Skip to content

sznajder/hep-calo-generative-modeling

 
 

Repository files navigation

hep-calo-generative-modeling

NYU data science project to simulate calorimeter readings using neural networks.

Charlie Guthrie (cdg356@nyu.edu), Israel Malkin (im965@nyu.edu), and Alex Pine (akp258@nyu.edu), under supervision from Prof. Kyle Cranmer (kyle.cranmer@nyu.edu).

Source Files

Name Description Location
Write-up A paper describing our models and experiments GenerativeModelsforHighEnergyPhysicsCalorimeters.pdf
Flat RNN Model RNN model. Requires Keras 2.0. scripts/rnn_flat/rnn_flat.py
MLP GAN Model Simple MLP Wasserstein GAN model. Requires Keras 1.5. scripts/spiral_gan/small.py
LAGAN Model Location-Aware Wasserstein GAN model. Requires Keras 1.5. scripts/spiral_gan/local.py
Data Loader Code to load and clean the calorimeter data. scripts/data_loader.py
Eval Code to generate evaluation charts given a file of generated images. scripts/eval.py

About

NYU project to simulate calorimeter readings from the Large Hadron Collider using neural networks

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published

Languages

  • Jupyter Notebook 97.4%
  • Python 2.5%
  • Other 0.1%