EfficientNet: Rethinking Model Scaling for Convolutional Neural Networks

Convolutional Neural Networks (ConvNets) are commonly developed at a fixed resource budget, and then scaled up for better accuracy if more resources are available.

Overview Structured Data Issues Contributors Activity

arxiv.org...v5.pdf

Is a

Technology

Technology attributes

Created/Discovered by

Related Industries

Related Technology

Deep learning

Other attributes

Parent Industry

‌

Image Classification

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.