[ICML2020] Normalized Loss Functions for Deep Learning with Noisy Labels
Code for ICML2020 Paper "Normalized Loss Functions for Deep Learning with Noisy Labels"
Python >= 3.6, PyTorch >= 1.3.1, torchvision >= 0.4.1, mlconfig
Check '*.yaml' file in the config folder for each experiment.
Example for 0.4 Symmetric noise rate with NCE+RCE loss
# CIFAR-10
$ python3 main.py --exp_name test_exp \
--noise_rate 0.4 \
--version nce+rce \
--config_path configs/cifar10/sym \
--seed 123
# CIFAR-100
$ python3 main.py --exp_name test_exp \
--noise_rate 0.4 \
--version nce+rce \
--config_path configs/cifar100/sym \
--seed 123
If you use this code in your work, please cite the accompanying paper:
@inproceedings{ma2020normalized,
title={Normalized Loss Functions for Deep Learning with Noisy Labels},
author={Ma, Xingjun and Huang, Hanxun and Wang, Yisen and Romano, Simone and Erfani, Sarah and Bailey, James},
booktitle={ICML},
year={2020}
}