YirongMao softmax_variants issues - Githubissues

YirongMao / softmax_variants

PyTorch code for softmax variants: center loss, cosface loss, large-margin gaussian mixture, COCOLoss, ring loss

252 stars 49 forks source link

issues

Newest

Newest Most commented Recently updated Oldest Least commented Least recently updated

Why is LGMLoss `margin_logits` is equivalent eq.17

#9 Chen-Dixi opened 2 years ago
0
LGM loss训练时出现loss为负值

#8 liuge1008 opened 3 years ago
0
ValueError: optimizer got an empty empty parameter list

#7 liuge1008 closed 3 years ago
0
In class LGMLoss(nn.Module), which is the implementation of the original paper the covariance was considered as identity matrix instead of a diagonal matrix

#6 kalyanainala opened 4 years ago
0
loss.data[0] gives an error instead use loss.item()

#5 kalyanainala closed 3 years ago
0
How to understand the centers shown in the LMCL_loss?

#4 taylover-pei opened 5 years ago
1
can you implement the affinity-loss

#3 ray-lee-94 closed 5 years ago
2
what is feat_dim in LGMLOSS_v0,How can I set it

#2 AnnaRose0226 opened 5 years ago
2
Margin in LMCL loss might have some problems.

#1 ZhangXLaurence closed 6 years ago
1