Skip to content

Setup

HH→bb̄ττ is installed and run like any FLAF analysis — clone the repository (with submodules), source env.sh, get a proxy, law index. The general procedure, including the prerequisites (CERN account, CVMFS, grid certificate, SSH keys) and what the first source env.sh builds, is in the FLAF installation guide.

This page lists what is specific to HH→bb̄ττ.

Clone

git clone --recursive git@github.com:cms-flaf/HH_bbtautau.git
cd HH_bbtautau
source env.sh

The --recursive flag is essential: HH→bb̄ττ pulls several physics submodules in addition to the shared FLAF and Corrections.

Analysis-specific submodules

Submodule Role
ClassicSVfit, SVfitTF SVfit reconstruction of the di-τ mass.
HHKinFit2 Kinematic fit of the HH system.
HHbtag HH-optimised b-jet tagging.
SyncTool Synchronisation/validation tooling.
StatInference, inference Datacards and combine-based limits/fits (shared).

These are wired into the CMSSW build automatically by env.sh (it symlinks HHbtag, the SVfit packages and HHKinFit2 into the CMSSW source tree). You do not set them up by hand.

DeepTau version

HH→bb̄ττ uses DeepTau for τ identification, and supports more than one version. Select it per run with a customisation:

--customisations deepTauVersion=2p5      # default is 2p1; pass this for 2.5

This is the one HH→bb̄ττ-specific customisation you will use most. See FLAF → Command arguments.

Version-naming convention

Because the DeepTau version is part of what a production means, HH→bb̄ττ encodes it in the --version label. The convention is:

VERSION_NAME = v<XX>_deepTau<YY>_<ZZZ>
  • XX — the anaTuple production number (v1, v2, …),
  • YY — the DeepTau version (2p1 or 2p5),
  • ZZZ — optional extra tags (e.g. _onlyTauTau, _Zmumu).

Using a consistent version name keeps DeepTau-2.1 and DeepTau-2.5 productions cleanly separated on storage.

Next