callen

@daviddao Many thanks. I have got the same problem. The accuracy is always 0.119 and the loss is about 2.3000. Can you help us to solve it?

Hello, @zengjianyou You need to pay attention to the initialization of the regress layer bias. The author customed his own initialization function.

@daerduoCarey Hi, Kaichun. Many thanks for your attention. I think the only difference is the data preprocessing between yours and mine. I suppose this situation caused by the two factors....