Code Repo for "Improving uranium oxide pathway discernment and generalizability using contrastive self-supervised learning", by Johnson et al. 2024 (link).
The SimCLR implementation draws from this repo: SimCLR Pytorch
run the experiments with
python baseline.py -c configs/cifar_basline.yamlPlease cite the paper with
@article{johnson_improving_2024,
title = {Improving uranium oxide pathway discernment and generalizability using contrastive self-supervised learning},
journal = {Computational Materials Science},
volume = {233},
pages = {112748},
year = {2024},
issn = {0927-0256},
doi = {https://doi.org/10.1016/j.commatsci.2023.112748},
url = {https://www.sciencedirect.com/science/article/pii/S0927025623007425},
author = {Jakob Johnson and Luther McDonald and Tolga Tasdizen},
keywords = {Nuclear forensics, Uranium ore concentrates, Self-supervised learning, Contrastive learning, Machine learning, Image analysis},
}
The repo is named for the Italian cyclist Gino Bartali