你也可以做 tf.reduce_mean(x, axis=[1,2])，特别是如果你的高度和宽度是未定义.

You could also do tf.reduce_mean(x, axis=[1,2]), specially if your height and width are not defined.

通常，在 CNN 中，张量具有 b, h, w, c 的形状，其中 b 是批大小，w和h对应的是宽高尺寸，c是通道数/过滤器数.

Typically, in a CNN, tensors have a shape of b, h, w, c where b is the batch size, w and h correspond to the width and height dimensions, and c is the number of channels/filters.

当你沿轴减少 [1,2] 时，你会减少张量的第一维和第二维(保持批量大小，以及通道/过滤器的数量)

When you reduce along the axis [1,2], you reduce on the first and second dimensions of the tensor (keeping the batch size, and the number of channels/filters)