pool3d

dragon.nn.pool3d(
  inputs,
  kernel_shape,
  strides,
  pads=0,
  padding='VALID',
  mode='max',
  global_pool=False,
  ceil_mode=False,
  data_format='NCHW',
  **kwargs
)[source]

Apply the 3d pooling.

  • Set mode for the specific pooling type, default is maxpool.
  • Use global_pool to apply the global pooling further.
  • If data_format is 'NCHW', excepts input shape \((N, C, D, H, W)\), and output shape is \((N, C, D_{\text{out}}, H_{\text{out}}, W_{\text{out}})\).
  • If data_format is 'NHWC', excepts input shape \((N, D, H, W, C)\), and output shape is \((N, D_{\text{out}}, H_{\text{out}}, W_{\text{out}}, C)\).
  • If padding is 'VALID', pads controls the explicit padding size. Otherwise, size are computed automatically use the given method.

Examples:

x = dragon.ones((1, 2, 2, 2, 2))
y = dragon.nn.pool3d(x, kernel_shape=2, strides=2)
assert y.shape == (1, 2, 1, 1, 1)
Parameters:
  • inputs (dragon.Tensor) – The input tensor.
  • kernel_shape (Union[int, Sequence[int]], required) – The shape of pooling window.
  • strides (Union[int, Sequence[int]], required) – The stride of pooling window.
  • pads (Union[int, Sequence[int]], optional, default=0) – The zero padding size.
  • padding (str, optional, default='VALID') – 'VALID', 'SAME', 'SAME_UPPER' or 'SAME_LOWER'.
  • mode (str, optional, default='max') – 'max' or 'avg'.
  • global_pool (bool, optional, default=False) – Apply the global pooling or not.
  • ceil_mode (bool, optional, default=False) – Ceil or floor the boundary.
  • data_format (str, optional, default='NCHW') – 'NCHW' or 'NHWC'.
Returns:

dragon.Tensor – The output tensor.