extraction

A framework to generate histograms, produce plots and datacards from NanoAOD files. Also storage of fitting scripts and other stuff.

This repository uses pre-commit hooks! Look into .pre-commit-config.yaml for the configuration.

Indices and tables

• :ref:`genindex`
• :ref:`modindex`

.. toctree::
   :maxdepth: 2

   source/config
   source/workflow
   source/python_scripts

Setup

1. Clone repository: git clone [email protected]:WbWbX/extraction.git
2. Source setup.sh to setup the required environments (found in env) and source the python environment.
3. Profit!

You can run buildDoc to update the documentation.

Workflow

The workflow is managed by luigi. Just check the configs and run lazy.sh.

1. Merge
2. Fill Histograms
3. Generate plots and datacards
4. Perform fit & plot results

Logs, plots and datacards can be found in the corresponding folder. Configuration is done in the config folder.

Folder structure

for data (WIP!):

<data folder (as definded in config/general)>
│
|-- 2016
│   |-- 2016_data_NanoAOD.root
│   |-- ...
│
|-- 2017
│   |-- 2017_data_NanoAOD.root
│   |-- ...
│
`-- 2018
    |-- 2018_data_NanoAOD.root
    |-- ...

and for MC:

<mc folder (as definded in config/general)>
│
|-- 2016
|   |-- signal_NanoAOD.root
|   |-- background_1_NanoAOD.root
|   |-- background_2_NanoAOD.root
|   |-- ...
|   |
|   |-- some region
|   |   |-- nom
|   |   |   |-- signal_hist.root
|   |   |   |--  background_1_hist.root
|   |   |   |--  background_2_hist.root
|   |   |   |--  ...
|   |   |
|   |   |-- some_systematicUP
|   |   |   |-- signal_hist.root
|   |   |   |--  background_1_hist.root
|   |   |   |--  background_2_hist.root
|   |   |   |--  ...
|   |   |
|   |   |-- some_systematicDOWN
|   |   |   |-- signal_hist.root
|   |   |   |--  background_1_hist.root
|   |   |   |--  background_2_hist.root
|   |   |   |--  ...
|   |   |
|   |   |-- <one folder per systematic variation (UP and DOWN)>
|   |   |   |--  ...
|   |   |
|   |   |--  ...
|   |
|   |-- <one folder per region>
|   |
|   |--  ...
|
|-- 2017
|   |-- ...
|
`-- 2018
    |-- ...