fit_2d_spline_trace

jwst.adaptive_trace_model.trace_model.fit_2d_spline_trace(flux, alpha, fit_scale=None, lrange=50, col_index=None, space_ratio=1.2, sigma_low=2.5, sigma_high=2.5, fit_iter=3, require_ngood=None, spline_bkpt=None, auto_ngood_factor=0.5, auto_bkpt_factor=2.0)

Create a trace model from spline fits to a single slit/slice image.

Image must be oriented so that wavelengths are along x-axis. Each column is fit separately, with a window to include nearby data.

Parameters:

fluxndarray

Input 2D flux image to fit.

alphandarray

Alpha coordinates for input flux.

fit_scalendarray, optional

Array of scale values to apply to the input flux before fitting.

lrangeint, optional

Local column range for data to include in the fit, to the left and right of each input column.

col_indexiterable or None, optional

Iterable or generator that produces column index values to fit. If provided, columns will be fit in the order specified. If not provided, columns will be fit left to right.

space_ratiofloat, optional

Maximum spacing ratio to allow fitting to continue. If the tenth-largest spacing in the input xvec is larger than the knot spacing by this ratio, then return None instead of attempting to fit the data.

sigma_lowfloat, optional

Low sigma threshold for iterative spline fit.

sigma_highfloat, optional

High sigma threshold for iterative spline fit.

fit_iterint, optional

Maximum number of iterations for spline fit.

require_ngoodint or None, optional

Minimum number of data points required to attempt a fit in a column.

spline_bkptint or None, optional

Number of spline breakpoints (knots).

auto_ngood_factorfloat, optional

If require_ngood is not provided, set it to this factor times the native spacing range in the input slit/slice.

auto_bkpt_factorfloat, optional

If spline_bkpt is not provided, set it to this factor times the native spacing range in the input slit/slice.

Returns:

splinesdict

Keys are column index numbers, values are dicts. Each dict contains a normalized model with BSpline value, a scale with a float value to scale to the data, and bounds with a 2-tuple of floating point values, corresponding to the lower and upper bounds of the alpha coordinates used in the fit. If a spline model could not be fit, the column index number is not present.