PyTorch Implementation of SNSCL (accepted to CVPR 2023).
Python 3.8, Pytorch 1.11, CUDA 11.1
Before running the code, you should create a fold '/dataset/{}' and download datasets from following links.
Stanford Dog: http://vision.stanford.edu/aditya86/ImageNetDogs/
Stanford Car: https://www.kaggle.com/datasets/jessicali9530/stanford-cars-dataset
Aircraft: https://www.robots.ox.ac.uk/~vgg/data/fgvc-aircraft/
Cub-200-2011: https://data.caltech.edu/records/65de6-vp158
python vanilla_w_SNSCL.py --dataset {dog, car, aircraft, cub}
--loss {ce_loss, APL, Asym, GCE, Sym, label_smooth, confPenalty}
--batch_size 64
--lr 0.002
--noise_type sym
--noise_r 0.2
--gpu 0
Some training records can be found in Google drive.
If you have any problem about our code, feel free to contact 1998v7@gmail.com
If you find the code useful, please consider citing our paper:
@inproceedings{wei2023fine,
title={Fine-grained classification with noisy labels},
author={Wei, Qi and Feng, Lei and Sun, Haoliang and Wang, Ren and Guo, Chenhui and Yin, Yilong},
booktitle={Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition},
pages={11651--11660},
year={2023}
}