Waveform noise adder

[moralejo]

First version of the component to add noise in MC waveforms, for adapting it to different observation conditions.
It adds at random N instances of waveforms from an NSB-only file to the waveforms of a showers file. It seems to reproduce well the noise distributions produced directly by sim_telarray:

Update: there was a mistake in the plots below (only one telescope was used, multiple times). After correcting this, it turns out that the noise padding produces slightly larger noise than what is produced by sim_telarray itself:

Test function still not ready, I open the draft PR to start discussion.

[moralejo]

The failures in the tests seems to be unrelated to this. Any idea of what is going on?

[maxnoe]

The connection timeout is unrelated and should go away be re-running, but this error is in code you added:

src/ctapipe/tests/test_traitlets_configurable.py::test_all_traitlets_configurable - AssertionError: Class ctapipe.image.modifications.WaveformModifier is missing .tag(config=True) for traitlets: {'total_noise'}

[moralejo]

The connection timeout is unrelated and should go away be re-running, but this error is in code you added:

src/ctapipe/tests/test_traitlets_configurable.py::test_all_traitlets_configurable - AssertionError: Class ctapipe.image.modifications.WaveformModifier is missing .tag(config=True) for traitlets: {'total_noise'}

Thanks, I had not noticed, because test ran well locally. Anyway it is not something that should be configured, it is something internally, so I just made a normal dict()

[maxnoe]

This looks good in general. However, at the moment, it's not usable via ctapipe-process.

I think there are two options:

• Adding it in ProcessorTool directly as its own step
• Adding it to it to the CameraCalibrator class in the r1 to dl0 step.

I think this does not really belong into the CameraCalibrator, so I would add it directly to the process tool.

Similar to how the ImageProcessor is using the ImageModifier here:

ctapipe/src/ctapipe/image/image_processor.py

Lines 217 to 219 in 04a1ec7

	for tel_id, dl1_camera in event.dl1.tel.items():
	if self.apply_image_modifier.tel[tel_id]:
	dl1_camera.image = self.modify(tel_id=tel_id, image=dl1_camera.image)

[maxnoe]

This leaves the question when and how to initialize it. Simplest would be adding a new boolean option at the processor level to enable it --add-waveform-noise for example.

The other option would be to always enable it, but make the default behavior a no-op.

I think I prefer the first option as it is more explicit.

[ctao-sonarqube]

Quality Gate passed

Issues
0 New issues
0 Fixed issues
0 Accepted issues

Measures
0 Security Hotspots
82.5% Coverage on New Code
0.0% Duplication on New Code

See analysis details on SonarQube