Conv2d — torch_conv2d • torch

Conv2d

Usage

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

Arguments

input: input tensor of shape $(minibatch, in\_channels, i H, i W)$
weight: filters of shape $(out\_channels, \frac{in\_channels}{groups}, k H, k W)$
bias: optional bias tensor of shape $(out\_channels)$ . 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, $in\_channels$ 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,.,.) = 
#>    0.0189   4.9447   0.6224   0.7597  -4.5645
#>    5.9175   7.0085   5.9416  -5.2371   8.0103
#>   -0.9964  -7.0508  -0.9801  -5.0388   4.4114
#>    5.4303  -5.1018   3.9502  16.1350  -3.8399
#>    0.9899  -3.3628  -6.8478   0.9403   3.0106
#> 
#> (1,2,.,.) = 
#>  -0.4150  1.9061  4.7993  4.0147 -2.4514
#>  -2.1454 -16.8381 -3.7747  6.1538 -3.3264
#>   2.5807 -12.0448  5.3367 -3.1636 -7.9218
#>   2.5600 -4.3728  8.6374  0.6756 -2.6467
#>  -3.3193 -8.9347  4.0515  2.1577  1.3617
#> 
#> (1,3,.,.) = 
#>   -7.2225   2.0309   5.5763  -5.5566  -4.7529
#>    0.1236   2.6598  -6.3783   7.5947  -4.1330
#>    6.4922  -1.9247   1.0046  -0.3332  10.4420
#>    6.7431  -0.1652   9.3561 -12.9978   1.6157
#>    1.6747  -8.7162   1.9752   5.6119  -7.2847
#> 
#> (1,4,.,.) = 
#>    7.6439  -2.1554  -2.2380   1.6979   5.8495
#>    8.1608   9.1738  -9.0112   1.5795   2.2012
#>   -3.5933  -0.0972  -7.7132  -6.2405  -4.9428
#>  -10.1969  10.3241   2.3233   2.7456   5.4896
#>   -1.1503  -5.4917  -0.2620  -1.1470  -4.7064
#> 
#> (1,5,.,.) = 
#>  -3.6470  2.4733  3.6185 -5.0308 -0.5395
#> ... [the output was truncated (use n=-1 to disable)]
#> [ CPUFloatType{1,8,5,5} ]