AdaptiveMaxPool2d

class brainpy.dnn.AdaptiveMaxPool2d(target_shape, channel_axis=-1, name=None, mode=None)

Adaptive two-dimensional maximum down-sampling.

Parameters:

• target_shape (Union[int, Sequence[int]]) – The target output shape.

• channel_axis (Optional[int]) – Axis of the spatial channels for which pooling is skipped. If None, there is no channel axis.

• name (Optional[str]) – The class name.

• mode (Optional[Mode]) – The computing mode.