Add decomposition for aten.native_batch_norm_backward op

[Shukla-Gaurav]

This commit adds decomposition for the aten.native_batch_norm_backward
op.

Signed-Off-by: Gaurav Shukla gaurav@nod-labs.com

[facebook-github-bot]

Hi @Shukla-Gaurav!

Thank you for your pull request and welcome to our community.

Action Required

In order to merge any pull request (code, docs, etc.), we require contributors to sign our Contributor License Agreement, and we don't seem to have one on file for you.

Process

In order for us to review and merge your suggested changes, 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.

Once the CLA is signed, our tooling will perform checks and validations. Afterwards, the pull request will be tagged with CLA signed. The tagging process may take up to 1 hour after signing. Please give it that time before contacting us about it.

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

[facebook-github-bot]

Thank you for signing our Contributor License Agreement. We can now accept your code for this (and any) Meta Open Source project. Thanks!

[Chillee]

I think there are still some tests failing. For example, I don't believe push_back is a method on python lists :) https://github.com/pytorch/functorch/pull/675/files#diff-8e71d16136c7deeb8104d8d5bb0dd794c208f38856f9ced3448ab02f3e34c43aR522

To run the tests, run

pytest test/test_ops.py::TestDecompositionOpInfoCUDA  -k batch_norm

and

pytest test/test_ops.py::TestDecompositionOpInfoCPU  -k batch_norm

[Shukla-Gaurav]

@Chillee 'pytest test/test_ops.py::TestDecompositionOpInfoCPU -k batch_norm` is always passing for me locally.
I do following steps to test:

1. pip install --pre torch torchvision-f https://download.pytorch.org/whl/nightly/cpu/torch_nightly.html --upgrade
2. python setup.py install
3. pytest test/test_ops.py::TestDecompositionOpInfoCPU -k batch_norm

Am I doing something wrong?

[Chillee]

@Shukla-Gaurav Do you mean that it always passes, regardless of whether your decomposition is correct? Or are you asking why the decompositions are passing even though the CI fails?

If it's the second one, then there's some failures that are unrelated to you :)

[Chillee]

@Shukla-Gaurav btw, is this mergeable?

[Shukla-Gaurav]

@Chillee Apologies for the delay. I will rebase and push. Thanks!

[Shukla-Gaurav]

@Chillee It looks like all the decompositions has been moved/removed?

[Chillee]

@Shukla-Gaurav Yes, can you re-open this PR on upstream PyTorch?

[Shukla-Gaurav]

@Shukla-Gaurav Yes, can you re-open this PR on upstream PyTorch?

Sure, will do that :)