bartorch.maps_to_kernels

bartorch.maps_to_kernels#

bartorch.maps_to_kernels()#

Coil sensitivities as k-space kernels: the centre of each map’s unitary spectrum.

What bartorch.linop.NoncartesianSense takes with kernels=True. Content outside the kernel’s band is lost; kernels_to_maps() gives the maps the kernels stand for, to measure that loss.

Parameters:

  • maps (torch.Tensor) – Sensitivities ([sets,] coils, *spatial).

  • size (int or tuple of int) – Kernel size per spatial axis, or one size for all.

  • ndim (int, default=None) – Number of spatial axes, for a single size over a bank with sets. By default the length of size, or every axis after the first.

Returns:

Kernels ([sets,] coils, *size).

Return type:

torch.Tensor