run of onnx shape inference after constant folding, since initializer may have changed and affected shape inference.

Differential Revision: [D26050881](https://our.internmc.facebook.com/intern/diff/D26050881)

[ghstack-poisoned]
BowenBao added a commit that referenced this pull request Jan 26, 2021
@BowenBao
Update on "[ONNX] Use parameter values in onnx shape inference (#49706)"
adddcc2 
Adds an additional run of onnx shape inference after constant folding, since initializer may have changed and affected shape inference.

Differential Revision: [D26050881](https://our.internmc.facebook.com/intern/diff/D26050881)

[ghstack-poisoned]
facebook-github-bot pushed a commit that referenced this pull request Jan 28, 2021
@BowenBao @facebook-github-bot
[ONNX] Use parameter values in onnx shape inference (#49706) (#50905)
1c9347c 
Summary:
Pull Request resolved: #50905

Adds an additional run of onnx shape inference after constant folding, since initializer may have changed and affected shape inference.

Test Plan: Imported from OSS

Reviewed By: pbelevich

Differential Revision: D26050881

Pulled By: SplitInfinity

fbshipit-source-id: 9e5d69c52b647133cd3a0781988e2ad1d1a9c09d
BowenBao added a commit to BowenBao/pytorch that referenced this pull request Jan 28, 2021
@BowenBao
[ONNX] Use parameter values in onnx shape inference (pytorch#49706)
e71e5a2 
Adds an additional run of onnx shape inference after constant folding, since initializer may have changed and affected shape inference.

ghstack-source-id: 8fecc35
Pull Request resolved: pytorch#50905

fix clang-format
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Reviewers

1 more reviewer

@neginraoof neginraoof neginraoof approved these changes

Reviewers whose approvals may not affect merge requirements

Assignees

No one assigned

Labels

cla signed oncall: jit Add this issue/PR to JIT oncall triage queue open source

Projects

None yet

Milestone

No milestone

Development

Successfully merging this pull request may close these issues.

4 participants

@BowenBao @facebook-github-bot @neginraoof @pytorchbot