PixelUnshuffle

class torch.nn.modules.pixelshuffle.PixelUnshuffle(downscale_factor)

Reverse the PixelShuffle operation.

Reverses the PixelShuffle operation by rearranging elements in a tensor of shape $(\ast ,C,H\times r,W\times r)$ to a tensor of shape $(\ast ,C\times r^2,H,W)$, where r is a downscale factor.

See the paper: Real-Time Single Image and Video Super-Resolution Using an Efficient Sub-Pixel Convolutional Neural Network by Shi et al. (2016) for more details.

Parameters

downscale_factor (int) – factor to decrease spatial resolution by

Shape:

• Input: $(\ast ,C_{in},H_{in},W_{in})$, where * is zero or more batch dimensions

• Output: $(\ast ,C_{out},H_{out},W_{out})$, where

$$C_{out}=C_{in}\times {\text{downscale\_factor}}^2$$

$$H_{out}=H_{in}÷\text{downscale\_factor}$$

$$W_{out}=W_{in}÷\text{downscale\_factor}$$

Examples:

>>> pixel_unshuffle = nn.PixelUnshuffle(3)
>>> input = torch.randn(1, 1, 12, 12)
>>> output = pixel_unshuffle(input)
>>> print(output.size())
torch.Size([1, 9, 4, 4])

extra_repr()

Return the extra representation of the module.

Return type

str

forward(input)

Runs the forward pass.

Return type

Tensor