SqueezeNext: Hardware-Aware Neural Network Design

Amir GholamiKiseok KwonBichen WuZizheng TaiXiangyu YuePeter JinSicheng ZhaoKurt Keutzer

   Papers with code   Abstract  PDF

One of the main barriers for deploying neural networks on embedded systems has been large memory and power consumption of existing neural networks. In this work, we introduce SqueezeNext, a new family of neural network architectures whose design was guided by considering previous architectures such as SqueezeNet, as well as by simulation results on a neural network accelerator... (read more)

Benchmarked Models

RANK
MODEL
REPO
CODE RESULT
PAPER RESULT
ε-REPRODUCED
BUILD
1
2.0-SqNxt-23v5
70.4%
--
2
2.0-SqNxt-23
69.4%
--
3
1.5-SqNxt-23v5
66.2%
--
4
1.5-SqNxt-23
65.1%
--
5
1.0-SqNxt-23v5
59.2%
--
6
1.0-SqNxt-23
57.5%
--