replace_bad_pixels

jwst.refpix.irs2_subtract_reference.replace_bad_pixels(data0, ngroups, ny, row)

Replace bad pixels.

Use cosine weighted interpolation to replace 0.0 values and bad pixels and gaps:

s[1] = nx
s[2] = ny
s[3] = ngroups

Parameters:

data0ndarray

Input data array

ngroupsint

Number of groups in input data array

nyint

Number of rows in input data array

rowint

Row definition:

row = (scipix_n + refpix_r + 2) * 512 // scipix_n + pad
# This means that if
scipix_n = 16
refpix_r = 4
pad = 8
# then
row = 712