第3章deep-nn的loss在不断降低，但是training_acc一直是0

[lihp11]

未发现代码有不对的地方，请求结果的可能解释==

[L1aoXingyu]

这是 pytorch 0.4 出来的问题，你把算acc里面的 .data[0] 全部换成 .item() 就可以了

…

On Aug 2, 2018, at 14:16, Frankie Lee ***@***.***> wrote: <https://user-images.githubusercontent.com/15975692/43565797-a198dafc-965e-11e8-8979-45fa5be6f133.png> 未发现代码有不对的地方，请求结果的可能解释== — You are receiving this because you are subscribed to this thread. Reply to this email directly, view it on GitHub <#21>, or mute the thread <https://github.com/notifications/unsubscribe-auth/ASWpwJO6akvvEmI7QG6w615svNeq1pQNks5uMplWgaJpZM4VrsYN>.