WR Plotter

Plotting tools for the WR analysis. Takes ROOT histogram files produced by the WrCoffea analyzer and generates stacked MC + data plots, DY comparison overlays, signal closure studies, and transfer factor plots.

Table of Contents

• Quick Start
  • Prerequisites
  • Making Stackplots
  • Output Locations
  • Filtering by Region
  • Filtering by Variable
  • Signal Overlays
  • Unblinding
• Command Reference
• Other Scripts
  • compare_dy.py
  • make_cutflow_table.py
  • signal_closure.py
  • transfer_factor_tt_tW.py
• Repository Structure
• Configuration
  • EOS / CERNBox Setup
  • Plot Settings YAML
• Getting Started

---

Quick Start

Prerequisites

Run at least RunIII2024Summer24 in the WrCoffea analyzer. The following files should exist:

rootfiles/Run3/2024/RunIII2024Summer24/WRAnalyzer_DYJets.root
rootfiles/Run3/2024/RunIII2024Summer24/WRAnalyzer_tt_tW.root
rootfiles/Run3/2024/RunIII2024Summer24/WRAnalyzer_Nonprompt.root
rootfiles/Run3/2024/RunIII2024Summer24/WRAnalyzer_Other.root
rootfiles/Run3/2024/RunIII2024Summer24/WRAnalyzer_EGamma.root
rootfiles/Run3/2024/RunIII2024Summer24/WRAnalyzer_Muon.root
rootfiles/Run3/2024/RunIII2024Summer24/WRAnalyzer_signal_WR4000_N2100.root
rootfiles/Run3/2024/RunIII2024Summer24/WRAnalyzer_signal_WR4000_N100.root

---

Making Stackplots

Plot all variables in all analysis regions:

python3 bin/make_stackplots.py --era RunIII2024Summer24 --local-plots

This produces resolved and boosted plots for all control and signal regions. Signal regions are blinded by default with a signal overlay shown instead of data.

List available eras:

python3 bin/make_stackplots.py --list-eras

Currently RunIISummer20UL18 and RunIII2024Summer24 are confirmed to work.

If you used --dir in the analyzer, pass the same subdirectory here:

python3 bin/make_stackplots.py --era RunIISummer20UL18 --dir my_directory --local-plots

---

Output Locations

With --local-plots, plots are saved locally:

plots/<Run>/<Year>/<Era>/<Region>_<Dataset>/<Variable>_<Region>.pdf

Example:

plots/Run3/2024/RunIII2024Summer24/resolved_dy_cr_EGamma/pt_leading_jet_resolved_dy_cr.pdf

Without --local-plots, plots are uploaded to EOS/CERNBox. See EOS / CERNBox Setup for configuration.

---

Filtering by Region

Plot a specific region with -r:

python3 bin/make_stackplots.py --era RunIII2024Summer24 -r resolved_dy_cr --local-plots

List available regions:

python3 bin/make_stackplots.py --era RunIII2024Summer24 --list-regions

---

Filtering by Variable

Plot a specific variable with -v:

python3 bin/make_stackplots.py --era RunIII2024Summer24 -r resolved_dy_cr -v pt_leading_jet --local-plots

Comma-separated lists work too:

python3 bin/make_stackplots.py --era RunIII2024Summer24 -v pt_leading_jet,pt_leading_lepton --local-plots

List available variables:

python3 bin/make_stackplots.py --list-variables

---

Signal Overlays

In signal regions, a default signal sample is overlaid automatically (depends on era and region topology). Override with -s:

# Single signal sample
python3 bin/make_stackplots.py --era RunIII2024Summer24 -s signal_WR6000_N3100 --local-plots

# Multiple signal samples (all overlaid on every SR)
python3 bin/make_stackplots.py --era RunIII2024Summer24 -s signal_WR4000_N2100,signal_WR4000_N100 --local-plots

Samples that don't have histograms for a given region are silently skipped.

---

Unblinding

Signal regions are blinded by default. Unblind with --unblind (safe for RunII):

python3 bin/make_stackplots.py --era RunIISummer20UL18 --unblind --local-plots

Run3 unblinding is blocked.

---

Command Reference

make_stackplots.py

Flag	Short	Arguments	Description
--era		<era_name>	Required. Era to process (e.g., RunIII2024Summer24)
--region	-r	<name>	Region(s) to plot. Repeat or comma-separate
--variable	-v	<name>	Variable(s) to plot. Repeat or comma-separate
--signal	-s	<sample>	Signal sample(s) to overlay on SR plots. Repeat or comma-separate
--local-plots			Save to plots/ instead of EOS
--unblind			Show data in signal regions
--dir		<subdir>	Subdirectory under input/output paths
--name		<suffix>	Append suffix to filenames
--plot-config	-c	<yaml>	Custom plot settings YAML
--variable-rebin			Use variable-width bins from YAML
--list-eras			List eras and exit
--list-regions			List regions for era and exit
--list-variables			List variables and exit

Examples

# All regions and variables
python3 bin/make_stackplots.py --era RunIII2024Summer24 --local-plots

# Single region, single variable
python3 bin/make_stackplots.py --era RunIII2024Summer24 -r resolved_dy_cr -v pt_leading_jet --local-plots

# Multiple regions and variables
python3 bin/make_stackplots.py --era RunIII2024Summer24 -r resolved_dy_cr,boosted_sr -v pt_leading_jet,mass_dilepton --local-plots

# Custom signal overlay
python3 bin/make_stackplots.py --era RunIII2024Summer24 -s signal_WR4000_N2100,signal_WR4000_N100 --local-plots

