Fix muon scripts

[rlopezcoto]

No description provided.

[morcuended]

I had a quick look and seems fine to me. Just x-check how you are using the image array in the lstchain_data_muon_analysis.py

[morcuended]

Why did you change this back to image[0] while in the rest of the script you corrected to just image?. In this script, does the image array still contain the two gains for each pixel?

[rlopezcoto]

you are right @morcuended. I should apply here gain selection...
by the way, for the data that you are analyzing in LP, are you applying the proper threshold for the gain selection? According to #200, we do not have a proper threshold due to the ctapipe bug

[rlopezcoto]

If you have the threshold set to 4094, then it means that the threshold is not properly applied.

For the muon script, I do not think it is a good idea to have a duplicated calibration for the regular data analysis and muon data analysis, therefore I expect the script lstchain_data_muon_analysis.py to be deprecated and run the analysis only on calibrated files. I'll mark this in a comment on the script as no-gain selection, but I'm not sure if it is worth it to solve it at the moment.

[morcuended]

These are the warning I'm getting:

WARNING: AstropyDeprecationWarning: channel selection is now performed at <= R1 calibration level [lstchain.reco.dl0_to_dl1]
WARNING: AstropyDeprecationWarning: gain selection is now performed at <= R1 calibration level [lstchain.calib.camera.calib]

[morcuended]

If you have the threshold set to 4094, then it means that the threshold is not properly applied.

For the muon script, I do not think it is a good idea to have a duplicated calibration for the regular data analysis and muon data analysis, therefore I expect the script lstchain_data_muon_analysis.py to be deprecated and run the analysis only on calibrated files. I'll mark this in a comment on the script as no-gain selection, but I'm not sure if it is worth it to solve it at the moment.

The config files I am using indicate:

"gain_selector": "ThresholdGainSelector",
 "gain_selector_config": {
   "threshold":  4094
 }
}

[morcuended]

So what threshold value should we use then?

[rlopezcoto]

then I think it will be affected... According to #200, the threshold should be 20.9
@vuillaut can you confirm?

[vuillaut]

For MC yes.

[moralejo]

The exact switching value is not so critical, around 20 should be ok, enough to have already a clear LG signal, but could also be taken up to 50 or so safely, I think. More important: the DL1 you were producing then had effectively only high gain, correct?

…

On Tue, 26 Nov 2019 at 13:11, Ruben Lopez-Coto ***@***.***> wrote: ***@***.**** commented on this pull request. ------------------------------ In lstchain/scripts/lstchain_data_muon_analysis.py <#232 (comment)> : > @@ -153,17 +153,17 @@ def main(): print("Event {}. Number of pixels above 10 phe: {}".format(event_id, np.size(image[0][image[0] > 10.]))) - if not tag_pix_thr(image): #default skipps pedestal and calibration events + if not tag_pix_thr(image[0]): #default skipps pedestal and calibration events then I think it will be affected... According to #200 <#200>, the threshold should be 20.9 @vuillaut <https://github.com/vuillaut> can you confirm? — You are receiving this because you are subscribed to this thread. Reply to this email directly, view it on GitHub <#232?email_source=notifications&email_token=ADMCCOXPE6I6SSWOWPYYFQDQVUG67A5CNFSM4JRQDCLKYY3PNVWWK3TUL52HS4DFWFIHK3DMKJSXC5LFON2FEZLWNFSXPKTDN5WW2ZLOORPWSZGOCM7Y5XQ#discussion_r350703786>, or unsubscribe <https://github.com/notifications/unsubscribe-auth/ADMCCOW5G52EOKL42AVZ5WTQVUG67ANCNFSM4JRQDCLA> .

-- Abelardo Moralejo Olaizola Institut de Física d'Altes Energies Tel : +34 931641662 Fax: +34 935811938 Avís - Aviso - Legal Notice - (LOPD) - http://legal.ifae.es

-- Avís - Aviso - Legal Notice - (LOPD) - http://legal.ifae.es <http://legal.ifae.es/>

[morcuended]

The exact switching value is not so critical, around 20 should be ok, enough to have already a clear LG signal, but could also be taken up to 50 or so safely, I think. More important: the DL1 you were producing then had effectively only high gain, correct?

Yes, I think so. I did not realize this before. Thus, I should set the threshold to 20.9 and reprocess everything again, right?

[morcuended]

Let's go ahead with these scripts

[rlopezcoto]

Thanks!