This is a modified version of the original CJLST/ZZAnalysis, using Bayesian event categorization instead.
See below for how to get the analysis environment installed. For instructions of how to train and use Bayesian classification, see the wiki instead.
# get a CMSSW environment
mkdir cmssw
cd cmssw
scram p CMSSW CMSSW_9_4_2
# run the checkout script
wget -O ${TMPDIR}/checkout_9X_bayes.csh https://raw.githubusercontent.com/philippwindischhofer/ZZAnalysis/experimental/checkout_9X_bayes.csh
cd $CMSSW_BASE/src
cmsenv
chmod u+x ${TMPDIR}/checkout_9X_bayes.csh
${TMPDIR}/checkout_9X_bayes.csh
# initial build
cd $CMSSW_BASE/src/ZZAnalysis/AnalysisStep/bin
scram b
# install Python requirements
cd $CMSSW_BASE/src/ZZAnalysis/AnalysisStep/test/Python/
pip install --user -r requirements.txt
To set up Tensorflow as default backend for Keras, create (or modify) the file ~/.keras/keras.json with the following content
{
"epsilon": 1e-07,
"floatx": "float32",
"image_data_format": "channels_last",
"backend": "tensorflow"
}
React
Typescript
Django
Laravel
Facebook
Microsoft
Google
Alibaba
D3
Tencent