feat: add inpaint-mode for lst binner #907
Conversation
hera_cal/lstbin_simple.py
Outdated
| mode. In this case, either choice is bad: to include them in the average is bad | ||
| because even though they may have been actually in-painted, whatever value they have | ||
| is clearly triggering the sigma-clipper and is therefore an outlier. On the other | ||
| hand, to ignore them is bad because the point of in-painting mode is to get | ||
| smoother spectra, and this negates that. So, it's best just to not do sigma-clipping | ||
| in inpainted mode. |
There was a problem hiding this comment.
Can there be a mode where inpainted data cannot get sigma-clipped but non-inpainted data can?
There was a problem hiding this comment.
Sure! I'll add that.
|
Also, what do you want to do about tests failing? |
hera_cal/lstbin_simple.py
Outdated
| return data, flags | ||
|
|
||
|
|
||
| def get_lst_mad( |
There was a problem hiding this comment.
This does both med and mad, so I feel like both should be in the name.
There was a problem hiding this comment.
Got it. Fixed that up.
There was a problem hiding this comment.
Just a few tiny changes, but otherwise good to go! Thanks for this @steven-murray!
Codecov ReportPatch coverage:
Additional details and impacted files@@ Coverage Diff @@
## main #907 +/- ##
==========================================
+ Coverage 97.17% 97.19% +0.01%
==========================================
Files 23 23
Lines 10282 10429 +147
==========================================
+ Hits 9992 10136 +144
- Misses 290 293 +3
Flags with carried forward coverage won't be shown. Click here to find out more.
☔ View full report in Codecov by Sentry. |
and some pep8
also pep8
This adds an "inpaint" mode for the LST-binner. In this mode, flagged data is assumed to have been in-painted to preserve spectral smoothness.