Short name describing what triggered the graph break
Non-function or method in subclass of torch.autograd.Function
Values or code snippet captured at the break point
call_apply {self} {args} {kwargs}
Explanation of why the graph break was triggered
Dynamo requires the forward attribute of a torch.autograd.Function subclass to be a standard Python function or method. Found type {type(fn).__name__} instead.
Hints on how to resolve the graph break
forward method is defined as a regular