Conv_transpose2d — torch_conv_transpose2d • torch

Conv_transpose2d

Usage

torch_conv_transpose2d(
  input,
  weight,
  bias = list(),
  stride = 1L,
  padding = 0L,
  output_padding = 0L,
  groups = 1L,
  dilation = 1L
)

Arguments

input: input tensor of shape $(minibatch, in\_channels, i H, i W)$
weight: filters of shape $(in\_channels, \frac{out\_channels}{groups}, k H, k W)$
bias: optional bias of shape $(out\_channels)$ . Default: NULL
stride: the stride of the convolving kernel. Can be a single number or a tuple (sH, sW). Default: 1
padding: dilation * (kernel_size - 1) - padding zero-padding will be added to both sides of each dimension in the input. Can be a single number or a tuple (padH, padW). Default: 0
output_padding: additional size added to one side of each dimension in the output shape. Can be a single number or a tuple (out_padH, out_padW). Default: 0
groups: split input into groups, $in\_channels$ should be divisible by the number of groups. Default: 1
dilation: the spacing between kernel elements. Can be a single number or a tuple (dH, dW). Default: 1

conv_transpose2d(input, weight, bias=NULL, stride=1, padding=0, output_padding=0, groups=1, dilation=1) -> Tensor

Applies a 2D transposed convolution operator over an input image composed of several input planes, sometimes also called "deconvolution".

See nn_conv_transpose2d() for details and output shape.

Examples

if (torch_is_installed()) {

# With square kernels and equal stride
inputs = torch_randn(c(1, 4, 5, 5))
weights = torch_randn(c(4, 8, 3, 3))
nnf_conv_transpose2d(inputs, weights, padding=1)
}
#> torch_tensor
#> (1,1,.,.) = 
#>    1.1253   5.4441   0.1970  -3.6803  -3.4119
#>    2.4322   2.1541  12.5961   6.5995   5.1548
#>   -0.6315  -6.8499  -3.0669  -1.1467  -1.7125
#>    1.9091  -2.0217   7.8142  -3.3610  -7.9866
#>   -3.9312  -4.6840   8.8192   5.3539  13.4730
#> 
#> (1,2,.,.) = 
#>   -4.5093   6.1076   3.1789   4.3431  -0.7988
#>   -4.7695 -11.7868  -1.9039  -3.8325   5.6265
#>   -9.8235  -3.9974   8.8578  14.6611   1.8173
#>    2.7077  -9.4768  -7.3567  -9.8555  -3.8919
#>   -1.9548  -1.2049   3.1667   9.8986   8.7437
#> 
#> (1,3,.,.) = 
#>   2.8846  1.9904 -2.0205  3.7424 -1.4139
#>  -2.3467 -7.5063  5.6216 -4.8747 -0.7202
#>   6.3255  6.5974 -3.7416  4.3378 -5.9021
#>   0.4539  5.2796 -2.6167  3.5263  4.8052
#>   3.7287  3.2542  1.0560  5.3785 -1.1284
#> 
#> (1,4,.,.) = 
#>   -6.3726  -2.3501   1.6128  11.2059   0.7719
#>   -3.2164  -4.2570  -1.9258   1.2205  -1.6253
#>   -9.2520   1.2597  -5.7226 -10.1837   2.9304
#>    5.2055  -5.9076  -0.5576   2.1778   5.3254
#>   -1.6957  -1.5328   5.0729   4.5368   4.7827
#> 
#> (1,5,.,.) = 
#>   0.4434  0.8264  2.0969 -2.7700  4.8120
#> ... [the output was truncated (use n=-1 to disable)]
#> [ CPUFloatType{1,8,5,5} ]