Question on the parameter beta

Xu-Jingyi / FedCorr

CVPR 2022: FedCorr: Multi-Stage Federated Learning for Label Noise Correction

35 stars 12 forks source link

Question on the parameter beta #3

Closed yaoxianze closed 7 months ago

yaoxianze commented 7 months ago

Hi, Jingyi. Many thanks for your excellent work. But i have a puzzle. parser.add_argument('--beta', type=float, default=0, help="coefficient for local proximal，0 for fedavg, 1 for fedprox, 5 for noise fl") I didn't find the change in the main.py where the oparameter beta has been fixed to 0. How do I proceed if I want to be able to do a comparative experiment?

yaoxianze commented 7 months ago

sorry to bother，i found the parameter in loacalupdate