Skip to contents

Conv2d

Usage

torch_conv2d(
  input,
  weight,
  bias = list(),
  stride = 1L,
  padding = 0L,
  dilation = 1L,
  groups = 1L
)

Arguments

input

input tensor of shape \((\mbox{minibatch} , \mbox{in\_channels} , iH , iW)\)

weight

filters of shape \((\mbox{out\_channels} , \frac{\mbox{in\_channels}}{\mbox{groups}} , kH , kW)\)

bias

optional bias tensor 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

implicit paddings on both sides of the input. Can be a single number or a tuple (padH, padW). Default: 0

dilation

the spacing between kernel elements. Can be a single number or a tuple (dH, dW). Default: 1

groups

split input into groups, \(\mbox{in\_channels}\) should be divisible by the number of groups. Default: 1

conv2d(input, weight, bias=NULL, stride=1, padding=0, dilation=1, groups=1) -> Tensor

Applies a 2D convolution over an input image composed of several input planes.

See nn_conv2d() for details and output shape.

Examples

if (torch_is_installed()) {

# With square kernels and equal stride
filters = torch_randn(c(8,4,3,3))
inputs = torch_randn(c(1,4,5,5))
nnf_conv2d(inputs, filters, padding=1)
}
#> torch_tensor
#> (1,1,.,.) = 
#>    5.1597   3.9603  -2.4935  -0.0889   0.7977
#>   -1.6765   7.7421  13.2688  -1.5230   8.0365
#>    3.6046  -1.0772  11.6349   5.6885  -0.5709
#>   -1.1548  -3.8150 -10.1069  -0.4077  -2.5187
#>   -7.5372  -0.4844   0.2251   2.6514  -0.7511
#> 
#> (1,2,.,.) = 
#>   -0.8449 -14.8820  -8.0421  -9.8562   4.9472
#>   -4.2979  -2.9638   6.7024  14.1146   1.8427
#>   -3.7069   7.4507 -10.9364  -0.8396  -7.4850
#>   -5.3940   2.9988  -5.6019   2.4148  -2.3016
#>    6.6855   4.0635   4.6500   1.1979   3.1503
#> 
#> (1,3,.,.) = 
#>   -0.6765  -7.8094   3.1173  -3.2594  -3.4862
#>   -0.6563  -8.1427  -7.3896  -8.6645   0.2167
#>    8.1900   0.5001  16.1295   1.7405   1.8729
#>   -3.3775   6.8973  -2.4682  10.3839   0.7046
#>   -2.5128   4.3628  -2.2759   1.0886  -1.9905
#> 
#> (1,4,.,.) = 
#>  -3.8900 -6.4439  1.6001  0.2653  0.7250
#>  -2.3296 -8.5687 -6.6549 -3.0899 -3.3301
#>   6.1453  3.4250 -0.9796  5.5812  2.3349
#>   2.3370  0.9344 -4.7569 -6.3572  2.1482
#>   5.9125 -0.7927 -0.5621  0.3140 -2.7526
#> 
#> (1,5,.,.) = 
#>   -1.0798   0.9808 -12.1057  -4.0144  -3.4264
#> ... [the output was truncated (use n=-1 to disable)]
#> [ CPUFloatType{1,8,5,5} ]