dhandl / mlkit

README

This is the framework for ongoing ML R&D studies for the ttbar+MET 1L analysis.

Getting started

Check your global config before with

git config --list

Set up your config properly:

git config --global user.name "Firstname Lastname"
git config --global user.email [email protected]

git clone ssh://[email protected]:7999/dhandl/stop1l-MLkit.git
cd stop1l-MLkit
source setup.sh

Transform .root files

To use python libraries for Machine Learning applications you first have transform your .root files into arrays of python dataformat. Open scripts/prepareSample.py and define a 'CUT' and the 'variables' you need for your analyis. Afterwards execute:

prepareSample.py <PATH_OF_ROOTFILES> <DESTINATION_PATH>

.root files are usually very large, hence the files are chunked in multiple files and afterwards transformed. You can delete the chunked .root files in your destination path if necessary.

Run Training

trainModel.py -a -d 

Miscellanea

Necessary python modules are loaded via an Anaconda based environment. Find a list of installed packages here: root, python, mkl, jupyter, numpy, scipy, matplotlib, scikit-learn, h5py, rootpy, root-numpy, pandas, scikit-image, seaborn, mkl-service, tqdm

Might be that root is not working properly. In this case try:

source <PATH_TO_YOUR_CONDA_ENVIRONMENT>/bin/thisroot.sh #to check your available environments try conda info --envs