added view_as

[k223kim]

Before submitting

• Was this discussed/approved via a Github issue? (no need for typos and docs improvements)
• Did you read the contributor guideline, Pull Request section?
• Did you make sure to update the docs?
• Did you write any new necessary tests?

What does this PR do?

Fixes #328. I am not sure if I should include the test case for view_as opinfos.py (cause there wasn't one for view, maybe it is similar to reshape?). Also, I added view_as to prims, although I am not sure if this is necessary. Let me know how I can improve!

PR review

Anyone in the community is free to review the PR once the tests have passed.
If we didn't discuss your PR in Github issues there's a high chance it will not be merged.

Did you have fun?

👊

[mruberry]

Hey @k223kim! Cool stuff! I made a couple minor comments

[mruberry]

Forgot to add -- yes, a test with an opinfo would be great! It doesn't have to be especially complicated

[k223kim]

Hi @mruberry! Thanks for the awesome review! Just some questions regarding this PR. I have added a test case for view_as but I am not sure if I should include supports_grad=True. If I do so, it throws the following errors:

FAILED thunder/tests/test_grad.py::test_phantom_grad_vs_torch_consistency_view_as_torch_cpu_bfloat16 - RuntimeError: One of the differentiated Tensors appears to not have been used in the graph. Set allow_unused=True if this is the desired behavior.

Do you think this is necessary?

[mruberry]

Hi @mruberry! Thanks for the awesome review! Just some questions regarding this PR. I have added a test case for view_as but I am not sure if I should include supports_grad=True. If I do so, it throws the following errors:

FAILED thunder/tests/test_grad.py::test_phantom_grad_vs_torch_consistency_view_as_torch_cpu_bfloat16 - RuntimeError: One of the differentiated Tensors appears to not have been used in the graph. Set allow_unused=True if this is the desired behavior.

Do you think this is necessary?

Nope, don't need to add supports_grad=True for now. fyi @IvanYashchuk

[mruberry]

Cool! Looks great with thorough tests

fyi @Borda and @t-vi to merge

[t-vi]

Thank you @k223kim @mruberry