pytorch/torch/_dynamo/variables
Matthias Reso 772db2a3ae Fix handling of torch.return_types in dynamo (#120826)
Handle quasi-namedtuples as a special case in dynamo

Fixes #120651

Pull Request resolved: https://github.com/pytorch/pytorch/pull/120826
Approved by: https://github.com/anijain2305
2024-02-29 22:11:35 +00:00
..
__init__.py Let torch dynamo inline torch.func.grad (#118407) 2024-02-28 20:05:00 +00:00
base.py [dynamo] Improve support for backwards hooks (#119525) 2024-02-10 01:14:03 +00:00
builder.py [Dynamo] Do not attempt to make nditer spawned arrays writable (#120868) 2024-02-29 07:49:59 +00:00
builtin.py [dynamo][refactor] Rename LIST_LENGTH to SEQUENCE_LENGTH, separate DICT_LENGTH (#120721) 2024-02-28 02:19:10 +00:00
constant.py Print the value of constants in __str__ (#119276) 2024-02-08 16:23:36 +00:00
ctx_manager.py [dynamo] Fix inference_mode context variable (#120830) 2024-02-29 17:10:06 +00:00
dicts.py [dynamo] Refactor reconstruct() not to return anything (#120150) 2024-02-17 17:13:41 +00:00
distributed.py Revert [dynamo] support group=None when rewriting collectives (#12018) (#120677) 2024-02-27 00:33:35 +00:00
functions.py Revert "[Dynamo] Remove deadcode: unwrapping script_if_tracing (#120444)" 2024-02-27 00:58:00 +00:00
higher_order_ops.py [Compiled Autograd] Introduce BackwardState capture (#120382) 2024-02-28 20:36:47 +00:00
iter.py Unify MYPYINDUCTOR and MYPY (#118432) 2024-01-27 17:23:20 +00:00
lazy.py Unify MYPYINDUCTOR and MYPY (#118432) 2024-01-27 17:23:20 +00:00
lists.py [dynamo] Refactor reconstruct() not to return anything (#120150) 2024-02-17 17:13:41 +00:00
misc.py [dynamo] Refactor reconstruct() not to return anything (#120150) 2024-02-17 17:13:41 +00:00
nn_module.py [Dynamo] Support lazy module with namedtuple/dict input (#119972) 2024-02-15 23:18:18 +00:00
optimizer.py [dynamo,optim] Use the actual sources from the parameters when tracing "params" in an optimizer (#118535) 2024-02-02 14:42:56 +00:00
sdpa.py [dynamo] Refactor reconstruct() not to return anything (#120150) 2024-02-17 17:13:41 +00:00
tensor.py [Compiled Autograd] Introduce BackwardState capture (#120382) 2024-02-28 20:36:47 +00:00
torch_function.py Fix handling none source in build_torch_function_fn (#119724) 2024-02-15 21:21:47 +00:00
torch.py [torch.export] Support is_compiling() flag for non-strict mode (#119602) 2024-02-29 05:52:51 +00:00
user_defined.py Fix handling of torch.return_types in dynamo (#120826) 2024-02-29 22:11:35 +00:00