juntang-zhuang Adabelief-Optimizer issues

juntang-zhuang / Adabelief-Optimizer

Repository for NeurIPS 2020 Spotlight "AdaBelief Optimizer: Adapting stepsizes by the belief in observed gradients"

BSD 2-Clause "Simplified" License

1.04k stars 108 forks source link

issues

Newest

Newest Most commented Recently updated Oldest Least commented Least recently updated

adapt to Tensorflow >= 2.11

#68 bertsky opened 1 week ago
0
loss become nan when beta1=0

#67 yojeep opened 4 months ago
0
AttributeError: 'AdaBeliefOptimizer' object has no attribute '_set_hyper'

#66 SamMohel opened 8 months ago
4
The problem of reproducing the result of ImageNet

#65 KaltsitI opened 1 year ago
4
Suppressing weight decoupling and rectification messages

#64 gunsodo opened 1 year ago
1
Update README.md

#63 yifan16 closed 2 years ago
0
Update README.md

#62 yifan16 closed 2 years ago
0
Inconsistent use of epsilon

#61 cossio closed 2 years ago
4
weight_decouple in adabelief tf

#60 YannPourcenoux closed 2 years ago
1
Tensorflow restoration issue

#59 soumen-ghosh closed 2 years ago
1
Some questions related to import adabelief

#58 HelloWorldLTY closed 2 years ago
2
Your method is just equivalent to SGD with a changable global learning rate.

#57 Yonghongwei closed 2 years ago
3
Inconsistent computation of weight_decay and grad_residual among pytorch versions

#56 sjscotti opened 2 years ago
5
Compatibility with warmup

#55 joihn closed 2 years ago
2
Question about SGD optimizer in LSTM experiments

#54 yunfei-teng closed 2 years ago
1
Changing init learning rate

#53 Kraut-Inferences closed 3 years ago
2
FileNotFoundError for ImageNet

#52 kchak31 closed 3 years ago
1
Documentation (at least for TF) and weight_decouple is not an option

#51 grofte opened 3 years ago
2
On imagenet accuracy result 70.08

#50 wyzjack closed 3 years ago
1
Fix Typo

#49 lorenzoprincipi closed 3 years ago
0
Why does g_t substract m_t, instead of m_{t-1} ?

#48 zxteloiv closed 3 years ago
1
MSVAG

#47 densechen closed 3 years ago
1
Implementation of pure keras

#46 liaoxuanzhi opened 3 years ago
6
Upgrade with Adas optimizer

#45 DaniyarM closed 3 years ago
3
Create LICENSE

#44 juntang-zhuang closed 3 years ago
0
Please add a license

#43 1e100 closed 3 years ago
1
fine-tune with bert models

#42 JaheimLee closed 3 years ago
2
Model load shows error message. ValueError: Unknown optimizer: AdaBeliefOptimizer

#41 damianospark closed 3 years ago
1
Imagenette baseline for AdaBelief

#40 tmabraham closed 3 years ago
4
Tf 0.3.0

#39 cryu854 closed 3 years ago
1
i use adabelief optimizer on fine-tune efficientb4 that acc is worse than Adam?

#38 daixiangzi closed 3 years ago
26
support for tensorflow 1.10+

#37 chenxinhua opened 3 years ago
8
update_tf_0.2.1 (add contributor)

#36 cryu854 closed 3 years ago
1
Remove in-place add of eps

#35 vpj closed 1 year ago
1
Tensorflow Implementation

#34 ManoharSai2000 closed 3 years ago
14
Do the weight decay before using grad

#33 vpj closed 3 years ago
13
Different usage of eps between "A quick look at the algorithm" and the code

#32 tatsuhiko-inoue closed 3 years ago
10
Should this work with Mixed precision training (AMP)

#31 Mut1nyJD closed 3 years ago
6
Update ImageNet default weight decay

#30 arrufat closed 3 years ago
4
what is details about the experiments for cifar-100

#29 XieBinghui closed 3 years ago
3
Remove some redundancies

#28 cryu854 closed 3 years ago
1
issues on AdaBlief-tensorflow

#27 dusk666 closed 3 years ago
7
raw results

#26 skyshoumeng closed 3 years ago
2
degenerated_to_sgd hyperparameter -- background and recommendations?

#25 evanatyourservice closed 3 years ago
2
Epsilon is important to Adaptive Optimizer

#24 yuanwei2019 closed 3 years ago
1
Is extra epsilon more important than belief?

#23 yasutoshi closed 3 years ago
4
Matlab implementation

#22 pcwhy closed 3 years ago
8
recommended experiments

#21 dvolgyes closed 3 years ago
1
Fix problem with sparse layers in tf0.1.0

#20 cryu854 closed 3 years ago
8
0.1.0 changes for ranger_adabelief

#19 bratao closed 3 years ago
6