issues
search
juntang-zhuang
/
Adabelief-Optimizer
Repository for NeurIPS 2020 Spotlight "AdaBelief Optimizer: Adapting stepsizes by the belief in observed gradients"
BSD 2-Clause "Simplified" License
1.05k
stars
108
forks
source link
issues
Newest
Newest
Most commented
Recently updated
Oldest
Least commented
Least recently updated
denom = (exp_avg_var.add_(group['eps']).sqrt() / math.sqrt(bias_correction2)).add_(group['eps'])
#18
yuanwei2019
closed
3 years ago
1
RangerAdaBelief setstate
#17
soloice
closed
3 years ago
2
Similarity to AdaHessian
#16
davda54
closed
3 years ago
7
fix deprecation warnings for pytorch 1.6+; streamline amsgrad option …
#15
pdimitrov-thoughtriver
closed
3 years ago
5
Fix a typo in README.md
#14
yueyericardo
closed
3 years ago
1
torch version requirement
#13
leonzgtee
closed
3 years ago
0
Make it compatible with tensorflow and keras
#12
cryu854
closed
3 years ago
23
Results on ImageNet with tuning weight decay
#11
XuezheMax
closed
3 years ago
11
Unstability in training in RNN
#10
bratao
closed
3 years ago
7
UserWarning: This overload of add_ is deprecated
#9
iiSeymour
closed
3 years ago
1
Performance vs AdamW
#8
iiSeymour
closed
3 years ago
10
keyerror exp_avg_var
#7
mcmingchang
closed
3 years ago
5
Unfair comparison on ImageNet?
#6
XuezheMax
closed
3 years ago
2
scripts for the toy examples?
#5
XuezheMax
closed
3 years ago
3
Debug prints in ranger-adabelief
#4
iiSeymour
closed
3 years ago
4
Question: How similar or dissimilar is this compared to Hypergradient Descent?
#3
muellerzr
closed
3 years ago
2
Tensorflow implementation doesn't work
#2
ben-arnao
closed
3 years ago
3
Rename readme.txt to readme.md
#1
RahulBhalley
closed
3 years ago
1
Previous