search

LiJunnan1992 / DivideMix

Code for paper: DivideMix: Learning with Noisy Labels as Semi-supervised Learning
MIT License
543 stars 84 forks source link

Lambda_u for CIFAR-100 on 40 asym noise #41

Open aldakata opened 2 years ago

aldakata commented 2 years ago

Hi, I haven't been able to find which hyper-parameters you use to train on CIFAR-100 with 40% asymmetric noise. Can you please tell me?

Thank you!

P.S: Awesome work!

LiJunnan1992 commented 2 years ago

Hi, we haven't performed experiments for CIFAR-100 with 40% asymmetric noise. You can find the values of lambda_u in table 7 of appendix. Thanks!

LanXiaoPang613 commented 1 year ago

hi Li, when you perform DMix on the clothing1M dataset, you say that for each epoch, only sample 1000 mini-batch samples, so whether i can understand that it never performs on the whole dataset?

Hi, we haven't performed experiments for CIFAR-100 with 40% asymmetric noise. You can find the values of lambda_u in table 7 of appendix. Thanks!