make_dataset.py - pre-processing of Monte-Carlo images and generating datasets
train.py and apply.py - training and test scripts for MSD based on pytorch implementation
smp.py - alternative script to train and test other network architectures. It is based on Lightning for training, Segmentation models for architectures and Albumentations for data augmentations.
univariate_pod.py - the first script for POD computation. It uses the output format of testing scripts.
multivariate_pod.py - more advanced script that computes POD not only as a function of defect size but also scattering-to-primary ratio.