Lightweight network PyTorch实现
MobileNets: Efficient Convolutional Neural Networks for Mobile Vision Applications
Google Inc
MobileNetV2: Inverted Residuals and Linear Bottlenecks
Google Inc.
Searching for MobileNetV3 Google AI, Google Brain
ShuffleNet: An Extremely Efficient Convolutional Neural Network for Mobile Devices Megvii Inc (Face++)
ShuffleNet V2: Practical Guidelines for Efficient CNN Architecture Design
1Megvii Inc (Face++) Tsinghua University
SqueezeNet:AlexNet-level accuracy with 50x fewer parameters and < 0.5MB Model Size DeepScale∗ & UC Berkeley Stanford University
Xception: Deep Learning with Depthwise Separable Convolutions Google, Inc.