joe-siyuan-qiao WeightStandardization issues

joe-siyuan-qiao / WeightStandardization

Standardizing weights to accelerate micro-batch training

546 stars 43 forks source link

issues

Newest

Newest Most commented Recently updated Oldest Least commented Least recently updated

style(readme): shorten pytorch implementation

#29 ClashLuke opened 2 years ago
0
WS with ConvTranspose2d?

#28 hiyyg opened 3 years ago
0
How to apply WS in video understanding?

#27 checheche222 closed 3 years ago
0
Implementation question in Tensorflow

#26 jbp70 opened 3 years ago
4
Possible typo in the paper and some other questions

#25 zhujiagang opened 4 years ago
0
consistent mean calculation with std calculation

#24 huangeddie opened 4 years ago
2
It just explodes!!!

#23 VCasecnikovs opened 4 years ago
3
How to load weights of Weight Standardization model?

#22 mobassir94 opened 4 years ago
0
Error with multi_gpu_model on Keras

#21 Xonxt opened 4 years ago
0
Network's performance decreases after adopting WS

#20 NoOneUST opened 4 years ago
5
WD or WD+GN with fp16

#19 vadik6666 opened 4 years ago
3
It seems that same method has been proposed before

#18 netw0rkf10w opened 4 years ago
2
AttributeError: 'module' object has no attribute 'reduce_mean'

#17 longmalongma closed 4 years ago
4
changing the position of epsilon

#16 implus opened 5 years ago
2
tf.math.reduce_mean(kernel, axis=[0, 1, 2],

#15 henbucuoshanghai closed 5 years ago
2
Can i apply the same method to 3D conv?

#14 seominseok0429 opened 5 years ago
5
Using BN+WS in ResNet50 similar or slightly worse than using BN only in ResNet50 for ImageNet classification

#13 niam06 closed 5 years ago
4
Possible Keras Implementation doesn't seem to work

#12 pgadosey closed 4 years ago
7
It seems to have no effect on this work.

#11 MoonWolf9067 opened 5 years ago
1
WS of Deformable Convolution.

#10 lizhen2017 opened 5 years ago
3
batch size and iteration

#9 pursueorigin opened 5 years ago
2
the loss is nan

#8 Maycbj opened 5 years ago
6
About pretrained model R101-GN-WS

#7 JrPeng closed 5 years ago
4
In my model, GN+WS is worse

#6 DrewdropLife closed 5 years ago
4
about train and inference

#5 lichunxiang93 opened 5 years ago
1
Quesions on accuracy gain of WS?

#4 huangh12 closed 5 years ago
2
In my test,gn+ws result is worse

#3 hust-kevin closed 5 years ago
2
Conv1d Version?

#2 kayuksel closed 5 years ago
5
Questions about the definition of WS and a possible typo in the paper.

#1 t-taniai closed 5 years ago
7