hujie-frank SENet issues

hujie-frank / SENet

Squeeze-and-Excitation Networks

Apache License 2.0

3.38k stars 837 forks source link

issues

Newest

Newest Most commented Recently updated Oldest Least commented Least recently updated

Does SENet need to expand the number of channels on the input?

#93 tanjia123456 opened 3 years ago
1
SE-WRN

#92 rederyang opened 3 years ago
0
could you provide the trained code and test code ?

#91 ghost opened 4 years ago
0
How to preprocess the image on pretrained model

#90 mancomao opened 4 years ago
2
Reproduce SE-VGG16 results

#89 eslambakr opened 4 years ago
0
About Inception-ResNet-v2 model and prototxt

#88 xueweili opened 4 years ago
0
Did anyone use SENet on Optical Flow features?

#87 tingweiwang opened 4 years ago
2
SE_Stage_5

#86 jerryqiang opened 4 years ago
0
about senet110 for cifar10

#85 benenlkyd opened 4 years ago
0
Train custom data

#84 Muhammad333 opened 4 years ago
1
Train SENet With Custom training data

#83 Muhammad333 closed 4 years ago
0
understanding about se structure

#82 jwnirvana closed 5 years ago
0
About SE-ResNet.

#81 zyyupup opened 5 years ago
0
SE is not working properly after 4th layer

#80 Fazlik995 opened 5 years ago
2
if data layer need scale?

#79 xuanyuyt closed 5 years ago
1
the scale value will all regression to 0.65

#78 ArtyZe closed 5 years ago
4
hello，i want ask a question

#77 yangninghua closed 5 years ago
0
Can SE blocks be used for regression tasks?

#76 Ericjiejie closed 5 years ago
1
Upload training version with accuracy/top1 layer

#75 adriacabeza closed 5 years ago
1
down and up weight initialization

#74 royhj closed 5 years ago
2
关于Repository的疑问和se_resnet50

#73 my-hello-world closed 5 years ago
1
Question on top-1 error

#72 Sirius083 closed 5 years ago
2
About the structure of SENet

#71 N-Kingsley opened 5 years ago
0
attention

#70 lilin19890401 closed 5 years ago
1
how to realize SE-VGG16 and its pretrained model

#69 Blue-Clean opened 5 years ago
0
SE-VGG-16

#68 SWHL opened 5 years ago
2
Check failed: ExactNumBottomBlobs() == bottom.size() (3 vs. 2) Axpy Layer takes 3 bottom blob(s) as input.

#67 rmxhhh closed 5 years ago
1
SeNet forward process remove squeeze-and-excitation architecture？

#66 chaipangpang closed 5 years ago
1
Different error rates for baseline Resnet-50 in Table 2 and Table (10,13)

#65 pratikm141 closed 5 years ago
1
Can you release the prototxt of se-mobilenet?

#64 Maobiao007 closed 5 years ago
4
About the label list of ilsvrc2017

#63 xhappy closed 5 years ago
1
solver.prototxt missing?

#62 mw66 closed 5 years ago
1
train loss not decay

#61 guoqiang01486 closed 5 years ago
2
Why SENet-ResNet-50 model size is much bigger than ResNet-50

#60 haikuoyao closed 6 years ago
2
SENet-154 layer name

#59 ZYYSny closed 5 years ago
1
data augmentation layer

#58 ZYYSny closed 5 years ago
0
unresolved external symbol "protected: virtual void __cdecl caffe::AxpyLayer

#57 aligoglos closed 5 years ago
1
test error

#56 xiaosuzhang closed 5 years ago
3
About Parameterized weights in S-E structure

#55 BOBrown closed 6 years ago
2
test problem about class

#54 elesun2018 closed 6 years ago
1
how to set mean value and std value?

#53 LiYijin closed 6 years ago
3
train

#52 zq130320339 closed 5 years ago
2
Have you try SE-Densenet?

#51 John1231983 closed 5 years ago
3
Enough memory for training but not enough memory for testing?

#50 SingL3 closed 6 years ago
1
SE-VGG

#49 langbin2014 closed 6 years ago
1
Training reproduce issue

#48 amsword closed 5 years ago
4
The loss is always 6.9 when I train the default SE-BN-Inception

#47 408550969 closed 6 years ago
0
convergence

#46 lzh990711 closed 6 years ago
0
top-1 accuracy is very low

#45 408550969 closed 6 years ago
5
Fixed TF => Torch error

#44 matthieudelaro closed 6 years ago
0