WZ D3PD analysis

How to get up and running

Get the code from SVN : https://svnweb.cern.ch/trac/atlasusr/browser/mbellomo/WZD3PDAnalysis/trunk

svn co svn+ssh://mbellomo@svn.cern.ch/reps/atlasusr/mbellomo/WZD3PDAnalysis/trunk

Setup doing (check ROOT setup is fine for you):

source setup.sh

Then each analysis program directory (see below) has its own Makefile.
Framework-like directories are:

Analysis programs are compile each in a dedicated directory. Each directory corresponds to a program using above classes to do the analysis.

Tools for event selection (D3PDSelection classes)

The structure foresees the coding of each analysis selection in a dedicated class (e.g. ZPrimeSelection.cxx). Each selection class derives from a base class (BaseSelection.cxx) which is then steered by a "selection manager" (EventSelections.cxx). The base class defines the common methods that can be used by each selection class. At least they need to be defined (even as dummy functions since they are declared as pure virtual in the base class).

These common methods are:

and they are used to define "histo flows" (for reco and truth) and fill histograms. The concept of a "histo flow" is similar to that of a "cut flow", a flow of histograms is filled after a given set of cuts. To understand how this works it must be understood first how the test of each cut is done and where the outcome of this test is saved.

Since is quite common to test many cuts and test many different cut flows (different set of cuts organized together) the step of testing a cut and putting cuts together to build a cut flow is disentangled. Therefore these three steps are defined:

(1) Cut testing : cuts are tested one independently of the other, outcome of test (passed or failed) is stored setting a bit 0 or 1 in a integer variable. Three types of cuts are defined now, event-level, single-muon and di-muon cuts. More need to be added for electrons. Cuts are added using functions like "addEvtCut(...)", "addMuonCut(...)", "addDiMuonCut(...)", ...

(2) Cut flow building : cuts are grouped together (by type) and multiple cut flows can be defined. Cut flows are defined using function "addCutFlow(...)". A cut flow can be made of different cut types.

(3) Histogram filling : the same set of histograms can be filled after different sequence of cuts. Multiple histo flow can be defined.

See selection classes for implementation examples.



Script to create input file list using ROOT protocol access to castor (better than rfio). Parameters are path to dir with datasets, dataset grep pattern and, optionally, the number of input file per list file.

Usage : make_inputlist_castor path pattern [NumOfFilePerList, default is ALL] = e.g. : make_inputlist_castor /castor/cern.ch/grid/atlas/atlaslocalgroupdisk/group10/phys-sm/mc10_7TeV PythiaWmunu_1Lepton= = e.g. : make_inputlist_castor /castor/cern.ch/grid/atlas/atlaslocalgroupdisk/group10/phys-sm/mc10_7TeV PythiaWmunu_1Lepton 100=


Script to submit skimming jobs on CERN T3 batch queues (atlascatlong)

-- MassimilianoBellomo - 30-Jun-2011

Edit | Attach | Watch | Print version | History: r3 < r2 < r1 | Backlinks | Raw View | WYSIWYG | More topic actions
Topic revision: r3 - 2011-07-03 - MassimilianoBellomo
    • Cern Search Icon Cern Search
    • TWiki Search Icon TWiki Search
    • Google Search Icon Google Search

    Main All webs login

This site is powered by the TWiki collaboration platform Powered by PerlCopyright & 2008-2019 by the contributing authors. All material on this collaboration platform is the property of the contributing authors.
Ideas, requests, problems regarding TWiki? Send feedback