/j,ddlZddlZejddZedejeddejedddejd ed ejfd Z d Z dS) NexportFRAGMENTzLaccess_subclass_inner_tensor(Tensor src_subclass_tensor, str attr) -> Tensoraccess_subclass_inner_tensorAutogradPythonsrc_subclass_tensorattrreturncddlm}||stdt|t ||d}|t |t jstd|d||S)Nr)is_traceable_wrapper_subclasszIExpected src_subclass_tensor to be a traceable wrapper subclass, but got z Attribute z% is not a tensor or doesn't exist in ) torch.utils._python_dispatchr AssertionErrortypegetattr isinstancetorchTensor RuntimeError)rr r vals \/home/longshao/multi-rider-rag/.venv/lib/python3.11/site-packages/torch/export/custom_ops.py_access_subclass_inner_tensorrsKJJJJJ ( ()< = =  3/00 3 3    %tT 2 2C {*S%,77{ Y Y YDW Y Y    Jc|dd\}}tj|}t||}t |dst d|d|j|i|S)a] Import a custom autograd function by string name and call it. This is pretty bad because: 1) There is no schema Ideally we should automatically wrap custom autograd functions with a custom op, but that is too much work because we need to schematize custom autograd functions. For now, we just hackily put it in the IR. .applyzExpected function class z to have 'apply' method)rsplit importlib import_modulerhasattrrr)function_cls_nameargskwargs module_name class_namemodule function_clss r._call_custom_autograd_function_in_pre_dispatchr($s066sA>>K $[ 1 1F6:..L < ) )  Q'8 Q Q Q    < t .v . ..r) rrlibraryLibrarylibdefineimplrstrrr(rrr0s  mHj11 R C7DDC7BB-0 \CBED$/////r