Why am I running the implementation McTruth script and not getting any numbers

Hi @jiuzhao , When a reconstructed particle has no match with the truth particle, then the TRUEP* information will give you nan. Are you always getting a nan? If not, then you will have to prune them with a cut e.g. mum_TRUEPZ!=TMath::QuiteNaN() or !TMath::IsNaN(mum_TRUEPZ).

Aside: Note that you can also specify an invalid value yourself using F.VALUE_OR functor. In-fact it is mandatory to specify this when using "interger" functors + F.MAP_INPUT e.g. TRUEID of the particle (FunTuple will complain if you don't for such a case). See example on how to do that.

Thank you very much for your advice. In fact, all truth outputs are like this I even got this result with those two lines of code in the example

Little surprising that there are no associated truth info being found at all. How many events do you run over? All the events in the DST file? And all of them have no truth info? I see you are running over sprucing output, is it in passthrough mode (i.e. Turbo) or do you have a dedicated Sprucing line with extra selection on top of Hlt2 line?

@amathad hi,I want to ask you how did you generate the.dst file, and I guess I made a mistake here

Isn't it as simple as that: https://lhcbdoc.web.cern.ch/lhcbdoc/moore/master/tutorials/hlt2_analysis.html#enabling-hlt2-output?

Note my question above. The relation table location depends on the type (options.process) of file you are running.

From the.xdigi file to the.dst file, there is the following code And there doesn't seem to be this line of code or other process process: 'Turbo'

Could you share your json file that you use for DV? Also can you set process:Hlt2 in the yaml file of DV and see what you get for truth info?

qee_zmumu_spruce.tck.json

Thanks. Also can you set process:Hlt2 in the yaml file of DV and see what you get for truth info?

okay. Can you remove process:Hlt2 from yaml and in the options file do configured_MCTruthAndBkgCatAlg(inputs=data, process='Hlt2')?

I using process:"Hlt2" in the yaml file but there are problems with that

ok ，I'll try

Very strange...Could you please put a reproducer somewhere public like lxplus with: DV options file, dst input and json file? Will try to have a look.

/afs/cern.ch/work/j/jiuzhao/public

thanks!

Could please tell me which version of Moore and DaVinci you have been using? master?

Moore version 53.6, DaVinciDev_v61r0

yep,I've only been on the job for two weeks

I can reproduce your error. I think there is problem with how the dst or json file is being produced. I have asked @nskidmor.

ok,thanks

mentioned in merge request !716 (merged)

mentioned in merge request !731 (merged)

The options at

https://gitlab.cern.ch/lhcb/DaVinci/-/blob/master/DaVinciExamples/python/DaVinciExamples/tupling/option_davinci_tupling_from_hlt2.py

worked for me as a minimum working example.

Also it looks like you are missing process=options.process in your configured_MCTruthAndBkgCatAlg which has led to empty / wrong output for me (related: #59 (closed))

Edit: oops, sorry, should have read above discussion more closely. Anyway hope the above link helps.

ok,thanks

@jiuzhao is the output of your HLT2 job an .mdf file?

The steps I took to generate a .dst file were from a .xidig file to a .mdf file and then from a .mdf file to a .dst file

MDF files cannot propagate MC info. Can you try the chain using only dst output?

ok,thanks!

Hi @nskidmor， After I changed the output format, I still did not save the desired information. The output information is in the log fileloghlt2_qee_example_persistreco.py

Hi @jiuzhao can you tell me exactly what you want to run and I will try from my end

@nskidmor I want to select events from the HLT2 high PT dimuon line (persist reco enabled) and then sprucing selection for Z->mu mu events. The MCtruth info will be saved.

OK, I see the hlt2 script above. Can you point me to the sprucing script

spruce_qee_example.py

First of all a lot of your lines are not running

# WARNING: Following 4 lines were automatically excluded:
# WARNING: Name of Line ---- list of found algos that are known to be not thread safe
# WARNING: Hlt2SingleHighPtMuonLine ---- [FilterDesktop('FilterDesktop')]
# WARNING: Hlt2SingleHighPtMuonPrescaleLine ---- [FilterDesktop('FilterDesktop')]
# WARNING: Hlt2SingleHighPtMuonIsoLine ---- [FilterDesktop('FilterDesktop')]
# WARNING: Hlt2SingleHighPtMuonNoMuIDLine ---- [FilterDesktop('FilterDesktop')]

we can fix that with exclude_incompatible=False

Im working through the rest of the chain now

I was able to get this working

config = run_moore(

TypeError: run_moore() got an unexpected keyword argument 'exclude_incompatible'

Here I assuming you have an up to date stack including DaVinci

Li_DV_options.yaml

Li_DV.py

Li_HLT2.py

Li_sprucing.py

In your Moore/ directory run

./run gaudirun.py Li_HLT2.py

then

./run gaudirun.py Li_sprucing.py

In your DaVinci/ directory run

./run lbexec Li_DV:main Li_DV_options.yaml

thanks!

Hi @jiuzhao , Saw your private message where you are encountering the following error:

HiveDataBrokerSvc                     FATAL in sysInitialize(): exception with tag=mapProducers is caught
HiveDataBrokerSvc                     ERROR mapProducers 	 
Unknown requested input by MCTruthAndBkgCatAlg/MCTruthAndBkgCatAlg#1 : '/Event/Spruce/SpruceQEE_ZToMuMuLine/Particles'.
You can set the OutputLevel of HiveDataBrokerSvc to DEBUG to get a list of inputs and outputs of every registered algorithm.
	 StatusCode=FAILURE
HiveDataBrokerSvc                     ERROR Exception stack trace

Which branch of DV are you using? Unfortunately I can only take a look at it later in the evening, however I see @nskidmor has already managed to make an nTuple, perhaps you can liaise with her.

Li_DV.pylog2Li_DV_options.yaml qee_zmumu_spruce.tck.json

I re-installed Stack and it's running successfully

Can this be closed at this point @jiuzhao? Else can you explain what's missing? Advance thanks.

ok

mentioned in issue Moore#458 (closed)

added Tuples label

closed