Framework for ttbar+1jet analysis
##Setup
cd /nfs/dust/cms/user/$USER
mkdir ttjetFWK
source $VO_CMS_SW_DIR/cmsset_default.sh export SCRAM_ARCH="slc6_amd64_gcc630"
cmsrel CMSSW_9_4_13_patch1 cd CMSSW_9_4_13_patch1/src cmsenv
git clone https://github.com/sewuchte/ttjet.git
cd ttjet
module load git/1.9
git lfs install --local
git lfs track "Anaconda2-4.4.0-Linux-x86_64.sh.psd"
git add .gitattributes
cd ..
bash /ttjet/Anaconda2-4.4.0-Linux-x86_64.sh
Answer yes to the terms of the license, then specify installation in a new location:
$PWD/anaconda2
if needed (ie it isn't added to the bashrc or you haven't sourced it): export PATH=$PATH:/nfs/dust/cms/user/$USER/ttjetFWK/anaconda2/bin cd ttjet
conda create -n ttjet -c nlesc root root_numpy keras pandas seaborn scikit-learn tensorflow matplotlib source activate ttjet conda install -c conda-forge lime
conda install -c nlesc xrootd
source $VO_CMS_SW_DIR/cmsset_default.sh cd .. export SCRAM_ARCH="slc6_amd64_gcc630"
(cmsrel CMSSW_9_4_13_patch1) cd /nfs/dust/cms/user/$USER/ttjetFWK/CMSSW_9_4_13_patch1/src cmsenv
pip install --upgrade --user uproot tensorflow keras
#everytime: Now everything is setup, when you login you will need to move to your working directory and initialise the environment again. If you didn't add anaconda to your bashrc you'll also have to manually add it to the $PATH All included in setup.sh
cd /nfs/dust/cms/user/$USER/ttjetFWK/ttjet #if needed: export PATH=$PATH:/nfs/dust/cms/user/$USER/ttjetFWK/anaconda2/bin source activate ttjet
alias sourceCMS="cd /nfs/dust/cms/user/$USER/ttjetFWK/CMSSW_9_4_13_patch1/src && cmsenv" alias gotoWorkDir="cd /nfs/dust/cms/user/$USER/CMSSW_9_4_13_patch1/src/ttjetFWK/ttjet"
#add to your .bashrc
alias cms="source $VO_CMS_SW_DIR/cmsset_default.sh" alias voms="voms-proxy-init --voms cms --valid 192:00" alias scramb="scram b -j 16" alias crabs="source $VO_CMS_SW_DIR/crab3/crab.sh"