python.控制流

dynamic_shape_if_guard

注意

标签: torch.dynamic-shape, python.control-flow

支持级别：支持

原始源代码：

import torch

结果:

ExportedProgram:
    class GraphModule(torch.nn.Module):
        def forward(self, arg0_1: "f32[3, 2, 2]"):
                cos: "f32[3, 2, 2]" = torch.ops.aten.cos.default(arg0_1);  arg0_1 = None
            return (cos,)

Graph signature: ExportGraphSignature(input_specs=[InputSpec(kind=<InputKind.USER_INPUT: 1>, arg=TensorArgument(name='arg0_1'), target=None, persistent=None)], output_specs=[OutputSpec(kind=<OutputKind.USER_OUTPUT: 1>, arg=TensorArgument(name='cos'), target=None)])
范围约束: {}

列表解包

注意

标签: python.数据结构, python.控制流程

支持级别：支持

原始源代码：

from typing import List

import torch

结果:

ExportedProgram:
    class GraphModule(torch.nn.Module):
        def forward(self, arg0_1: "f32[3, 2]", arg1_1: "i64[]", arg2_1: "i64[]"):
                add: "f32[3, 2]" = torch.ops.aten.add.Tensor(arg0_1, arg1_1);  arg0_1 = arg1_1 = None
            return (add,)

Graph signature: ExportGraphSignature(input_specs=[InputSpec(kind=<InputKind.USER_INPUT: 1>, arg=TensorArgument(name='arg0_1'), target=None, persistent=None), InputSpec(kind=<InputKind.USER_INPUT: 1>, arg=TensorArgument(name='arg1_1'), target=None, persistent=None), InputSpec(kind=<InputKind.USER_INPUT: 1>, arg=TensorArgument(name='arg2_1'), target=None, persistent=None)], output_specs=[OutputSpec(kind=<OutputKind.USER_OUTPUT: 1>, arg=TensorArgument(name='add'), target=None)])
Range constraints: {}

静态循环

注意

标签: python.控制流

支持级别：支持

原始源代码：

import torch

结果:

```html

ExportedProgram:
    class GraphModule(torch.nn.Module):
        def forward(self, arg0_1: "f32[3, 2]"):
                add: "f32[3, 2]" = torch.ops.aten.add.Tensor(arg0_1, 0)
            add_1: "f32[3, 2]" = torch.ops.aten.add.Tensor(arg0_1, 1)
            add_2: "f32[3, 2]" = torch.ops.aten.add.Tensor(arg0_1, 2)
            add_3: "f32[3, 2]" = torch.ops.aten.add.Tensor(arg0_1, 3)
            add_4: "f32[3, 2]" = torch.ops.aten.add.Tensor(arg0_1, 4)
            add_5: "f32[3, 2]" = torch.ops.aten.add.Tensor(arg0_1, 5)
            add_6: "f32[3, 2]" = torch.ops.aten.add.Tensor(arg0_1, 6)
            add_7: "f32[3, 2]" = torch.ops.aten.add.Tensor(arg0_1, 7)
            add_8: "f32[3, 2]" = torch.ops.aten.add.Tensor(arg0_1, 8)
            add_9: "f32[3, 2]" = torch.ops.aten.add.Tensor(arg0_1, 9);  arg0_1 = None
            return (add, add_1, add_2, add_3, add_4, add_5, add_6, add_7, add_8, add_9)

Graph signature: ExportGraphSignature(input_specs=[InputSpec(kind=<InputKind.USER_INPUT: 1>, arg=TensorArgument(name='arg0_1'), target=None, persistent=None)], output_specs=[OutputSpec(kind=<OutputKind.USER_OUTPUT: 1>, arg=TensorArgument(name='add'), target=None), OutputSpec(kind=<OutputKind.USER_OUTPUT: 1>, arg=TensorArgument(name='add_1'), target=None), OutputSpec(kind=<OutputKind.USER_OUTPUT: 1>, arg=TensorArgument(name='add_2'), target=None), OutputSpec(kind=<OutputKind.USER_OUTPUT: 1>, arg=TensorArgument(name='add_3'), target=None), OutputSpec(kind=<OutputKind.USER_OUTPUT: 1>, arg=TensorArgument(name='add_4'), target=None), OutputSpec(kind=<OutputKind.USER_OUTPUT: 1>, arg=TensorArgument(name='add_5'), target=None), OutputSpec(kind=<OutputKind.USER_OUTPUT: 1>, arg=TensorArgument(name='add_6'), target=None), OutputSpec(kind=<OutputKind.USER_OUTPUT: 1>, arg=TensorArgument(name='add_7'), target=None), OutputSpec(kind=<<span class

static_if

注意

标签: python.控制流

支持级别：支持

原始源代码：

import torch

结果:

ExportedProgram:
    class GraphModule(torch.nn.Module):
        def forward(self, arg0_1: "f32[3, 2, 2]"):
                ones: "f32[1, 1, 1]" = torch.ops.aten.ones.default([1, 1, 1], device = device(type='cpu'), pin_memory = False)
            add: "f32[3, 2, 2]" = torch.ops.aten.add.Tensor(arg0_1, ones);  arg0_1 = ones = None
            return (add,)

Graph signature: ExportGraphSignature(input_specs=[InputSpec(kind=<InputKind.USER_INPUT: 1>, arg=TensorArgument(name='arg0_1'), target=None, persistent=None)], output_specs=[OutputSpec(kind=<OutputKind.USER_OUTPUT: 1>, arg=TensorArgument(name='add'), target=None)])
Range constraints: {}