Lepton corrections for Full2024v15 campaign #70

BlancoFS · 2025-06-17T06:35:43Z

Hi all,

The PR introduced all the needed changes to process a Full2024v15 campaign with the basic leptonic corrections applied. It basically introduces the muon, electron, and trigger scale factors. Additionally, it adds some steps to run and test the production.

More importantly, I noticed a bug in the JetSelMask module. The runModule function was not correctly declared due to a typo, which made the jet selection not to be applied in the postprocessing steps. This issue can potentially affect previous campaigns (@NTrevisani @mlizzo @squinto5, please check and confirm if this is the case). Additionally, since the jetId is not available in nanoAODv15, the module is adapted to evaluate the selector using correctionlib.

Most probably, there are conflicts in the PR due to the Steps_cfg.py and JetMaker_cfg.py files and they need to be revisited. They are present due to some work in parallel in the master branch. To be fixed.

Bests,
Sergio

NTrevisani · 2025-10-15T11:11:31Z

Hi @BlancoFS, thank you again for preparing this PR. Could you have a look at the conflict in Steps_cfg.py, before we start reviewing the rest of the changes?

BlancoFS · 2025-10-16T07:46:49Z

Hi @NTrevisani, the conflict has been solved

NTrevisani · 2025-10-20T12:14:19Z

Hi @NTrevisani, the conflict has been solved

Thank you @BlancoFS. This includes also the conflicts to Steps_cfg.py and JetMaker_cfg.py that you were mentioning in the PR description, or were they fixed in previous commits?

BlancoFS · 2025-10-20T13:12:11Z

Hi @NTrevisani, I believe those files don't matter that much, since from the time I wrote this PR the corrections have evolved and a second PR will be needed to put everything in place

acalandr · 2025-10-20T13:30:29Z

Hi @BlancoFS, sorry if I missed the detail in the discussion - are you referring to a second PR which is going to be the one relevant for the 2024 production that needs to be merged to start the processing on 2024? Would it be possible to either use this PR which has already been opened or could we open another PR with the full set of material to be merged rather than having two PR's? What's the timescale for it so that we can start organising for the work sharing on the 2024 production? Thanks a lot!

BlancoFS · 2025-10-20T14:37:16Z

Hi @acalandr, yes, you are more or less right.

The point is the following. On June, I opened this PR with the material that we had at that time. However, some developments, MC samples, etc. started to appear, so we opened another branch in which we could continue working on. This is the Sergio-ifca-2024 branch. We did it in that way to avoid pushing again and again to an open PR since it would have lead to an infitine review. Therefore, now we have two branches that include 2024, but one (this one) is clearly out-of-date and it needs to be updated with the developments from Sergio-ifca-2024 branch.

If I recall correctly, the Sergio-ifca-2024 branch should contain all the developments to be integrated in this PR plus additional material. We can decide whether to close this PR and open a new one to the latest version of 2024 production, or to merge this one and open another one later.

I was planing to update the Sergio-ifca-2024 branch this week with some missing lepton corrections (2024 muon scale and smearing, and electron Reco SF), so maybe we can finish the review of the new modules introduced here and next week we can push for the next PR to be done.

squinto5 · 2025-10-22T10:06:25Z

mkShapesRDF/processor/data/JetMaker_cfg.py

+    'Full2024v15': {
+        "vetomap": frameworkPath + "/processor/data/jsonpog-integration/POG/JME/2024_Winter24/jetvetomaps.json.gz",
+        "vetokey": "Winter24Prompt2024BCDEFGHI_V1",
+        "JEC": "Winter24Prompt24_V3_MC",


Hi @BlancoFS , this key now should be changed to the newest version: Summer24Prompt24_V1, if I am not mistaken. Also, why are we not using JEC for data like the other eras?

squinto5 · 2025-10-22T10:08:00Z

mkShapesRDF/processor/data/JetMaker_cfg.py

    },
+    'Full2024v15': {
+        "vetomap": frameworkPath + "/processor/data/jsonpog-integration/POG/JME/2024_Winter24/jetvetomaps.json.gz",
+        "vetokey": "Winter24Prompt2024BCDEFGHI_V1",


Change key to Summer24Prompt24_RunBCDEFGHI_V1

squinto5 · 2025-10-22T10:09:27Z

mkShapesRDF/processor/data/JetMaker_cfg.py

I guess everywhere now we can change from winter24 to summer24

squinto5 · 2025-10-22T10:16:26Z

mkShapesRDF/processor/framework/Steps_cfg.py

+            "puW",
+            "formulasMCnoSF",
+            "l2Kin",
+            "fakeSel",


If I remember correctly we were discussing a bit ago with @NTrevisani of changing the cuts of the fakeSel.py modules. I don't rememer what are the modifications needed, maybe @NTrevisani can help?

