This is a set of tutorials for the Machine Learning Hands-on Advanced Tutorial Session (HATS). They are intended to show you how to build machine learning models in python (Keras
/TensorFlow
) and use them in your ROOT
-based analyses. We will build event-level classifiers for differentiating VBF Higgs and standard model background 4 muon events and jet-level classifiers for differentiating boosted W boson jets from QCD jets.
a-dataset-and-plot.ipynb
: reading/writing datasets fromROOT
files withuproot
and plotting withmatplotlib
b-dense.ipynb
: building, training, and evaluating a fully connected (dense) neural network inKeras
b.1-dense-pytorch.ipynb
: preprocessing CMS open data to build jet-images (optional)c-conv2d.ipynb
: preprocessing, building, training, and evaluating a 2D convolutional neural network inKeras
We will be setting up the environment using Miniconda wity Python3. This is wrapped in a Docker container for easy deployment.
# currently it's setup for Mac, change the file if you want to run on Linux
source install_miniconda3.sh
# setup conda environmeent and install needed packages
source setup.sh
On its way -- need some help from Maria
You can launch notebooks in Binder for quick tests, but note this is not for resource-intensive computing:
The accompanying lecture is here