[clas12_rgk] [Hybrid baryons] [EXTERNAL] new rgk trains (clara 5.0.2 coatjava 10.0.2)

Izzy Illari izzy at jlab.org
Tue Aug 1 16:28:34 EDT 2023


Dear all,

In head of tomorrow's meeting I looked at the ratios discussed by Daniel 
and Volker at last meeting and in emails... I have attached the slides 
here. I took ratios of missing mass squared distributions with focus on 
eta mass range to compare pass2 with denoise/pass 1, pass2 no 
denoise/pass 1, and then pass2 with denoise/pass 2 without denoise. I 
also plotted the momentum distributions of the electrons and protons in 
my final states and the ratios of those as well.

I see increased number of events from pass 1 to pass2, but not quite a 
huge improvement from pass 2 with no denoise to pass 2 with denoise.

My question is the same as FX's: should there be anything else 
investigated in order to wrap this up?

All the best,

Izzy

On 8/1/23 10:30 AM, Francois-Xavier Girod via clas12_rgk wrote:
> Dear all
>
> I compared train output under
> /volatile/clas12/rg-k/production/recon/fall2018/torus+1/6535MeV/pass1/v7/
> which I believe is the "old" pass 1 with
> /volatile/clas12//rg-k/production/recon/fall2018/torus+1/6535MeV/pass1/v19/denoise_1002/train/
> which should be the "denoise pass 2" (?)
>
> Here are the numbers I obtained
>
> DVCS pass 1 FD = 6564 , CD = 18318
> DVCS denoise FD = 8135 , CD = 31159
> DVCS gains FD = 23.9% , CD = 70.1%
>
> pi0 pass 1 FD = 4952 , CD = 12395
> pi0 denoise FD = 6606 , CD = 21126
> pi0 gain FD = 33.4% , CD = 70.4%
>
> phi pass 1 FD = 51.5
> phi denoise FD = 93.0
> phi gain FD = 80.6%
>
> Please let me know if there are additional comparisons or 
> investigation I can perform to wrap this up
>
> Best regards
> FX
>
>
> On Tue, Aug 1, 2023 at 10:10 AM Annalisa D'Angelo 
> <annalisa.dangelo at roma2.infn.it> wrote:
>
>     Dear Lucilla,
>
>     can you please point at the pass1 trains, please.
>
>
>     Thank you in advance
>     Annalisa
>
>     -------- Messaggio Inoltrato --------
>     Oggetto: 	Re: [Hybrid baryons] [EXTERNAL] new rgk trains (clara
>     5.0.2 coatjava 10.0.2)
>     Data: 	Tue, 1 Aug 2023 09:42:08 -0400
>     Mittente: 	Francois-Xavier Girod via Hybrid_baryons
>     <hybrid_baryons at jlab.org> <mailto:hybrid_baryons at jlab.org>
>     Rispondi-a: 	Francois-Xavier Girod <fxgirod at jlab.org>
>     <mailto:fxgirod at jlab.org>, hybrid_baryons at jlab.org
>     A: 	Annalisa D'Angelo <annalisa at jlab.org> <mailto:annalisa at jlab.org>
>     CC: 	hybrid_baryons at jlab.org
>
>
>
>     Dear Annalisa, all
>
>     Could someone point me out to the directories I should compare then?
>     It will not take long to run my code again but I would rather not
>     guess the files locations
>
>     Thank you
>
>     Best regards
>     FX
>
>     On Tue, Aug 1, 2023 at 5:39 AM Annalisa D'Angelo
>     <annalisa at jlab.org> <mailto:annalisa at jlab.org> wrote:
>>     Dear FX,
>>
>>     thank you very much!
>>
>>     It would be great if you could could compare results among pass1 and
>>     pass2 with denoising. Please tell us if you need and any additional
>>     input from us.
>>
>>     Annalisa
>>
>>     Il 31/07/23 17:22, Francois-Xavier Girod via Hybrid_baryons ha scritto:
>>>     Dear all
>>>
>>>     I also analyzed the exclusive phi events since they have 4 charged
