billtiger

I also encounter this problem about when lr>0.05,after a moment the total loss is nan,so i use the learning rate start at 0.02,i don't know why?can you solve it? At...

@HsuTzuJen thank you for your reply! but I think the learning rate can't be set at 0.1 is abnormal,and this will result low accuracy!

@HsuTzuJen how about the prelu and leaky_relu influence the accuracy? i find the the original paper authors are use PReLU,and you and me are all used leaky_relu, and i don't...

Fisrtly,thank you for sharing the loss code and model architecture,it's a very nice work!! In you paper,in the table 5,you only compare the margin value of arcface loss!  I...