search

chenchongthu / JNSKR

This is our implementation of JNSKR: Jointly Non-Sampling Learning for Knowledge Graph Enhanced Recommendation (SIGIR 2020)
28 stars 5 forks source link

关于系数的计算 #3

Closed HuaYZhao closed 4 years ago

HuaYZhao commented 4 years ago

在https://github.com/chenchongthu/JNSKR/blob/master/Model/utility/our_helper.py#L59这一行中 ck[i] = c1 * np.power(m[i], p) / tem 它是否应该是 ck[i] = c1 * np.power(mk[i], p) / tem

chenchongthu commented 4 years ago

感谢！我们刚才检查了一下，确实应该是ck[i] = c1 * np.power(mk[i], p) / tem ！不过对结果影响甚微😆，我们现在将代码更改过来了，再次感谢！

HuaYZhao commented 4 years ago

不客气