z-hXu ReCU issues - Githubissues

z-hXu / ReCU

Pytorch implementation of our paper accepted by ICCV 2021 -- ReCU: Reviving the Dead Weights in Binary Neural Networks http://arxiv.org/abs/2103.12369

39 stars 8 forks source link

issues

Newest

Newest Most commented Recently updated Oldest Least commented Least recently updated

a little question about information entropy

#12 XA23i opened 1 year ago
1
evaluation error

#11 SoyedTuhinAhmed opened 1 year ago
1
Different shortcut design between cifar and imagenet in ResNet?

#10 kriskrisliu opened 2 years ago
0
How does torch.clamp() reviving dead weights?

#9 YKrisLiu opened 2 years ago
1
When loading (official) pretrained weights it results in "unexpected kyes in state_dict" (which implies resnet architecture mismatches with pretrained weights)

#8 kriskrisliu closed 2 years ago
2
Why weights of intermedium layers are not binary(+1 or -1)?

#7 kriskrisliu closed 2 years ago
1
Why divide the std of activations only during training, while not in the inference mode?

#6 talenz closed 2 years ago
3
ReActNet architecture

#5 CosimoRulli opened 2 years ago
1
关于clamp

#4 WhereKey closed 3 years ago
1
Q_tau 初始化是无穷大

#3 PeiqinSun closed 2 years ago
5
为什么在论文中不比较与ReActNet基于mobilenet-v1提出的新网络结构在应用ReCU后的效果呢?

#2 PeiqinSun opened 3 years ago
1
论文里公式(12)

#1 songyonger closed 3 years ago
2