nabu.processing.fftshift module

class nabu.processing.fftshift.FFTshiftBase(shape, dtype, dst_dtype=None, axes=None, **backend_options)[source]

Bases: object

Parameters:

  • shape (tuple) – Array shape - can be 1D or 2D. 3D is not supported.

  • dtype (str or numpy.dtype) – Data type, eg. “f”, numpy.complex64, …

  • dst_dtype (str or numpy.dtype) – Output data type. If not provided (default), the shift is done in-place.

  • axes (tuple, optional) – Axes over which to shift. Default is None, which shifts all axes.

  • backend_options – named arguments to pass to CudaProcessing or OpenCLProcessing

KernelCls

alias of BaseClassError

ProcessingCls

alias of BaseClassError

dtype_to_ctype

alias of BaseClassError

backend = 'none'
fftshift(arr, dst=None)[source]
ifftshift(arr, dst=None)[source]
class nabu.processing.fftshift.OpenCLFFTshift(shape, dtype, dst_dtype=None, axes=None, **backend_options)[source]

Bases: FFTshiftBase

Parameters:

  • shape (tuple) – Array shape - can be 1D or 2D. 3D is not supported.

  • dtype (str or numpy.dtype) – Data type, eg. “f”, numpy.complex64, …

  • dst_dtype (str or numpy.dtype) – Output data type. If not provided (default), the shift is done in-place.

  • axes (tuple, optional) – Axes over which to shift. Default is None, which shifts all axes.

  • backend_options – named arguments to pass to CudaProcessing or OpenCLProcessing

KernelCls

alias of OpenCLKernel

ProcessingCls

alias of OpenCLProcessing

dtype_to_ctype(dtype)
backend = 'opencl'