# Unblind RunII with custom directory
python3 bin/make_stackplots.py --era RunIISummer20UL18 --unblind --dir my_analysis --local-plots

---

Other Scripts

compare_dy.py

Compare DYJets histograms in three modes:

# LO vs NLO within one era
python3 bin/compare_dy.py --mode lo-nlo --era RunIII2024Summer24

# 2024 NLO mll-binned vs 2022 LO HT-binned
python3 bin/compare_dy.py --mode mll-vs-ht --era 2022

# Compare DYJets between two eras
python3 bin/compare_dy.py --mode cross-era --era RunIII2024Summer24 --ref-era RunIISummer20UL18

make_cutflow_table.py

Generate a LaTeX cutflow table from analyzer output:

python3 make_cutflow_table.py --era RunIII2024Summer24

signal_closure.py

Run2 vs Run3 signal closure study (in scripts/):

python3 scripts/signal_closure.py

transfer_factor_tt_tW.py

Compute SR / flavor-CR transfer factors for tt+tW (in scripts/):

python3 scripts/transfer_factor_tt_tW.py

---

Repository Structure

WR_Plotter/
├── bin/                              # Production CLI scripts
│   ├── make_stackplots.py            #   Stacked MC + data plots
│   └── compare_dy.py                 #   DY comparison overlays (LO/NLO, cross-era)
├── scripts/                          # One-off analysis scripts
│   ├── signal_closure.py             #   Run2 vs Run3 signal closure
│   └── transfer_factor_tt_tW.py      #   SR/CR transfer factors
├── wrplotter/                        # Core library
│   ├── config.py                     #   Load lumi, kfactors, plot settings
│   ├── io.py                         #   File I/O, EOS upload, repo_root()
│   ├── regions.py                    #   Analysis region definitions
│   ├── variables.py                  #   Physics variable definitions
│   ├── sample_groups.py              #   Sample grouping and styling
│   ├── histo.py                      #   Histogram loading and rebinning (high-level)
│   ├── histogram_utils.py            #   Histogram rebinning and manipulation (low-level)
│   ├── plotting_helpers.py           #   Matplotlib/mplhep CMS plot formatting
│   └── cli_utils.py                  #   CLI helpers (parse_multi, setup_logging)
├── data/                             # Configuration files
│   ├── lumi.json                     #   Luminosity, run, year, CoM per era
│   ├── kfactors.yaml                 #   MC scale factors
│   ├── plot_settings/                #   Per-era rebin/xlim/ylim YAML configs
│   │   ├── RunIII2024Summer24.yaml
│   │   ├── RunIISummer20UL18.yaml
│   │   └── ...
│   └── sample_groups/                #   Per-era sample grouping and colors
│       ├── base.yaml
│       ├── RunIII2024Summer24.yaml
│       └── ...
├── tests/                            # Unit tests (pytest)
│   ├── test_config.py
│   ├── test_regions.py
│   ├── test_histogram_utils.py
│   ├── test_plotting_helpers.py
│   └── test_cli_utils.py
├── test/                             # Development/validation studies
│   ├── mll_study/                    #   Dilepton mass optimization
│   └── ...                           #   Cross-era comparisons, SF validation
├── rootfiles/                        # Input ROOT histograms (from analyzer)
│   └── <Run>/<Year>/<Era>/           #   e.g., Run3/2024/RunIII2024Summer24/
├── plots/                            # Output plots (created by --local-plots)
├── make_cutflow_table.py             # Cutflow LaTeX table generator
├── pytest.ini
├── requirements.txt
└── README.md

---

Configuration

EOS / CERNBox Setup

Without --local-plots, plots upload to /eos/user/<first-char>/<username>/.... The EOS username defaults to $USER. If your CERN username differs from your local login (e.g., LPC username bjackson but CERN username wijackso), set one of these in ~/.bashrc:

# Option 1: CERN username (builds path as first-char/username)
export EOSUSER=wijackso        # -> /eos/user/w/wijackso/...

# Option 2: Full path segment
export EOSUSER_PATH=w/wijackso # -> /eos/user/w/wijackso/...

# Option 3: Override the entire EOS root
export EOS_BASE=/eos/user/w/wijackso

Additional environment variables:

Variable	Description
EOS_ENDPOINT	xrdfs/xrdcp hostname (default: eosuser.cern.ch)
FORCE_EOS	Set to 1 to use EOS even if /eos is not mounted
FORCE_LOCAL	Set to 1 to always write locally instead of EOS

Plot Settings YAML

Each era has a YAML file in data/plot_settings/ controlling rebinning and axis ranges per region and variable. Example:

wr_resolved_flavor_cr:
  pt_leading_jet:
    rebin: 4
    xlim: [0, 600]
    ylim: [1, 1e6]

Override with --plot-config <path>.

---

Getting Started

If you cloned WrCoffea and the WR_Plotter submodule is empty:

git submodule update --init --recursive

Or clone with --recursive:

git clone --recursive git@github.com:UMN-CMS/WrCoffea.git

Create a branch in the submodule:

cd WR_Plotter
git checkout -b branch_name
git push -u origin branch_name

Install dependencies:

python3 -m pip install -r requirements.txt

Grid Proxy

voms-proxy-init --rfc --voms cms -valid 192:00

ROOT (LCG)

At LPC:

source /cvmfs/sft.cern.ch/lcg/views/LCG_106/x86_64-el9-gcc13-opt/setup.sh

At UMN:

source /cvmfs/sft.cern.ch/lcg/views/LCG_104/x86_64-centos8-gcc11-opt/setup.sh

Running Tests

cd WR_Plotter
python -m pytest -v