I think the idea is to drop the mtw1 so that the same samples can be used also to produce the prompt rates:
https://github.com/latinos/PlotsConfigurationsRun3/blob/main/FakeRate/Full2018_v9/cuts.py#L11

More in general, we are not using MET_pt.

In the end, we may want to keep only PuppiMET_pt < 20, which is used in all cuts we use to estimate fake and prompt rates.

I leave this here for reference, even if we decide to close this PR and keep only the next one.

squinto5 · 2025-10-22T10:17:02Z

mkShapesRDF/processor/framework/Steps_cfg.py

+            "l3Kin",
+            "trigData",
+            "formulasDATA",
+            "fakeSel",


squinto5 · 2025-10-22T10:28:37Z

Hi everyone, I am leaving this here as a point of discussion which is not strictly related to this MR but more to the general utilization of the framework. I think that we are still using a version of the JMECalculator tool which was modified by Mattia to do our studies on the "horns". I am not entirely sure but I think that now the official tool follows our recommendations of not applying smearing to jets which are not gen-matched. Maybe we can try with a small batch of events to verify this and start using the official tool so that we can have an easier object review. It should be sufficient to uncomment this line and comment the following. The test can be done locally running for example this sample. (maybe test for all eras?)

BlancoFS · 2025-10-22T10:41:32Z

Hi @squinto5, thank you for all the comments and suggestions. Most of them were already implemented, but in the Sergio-ifca-2024 branch. That's exactly what I stated yesterday. The code is not updated in this branch since it has been froozen for a while due to the opened PR. We continued the developments in Sergio-ifca-2024, where everything should be up-to-date. That's also why I was wondering about doing this PR and continue to inmediatly to the next one. There are other bug-fixes and changes not mentioned here but important in the new branch.

About the jets, I discussed a couple of months ago with Mattia about some changes that were incompatible with his version of CMSJMECalculator. In the last version of JECs for 2023, the input variables changed. At the time I tested the code, the central version of CMSJMECalculator was not supporting that change, so I commit the needed changes to my fork of the CMSJMECalculator. In the latest version of the code, this is the CMSJMECalculator that is employed. It's probably worth checking that if it's correct but, anyhow, we checked that this version is working properly up to Summer24.

BlancoFS added 3 commits June 17, 2025 08:20

Full2024v15 studies and first corrections

2fc51cb

Fix conflicts

c5fd521

Re-add JEC for Data

f49c768

Merge branch 'master' into Sergio-patch-2024

2991f66

squinto5 reviewed Oct 22, 2025

View reviewed changes

mkShapesRDF/processor/data/JetMaker_cfg.py

Copy link

squinto5 Oct 22, 2025

Choose a reason for hiding this comment

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

I guess everywhere now we can change from winter24 to summer24

squinto5 reviewed Oct 22, 2025

View reviewed changes

mkShapesRDF/processor/framework/Steps_cfg.py

"l3Kin",

"trigData",

"formulasDATA",

"fakeSel",

Copy link

squinto5 Oct 22, 2025 •

edited

Loading

Choose a reason for hiding this comment

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

see above

BlancoFS mentioned this pull request Nov 4, 2025

Configuration for 2024, 2023, and 2022 productions #76

Open

Lepton corrections for Full2024v15 campaign #70

Are you sure you want to change the base?

Lepton corrections for Full2024v15 campaign #70

Uh oh!

Conversation

BlancoFS commented Jun 17, 2025

Uh oh!

NTrevisani commented Oct 15, 2025

Uh oh!

BlancoFS commented Oct 16, 2025

Uh oh!

NTrevisani commented Oct 20, 2025

Uh oh!

BlancoFS commented Oct 20, 2025

Uh oh!

acalandr commented Oct 20, 2025

Uh oh!

BlancoFS commented Oct 20, 2025

Uh oh!

squinto5 Oct 22, 2025

Choose a reason for hiding this comment

Uh oh!

squinto5 Oct 22, 2025

Choose a reason for hiding this comment

Uh oh!

squinto5 Oct 22, 2025

Choose a reason for hiding this comment

Uh oh!

squinto5 Oct 22, 2025

Choose a reason for hiding this comment

Uh oh!

NTrevisani Oct 22, 2025

Choose a reason for hiding this comment

Uh oh!

squinto5 Oct 22, 2025 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Choose a reason for hiding this comment

Uh oh!

squinto5 commented Oct 22, 2025 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Uh oh!

BlancoFS commented Oct 22, 2025

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

5 participants

squinto5 Oct 22, 2025 •

edited

Loading

squinto5 commented Oct 22, 2025 •

edited

Loading