Datasets and Metadata#

The OneStopCoffea Analyzer separates dataset definitions and era-specific configurations from the core code. This allows for flexible management of different data-taking periods and MC campaigns.

The configuration is loaded from two locations:

  1. Eras: analyzer_resources/eras/

  2. Datasets: analyzer_resources/datasets/

Eras#

Era files (e.g., 2018.yaml) define global metadata and corrections for a specific data-taking period.

Key fields include:

  • name: The era name (e.g., ‘2018’).

  • lumi: Integrated luminosity in fb^-1.

  • trigger_names: Mapping of abstract trigger names (e.g., HT) to concrete HLT paths.

  • jet_corrections: Paths to JEC/JER files and definitions of systematics.

  • json_pog: Paths to JSON POG correction files.

See analyzer_resources/eras/ for examples.

Datasets#

Dataset files are organized by era (e.g., analyzer_resources/datasets/2018/). They group related samples (like different HT bins of QCD) into a single logical dataset.

Each file is a YAML list containing one or more dataset definitions.

Common fields:

  • name: Unique identifier for the dataset.

  • title: Human-readable title.

  • sample_type: Data or MC.

  • era: The era this dataset belongs to.

  • samples: A list of specific file sets (DAS paths or file lists).

Data Example#

- name: data_JetHT_2018
  title: Data
  sample_type: Data
  era: '2018'
  samples:
    - name: Run2018A-UL2018_MiniAODv2_NanoAODv9-v2
      n_events: 171484635
      das_path: /JetHT/Run2018A-UL2018_MiniAODv2_NanoAODv9-v2/NANOAOD
    - name: Run2018B-UL2018_MiniAODv2_NanoAODv9-v1
      n_events: 78255208
      das_path: /JetHT/Run2018B-UL2018_MiniAODv2_NanoAODv9-v1/NANOAOD
    - name: Run2018C-UL2018_MiniAODv2_NanoAODv9-v1
      n_events: 70027804
      das_path: /JetHT/Run2018C-UL2018_MiniAODv2_NanoAODv9-v1/NANOAOD
    - name: Run2018D-UL2018_MiniAODv2_NanoAODv9-v2
      n_events: 356976276
      das_path: /JetHT/Run2018D-UL2018_MiniAODv2_NanoAODv9-v2/NANOAOD

Background MC Example#

- name: wjets_to_qq_2018
  title: WJetsQQ
  sample_type: MC
  era: '2018'
  samples:
    - name: WJetsToQQ_HT-200to400_TuneCP5_13TeV-madgraphMLM-pythia8
      n_events: 14494966
      das_path: /WJetsToQQ_HT-200to400_TuneCP5_13TeV-madgraphMLM-pythia8/RunIISummer20UL18NanoAODv9-106X_upgrade2018_realistic_v16_L1v1-v2/NANOAODSIM
      x_sec: 2546000.0
    - name: WJetsToQQ_HT-400to600_TuneCP5_13TeV-madgraphMLM-pythia8
      n_events: 9335298
      das_path: /WJetsToQQ_HT-400to600_TuneCP5_13TeV-madgraphMLM-pythia8/RunIISummer20UL18NanoAODv9-106X_upgrade2018_realistic_v16_L1v1-v2/NANOAODSIM
      x_sec: 276500.0
    - name: WJetsToQQ_HT-600to800_TuneCP5_13TeV-madgraphMLM-pythia8
      n_events: 13633226
      das_path: /WJetsToQQ_HT-600to800_TuneCP5_13TeV-madgraphMLM-pythia8/RunIISummer20UL18NanoAODv9-106X_upgrade2018_realistic_v16_L1v1-v2/NANOAODSIM
      x_sec: 59250.0
    - name: WJetsToQQ_HT-800toInf_TuneCP5_13TeV-madgraphMLM-pythia8
      n_events: 13581343
      das_path: /WJetsToQQ_HT-800toInf_TuneCP5_13TeV-madgraphMLM-pythia8/RunIISummer20UL18NanoAODv9-106X_upgrade2018_realistic_v16_L1v1-v2/NANOAODSIM
      x_sec: 28750.0

Metadata Access#

In an AnalyzerModule, the columns.metadata dictionary provides access to the merged configuration:

  1. Era Metadata: Everything from the era YAML file is available under the era key.

    # Access trigger mapping defined in 2018.yaml
trigger_name = columns.metadata["era"]["trigger_names"]["HT"]

  2. Dataset Metadata: Top-level keys like sample_type and x_sec come from the dataset definition.

    if columns.metadata["sample_type"] == "MC":
    weight = columns.metadata["x_sec"]