Short name describing what triggered the graph break
torch.cond: improper predicate
Values or code snippet captured at the break point
str(pred)
Explanation of why the graph break was triggered
Expected pred to be a bool or a boolean tensor with a single item but got {str(type(pred))} with original python type {str(pred.python_type())}.
Hints on how to resolve the graph break
torch.compile call, or by using torch.compiler.set_stance("force_eager").