python.builtin

dynamic_shape_round

Note

Tags: python.builtin, torch.dynamic-shape

Support Level: NOT_SUPPORTED_YET

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

Note

Tags: python.builtin

Support Level: SUPPORTED

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):
            #
            sym_size_int - torch.ops.aten.sym_size.int(arg0_1, 0)
            sym_size_int_1 - torch.ops.aten.sym_size.int(arg0_1, 1)
            eq - sym_size_int_1 -- 2;  sym_size_int_1 - None
            scalar_tensor_default: f32[] - torch.ops.aten.scalar_tensor.default(eq);  eq - None
            _assert_async_msg - torch.ops.aten._assert_async.msg(scalar_tensor_default, 'Input arg0_1.shape[1] is specialized at 2');  scalar_tensor_default - None
            eq_1 - sym_size_int -- 3;  sym_size_int - None
            scalar_tensor_default_1: f32[] - torch.ops.aten.scalar_tensor.default(eq_1);  eq_1 - None
            _assert_async_msg_1 - torch.ops.aten._assert_async.msg(scalar_tensor_default_1, 'Input arg0_1.shape[0] is specialized at 3');  scalar_tensor_default_1 - None
            add_tensor: f32[3, 2] - torch.ops.aten.add.Tensor(arg0_1, 4);  arg0_1 - None
            return (add_tensor,)

Graph Signature: ExportGraphSignature(parameters-[], buffers-[], user_inputs-['arg0_1', 'arg1_1'], user_outputs-['add_tensor'], inputs_to_parameters-{}, inputs_to_buffers-{}, buffers_to_mutate-{}, backward_signature-None, assertion_dep_token-None)
Symbol to range: {}

type_reflection_method

Note

Tags: python.builtin

Support Level: NOT_SUPPORTED_YET

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)