Dqm update - using standard ctapipe tools

[hashkar]

When using --method GlobalPeakWindowSum --extractor_kwargs '{"window_width":16,"window_shift":4}'
In chargeIntegration:
Line 108 - 117
The way Broken Pixles is computed does not hide the 21 pixels hidden by the camera shutter. In the display the values are still there.
Alternatively using self.pixelBAD = evt.mon.tel[0].pixel_status.hardware_failing_pixels breaks the code at line 129 or 139

[codecov]

Codecov Report

Attention: Patch coverage is 24.00000% with 57 lines in your changes missing coverage. Please review.

Project coverage is 47.51%. Comparing base (b306f30) to head (649bf8d).
Report is 31 commits behind head on main.

Files with missing lines	Patch %	Lines
src/nectarchain/dqm/charge_integration.py	22.64%	41 Missing ⚠️
src/nectarchain/dqm/start_dqm.py	0.00%	16 Missing ⚠️

Additional details and impacted files

@@            Coverage Diff             @@
##             main     #154      +/-   ##
==========================================
+ Coverage   40.76%   47.51%   +6.74%     
==========================================
  Files          65       64       -1     
  Lines        4452     4590     +138     
==========================================
+ Hits         1815     2181     +366     
+ Misses       2637     2409     -228     

☔ View full report in Codecov by Sentry.
📢 Have feedback on the report? Share it here.

[jlenain]

Hi @hashkar !

Thanks for this PR. The fact that codecov fails is not so worrisome. The other tests pass now.

On the issue you raised in your first comment concerning runs with missing pixels, could you elaborate a bit what is failing where?

When @guillaumegrolleron implemented the charge extractors layer in nectarchain, this was handled gracefully and transparently without having to hardcode any masked pixel pattern.

[hashkar]

@jlenain issues of the first comment are fixed now

[jlenain]

Thanks @hashkar !
It all seems to work fine! Maybe a point I would like to discuss with you, @frnbrun and @vmarandon : I understand the default behavior is using a LocalPeakWindowSum which is 16 ns wide and with 4 ns leading offset. Do we want to use another default setting?

[hashkar]

Thanks @hashkar ! It all seems to work fine! Maybe a point I would like to discuss with you, @frnbrun and @vmarandon : I understand the default behavior is using a LocalPeakWindowSum which is 16 ns wide and with 4 ns leading offset. Do we want to use another default setting?

@jlenain I switched the default from LocalPeakWindowSum to GlobalPeakWinfowSum in the start_dqm.py script

[jlenain]

Thank you, @hashkar. I wasn't calling for a change, but rather a discussion with @vmarandon , @frnbrun and you on what would be the best strategy for regular DQM runs.