google-research augmix issues

google-research / augmix

AugMix: A Simple Data Processing Method to Improve Robustness and Uncertainty

Apache License 2.0

979 stars 157 forks source link

issues

Newest

Newest Most commented Recently updated Oldest Least commented Least recently updated

Bump torch from 1.2.0 to 2.2.0

#29 dependabot[bot] opened 4 months ago
0
torch version 1.2.0 not compatible with GPU version

#28 arnuhu08 opened 8 months ago
1
Nan loss for ResNext backbone trained on cifar 100

#27 devavratTomar opened 10 months ago
1
none

#26 Try1234567 opened 1 year ago
0
Slurm

#25 shashankskagnihotri closed 1 year ago
1
Use combined with Object Detection

#24 ver0z opened 2 years ago
0
command for imagenet evaluation

#23 Sarimuko opened 2 years ago
0
augmentations used in augmix

#22 kiranchari closed 3 years ago
3
Corruption acc. of a Resnet50 trained on cifar10 with augmix

#21 kiranchari closed 3 years ago
2
Jensen–Shannon divergence

#20 LamyaMohaned closed 3 years ago
2
mean and std for every channel is 0.5 instead of (0.5071, 0.4867. 0.4408) as mean and (0.2657, 0.2565, 0.2761) as std

#19 shashankskagnihotri closed 3 years ago
2
Issue loading pretrained weights

#18 humzaiqbal closed 3 years ago
3
KL divergenve

#17 MADONOKOUKI closed 3 years ago
1
What about the performance on regular classification problems?

#16 JiyueWang closed 3 years ago
5
There is a big gap between the results of the code and the results in the paper on CIFAR10-C

#15 LinusWu closed 4 years ago
2
Fix of randomized depth each width

#14 yumion closed 4 years ago
4
`depth` is constant each `width`

#13 yumion closed 4 years ago
2
A question in "third_party/WideResNet_pytorch/wideresnet.py out = self.relu2(self.bn2(self.conv1(out)))

#12 longlongvip closed 4 years ago
2
Reproduce of Baseline

#11 LeeDoYup closed 4 years ago
0
Dev/test the best

#10 LeeDoYup closed 4 years ago
4
Repeated Evaluation in ImageNet

#9 LeeDoYup closed 4 years ago
1
Testing with the best model

#8 LeeDoYup closed 4 years ago
4
Can you share any acc information per epoch?

#7 seominseok0429 closed 4 years ago
3
Little difference between with/without JSD loss

#6 xiangyu19 closed 4 years ago
1
CIFAR-10/ImageNet-P code

#5 moskomule closed 4 years ago
2
The CIFAR-10-C dataset are not normalized

#4 Newbeeyoung closed 4 years ago
1
ImageNet hparams

#3 rwightman closed 4 years ago
3
I found an error

#2 minjieharmo closed 4 years ago
2
About the loss function

#1 WdBlink closed 4 years ago
1