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,.,.) = 
#>   2.4975  0.7018 -4.6872 -6.0053  8.9148
#>  -3.9835 -0.0577 -5.2685  4.8591 -2.4998
#>   1.0087 -3.5754 -6.7111 -4.4273  6.8516
#>  -4.6899  4.3283 -1.2895 -5.7197 -0.3994
#>  -10.3614  4.5721 -3.1797 -1.5449  2.2693
#> 
#> (1,2,.,.) = 
#>   7.6083  5.9991 -4.0038 -0.8082 -2.2552
#>   1.4880 -9.6177  2.6376 -5.2796  4.5499
#>   1.6570 -0.0920  7.3322 -10.5819  8.7521
#>   8.6734 -13.4999  2.2524  8.5251 -0.2070
#>  -9.3134 -5.6651  7.4942 -5.0487 -6.8508
#> 
#> (1,3,.,.) = 
#>  -7.2925 -1.7028  7.0622 -3.4469 -4.5846
#>   2.2165 -8.8995 -1.2713 -2.6526 -2.3976
#>  -5.7185 -5.6672 -1.9402  4.6768  2.8479
#>  -3.5815 -4.0945 -2.7768  2.9213  2.9248
#>  -3.1693  2.9927  1.4101  2.8391 -0.4298
#> 
#> (1,4,.,.) = 
#>  -2.3415  2.0156 -3.1422 -0.3747 -0.4778
#>  -2.2533  6.8492  2.7716 -4.1783 -1.0009
#>   0.5770  1.0754 -9.5410  3.4406 -2.3913
#>  -0.6745 -0.4763  2.6882 -0.2782  2.1338
#>  -4.6347 -5.2201  0.3038  1.2836 -0.2038
#> 
#> (1,5,.,.) = 
#>   -7.9878   4.6672  -5.0157   9.0481   2.7546
#> ... [the output was truncated (use n=-1 to disable)]
#> [ CPUFloatType{1,8,5,5} ]