DiscreteFourierTransformBase¶

class odl.trafos.fourier.DiscreteFourierTransformBase(*args, **kwargs)[source]¶

Bases: Operator

Base class for discrete fourier transform classes.

Attributes:

adjoint: Adjoint transform, equal to the inverse.
axes: Axes along the FT is calculated by this operator.
domain: Set of objects on which this operator can be evaluated.
halfcomplex: Return True if the last transform axis is halved.
impl: Backend for the FFT implementation.
inverse: Inverse Fourier transform.
is_functional: True if this operator's range is a Field.
is_linear: True if this operator is linear.
range: Set in which the result of an evaluation of this operator lies.
sign: Sign of the complex exponent in the transform.

Methods

`__call__`(x[, out])	Return `self(x[, out, **kwargs])`.
`clear_fftw_plan`()	Delete the FFTW plan of this transform.
`derivative`(point)	Return the operator derivative at `point`.
`init_fftw_plan`([planning_effort])	Initialize the FFTW plan for this transform for later use.
`norm`([estimate])	Return the operator norm of this operator.

__init__(inverse, domain, range=None, axes=None, sign='-', halfcomplex=False, impl=None)[source]¶

Initialize a new instance.

All parameters are given according to the specifics of the forward transform. The inverse parameter is used to control conversions for the inverse transform.

Parameters:

inversebool: If True, the inverse transform is created, otherwise the forward transform.
domainDiscretizedSpace: Domain of the Fourier transform. If its DiscretizedSpace.exponent is equal to 2.0, this operator has an adjoint which is equal to the inverse.
rangeDiscretizedSpace, optional: Range of the Fourier transform. If not given, the range is determined from domain and the other parameters as a uniform_discr with exponent unit cell size and exponent p / (p - 1) (read as 'inf' for p=1 and 1 for p='inf').
axesint or sequence of ints, optional: Dimensions in which a transform is to be calculated. None means all axes.
sign{'-', '+'}, optional: Sign of the complex exponent.
halfcomplexbool, optional: If True, calculate only the negative frequency part along the last axis in axes for real input. This reduces the size of the range to floor(N[i]/2) + 1 in this axis i, where N is the shape of the input arrays. Otherwise, calculate the full complex FFT. If dom_dtype is a complex type, this option has no effect.
impl{'numpy', 'pyfftw', None}, optional: Backend for the FFT implementation. The 'pyfftw' backend is faster but requires the pyfftw package. None selects the fastest available backend.