gusye1234 / LightGCN-PyTorch

The PyTorch implementation of LightGCN
871 stars 230 forks source link

Hi,an issue about th hyper-parameter alpha #23

Closed tbontbiaq closed 2 years ago

tbontbiaq commented 2 years ago

image in

image

Hi,in the paper:LightGCN which is published in SIGIR ,2020.The formula 4 means k layer embedding should multiple a weight 1/(k+1),but in your code ( model.py ,function computer), different layer embedding share the same weight. Is there anything wrong? 您好:论文里公式4,第k层的embedding是需要乘一个系数1/(k+1)再加和。但代码model.py的computer函数里,是直接将若干层的embedding取均值。请问这是什么情况?

Yukayo commented 2 years ago

他这个1/k+1的系数不就是所有人的平均么