Deploying deep learning models on mobile devices draws more and more attention recently. However, designing an efficient inference engine on devices is under the great challenges of model compatibility, device diversity, and resource limitation. To deal with these challenges, we propose Mobile Neural Network (MNN), a universal and efficient inference engine tailored to mobile applications. In this paper, the contributions of MNN include: (1) presenting a mechanism called pre-inference that manages to conduct runtime optimization; (2)deliveringthorough kernel optimization on operators to achieve optimal computation performance; (3) introducing backend abstraction module which enables hybrid scheduling and keeps the engine lightweight. Extensive benchmark experiments demonstrate that MNN performs favorably against other popular lightweight deep learning frameworks. MNN is available to public at: https://github.com/alibaba/MNN.
@article{arxiv.2002.12418,
title = {MNN: A Universal and Efficient Inference Engine},
author = {Xiaotang Jiang and Huan Wang and Yiliu Chen and Ziqi Wu and Lichuan Wang and Bin Zou and Yafeng Yang and Zongyang Cui and Yu Cai and Tianhang Yu and Chengfei Lv and Zhihua Wu},
journal= {arXiv preprint arXiv:2002.12418},
year = {2020}
}