Author: Claire Savard
Email: [email protected]
In run_ML_analysis.ipynb, I create and train a neural network and a gradient boosted decision tree for the CMS particle quality clsssification task. This script shows a few metrics that I use to compare the performances of these 2 classifiers and against a set of physics cuts used by some of the CMS community.
Before running, you will need to install:
- *jupyter notebook (https://jupyter.org/)
- *scikit-learn (https://scikit-learn.org/stable/install.html)
- *keras (https://keras.io/#installation)
- uproot (https://pypi.org/project/uproot/)
*I suggest you install anaconda (https://www.anaconda.com/distribution/) which will install all packages 1-3 necessary from python.
You can also run this as a python (.py) file if your prefer that to a jupyter notebook. To do that, you need to create a .py file and copy and paste the code into it, then you can run it using "python .py".