AdaptiveAvgPool2D

Description

Performs 2D adaptive average pooling calculation on the input.

Input

x: tensor of type float16 or float, with shape [N, C, H_x, W_x]. The data format can be ND.

Attribute

output_size: list of two ints, specifying the output H_y and W_y shape sizes.

Output

y: output tensor of the identical data type as input x, with shape [N, C, H_y, W_y]. The data format can be ND.

Constraints

None

ONNX Opset Support

Opset v8/v9/v10/v11/v12/v13/v14/v15/v16