>>>     particles in the forward detectors
>>>     exclu phi integrated peak estimated events:
>>>     no denoise : 75.7
>>>     denoise : 93.0
>>>     gain : 22.8%
>>>
>>>     (raw spectra without fitting the peak gives: no denoise 1058, denoise
>>>     1181, gain 11.6%)
>>>
>>>     Best regards
>>>     FX
>>>
>>>     On Mon, Jul 31, 2023 at 9:55 AM Francois-Xavier Girod<fxgirod at jlab.org>  <mailto:fxgirod at jlab.org>  wrote:
>>>>     Dear all
>>>>
>>>>     Here are some deep exclusive reactions comparison results between "6.5
>>>>     GeV no denoise" and "6.5 GeV denoise"
>>>>
>>>>     DVCS no denoise FD = 7845 , CD = 30260
>>>>     DVCS denoise FD = 8135 , CD = 31159
>>>>     DVCS gains FD = 3.7% , CD = 3.0%
>>>>
>>>>     pi0 no denoise FD = 6396 , CD = 20721
>>>>     pi0 denoise FD = 6606 , CD = 21126
>>>>     pi0 gain FD = 3.3% , CD = 1.9%
>>>>
>>>>     skim20 "pip pim pi0" which should contain exclusive omega events is
>>>>     not working for some reason I think
>>>>
>>>>     Best regards
>>>>     FX
>>>>
>>>>
>>>>     On Thu, Jul 27, 2023 at 5:12 PM Lucilla Lanza via Hybrid_baryons
>>>>     <hybrid_baryons at jlab.org>  <mailto:hybrid_baryons at jlab.org>  wrote:
>>>>>     Dear all,
>>>>>
>>>>>     You can find the new rgk trains (clara 5.0.2 coatjava 10.0.2) at:
>>>>>
>>>>>     6.5 GeV denoise (5901,5906,5913,5920,5986,5990):
>>>>>
>>>>>     /volatile/clas12//rg-k/production/recon/fall2018/torus+1/6535MeV/pass1/v19/denoise_1002/train/
>>>>>
>>>>>     6.5 GeV no denoise (5901,5906,5913,5920,5986,5990):
>>>>>
>>>>>     /volatile/clas12//rg-k/production/recon/fall2018/torus+1/6535MeV/pass1/v19/no_denoise_1002///train/
>>>>>
>>>>>     7.5 GeV no denoise (5682,5696,5700,5708,5734,5791):
>>>>>
>>>>>     /volatile/clas12//rg-k/production/recon/fall2018/torus+1/7546MeV/pass1/v19/no_denoise_1002///train/
>>>>>
>>>>>     Please let me know if you have any comment or question.
>>>>>     Best regards,
>>>>>
>>>>>         Lucilla
>>>>>
>>>>>
>>>>>
>>>>>>>     Il 2023-07-25 22:06 Raffaella De Vita ha scritto:
>>>>>>>>     Dear Lucilla,
>>>>>>>>
>>>>>>>>     Daniel noticed the REC::Scintillator banks in the skims seem to be
>>>>>>>>     corrupted. I did some investigation and found that the version of the
>>>>>>>>     train that was run is based on an old coatjava version where the
>>>>>>>>     REC::Scintillator bank was different than what it is now. This is
>>>>>>>>     likely the cause of the problem.
>>>>>>>>
>>>>>>>>     We verified the skims can be recovered (with no need to rerun the
>>>>>>>>     trains), replacing the bank schema in the files. This can be done
>>>>>>>>     running the following command for each of the skim files;
>>>>>>>>
>>>>>>>>     hipo-utils -update -d
>>>>>>>>     /group/clas12/packages/clara/5.0.2_10.0.2/plugins/clas12/etc/bankdefs/hipo4
>>>>>>>>     -o recovered-skim.hipo original-skim.hipo
>>>>>>>>
>>>>>>>>     which basically restores the correct schema.
>>>>>>>>
>>>>>>>>     My suggestion is to create a new directory where to write the
>>>>>>>>     recovered files with the same directory structure as the train
>>>>>>>>     output.
>>>>>>>>     Let me know if you have any questions.
>>>>>>>>
>>>>>>>>     Best regards,
>>>>>>>>           Raffaella
>>>>>>>>
>>>>>>>>
