C2D icon indicating copy to clipboard operation
C2D copied to clipboard
verified publisher icon ContrastToDivide

Lambda_u for CIFAR-100 on 40 asym noise

Open aldakata opened this issue 3 years ago • 1 comments

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!

aldakata avatar Mar 25 '22 13:03 aldakata

Hi! Sorry for late reply. We ran sweep for optimal value of lambda similarly to DivideMix, and the values for 40% asymmetric noise were 150 for CIFAR-100 (50 gave similar results, so there shouldn't be much sensitivity) and 0 for CIFAR-10.

Randl avatar May 26 '22 07:05 Randl