search

yusanshi / news-recommendation

Implementations of some methods in news recommendation.
MIT License
241 stars 50 forks source link

loss问题 #15

Closed zhuchangjiang closed 3 years ago

zhuchangjiang commented 3 years ago

your code: y = torch.zeros(len(y_pred)).long().to(device) loss = criterion(y_pred, y)

这个y=[0,0,0,0,0,0,0,0] 这个y是不是错了

zhuchangjiang commented 3 years ago

y不是 应该这样么[1,0,0,0,0,0,0,0]

zhuchangjiang commented 3 years ago

懂了。