DeepLGR
In this study, we have revisited the CNN approaches for citywide crowd flow analytics and identified the drawbacks of existing arts. To overcome these challenges, we present a new framework named DeepLGR.
This is an easy implement of DeepLGR using Pytorch 1.1.0, tested on Ubuntu 16.04 with a RTX 2080Ti GPU.
Paper
If you find our code useful for your research, please cite our paper (preprint at arxiv):
@inproceedings{liang2020revisit,
title={Revisiting Convolutional Neural Networks for Citywide Crowd Flow Analytics},
author={Liang, Yuxuan and Ouyang, Kun and Wang, Yiwei and Liu, Ye and Zhang, Junbo and Zheng, Yu and Rosenblum, David S.},
booktitle={The European Conference on Machine Learning and Principles and Practice of Knowledge Discovery in Databases},
year={2020},
organization={Springer}
}
License
DeepLGR is released under the MIT License (refer to the LICENSE file for details).