Ryan Guo
|
0d4dbfd9ed
|
[dynamo] Support torch.Tensor._make_subclass and tracing through tensor subclass __new__ (#149483)
This builds off the previous patch in the stack, and fully fixes
https://github.com/huggingface/diffusers/issues/10795.
Essentially, tensor subclass in the issue uses
`torch.Tensor._make_subclass`, which has a pretty simple shallow-copy
plus type change semantics, as far as Dynamo is concerned. So this patch
adds a polyfill for it.
As a result, this allows us to trace through many user-defined `__new__`
in tensor subclass (it's similar to how we trace through user-defined
`__new__` for `UserDefinedClassVariable`), so this patch also faithfully
trace through these `__new__` methods.
Differential Revision: [D71906139](https://our.internmc.facebook.com/intern/diff/D71906139)
Pull Request resolved: https://github.com/pytorch/pytorch/pull/149483
Approved by: https://github.com/zou3519, https://github.com/mlazos
ghstack dependencies: #149482
|
2025-04-02 20:56:52 +00:00 |
|
Ryan Guo
|
203e1d681d
|
[dynamo] Support torch.Tensor._make_subclass and tracing through tensor subclass __new__ (#149483)
This builds off the previous patch in the stack, and fully fixes
https://github.com/huggingface/diffusers/issues/10795.
Essentially, tensor subclass in the issue uses
`torch.Tensor._make_subclass`, which has a pretty simple shallow-copy
plus type change semantics, as far as Dynamo is concerned. So this patch
adds a polyfill for it.
As a result, this allows us to trace through many user-defined `__new__`
in tensor subclass (it's similar to how we trace through user-defined
`__new__` for `UserDefinedClassVariable`), so this patch also faithfully
trace through these `__new__` methods.
Differential Revision: [D71906139](https://our.internmc.facebook.com/intern/diff/D71906139)
Pull Request resolved: https://github.com/pytorch/pytorch/pull/149483
Approved by: https://github.com/zou3519, https://github.com/mlazos
ghstack dependencies: #149482
|
2025-04-02 17:05:19 +00:00 |
|