conv2

dnois.fourier.conv2(f1: Tensor, f2: Tensor, dx: float | Tensor = None, dy: float | Tensor = None, dims: tuple[int, int] = (-2, -1), out: Literal['full', 'same', 'valid'] = 'full', padding: int | tuple[int, int] | str = 'linear', simpson: bool = False, real: bool = None) → Tensor

2D version of conv().

See also

conv() and dconv2().

Parameters:

  • f1 (Tensor) – The first array \(f_1\).

  • f2 (Tensor) – The second array \(f_2\).

  • dx – Sampling spacing in x direction i.e. the second dimension, either a float, a 0D tensor, or a tensor broadcastable with f1 and f2. Default: omitted.

  • dy – Sampling spacing in y direction i.e. the first dimension, similar to dx. Default: identical to dx.

  • dims (tuple[int, int]) – The dimensions to be convolved. Default: (-2,-1).

  • out (str) – See conv().

  • padding (int, tuple[int, int] or str) – See conv().

  • simpson (bool) – Whether to apply Simpson’s rule. Default: True.

  • real (bool) – See conv().

Type:

float or Tensor

Type:

float or Tensor

Returns:

Convolution between f1 and f2. Complex if either f1 or f2 is complex or real=False, real-valued otherwise.

Return type:

Tensor