EfficientNet: Rethinking Model Scaling for Convolutional Neural Networks

ICML 2019 • Mingxing Tan • Quoc V. Le

Convolutional Neural Networks (ConvNets) are commonly developed at a fixed resource budget, and then scaled up for better accuracy if more resources are available. In this paper, we systematically study model scaling and identify that carefully balancing network depth, width, and resolution can lead to better performance... (read more)

PDF Abstract

Tasks
Add Remove

Evaluation results from the paper
Add Remove

 SOTA for Image Classification on Stanford Cars (using extra training data)

     Get a GitHub badge
Task Dataset Model Metric name Metric value Global rank Uses extra
training data 		Compare
Image Classification CIFAR-10 EfficientNet Percentage correct 98.9 # 2
See all
Image Classification CIFAR-100 EfficientNet Percentage correct 91.7 # 1
See all
Image Classification Flowers-102 EfficientNet Accuracy 98.8% # 1
See all
Image Classification ImageNet EfficientNet-B3 Top 1 Accuracy 81.1% # 19
See all
Image Classification ImageNet EfficientNet-B3 Top 5 Accuracy 95.5% # 17
See all
Image Classification ImageNet EfficientNet-B3 Number of params 12M # 1
See all
Image Classification ImageNet EfficientNet-B2 Top 1 Accuracy 79.8% # 26
See all
Image Classification ImageNet EfficientNet-B2 Top 5 Accuracy 94.9% # 21
See all
Image Classification ImageNet EfficientNet-B2 Number of params 9.2M # 1
See all
Image Classification ImageNet EfficientNet-B1 Top 1 Accuracy 78.8% # 35
See all
Image Classification ImageNet EfficientNet-B1 Top 5 Accuracy 94.4% # 31
See all
Image Classification ImageNet EfficientNet-B1 Number of params 7.8M # 1
See all
Image Classification ImageNet EfficientNet-B6 Top 1 Accuracy 84.0% # 7
See all
Image Classification ImageNet EfficientNet-B6 Top 5 Accuracy 96.9% # 7
See all
Image Classification ImageNet EfficientNet-B6 Number of params 43M # 1
See all
Image Classification ImageNet EfficientNet-B7 Top 1 Accuracy 84.4% # 4
See all
Image Classification ImageNet EfficientNet-B7 Top 5 Accuracy 97.1% # 5
See all
Image Classification ImageNet EfficientNet-B7 Number of params 66M # 1
See all
Image Classification ImageNet EfficientNet-B0 Top 1 Accuracy 76.3% # 52
See all
Image Classification ImageNet EfficientNet-B0 Top 5 Accuracy 93.2% # 44
See all
Image Classification ImageNet EfficientNet-B0 Number of params 5.3M # 1
See all
Image Classification ImageNet EfficientNet-B5 Top 1 Accuracy 83.3% # 11
See all
Image Classification ImageNet EfficientNet-B5 Top 5 Accuracy 96.7% # 9
See all
Image Classification ImageNet EfficientNet-B5 Number of params 30M # 1
See all
Image Classification ImageNet EfficientNet-B4 Top 1 Accuracy 82.6% # 15
See all
Image Classification ImageNet EfficientNet-B4 Top 5 Accuracy 96.3% # 12
See all
Image Classification ImageNet EfficientNet-B4 Number of params 19M # 1
See all
Image Classification Stanford Cars EfficientNet Accuracy 94.7% # 1
See all
Contact us on: [email protected]. Papers With Code is a free resource supported by Atlas ML.
Terms Privacy Cookies policy