Closed cylammarco closed 3 years ago
Tracing is now working it seems, however, extraction is still a problem. Should employ a robust polynomial fit instead of using gaussian for optimal extraction in the case of Horne86 (and in general). Should consider using: https://www.statsmodels.org/dev/generated/statsmodels.nonparametric.smoothers_lowess.lowess.html
Saturated sources can be extracted to give finite values, we are masking the saturated pixels and weighting the extraction by the tails of the LSF only. This is different from attempting to reconstructing the LSF profile which may produce a more usable spectrum, that would be out of the scoop of this software at the moment.
ap_trace()
can trace a saturated spectrum, however, the LSF is not usable.