MnasNet: Platform-Aware Neural Architecture Search for Mobile

Mingxing TanBo ChenRuoming PangVijay VasudevanMark SandlerAndrew HowardQuoc V. Le

   Papers with code   Abstract  PDF

Designing convolutional neural networks (CNN) for mobile devices is challenging because mobile models need to be small and fast, yet still accurate. Although significant efforts have been dedicated to design and improve mobile CNNs on all dimensions, it is very difficult to manually balance these trade-offs when there are so many architectural possibilities to consider... (read more)

Benchmarked Models

RANK
MODEL
REPO
CODE RESULT
PAPER RESULT
ε-REPRODUCED
BUILD
3
MnasNet-A1
73.1%
75.2%
5
MnasNet-A1
(depth multiplier=0.5)
67.2%
68.9%