ConvTranspose3d(in_channels: int, out_channels: int, kernel_size: Union[int, Tuple[int, int, int]], stride: Union[int, Tuple[int, int, int]] = 1, padding: Union[int, Tuple[int, int, int]] = 0, output_padding: Union[int, Tuple[int, int, int]] = 0, groups: int = 1, bias: bool = True, dilation: Union[int, Tuple[int, int, int]] = 1, padding_mode: str = 'zeros')¶
Applies a 3D transposed convolution operator over an input image composed of several input planes. The transposed convolution operator multiplies each input value element-wise by a learnable kernel, and sums over the outputs from all input feature planes.
This module can be seen as the gradient of Conv3d with respect to its input. It is also known as a fractionally-strided convolution or a deconvolution (although it is not an actual deconvolution operation).
stridecontrols the stride for the cross-correlation.
paddingcontrols the amount of implicit zero-paddings on both sides for
dilation * (kernel_size - 1) - paddingnumber of points. See note below for details.
output_paddingcontrols the additional size added to one side of the output shape. See note below for details.
dilationcontrols the spacing between the kernel points; also known as the à trous algorithm. It is harder to describe, but this link has a nice visualization of what
groupscontrols the connections between inputs and outputs.
out_channelsmust both be divisible by
groups. For example,
At groups=1, all inputs are convolved to all outputs.
At groups=2, the operation becomes equivalent to having two conv layers side by side, each seeing half the input channels, and producing half the output channels, and both subsequently concatenated.
in_channels, each input channel is convolved with its own set of filters (of size ).
output_paddingcan either be:
int– in which case the same value is used for the depth, height and width dimensions
tupleof three ints – in which case, the first int is used for the depth dimension, the second int for the height dimension and the third int for the width dimension
Depending of the size of your kernel, several (of the last) columns of the input might be lost, because it is a valid cross-correlation, and not a full cross-correlation. It is up to the user to add proper padding.
paddingargument effectively adds
dilation * (kernel_size - 1) - paddingamount of zero padding to both sizes of the input. This is set so that when a
ConvTranspose3dare initialized with same parameters, they are inverses of each other in regard to the input and output shapes. However, when
stride > 1,
Conv3dmaps multiple input shapes to the same output shape.
output_paddingis provided to resolve this ambiguity by effectively increasing the calculated output shape on one side. Note that
output_paddingis only used to find output shape, but does not actually add zero-padding to output.
In some circumstances when using the CUDA backend with CuDNN, this operator may select a nondeterministic algorithm to increase performance. If this is undesirable, you can try to make the operation deterministic (potentially at a performance cost) by setting
torch.backends.cudnn.deterministic = True. Please see the notes on Reproducibility for background.
in_channels (int) – Number of channels in the input image
out_channels (int) – Number of channels produced by the convolution
groups (int, optional) – Number of blocked connections from input channels to output channels. Default: 1
bias (bool, optional) – If
True, adds a learnable bias to the output. Default:
>>> # With square kernels and equal stride >>> m = nn.ConvTranspose3d(16, 33, 3, stride=2) >>> # non-square kernels and unequal stride and with padding >>> m = nn.ConvTranspose3d(16, 33, (3, 5, 2), stride=(2, 1, 1), padding=(0, 4, 2)) >>> input = torch.randn(20, 16, 10, 50, 100) >>> output = m(input)