Pytorch implementation of Cut-Thumbnail in the paper Cut-Thumbnail:A Novel Data Augmentation for Convolutional Neural Network.

python train.py -d [datasetlocation] --depth 50 --mode mst --size 112 --lam 0.25 --participation_rate 0.8

python train.py -d [datasetlocation] --depth 50 --mode st --size 112 --lam 0.25 --participation_rate 0.8

@inproceedings{xie20cut-thumbnail,
    author = {Xie, Tianshu and Cheng, Xuan and Wang, Xiaomin and Liu, Minghui and Deng, Jiali and Zhou, Tao and Liu, Ming},
    title = {Cut-Thumbnail: A Novel Data Augmentation for Convolutional Neural Network},
    year = {2021},
    isbn = {9781450386517},
    publisher = {Association for Computing Machinery},
    address = {New York, NY, USA},
    url = {https://doi.org/10.1145/3474085.3475302},
    doi = {10.1145/3474085.3475302},
    booktitle = {Proceedings of the 29th ACM International Conference on Multimedia},
    pages = {1627–1635},
    numpages = {9},
    location = {Virtual Event, China},
    series = {MM '21}
}