espnet2.gan_svs.avocodo.avocodo.get_padding

source

espnet2.gan_svs.avocodo.avocodo.get_padding(kernel_size, dilation=1)

Calculate the padding needed for a convolution operation.

This function computes the amount of padding required to maintain the spatial dimensions of the input tensor when applying a convolution with a specified kernel size and dilation.

• Parameters:
  • kernel_size (int) – The size of the convolution kernel.
  • dilation (int , optional) – The dilation rate for the convolution. Default is 1, which means no dilation.
• Returns: The calculated padding value.
• Return type: int

Examples

>>> get_padding(3)
1
>>> get_padding(5, dilation=2)
4
>>> get_padding(7, dilation=3)
9