mindspore.ops.GridSampler3D

class mindspore.ops.GridSampler3D(interpolation_mode='bilinear', padding_mode='zeros', align_corners=False)[source]

Given an input and a grid, the output is calculated using the input values and pixel positions in the grid. Only volume (5-D) input is supported.

Warning

This is an experimental API that is subject to change or deletion.

Refer to mindspore.ops.grid_sample() for more details.

Parameters

interpolation_mode (str, optional) – An optional string specifying the interpolation method. The optional values are “bilinear” or “nearest”. Default: “bilinear”.
padding_mode (str, optional) – An optional string specifying the pad method. The optional values are “zeros”, “border” or “reflection”. Default: “zeros”.
align_corners (bool, optional) – An optional bool specifying alignment method. If set to True, the extrema (-1 and 1) are considered as referring to the center points of the input’s corner pixels. If set to False, they are instead considered as referring to the corner points of the input’s corner pixels, making the sampling more resolution agnostic. Default: False.

Inputs:

input_x (Tensor) - A 5-D tensor with dtype of float32 or float64 and shape of \((N, C, D_{in}, H_{in}, W_{in})\).
grid (Tensor) - A 5-D tensor whose dtype is the same as input_x and whose shape is \((N, D_{out}, H_{out}, W_{out}, 3)\).

Outputs:

A 5-D Tensor whose dtype is the same as input_x and whose shape is \((N, C, D_{out}, H_{out}, W_{out})\).

Supported Platforms:

Ascend GPU CPU

Examples

>>> gridsampler = ops.GridSampler3D(interpolation_mode='bilinear', padding_mode='zeros', align_corners=True)
>>> input_x = Tensor(np.arange(32).reshape((2, 2, 2, 2, 2)).astype(np.float32))
>>> grid = Tensor(np.arange(-0.2, 1, 0.1).reshape((2, 2, 1, 1, 3)).astype(np.float32))
>>> output = gridsampler(input_x, grid)
>>> print(output)
[[[[[ 3.3     ]]
   [[ 4.35    ]]]
  [[[11.300001]]
   [[12.349999]]]]
 [[[[21.4     ]]
   [[22.449999]]]
  [[[29.4     ]]
   [[30.449999]]]]]