depthwise_conv2d

dragon.vm.torch.nn.functional.depthwise_conv2d(
  input,
  weight,
  bias=None,
  stride=1,
  padding=0,
  dilation=1
)[source]

Apply the 2d depthwise convolution to input.

Parameters:
  • input (dragon.vm.torch.Tensor) – The input tensor.
  • weight (dragon.vm.torch.Tensor) – The weight tensor.
  • bias (dragon.vm.torch.Tensor, optional) – The optional bias.
  • stride (Sequence[int], default=1) – The stride of sliding window.
  • padding (Sequence[int], default=0) – The zero-padding size.
  • dilation (Sequence[int], default=1) – The rate of dilated kernel.
Returns:

dragon.vm.torch.Tensor – The output tensor.