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


def dynamic_shape_map(xs, y):
    """
    functorch map() maps a function over the first tensor dimension.
    """

    def body(x, y):
        return x + y

    return map(body, xs, y)

Result:

ExportedProgram:
    class GraphModule(torch.nn.Module):
        def forward(self, l_xs_: "f32[3, 2]", l_y_: "f32[2]"):
                body_graph_0 = self.body_graph_0
            map_impl = torch.ops.higher_order.map_impl(body_graph_0, 1, l_xs_, l_y_);  body_graph_0 = l_xs_ = l_y_ = 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='l_xs_'), target=None), InputSpec(kind=<InputKind.USER_INPUT: 1>, arg=TensorArgument(name='l_y_'), target=None)], output_specs=[OutputSpec(kind=<OutputKind.USER_OUTPUT: 1>, arg=TensorArgument(name='getitem'), target=None)])
Range constraints: {}
Equality constraints: []