Shortcuts

torch.map

dynamic_shape_map

Note

Tags: torch.dynamic-shape, torch.map

Support Level: SUPPORTED

Original source code:

import torch

from functorch.experimental.control_flow import map


class DynamicShapeMap(torch.nn.Module):
    """
    functorch map() maps a function over the first tensor dimension.
    """

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

    def forward(self, xs, y):
        def body(x, y):
            return x + y

        return map(body, xs, y)

Result:

ExportedProgram:
    class GraphModule(torch.nn.Module):
        def forward(self, arg0_1: "f32[3, 2]", arg1_1: "f32[2]"):
                body_graph_0 = self.body_graph_0
            map_impl = torch.ops.higher_order.map_impl(body_graph_0, [arg0_1], [arg1_1]);  body_graph_0 = arg0_1 = arg1_1 = None
            getitem: "f32[3, 2]" = map_impl[0];  map_impl = None
            return (getitem,)

        class <lambda>(torch.nn.Module):
            def forward(self, arg0_1: "f32[2]", arg1_1: "f32[2]"):
                        add: "f32[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)], output_specs=[OutputSpec(kind=<OutputKind.USER_OUTPUT: 1>, arg=TensorArgument(name='getitem'), 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