torchcvnn.transforms.FFT2

class torchcvnn.transforms.FFT2(axis: Tuple[int, ...] = (-2, -1))

Applies 2D Fast Fourier Transform (FFT) to the input. This transform computes the 2D FFT along specified dimensions of the input array/tensor. It applies FFT2 and shifts zero-frequency components to the center.

Args

axisTuple[int, …], optional

The axes over which to compute the FFT. Default is (-2, -1).

Returns

numpy.ndarray or torch.Tensor

The 2D Fourier transformed input with zero-frequency components centered. Output has the same shape as input.

Notes

• Transform is applied along specified dimensions (axis).

__init__(axis: Tuple[int, ...] = (-2, -1))

Methods

__init__([axis])