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.
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
Remove backward ops for cuDNN transposed convolution #69902
Conversation
[ghstack-poisoned]
CI Flow Status⚛️ CI FlowRuleset - Version:
You can add a comment to the PR and tag @pytorchbot with the following commands: # ciflow rerun, "ciflow/default" will always be added automatically
@pytorchbot ciflow rerun
# ciflow rerun with additional labels "-l <ciflow/label_name>", which is equivalent to adding these labels manually and trigger the rerun
@pytorchbot ciflow rerun -l ciflow/scheduled -l ciflow/slow For more information, please take a look at the CI Flow Wiki. |
🔗 Helpful links
💊 CI failures summary and remediationsAs of commit 67f76a9 (more details on the Dr. CI page):
🕵️ 1 new failure recognized by patternsThe following CI failures do not appear to be due to upstream breakages: linux-bionic-py3.6-clang9 / test (xla, 1, 1, linux.2xlarge) (1/1)Step: "Test" (full log | diagnosis details | 🔁 rerun)
|
@jbschlosser has imported this pull request. If you are a Facebook employee, you can view this diff on Phabricator. |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
I think we should re-use the other cudnn backward op. Otherwise looks good.
@@ -4763,6 +4763,22 @@ std::tuple<Tensor, Tensor> _cudnn_convolution_backward( | |||
return result; | |||
} | |||
|
|||
std::tuple<Tensor, Tensor> _cudnn_convolution_transpose_backward( |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Can we re-use the op from above with an extra boolean argument for transpose?
Differential Revision: [D33093795](https://our.internmc.facebook.com/intern/diff/D33093795) [ghstack-poisoned]
ghstack-source-id: 23aaaabb4ad0e579be50e44abd804fc5c271d895 Pull Request resolved: #69902
@jbschlosser has imported this pull request. If you are a Facebook employee, you can view this diff on Phabricator. |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
SGTM
Differential Revision: [D33093795](https://our.internmc.facebook.com/intern/diff/D33093795) [ghstack-poisoned]
ghstack-source-id: 46cf61dd8e77d3c82fcecf4bc4cca5b81248c742 Pull Request resolved: #69902
@jbschlosser has imported this pull request. If you are a Facebook employee, you can view this diff on Phabricator. |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Good catch
Differential Revision: [D33093795](https://our.internmc.facebook.com/intern/diff/D33093795) [ghstack-poisoned]
@jbschlosser has imported this pull request. If you are a Facebook employee, you can view this diff on Phabricator. |
Differential Revision: [D33093795](https://our.internmc.facebook.com/intern/diff/D33093795) [ghstack-poisoned]
@jbschlosser has imported this pull request. If you are a Facebook employee, you can view this diff on Phabricator. |
Differential Revision: [D33093795](https://our.internmc.facebook.com/intern/diff/D33093795) [ghstack-poisoned]
ghstack-source-id: 5acebf79e6819baab2b8086359327c9f19dee081 Pull Request resolved: #69902
@jbschlosser has imported this pull request. If you are a Facebook employee, you can view this diff on Phabricator. |
Stack from ghstack:
Differential Revision: D33093795