SqueezeNet: AlexNet-level accuracy with 50x fewer parameters and <0.5MB model size | Bytez

b

DiscoverModelsSearch

SqueezeNet: AlexNet-level accuracy with 50x fewer parameters and <0.5MB model size

2016·arXiv