Pytorch implementation of the paper "Debiased Sample Selection for Combating Noisy Labels"
Our framework mainly contains two hyperparameters, i.e., the number of experts
We set
For CIFAR-10/100 with symmetric or instance-dependent label noise
python Train_cifar.py --dataset ['cifar10', 'cifar100']
--batch_size 64
--noise_mode sym
--r 0.2
--cls_num 4
--beta 3
--gpuid 0
For CIAFAR-10N with worst, random 1/2/3.
python Train_cifarN.py --noise_mode ['worse_label', 'random_label1', 'random_label2', 'random_label3']
--cls_num 4
--beta 3
--gpuid 0