DeepIST aims to predict the travel time of a given path (i.e., a sequence of road segments) in a road network. Please refer the paper here.
Download the Images and the training file and put it into the same folder as the code
Write [python main.py porto_training.txt] in console
MAKE SURE YOU RUN THE FILE IN SERVER HAVING ACCESS TO GOOD MEMORY, HDD AND GTX CPU WITH CUDA INSTALLED
If you find DeepIST useful for your research, please cite the following paper:
@inproceedings{fu2019deepist,
title={DeepIST: Deep Image-based Spatio-Temporal Network for Travel Time Estimation},
author={Fu, Tao-yang and Lee, Wang-Chien},
booktitle={Proceedings of the 28th ACM International Conference on Information and Knowledge Management},
pages={69--78},
year={2019},
organization={ACM}
}
Please send any questions you might have about the code and/or the algorithm to txf225@cse.psu.edu or csiegoat@gmail.com.