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,.,.) = 
#>   -5.9030   2.3719   3.2671  -6.3717   4.9661
#>   -9.3490  -2.2151   2.7751  -1.1339   1.3596
#>   -3.3151   2.1332  11.8538  -1.7507  -3.1123
#>    1.8214  -5.5401   5.5321  -2.6120   4.0258
#>   -7.6614   4.7752   2.8485   7.5207  -3.6628
#> 
#> (1,2,.,.) = 
#>    5.2193   2.6121  -1.2951   5.8279 -10.5490
#>    1.0212   3.0566   5.3418  -0.5528 -19.0809
#>   -2.4759   5.2707  -1.3049 -11.1098  -0.6705
#>   -0.8664   7.7155  -7.3916  13.1146   3.6660
#>    3.2671  -6.9516   8.2204   4.8681  -4.0088
#> 
#> (1,3,.,.) = 
#>   -8.2957   5.5987   3.3467  -3.6443   1.0645
#>    8.2492  -3.3375  -6.9863  -6.0146  -8.4941
#>    3.4540  -3.0300  -5.0501   5.6115   1.7198
#>    7.3158  -2.3231   0.1430  -0.7191  10.1638
#>   -0.5320  -0.4238  -2.4003   2.9976   3.9288
#> 
#> (1,4,.,.) = 
#>    8.2648   3.0973  10.2399   3.8716  -6.8306
#>    1.4585   3.1623  -1.6594   3.6279   1.1162
#>    2.4186  -4.9536   4.4081  -3.1237  -8.9828
#>   -7.3744  -2.3638   0.2175   8.9855   1.6173
#>    1.3701   3.9888   3.5332   2.9809   0.5901
#> 
#> (1,5,.,.) = 
#>  -3.8535 -3.2569  7.7058 -1.4514 -1.5325
#> ... [the output was truncated (use n=-1 to disable)]
#> [ CPUFloatType{1,8,5,5} ]