GlobalMaxPool#

GlobalMaxPool - 1#

Version

name: GlobalMaxPool (GitHub)
domain: main
since_version: 1
function:
support_level: SupportType.COMMON
shape inference: True

This version of the operator has been available since version 1.

Summary

Inputs

X (heterogeneous) - T:

Outputs

Y (heterogeneous) - T:

Type Constraints

T in ( tensor(double), tensor(float), tensor(float16) ): Constrain input and output types to float tensors.

Examples

default

import numpy as np
import onnx

node = onnx.helper.make_node(
    "GlobalMaxPool",
    inputs=["x"],
    outputs=["y"],
)
x = np.random.randn(1, 3, 5, 5).astype(np.float32)
y = np.max(x, axis=tuple(range(2, np.ndim(x))), keepdims=True)
expect(node, inputs=[x], outputs=[y], name="test_globalmaxpool")

_globalmaxpool_precomputed

import numpy as np
import onnx

node = onnx.helper.make_node(
    "GlobalMaxPool",
    inputs=["x"],
    outputs=["y"],
)
x = np.array(
    [
        [
            [
                [1, 2, 3],
                [4, 5, 6],
                [7, 8, 9],
            ]
        ]
    ]
).astype(np.float32)
y = np.array([[[[9]]]]).astype(np.float32)
expect(node, inputs=[x], outputs=[y], name="test_globalmaxpool_precomputed")