/NtupleTools

Tools for making CMS3 Ntuples (see NtupleMaker for the ntuple making code)

Primary LanguagePython

NtupleTools

Old readme is left below, but basically all of the old functionality is contained within

Old readme below

#NtupleTools

This repository is used to facilitate running the NtupleMaker directory at scale. This allows CMS3 Ntuples to be quickly made.

###AutoTupler

  • Uses crab to run the NtupleMaker. Need to know the published dataset name

###AutoTwopler

  • Uses crab to run the NtupleMaker. MkII of the AutoTupler (hence the "Two" in AutoTwopler. Get it?).

###CMS3withCondor

  • Uses condor to run the NtupleMaker. Need to know the name and location of the MINIAOD file (check DAS), starting with /store....
  • Necessary if data has not been published (produced centrally)

###dataTuple

  • Runs the NtupleMaker on files as the files become available. Uses CMS3withCondor to do this

###checkCMS3

  • Called by the AutoTupler or independently, this runs a few checks on finished CMS3 files to make sure all is OK

###condorMergingTools

  • Merges and does the post-processing for CMS3 jobs. Called by all three methods.

###sampleParser

  • This is just a cheat sheet that allows us to parse the huge e-mails announcing new samples and returns only the potentially useful samples.

###subscribedDatasets

  • checks which samples are valid but not present at T2_US_UCSD

###sweepRoot

  • runs various checks on one CMS3 file for validity. Called by many of the other directories.

###validation

  • allows the user to make before-and-after comparisons (or single release plot dumps) to verify that our NtupleMaker is not buggy