ScalingFactorComputer

class jwst.background.background_sub_wfss.ScalingFactorComputer(p=1.0, maxiter=5, delta_rms_thresh=0, dispersion_axis=None)[source]

Handle computation of scaling factor.

Parameters:

pfloat, optional: Percentile for sigma clipping on both low and high ends per iteration, default 1.0. For example, with p=2.0, the middle 96% of the data is kept.
maxiterint, optional: Maximum number of iterations for outlier rejection. Default 5.
delta_rms_threshfloat, optional: Stopping criterion for outlier rejection; stops when the rms residuals change by less than this fractional threshold in a single iteration. For example, assuming delta_rms_thresh=0.1 and a residual RMS of 100 in iteration 1, the iteration will stop if the RMS residual in iteration 2 is greater than 90. Default 0.0, i.e., ignore this and only stop at maxiter.
dispersion_axisint, optional: The index to select the along-dispersion axis. Used to compute the RMS residual, so must be set if rms_thresh > 0. Default None.

Methods Summary

`__call__`(sci, bkg, var[, mask])	Call function for the class.
`err_weighted_mean`(sci, bkg, var)	Remove any var=0 values, which can happen for real data.

Methods Documentation

__call__(sci, bkg, var, mask=None)[source]

Call function for the class.

Parameters:

scindarray: The science data.
bkgndarray: The reference background model.
varndarray: Total variance (error squared) of the science data.
maskndarray[bool], optional: Initial mask to be applied to the data, True where bad. Typically, this would mask out the real sources in the data.

Returns:

float: Scaling factor that minimizes sci - factor*bkg, taking into account residuals and outliers.
ndarray[bool]: Outlier mask generated by the iterative clipping procedure.

err_weighted_mean(sci, bkg, var)[source]

Remove any var=0 values, which can happen for real data.

Parameters:

Returns: