Efficiency Computation

Efficiency Computation

Motivation

The computation of efficiencies needed some updating to account for the wealth of new MC samples that we have. The previous incarnation was rather simplified.

MR description

There are several new scripts which each in turn produce efficiency tables for the various stages of the analysis:

• Generator Level: GetGenStats.py
• Stripping Level: GetDVStats.py
• Selection Level: GetSelStats.py

There is then a script which will merge them together: GetTotalEff.py

There are then also some scripts which compute the luminosities from the tuples:

• GetRun1Lumi.py
• GetRun2Lumi.py
• GatherLumis.py

There is also some code (and instructions) for producing the generator level stats pages if needed (but this has been done now).

The output efficiency tables are stored in both .md and .json format.

How-to

There is no snakefile. The Lumi scripts only need to be re-run if the data tuples change. In principle the Generator level and Stripping level scripts should not need to be re-run. The Selection level script will need to be re-run if selection criteria change. N.B. There is one cut hard coded (the visible mass of the B on the D0 mode tuples) All of the details are documented in a README.md

Clarity

• Complete: README.md exists, code is commented, headers are in place and variables are self-explanatory. I have spent time making sure that this is very clear and ready out of the box. This is the gold standard.

Checks

Have been through and checked this. Ironed out a few bugs and investigated year-on-year differences using the GetTrigEffs.py script which breaks down the trigger efficiencies into stages. See L0 is responsible and provide some discussion on this in the ANA.

Check-list

Before submitting MR:

• Opened issue (#48 (closed)) to outline how does the ANA need to describe the accomplished improvements implemented with this MR.

/cc @bldelane