>>>>>>>>>     On 24 Jul 2023, at 09:28, Lucilla Lanza
>>>>>>>>>     <lucilla.lanza at roma2.infn.it>  <mailto:lucilla.lanza at roma2.infn.it>  wrote:
>>>>>>>>>
>>>>>>>>>     Dear all,
>>>>>>>>>
>>>>>>>>>     the rgk dst files with (5901,5906,5913,5920,5986,5990) and without
>>>>>>>>>     (5682,5696,5700,5708,5734,5791,5901,5906,5913,5920,5986,5990)
>>>>>>>>>     denoising applied are available for comparison (coatjava 10.0.2,
>>>>>>>>>     timestamp 07/15/2023-03:00:00).
>>>>>>>>>
>>>>>>>>>     No denoising:
>>>>>>>>>
>>>>>>>>>     /volatile/clas12/rg-k/production/recon/fall2018/torus+1/6535MeV/pass0/v0_cook_no_denoise_new/dst/recon/
>>>>>>>>>
>>>>>>>>>     Denoising:
>>>>>>>>>
>>>>>>>>>     /volatile/clas12/rg-k/production/recon/fall2018/torus+1/6535MeV/pass0/v0_cook_denoise_new/dst/recon/
>>>>>>>>>
>>>>>>>>>     You can find the corresponding skims with and without denoising
>>>>>>>>>     applied at:
>>>>>>>>>
>>>>>>>>>     6.5 GeV denoise (5901,5906,5913,5920,5986,5990):
>>>>>>>>>
>>>>>>>>>     /volatile/clas12//rg-k/production/recon/fall2018/torus+1/6535MeV/pass1/v19/denoise/train/
>>>>>>>>>
>>>>>>>>>     7.5 GeV no denoise (5682,5696,5700,5708,5734,5791):
>>>>>>>>>
>>>>>>>>>     /volatile/clas12//rg-k/production/recon/fall2018/torus+1/7546MeV/pass1/v19/no_denoise/train/
>>>>>>>>>
>>>>>>>>>     6.5 GeV no denoise (5901,5906,5913,5920,5986,5990):
>>>>>>>>>
>>>>>>>>>     /volatile/clas12//rg-k/production/recon/fall2018/torus+1/6535MeV/pass1/v19/no_denoise/train/
>>>>>>>>>
>>>>>>>>>     Please let me know if you have any comment or question.
>>>>>>>>>     Best regards,
>>>>>>>>>
>>>>>>>>>     Lucilla
>>>>>>>>>
>>>>>>>>>
>>>>>     _______________________________________________
>>>>>     Hybrid_baryons mailing list
>>>>>     Hybrid_baryons at jlab.org
>>>>>     https://mailman.jlab.org/mailman/listinfo/hybrid_baryons
>>>     _______________________________________________
>>>     Hybrid_baryons mailing list
>>>     Hybrid_baryons at jlab.org
>>>     https://mailman.jlab.org/mailman/listinfo/hybrid_baryons
>>
>>     --
>>     ================================================
>>     Prof. Annalisa D'Angelo
>>     Dip. Fisica, Universita' di Roma "Tor Vergata"
>>     Istituto Nazionale di Fisica Nucleare
>>     Sezione di Roma Tor Vergata, Rome Italy
>>     email:annalisa.dangelo at roma2.infn.it
>>     Jefferson Laboratory, Newport News, VA USA
>>     Email:annalisa at jlab.org
>>     Tel: + 39 06 72594562
>>
>
>     _______________________________________________
>     Hybrid_baryons mailing list
>     Hybrid_baryons at jlab.org
>     https://mailman.jlab.org/mailman/listinfo/hybrid_baryons
>
>
> _______________________________________________
> clas12_rgk mailing list
> clas12_rgk at jlab.org
> https://mailman.jlab.org/mailman/listinfo/clas12_rgk
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://mailman.jlab.org/pipermail/clas12_rgk/attachments/20230801/595ad919/attachment-0001.html>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: CompareDenoise_1Aug2023.pptx
Type: application/vnd.openxmlformats-officedocument.presentationml.presentation
Size: 692889 bytes
Desc: not available
URL: <https://mailman.jlab.org/pipermail/clas12_rgk/attachments/20230801/595ad919/attachment-0001.pptx>


More information about the clas12_rgk mailing list