python.builtin¶
dynamic_shape_round¶
Original source code:
import torch
from torch._export import dynamic_dim
x = torch.ones(3, 2)
dynamic_constraint = dynamic_dim(x, 0)
def dynamic_shape_round(x):
"""
Calling round on dynamic shapes is not supported.
"""
return x[: round(x.shape[0] / 2)]
Result:
Unsupported: Calling round() on symbolic value is not supported. You can use floor() to implement this functionality
tensor_setattr¶
Original source code:
import torch
def tensor_setattr(x, attr):
"""
setattr() call onto tensors is not supported.
"""
setattr(x, attr, torch.randn(3, 2))
return x + 4
Result:
ExportedProgram:
class GraphModule(torch.nn.Module):
def forward(self, arg0_1: f32[3, 2], arg1_1):
#
add: f32[3, 2] = torch.ops.aten.add.Tensor(arg0_1, 4); arg0_1 = None
return (add,)
Graph Signature: ExportGraphSignature(parameters=[], buffers=[], user_inputs=['arg0_1', 'arg1_1'], user_outputs=['add'], inputs_to_parameters={}, inputs_to_buffers={}, buffers_to_mutate={}, backward_signature=None, assertion_dep_token=None)
Symbol to range: {}
type_reflection_method¶
Original source code:
import torch
class A:
@classmethod
def func(cls, x):
return 1 + x
def type_reflection_method(x):
"""
type() calls on custom objects followed by method calls are not allowed
due to its overly dynamic nature.
"""
a = A()
return type(a).func(x)
Result:
Unsupported: Can't call type() on generated custom object. Please use __class__ instead
You can rewrite the example above to something like the following:
def type_reflection_method_rewrite(x):
"""
Custom object class methods will be inlined.
"""
return A.func(x)
