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 \((\mbox{minibatch} , \mbox{in\_channels} , iH , iW)\)
weight: filters of shape \((\mbox{in\_channels} , \frac{\mbox{out\_channels}}{\mbox{groups}} , kH , kW)\)
bias: optional bias of shape \((\mbox{out\_channels})\). 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, \(\mbox{in\_channels}\) 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,.,.) = 
#>   -4.0098  -5.0313   6.8105  -1.0021   5.3903
#>   -3.2639  -4.7224  -3.3878  -1.1967   4.0446
#>   10.0232  -4.4150  -8.6390   3.8996   1.6335
#>    0.9296   2.9197  -2.3360  -1.6417  -2.7890
#>   -4.6193  11.4419  -7.5627   2.9798  -1.7773
#> 
#> (1,2,.,.) = 
#>   -0.1793   3.5783   4.7631   5.9809   2.0518
#>    1.5537  -4.2509 -14.4141  -4.8391   0.9355
#>   -6.1943   7.3959   5.5308  -3.1655  -8.4654
#>    3.2452 -12.2044  -4.8781   3.6186  10.5318
#>    2.8551  -1.6354   1.9752   7.5736  -8.7246
#> 
#> (1,3,.,.) = 
#>   0.6693 -11.2314  9.6125 -3.4877  1.9307
#>  -0.6870  5.4097 -4.4927 -1.0477 -7.2388
#>  -10.4915  6.4477  5.2581 -2.3866 -0.1821
#>  -0.8842  3.5092  0.9792 -2.6495  6.7911
#>  -3.2045 -4.5648  8.2479  3.4012 -1.5479
#> 
#> (1,4,.,.) = 
#>   8.1714  3.0506 -0.7998 -2.1986 -6.7395
#>  -7.7921  0.8190 -2.3912 -1.0331  3.8875
#>  -6.4955 -1.6437  2.8152 -5.6818 -9.8899
#>   2.1046 -14.9640 -0.2641  2.2211  9.4924
#>   2.3449  1.2507 -7.8120  7.2972 -3.8666
#> 
#> (1,5,.,.) = 
#>  -3.2535 -2.4848  5.6120 -8.4186  2.4904
#> ... [the output was truncated (use n=-1 to disable)]
#> [ CPUFloatType{1,8,5,5} ]