AdamW優化算法 筆記

https://www.jiqizhixin.com/articles/2018-07-03-14 例子： https://github.com/ShikamaruZhang/AdamW optim_adam = torch.optim.Adam(net_Adam.parameters(), lr=LR, betas=(0.9, 0.99), weight_decay = WD) optim_W

>>阅读原文<<