Running the analysis¶

The pipeline — anaTuples, observables, histograms, plots — is the standard FLAF chain; follow the FLAF full-workflow walkthrough for the commands (InputFileTask → … → HistPlotTask). This page collects what is specific to H→μμ.

ERA=Run3_2022
VER=dev

law run FLAF.Analysis.tasks.HistPlotTask \
  --era $ERA --version $VER --workflow local --branches 0 --test 1000

No statistical-inference stage here¶

Unlike the HH analyses, H→μμ does not include a statistical-inference step in this repository — there is no StatInference/inference submodule, and the pipeline ends at histograms/plots. Any interpretation is done with separate tooling outside this repository.

Categories¶

H→μμ is split into production-mode categories (e.g. VBF- and ggH-enriched selections). Category and channel selection is driven by config/global.yaml; adjust it there or via your user_custom.yaml.

Running all eras¶

H→μμ targets every Run 3 era. One law run covers several eras with --era Run3_2022,Run3_2022EE,... (see FLAF → Eras), or run a stage per era; in CI, list them in H_mumu_eras (see FLAF → Integration pipeline).

Choosing which variables to histogram¶

As for any FLAF analysis, the variable set is controlled by the variables: list in user_custom.yaml (or --variables):

variables:
  - mu1_pt
  - m_mumu

Lower-case CI process names

H→μμ's CI process names are lower-case (custom_CI_signal, custom_CI_background_TT, custom_CI_background_DY, custom_CI_data) — unlike the capitalised names in the HH analyses. Use the exact name from config/processes.yaml when passing --process.

There are two CI backgrounds: custom_CI_background_TT is one t̄t dataset (unstitched, like the real TT process) and custom_CI_background_DY is one DY dataset carrying the same stitcher the era's DY process uses — DYMllStitcher for 2022–2023BPix, the plain single-flavour MCStitcher for 2024 onwards. That is what runs the stitching over the whole anaTuple → merge chain in CI; keep them in step with the real processes.

Blinded mass window¶

The plots of m_mumu and its variants (the m_mumu(_.*)? entry of config/plot/histograms.yaml) show no data between 115 and 135 GeV, in every region: blind_ranges: [ 115, 135 ] hides the data points of every bin that overlaps the window, in the plot and in the ratio panel. It acts on the m_mumu axis only: events inside the window still enter the plots of the other variables in the regions that select them (Signal_Fit, Signal_ext, mass_inclusive), the NN outputs included, unless those variables are blinded themselves (blind: true). The key is described under FLAF → HistPlotTask.