xlnstorch.csrc.avg_pool1d_forward

xlnstorch.csrc.avg_pool1d_forward#

xlnstorch.csrc.avg_pool1d_forward(x, kernel_size, base, stride, padding, ceil_mode, count_include_pad)#

Perform LNS 1D average pooling.

Parameters:

  • x (torch.Tensor) – The input tensor in LNS format (torch.int64).

  • kernel_size (int) – The size of the pooling kernel.

  • base (torch.Tensor) – The base for the LNS representation.

  • stride (int, optional) – The stride of the pooling operation. Defaults to kernel_size if None.

  • padding (int) – The padding applied to the input tensor.

  • ceil_mode (bool) – If True, will use ceil instead of floor to compute the output shape.

  • count_include_pad (bool) – If True, will include the zero-padding in the averaging calculation.

Returns:

The output tensor in LNS format (torch.int64) after average pooling.

Return type:

torch.Tensor

Contents