torch.nn.functional.affine_grid¶

torch.nn.functional.
affine_grid
(theta, size, align_corners=None)[source]¶ Generates a 2D or 3D flow field (sampling grid), given a batch of affine matrices
theta
.Note
This function is often used in conjunction with
grid_sample()
to build Spatial Transformer Networks . Parameters
theta (Tensor) – input batch of affine matrices with shape ($N \times 2 \times 3$) for 2D or ($N \times 3 \times 4$) for 3D
size (torch.Size) – the target output image size. ($N \times C \times H \times W$ for 2D or $N \times C \times D \times H \times W$ for 3D) Example: torch.Size((32, 3, 24, 24))
align_corners (bool, optional) – if
True
, consider1
and1
to refer to the centers of the corner pixels rather than the image corners. Refer togrid_sample()
for a more complete description. A grid generated byaffine_grid()
should be passed togrid_sample()
with the same setting for this option. Default:False
 Returns
output Tensor of size ($N \times H \times W \times 2$)
 Return type
output (Tensor)
Warning
When
align_corners = True
, the grid positions depend on the pixel size relative to the input image size, and so the locations sampled bygrid_sample()
will differ for the same input given at different resolutions (that is, after being upsampled or downsampled). The default behavior up to version 1.2.0 wasalign_corners = True
. Since then, the default behavior has been changed toalign_corners = False
, in order to bring it in line with the default forinterpolate()
.Warning
When
align_corners = True
, 2D affine transforms on 1D data and 3D affine transforms on 2D data (that is, when one of the spatial dimensions has unit size) are illdefined, and not an intended use case. This is not a problem whenalign_corners = False
. Up to version 1.2.0, all grid points along a unit dimension were considered arbitrarily to be at1
. From version 1.3.0, underalign_corners = True
all grid points along a unit dimension are considered to be at`0
(the center of the input image).