bartorch.fft

Contents

bartorch.fft#

bartorch.fft()#

Fourier transform along axes, centred by default.

Centred, the zero frequency and the image origin are both at index n // 2. BART’s fft.

Parameters:

  • input (torch.Tensor) – Array to transform.

  • axes (int or tuple of int) – Axes to transform, as indices into input.shape.

  • inverse (bool, default=False) – Transform with the positive exponent.

  • unitary (bool, default=False) – Scale by one over the square root of the transformed size; otherwise unnormalized.

  • uncentred (bool, default=False) – Keep the zero frequency and the origin at index zero.

Returns:

The shape of input.

Return type:

torch.Tensor

Examples

>>> spectrum = fft(image, axes=(-2, -1), unitary=True)
Contents