ConvTranspose3d#

classtorch.nn.modules.conv.ConvTranspose3d(in_channels,out_channels,kernel_size,stride=1,padding=0,output_padding=0,groups=1,bias=True,dilation=1,padding_mode='zeros',device=None,dtype=None)[source]#

Applies a 3D transposed convolution operator over an input image composed of several inputplanes.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 ora deconvolution (although it is not an actual deconvolution operation as it doesnot compute a true inverse of convolution). For more information, see the visualizationshere and theDeconvolutional Networks paper.

This module supportsTensorFloat32.

On certain ROCm devices, when using float16 inputs this module will usedifferent precision for backward.

• stride controls the stride for the cross-correlation.

• padding controls the amount of implicit zero padding on bothsides fordilation*(kernel_size-1)-padding number of points. See notebelow for details.

• output_padding controls the additional size added to one sideof the output shape. See note below for details.

• dilation controls the spacing between the kernel points; also known as the à trous algorithm.It is harder to describe, but the linkhere has a nice visualization of whatdilation does.

• groups controls the connections between inputs and outputs.in_channels andout_channels must both be divisible bygroups. For example,

  • At groups=1, all inputs are convolved to all outputs.

  • At groups=2, the operation becomes equivalent to having two convlayers side by side, each seeing half the input channelsand producing half the output channels, and both subsequentlyconcatenated.

  • At groups=in_channels, each input channel is convolved withits own set of filters (of size$\frac{\text{out\_channels}}{\text{in\_channels}}$in_channelsout_channels​).

The parameterskernel_size,stride,padding,output_paddingcan either be:

• a singleint – in which case the same value is used for the depth, height and width dimensions

• atuple of three ints – in which case, the firstint is used for the depth dimension,the secondint for the height dimension and the thirdint for the width dimension

Note

Thepadding argument effectively addsdilation*(kernel_size-1)-paddingamount of zero padding to both sizes of the input. This is set so thatwhen aConv3d and aConvTranspose3dare initialized with same parameters, they are inverses of each other inregard to the input and output shapes. However, whenstride>1,Conv3d maps multiple input shapes to the same outputshape.output_padding is provided to resolve this ambiguity byeffectively increasing the calculated output shape on one side. Notethatoutput_padding is only used to find output shape, but doesnot actually add zero-padding to output.

Note

In some circumstances when given tensors on a CUDA device and using 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 settingtorch.backends.cudnn.deterministic=True. SeeReproducibility for more information.

Parameters

• in_channels (int) – Number of channels in the input image

• out_channels (int) – Number of channels produced by the convolution

• kernel_size (int ortuple) – Size of the convolving kernel

• stride (int ortuple,optional) – Stride of the convolution. Default: 1

• padding (int ortuple,optional) –dilation*(kernel_size-1)-padding zero-paddingwill be added to both sides of each dimension in the input. Default: 0

• output_padding (int ortuple,optional) – Additional size added to one sideof each dimension in the output shape. Default: 0

• groups (int,optional) – Number of blocked connections from input channels to output channels. Default: 1

• bias (bool,optional) – IfTrue, adds a learnable bias to the output. Default:True

• dilation (int ortuple,optional) – Spacing between kernel elements. Default: 1

Shape:

• Input:$(N,C_{in},D_{in},H_{in},W_{in})$(N,Cin​,Din​,Hin​,Win​) or$(C_{in},D_{in},H_{in},W_{in})$(Cin​,Din​,Hin​,Win​)

• Output:$(N,C_{out},D_{out},H_{out},W_{out})$(N,Cout​,Dout​,Hout​,Wout​) or$(C_{out},D_{out},H_{out},W_{out})$(Cout​,Dout​,Hout​,Wout​), where

$$D_{out}=(D_{in}-1)\times \text{stride}[0]-2\times \text{padding}[0]+\text{dilation}[0]\times (\text{kernel\_size}[0]-1)+\text{output\_padding}[0]+1$$Dout​=(Din​−1)×stride[0]−2×padding[0]+dilation[0]×(kernel_size[0]−1)+output_padding[0]+1

$$H_{out}=(H_{in}-1)\times \text{stride}[1]-2\times \text{padding}[1]+\text{dilation}[1]\times (\text{kernel\_size}[1]-1)+\text{output\_padding}[1]+1$$Hout​=(Hin​−1)×stride[1]−2×padding[1]+dilation[1]×(kernel_size[1]−1)+output_padding[1]+1

$$W_{out}=(W_{in}-1)\times \text{stride}[2]-2\times \text{padding}[2]+\text{dilation}[2]\times (\text{kernel\_size}[2]-1)+\text{output\_padding}[2]+1$$Wout​=(Win​−1)×stride[2]−2×padding[2]+dilation[2]×(kernel_size[2]−1)+output_padding[2]+1

Variables

• weight (Tensor) – the learnable weights of the module of shape$(\text{in\_channels},\frac{\text{out\_channels}}{\text{groups}},$(in_channels,groupsout_channels​,$\text{kernel\_size[0]},\text{kernel\_size[1]},\text{kernel\_size[2]})$kernel_size[0],kernel_size[1],kernel_size[2]).The values of these weights are sampled from$\mathcal{U}(-\sqrt{k},\sqrt{k})$U(−k​,k​) where$k=\frac{groups}{C_{\text{out}}\ast {\prod }_{i=0}^2\text{kernel\_size}[i]}$k=Cout​∗∏i=02​kernel_size[i]groups​

• bias (Tensor) – the learnable bias of the module of shape (out_channels)Ifbias isTrue, then the values of these weights aresampled from$\mathcal{U}(-\sqrt{k},\sqrt{k})$U(−k​,k​) where$k=\frac{groups}{C_{\text{out}}\ast {\prod }_{i=0}^2\text{kernel\_size}[i]}$k=Cout​∗∏i=02​kernel_size[i]groups​

Examples:

>>># 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)