Shortcuts

python.data-structure

dictionary

Note

Tags: python.data-structure

Support Level: SUPPORTED

Original source code:

# mypy: allow-untyped-defs
import torch



class Dictionary(torch.nn.Module):
    """
    Dictionary structures are inlined and flattened along tracing.
    """
    def __init__(self):
        super().__init__()

    def forward(self, x, y):
        elements = {}
        elements["x2"] = x * x
        y = y * elements["x2"]
        return {"y": y}

Result:

ExportedProgram:
    class GraphModule(torch.nn.Module):
        def forward(self, x: "f32[3, 2]", y: "i64[]"):
                mul: "f32[3, 2]" = torch.ops.aten.mul.Tensor(x, x);  x = None

                mul_1: "f32[3, 2]" = torch.ops.aten.mul.Tensor(y, mul);  y = mul = None
            return (mul_1,)

Graph signature: ExportGraphSignature(input_specs=[InputSpec(kind=<InputKind.USER_INPUT: 1>, arg=TensorArgument(name='x'), target=None, persistent=None), InputSpec(kind=<InputKind.USER_INPUT: 1>, arg=TensorArgument(name='y'), target=None, persistent=None)], output_specs=[OutputSpec(kind=<OutputKind.USER_OUTPUT: 1>, arg=TensorArgument(name='mul_1'), target=None)])
Range constraints: {}

fn_with_kwargs

Note

Tags: python.data-structure

Support Level: SUPPORTED

Original source code:

# mypy: allow-untyped-defs
import torch



    ),
    tags={"python.data-structure"},
    support_level=SupportLevel.SUPPORTED,
)
class FnWithKwargs(torch.nn.Module):
    """
    Keyword arguments are not supported at the moment.
    """
    def __init__(self):
        super().__init__()

    def forward(self, pos0, tuple0, *myargs, mykw0, **mykwargs):
        out = pos0
        for arg in tuple0:
            out = out * arg
        for arg in myargs:
            out = out * arg
        out = out * mykw0
        out = out * mykwargs["input0"] * mykwargs["input1"]
        return out

Result:

ExportedProgram:
    class GraphModule(torch.nn.Module):
        def forward(self, pos0: "f32[4]", tuple0_0: "f32[4]", tuple0_1: "f32[4]", myargs_0: "f32[4]", myargs_1: "f32[4]", mykw0: "f32[4]", input0: "f32[4]", input1: "f32[4]"):
                mul: "f32[4]" = torch.ops.aten.mul.Tensor(pos0, tuple0_0);  pos0 = tuple0_0 = None
            mul_1: "f32[4]" = torch.ops.aten.mul.Tensor(mul, tuple0_1);  mul = tuple0_1 = None

                mul_2: "f32[4]" = torch.ops.aten.mul.Tensor(mul_1, myargs_0);  mul_1 = myargs_0 = None
            mul_3: "f32[4]" = torch.ops.aten.mul.Tensor(mul_2, myargs_1);  mul_2 = myargs_1 = None

                mul_4: "f32[4]" = torch.ops.aten.mul.Tensor(mul_3, mykw0);  mul_3 = mykw0 = None

                mul_5: "f32[4]" = torch.ops.aten.mul.Tensor(mul_4, input0);  mul_4 = input0 = None
            mul_6: "f32[4]" = torch.ops.aten.mul.Tensor(mul_5, input1);  mul_5 = input1 = None
            return (mul_6,)

Graph signature: ExportGraphSignature(input_specs=[InputSpec(kind=<InputKind.USER_INPUT: 1>, arg=TensorArgument(name='pos0'), target=None, persistent=None), InputSpec(kind=<InputKind.USER_INPUT: 1>, arg=TensorArgument(name='tuple0_0'), target=None, persistent=None), InputSpec(kind=<InputKind.USER_INPUT: 1>, arg=TensorArgument(name='tuple0_1'), target=None, persistent=None), InputSpec(kind=<InputKind.USER_INPUT: 1>, arg=TensorArgument(name='myargs_0'), target=None, persistent=None), InputSpec(kind=<InputKind.USER_INPUT: 1>, arg=TensorArgument(name='myargs_1'), target=None, persistent=None), InputSpec(kind=<InputKind.USER_INPUT: 1>, arg=TensorArgument(name='mykw0'), target=None, persistent=None), InputSpec(kind=<InputKind.USER_INPUT: 1>, arg=TensorArgument(name='input0'), target=None, persistent=None), InputSpec(kind=<InputKind.USER_INPUT: 1>, arg=TensorArgument(name='input1'), target=None, persistent=None)], output_specs=[OutputSpec(kind=<OutputKind.USER_OUTPUT: 1>, arg=TensorArgument(name='mul_6'), target=None)])
Range constraints: {}

list_contains

Note

Tags: python.data-structure, python.assert, torch.dynamic-shape

Support Level: SUPPORTED

Original source code:

# mypy: allow-untyped-defs
import torch



class ListContains(torch.nn.Module):
    """
    List containment relation can be checked on a dynamic shape or constants.
    """
    def __init__(self):
        super().__init__()

    def forward(self, x):
        assert x.size(-1) in [6, 2]
        assert x.size(0) not in [4, 5, 6]
        assert "monkey" not in ["cow", "pig"]
        return x + x

Result:

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

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

list_unpack

Note

Tags: python.data-structure, python.control-flow

Support Level: SUPPORTED

Original source code:

# mypy: allow-untyped-defs
from typing import List

import torch



class ListUnpack(torch.nn.Module):
    """
    Lists are treated as static construct, therefore unpacking should be
    erased after tracing.
    """

    def __init__(self):
        super().__init__()

    def forward(self, args: List[torch.Tensor]):
        """
        Lists are treated as static construct, therefore unpacking should be
        erased after tracing.
        """
        x, *y = args
        return x + y[0]

Result:

ExportedProgram:
    class GraphModule(torch.nn.Module):
        def forward(self, args_0: "f32[3, 2]", args_1: "i64[]", args_2: "i64[]"):
                add: "f32[3, 2]" = torch.ops.aten.add.Tensor(args_0, args_1);  args_0 = args_1 = None
            return (add,)

Graph signature: ExportGraphSignature(input_specs=[InputSpec(kind=<InputKind.USER_INPUT: 1>, arg=TensorArgument(name='args_0'), target=None, persistent=None), InputSpec(kind=<InputKind.USER_INPUT: 1>, arg=TensorArgument(name='args_1'), target=None, persistent=None), InputSpec(kind=<InputKind.USER_INPUT: 1>, arg=TensorArgument(name='args_2'), target=None, persistent=None)], output_specs=[OutputSpec(kind=<OutputKind.USER_OUTPUT: 1>, arg=TensorArgument(name='add'), target=None)])
Range constraints: {}

Docs

Access comprehensive developer documentation for PyTorch

View Docs

Tutorials

Get in-depth tutorials for beginners and advanced developers

View Tutorials

Resources

Find development resources and get your questions answered

View Resources