Function avg_pool2d

imperative/python/megengine/functional/nn.py:715–762

r"""Applies 2D average pooling over an input tensor. Refer to :class:`~.AvgPool2d` for more information. Args: inp: input tensor of shape :math:`(N, C, H_{\text{in}}, W_{\text{in}})` . kernel_size: size of the window used to calculate the average value. stride: stri

(
    inp: Tensor,
    kernel_size: Union[int, Tuple[int, int]],
    stride: Optional[Union[int, Tuple[int, int]]] = None,
    padding: Union[int, Tuple[int, int]] = 0,
    mode: str = "average_count_exclude_padding",
)

def avg_pool2d(
    inp: Tensor,
    kernel_size: Union[int, Tuple[int, int]],
    stride: Optional[Union[int, Tuple[int, int]]] = None,
    padding: Union[int, Tuple[int, int]] = 0,
    mode: str = "average_count_exclude_padding",
) -> Tensor:
    r"""Applies 2D average pooling over an input tensor.

    Refer to :class:`~.AvgPool2d` for more information.

    Args:
        inp: input tensor of shape :math:`(N, C, H_{\text{in}}, W_{\text{in}})` .
        kernel_size: size of the window used to calculate the average value.
        stride: stride of the window. Default value is ``kernel_size``.
        padding: implicit zero padding added on both sides. Default: 0.
        mode: whether to include the padding values while calculating the average, set
            to "average" will do counting.
            Default: "average_count_exclude_padding"

    Returns:
        output tensor of shape :math:`(N, C, H_{\text{out}}, W_{\text{out}})`.

    Examples:
        >>> import numpy as np
        >>> inp = Tensor(np.arange(1 * 1 * 3 * 4).astype(np.float32).reshape(1, 1, 3, 4))
        >>> F.avg_pool2d(inp, kernel_size=2, stride=2, padding=[1,0], mode="average")
            Tensor([[[[0.25 1.25]
             [6.5  8.5 ]]]], device=xpux:0)
    """
    if stride is None:
        stride = kernel_size
    window_h, window_w = expand_hw(kernel_size)
    stride_h, stride_w = expand_hw(stride)
    padding_h, padding_w = expand_hw(padding)

    op = builtin.Pooling(
        window_h=window_h,
        window_w=window_w,
        stride_h=stride_h,
        stride_w=stride_w,
        pad_h=padding_h,
        pad_w=padding_w,
        mode=mode,
        strategy=get_execution_strategy(),
    )
    (output,) = apply(op, inp)
    return output


def adaptive_max_pool2d(