WidthEstimate2D¶
-
turbustat.statistics.
WidthEstimate2D
(inList, method='contour', noise_ACF=0, diagnosticplots=False, brunt_beamcorrect=True, beam_fwhm=None, spatial_cdelt=None, **fit_kwargs)[source] [edit on github]¶ Estimate spatial widths from a set of autocorrelation images.
Warning
Error estimation is not implemented for
interpolate
orxinterpolate
.Parameters: - inList: {list of 2D `~numpy.ndarray`s, 3D `~numpy.ndarray}
The list of autocorrelation images.
- method: {‘contour’, ‘fit’, ‘interpolate’, ‘xinterpolate’}, optional
The width estimation method to use.
contour
fits an ellipse to the 1/e contour about the peak.fit
fits a 2D Gaussian to the peak.interpolate
andxinterpolate
both estimate the 1/e level from interpolating the data onto a finer grid near the center.xinterpolate
first fits a 2D Gaussian to estimate the radial distances about the peak.- noise_ACF: {float, 2D `~numpy.ndarray`}, optional
The noise autocorrelation function to subtract from the autocorrelation images. This is typically produced by the last few eigenimages, whose structure should consistent of irreducible noise.
- diagnosticplots: bool, optional
Show diagnostic plots for the first 9 autocorrelation images showing the goodness of fit (for the gaussian estimator) or ??? (presently nothing) for the others.
- brunt_beamcorrect : bool, optional
Apply the beam correction. When enabled, the beam size must be given.
- beam_fwhm : None or astropy.units.Quantity
The FWHM beam width in angular units. Must be given when using
brunt_beamcorrect
.- spatial_cdelt : {None, astropy.units.Quantity}, optional
The angular scale of a pixel in the given data. Must be given when using brunt_beamcorrect.
- fit_kwargs : dict, optional
Used when method is ‘contour’. Passed to
turbustat.statistics.stats_utils.EllipseModel.estimate_stderrs
.
Returns: - scales : array
The array of estimated scales with length len(inList) or the 0th dimension size if
inList
is a 3D array.- scale_errors : array
Uncertainty estimations on the scales.