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
Fix gradcheck failing for OpInfo with TensorList inputs #53418
Fix gradcheck failing for OpInfo with TensorList inputs #53418
Conversation
[ghstack-poisoned]
💊 CI failures summary and remediationsAs of commit 770181a (more details on the Dr. CI page):
🕵️ 2 new failures recognized by patternsThe following CI failures do not appear to be due to upstream breakages: pytorch_linux_xenial_py3_6_gcc5_4_build (1/2)Step: "(Optional) Merge target branch" (full log | diagnosis details | 🔁 rerun)
|
This PR adds a workaround to OpInfo tests for ops that take TensoList inputs which were failing gradcheck (see #51996). It also updates the *stack ops to pass in a list of Tensors as input to SampleInput to demonstrate that the tests now pass without the need for defining a lambda in op. [ghstack-poisoned]
This PR adds a workaround to OpInfo tests for ops that take TensoList inputs which were failing gradcheck (see #51996). It also updates the *stack ops to pass in a list of Tensors as input to SampleInput to demonstrate that the tests now pass without the need for defining a lambda in op. Differential Revision: [D26860788](https://our.internmc.facebook.com/intern/diff/D26860788) [ghstack-poisoned]
This PR adds a workaround to OpInfo tests for ops that take TensoList inputs which were failing gradcheck (see #51996). It also updates the *stack ops to pass in a list of Tensors as input to SampleInput to demonstrate that the tests now pass without the need for defining a lambda in op. Differential Revision: [D26860788](https://our.internmc.facebook.com/intern/diff/D26860788) [ghstack-poisoned]
This PR adds a workaround to OpInfo tests for ops that take TensoList inputs which were failing gradcheck (see #51996). It also updates the *stack ops to pass in a list of Tensors as input to SampleInput to demonstrate that the tests now pass without the need for defining a lambda in op. Differential Revision: [D26860788](https://our.internmc.facebook.com/intern/diff/D26860788) [ghstack-poisoned]
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.
lgtm
This PR adds a workaround to OpInfo tests for ops that take TensoList inputs which were failing gradcheck (see #51996). It also updates the *stack ops to pass in a list of Tensors as input to SampleInput to demonstrate that the tests now pass without the need for defining a lambda in op. Differential Revision: [D26860788](https://our.internmc.facebook.com/intern/diff/D26860788) [ghstack-poisoned]
Codecov Report
@@ Coverage Diff @@
## gh/heitorschueroff/54/base #53418 +/- ##
==============================================================
+ Coverage 77.36% 77.43% +0.07%
==============================================================
Files 1879 1869 -10
Lines 183270 182394 -876
==============================================================
- Hits 141782 141242 -540
+ Misses 41488 41152 -336 |
This PR adds a workaround to OpInfo tests for ops that take TensoList inputs which were failing gradcheck (see #51996). It also updates the *stack ops to pass in a list of Tensors as input to SampleInput to demonstrate that the tests now pass without the need for defining a lambda in op. Differential Revision: [D26860788](https://our.internmc.facebook.com/intern/diff/D26860788) [ghstack-poisoned]
I really like this change and the design looks good. I've asked @IvanYashchuk to weigh in on removing the two OpInfo classes for linalg operations. Although this PR can replace them in the current build, I think he may have plans for them. Looks like the test failures are just mypy issues. |
This PR resolves #50837. |
This PR adds a workaround to OpInfo tests for ops that take TensoList inputs which were failing gradcheck (see #51996). It also updates the *stack ops to pass in a list of Tensors as input to SampleInput to demonstrate that the tests now pass without the need for defining a lambda in op. Differential Revision: [D26860788](https://our.internmc.facebook.com/intern/diff/D26860788) [ghstack-poisoned]
This PR adds a workaround to OpInfo tests for ops that take TensoList inputs which were failing gradcheck (see #51996). It also updates the *stack ops to pass in a list of Tensors as input to SampleInput to demonstrate that the tests now pass without the need for defining a lambda in op. Differential Revision: [D26860788](https://our.internmc.facebook.com/intern/diff/D26860788) [ghstack-poisoned]
Stack from ghstack:
This PR adds a workaround to OpInfo tests for ops that take TensoList inputs which were failing gradcheck (see #51996). It also updates the *stack ops to pass in a list of Tensors as input to SampleInput to demonstrate that the tests now pass without the need for defining a lambda in op.
Differential Revision: D26860788