RandAugment: Practical automated data augmentation with a reduced search space

30 Sep 2019Ekin D. CubukBarret ZophJonathon ShlensQuoc V. Le

Recent work has shown that data augmentation has the potential to significantly improve the generalization of deep learning models. Recently, automated augmentation strategies have led to state-of-the-art results in image classification and object detection... (read more)

PDF Abstract

Results from the Paper


 SOTA for Image Classification on CIFAR-10 (Top 1 Accuracy metric )

     Get a GitHub badge
TASK DATASET MODEL METRIC NAME METRIC VALUE GLOBAL RANK RESULT LEADERBOARD
Image Classification CIFAR-10 PyramidNet Top 1 Accuracy 98.5 # 1
Image Classification ImageNet EfficientNet-B7 (RandAugment) Top 5 Accuracy 97.2% # 14
Number of params 66M # 21
Image Classification ImageNet EfficientNet-B8 (RandAugment) Top 1 Accuracy 85.4% # 13
Image Classification ImageNet EfficientNet-B7 Top 1 Accuracy 85% # 18
Image Classification SVHN WideResNet-28-10 Percentage Error 1.0 # 1