GlobalAveragePool

Description

Pass an input Tensor and apply average pooling to the values in each channel.

Input

One input:

x: input tensor of type float16 or float. Must be in ND, NCHW, or NCDHW format. Has at least 4 dimensions and at most 5 dimensions.

Output

One output:

y: output tensor. The data type and format must be the same as those of x. The shape of y is the same as that of x. The first two dimensions of y are the same as those of x, and the other spatial dimensions are compressed to 1. The first two dimensions are the same as those of x, and the last dimension is 1.

Constraints

None.

ONNX Opset Support

Opset v8/v9/v10/v11/v12/v13/v14/v15/v16/v17/v18