Conv2d

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,.,.) = 
#>  -6.2293  3.6368 -6.9950 -3.2092  4.0164
#>   4.4499  5.6002  5.9233  0.9858  4.2309
#>  -5.9799 -12.2515 -1.1408  5.2784  4.7481
#>   3.5944  0.3573  1.3755 -0.4545 -0.4302
#>   4.4743  3.5464  2.9807  3.5504  3.6378
#> 
#> (1,2,.,.) = 
#>  -14.1458  -2.4284  -1.6412  -4.7466  -0.5833
#>   -2.7556  -2.9063 -13.3437  -1.0216   0.0556
#>    8.1979  -9.5774  -2.3079  -9.8105  -6.1516
#>   -3.3297   1.0058   3.2910  -4.9661 -10.2056
#>   -2.9389   7.1504  -4.6618   3.7112   0.9689
#> 
#> (1,3,.,.) = 
#>   2.0213  4.8348 -0.0753 -5.5559  2.1641
#>   7.3397  5.3186 -0.5203 -10.4046 -2.4127
#>  -2.8926 -5.7183  7.3088  1.4437 -0.2859
#>  -0.1150  1.8809 -4.1334  4.5544  1.0229
#>  -4.4993  0.7790 -2.1356  1.6472 -3.0782
#> 
#> (1,4,.,.) = 
#>   -8.0488   0.3474   9.3235  -1.6900  -3.2958
#>   -8.0323  -9.7287  -6.8925   4.5854  10.4326
#>   10.2315 -12.7851  -6.9631 -11.5428  -1.7736
#>    1.9657   0.0843  -5.3749  -8.2050  -2.8523
#>   -1.6009   3.4971   2.5930   0.7100   5.0198
#> 
#> (1,5,.,.) = 
#>   -3.1366   2.1507  -1.2642   1.5051  -3.3853
#> ... [the output was truncated (use n=-1 to disable)]
#> [ CPUFloatType{1,8,5,5} ]