Skip to content

Try 2024 - #25

Merged
ludovicomori merged 7 commits into
LPC-HH:mainfrom
jinwang137:try_2024
Jun 16, 2026
Merged

ludovicomori merged 7 commits into
LPC-HH:mainfrom
jinwang137:try_2024

Conversation

@jinwang137

Copy link
Copy Markdown
Contributor

Hello, I have submitted the updates for the 2024 samples.

Please note:

In Some code:
(only for 2024) Not all samples have been fully processed yet, so this is temporarily commented out. If running 2022 + 2023, please uncomment.

The reason for this is that we will include these for 2024, rather than being unable to include them, so I did not make it a selectable option.

@ludovicomori ludovicomori left a comment

Copy link
Copy Markdown
Collaborator

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Great work. Just a couple comments, let me know your thoughts!

Comment thread src/boostedhh/processors/corrections.py Outdated
# correction: Non-zero value for (eta, phi) indicates that the region is vetoed
cset = correctionlib.CorrectionSet.from_file(get_pog_json("jetveto", year))
#jin20260406
print(list(cset.keys()))

Copy link
Copy Markdown
Collaborator

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Maybe we can remove this, probably was used in debugging.

Copy link
Copy Markdown
Contributor Author

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

sure we can

Comment thread src/boostedhh/hh_vars.py Outdated
Comment thread src/boostedhh/hh_vars.py Outdated
"ggfbbtt-kl0p00": "GluGlutoHHto2B2Tau_kl-0p00_kt-1p00_c2-0p00",
"ggfbbtt-kl2p45": "GluGlutoHHto2B2Tau_kl-2p45_kt-1p00_c2-0p00",
"ggfbbtt-kl5p00": "GluGlutoHHto2B2Tau_kl-5p00_kt-1p00_c2-0p00",

Copy link
Copy Markdown
Collaborator

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

I think it's ok to leave the BSM signals uncommented, since it just means that whoever will run the skimmer next will be able to produce also the BSM samples. Or are there other reasons why we cannot run the skimmer over these samples right now?

Copy link
Copy Markdown
Contributor Author

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Sorry, it’s just that I haven’t added the BSM signals for 2024 yet, so running the postprocessing will currently result in errors. I will add those signals later today.

@ludovicomori
ludovicomori merged commit 868d7e5 into LPC-HH:main Jun 16, 2026
3 checks passed
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

2 participants