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,.,.) = 
#>  -3.7738 -1.2083 -8.1065  1.3506  5.4331
#>  -2.6131  2.3640  1.2309  1.4801 -5.3338
#>  -5.0187 -2.4033  0.5710 -1.6880  7.3623
#>  -1.5215  1.5235  0.0348  7.3801 -1.7101
#>  -8.5764 -2.9294  4.9842  3.5400 -1.8436
#> 
#> (1,2,.,.) = 
#>  -4.6851  1.3777 -1.4969 -11.3123  6.7215
#>  -1.5802  8.2403  8.6541  3.3405  1.9431
#>  -2.2322  8.4269  1.9300  0.1297 -2.2860
#>   8.3438 -7.1255 -1.8514 -2.1738  7.8578
#>   0.0024 -1.7037  0.2490 -2.8967  1.0559
#> 
#> (1,3,.,.) = 
#>   2.5347  7.3605  0.8325  7.3045 -2.8945
#>   2.2124  5.4812  7.6731 -4.5134 -1.1054
#>   1.7022 -7.9886 -2.7748 -3.1381  7.6821
#>  -14.2288  0.4543  2.5085  2.6854 -6.7399
#>  -4.9975 -1.9640 -1.2460  0.0871  5.4244
#> 
#> (1,4,.,.) = 
#>  -4.2018  2.5438  3.7225  6.2680  6.3939
#>   0.4814  1.8399  2.2175 -1.8820 -11.3187
#>   1.7938 -1.3694 -9.8122  3.0611  5.0931
#>   7.7106  0.4313  3.9273 -3.4084  1.6744
#>   6.5445 -1.2279 -2.0083 -1.1604 -0.0577
#> 
#> (1,5,.,.) = 
#>   0.0969  0.6377  4.5573  5.0655  1.0746
#> ... [the output was truncated (use n=-1 to disable)]
#> [ CPUFloatType{1,8,5,5} ]