Conversation
ArmelleJB
commented
Dec 31, 2025
ArmelleJB
commented
- save the ID of the identified bad pixels (instead of their position in the list),
- fix format issues when using LocalPeakWindowSum or GlobalPeakWindowSum methods
- add a parameter for the number of samples used for the pedestal estimation
- add simple check on the location of the peak and discard events that are too close from the beginning of the trace, to safely estimate the pedestal
- happy new year !
Codecov Report❌ Patch coverage is
Additional details and impacted files@@ Coverage Diff @@
## main #245 +/- ##
==========================================
+ Coverage 51.98% 52.14% +0.15%
==========================================
Files 80 80
Lines 6713 6680 -33
==========================================
- Hits 3490 3483 -7
+ Misses 3223 3197 -26 ☔ View full report in Codecov by Sentry. 🚀 New features to boost your workflow:
|
| apply_integration_correction=self.charge_integration_correction, | ||
| ) | ||
| amp_int_per_pix_per_event = integrator( | ||
| wfs_pedsub, 0, None, np.invert(self.__bad_pixels_mask) |
There was a problem hiding this comment.
The 0 in second argument hardcodes the fact that only data from the NectarCAM QM can be analysed. Instead, a tel_id should be provided, to allow the analysis of any NectarCAM camera.
|
This PR currently conflicts with #242 |
…ber of samples used for the pedestal estimation, fix bugs when using LocalPeakWindowSum or GlobalPeakWindowSum methods
…ose from the beginning of the trace, to safely estimate the pedestal
…ose from the beginning of the trace, to safely estimate the pedestal
…ose from the beginning of the trace, to safely estimate the pedestal
…caused the flatfield script to fail during the second pass in Armelle's script. Fixed a few typos as well.
Just rebased this PR from |
|
This PR still undoes the possibility to use in input the pedestal (resp. gain) values from the output files of pedestal (resp. gain) calibration tools, introduced in #226. These improvements should not disappear. |
|
Could we please close this, and work from #250 instead ? |
2 participants
