search

YunYang1994 / SphereFace

🍑 TensorFlow Code for CVPR 2017 paper "SphereFace: Deep Hypersphere Embedding for Face Recognition"
MIT License
113 stars 28 forks source link

what is the meaning of ff, f, l ? #2

Open qiyang77 opened 5 years ago

qiyang77 commented 5 years ago

hello, I am a bit of confuse about reading this code about calculating A-softloss

sycophant-stone commented 5 years ago

you'd better refer the Large-Margin Softmax Loss for Convolutional Neural network, which explains the lambda's meanings.

sycophant-stone commented 5 years ago

but, here,the code. l, f, ff is constant.which isn't right. l = 0. f = 1.0/(1.0+l) ff = 1.0 - f

the l should depends at the global_steps.

jsdtwry commented 4 years ago

Hello, I change the "l" crosspond to the training step, and let angular-softmax with a very small part at beginning and about 0.1 ratio at the end. however, it seems strange, the loss decrease normally at beginning and start increase at some training step, with a-softmax increase. and the acc decrease too, have anyone meet this problem? If anyone have some trick about the raito decay? ps: I have fixed the code to avoid gradient explosion problem.