Skip to content

[reland][DNNL]:enable conv3d #40691

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Jump to bottom
Closed
wants to merge 4 commits into from
Closed

[reland][DNNL]:enable conv3d #40691

wants to merge 4 commits into from

Conversation

XiaobingSuper
Copy link
Collaborator

@XiaobingSuper XiaobingSuper commented Jun 29, 2020 •
edited
Loading

Stack from ghstack:

Differential Revision: D22296548

@XiaobingSuper
[reland][DNNL]:enable conv3d
42a87e2 
[ghstack-poisoned]
XiaobingSuper added a commit that referenced this pull request Jun 29, 2020
@XiaobingSuper
[reland][DNNL]:enable conv3d
c23d3e3 
ghstack-source-id: 1fdd586
Pull Request resolved: #40691
@XiaobingSuper
Copy link
Collaborator Author

#35662 was reverted, the failed case is pytorch_paralleltbb_linux_xenial_py3_6_gcc5_4_test, but it can't be reproduced locally, need to check the platform details of this failed case.

@colesbury colesbury added the triaged This issue has been looked at a team member, and triaged and prioritized into an appropriate module label Jun 30, 2020
@VitalyFedyunin
Copy link
Contributor

It is still failing tests (see #40735) please review tolerance and adjust affected tests.

XiaobingSuper added a commit that referenced this pull request Jul 1, 2020
@XiaobingSuper
[reland][DNNL]:enable conv3d
937b94a 
ghstack-source-id: 11c5a35
Pull Request resolved: #40691
@XiaobingSuper
Copy link
Collaborator Author

@VitalyFedyunin, for the conv3d failed cases, DNNL using different algorithm for group conv3d backward(gemm) and for non-group conv3d backward (jit), their results may have a little difference, So I change the relative tolerance from 0 to 1e-5.

@dr-ci
Copy link

dr-ci bot commented Jul 1, 2020 •
edited
Loading

💊 CI failures summary and remediations

As of commit 097e253 (more details on the Dr. CI page):

💚 💚 Looks good so far! There are no failures yet. 💚 💚

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 11 times.

XiaobingSuper added a commit that referenced this pull request Jul 1, 2020
@XiaobingSuper
[reland][DNNL]:enable conv3d
485215b 
ghstack-source-id: 84234e9
Pull Request resolved: #40691
@XiaobingSuper
Copy link
Collaborator Author

@VitalyFedyunin

This was referenced Jul 6, 2020
Closed
Closed
@XiaobingSuper
Copy link
Collaborator Author

@VitalyFedyunin , could help see the internal failed case?

@VitalyFedyunin
Copy link
Contributor

Looks like rtol in internal tests, I plan to fix/land next few days.

@facebook-github-bot
Copy link
Contributor

@VitalyFedyunin merged this pull request in b48ee17.

@facebook-github-bot facebook-github-bot deleted the gh/xiaobingsuper/21/head branch July 19, 2020 14:18
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Reviewers

@zhaojuanmao zhaojuanmao Awaiting requested review from zhaojuanmao

@VitalyFedyunin VitalyFedyunin Awaiting requested review from VitalyFedyunin

Assignees
No one assigned
Labels
Merged open source triaged This issue has been looked at a team member, and triaged and prioritized into an appropriate module
Projects
None yet
Milestone
No milestone
Development

Successfully merging this pull request may close these issues.
6 participants
@XiaobingSuper @VitalyFedyunin @facebook-github-bot @colesbury @pytorchbot @mruberry