search

wy1iu / LargeMargin_Softmax_Loss

Implementation for <Large-Margin Softmax Loss for Convolutional Neural Networks> in ICML'16.
Other
350 stars 115 forks source link

Computation of k value from eq. (6) #19

Closed MaggiePas closed 7 years ago

MaggiePas commented 7 years ago

Hello and thank you for the code and the awesome paper!

I haven't fully understood how you compute the value of k from the Equation (6) in the paper. Could you please provide some small explanation? I have followed luoyetx implementation for the computation of k for mxnet but in most cases I get k=0, even for m>1 in MNIST and I think that potentially it's a problem that does not allow my implementation in tensorflow to reach the high accuracy you have.

Thanks a lot in advance, Magda

wy1iu commented 7 years ago

k decides which range \phi(\theta) is in. It is simply a formulation for the multi-range function \phi(\theta). Also, it is okay for k to be 0. In fact, if m=1, k can only be 0.

yxchng commented 6 years ago

@MaggiePas have you figured out? I am trying to decipher where this code compute k too