Unsupervised machine learning for SUSY signal detection
- SOM:
- SOM_theano.py: class file for a SOM object
- SOM_run.py: the script used to train a SOM
- SOM_test.py: the script used to test a SOM
- NADE:
- OrderlessNADE.py: the wrap for an OrderlessNADE object. "main" method has code for training and testing
- the rest: support for OrderlessNADE obtained from the original NADE script
- Datasets:
- DatasetProcessor.py: a script that selects the data according to specified cuts
- FileProcessor.py: a script that does stage 1 selection, calculates the Razor variables and converts .root files into .hdf5
- localtest.py: testing script for my local directory
- move.py: a script that copies all the .root files from eos to my directory on lxplus
- Razor:
- Razor.py: a script that bins the razor plane according to the data and performs the likelihood test
- ShowData.py: A plotting script for show the distribution in each variable