Current Visitors:


LibCity: An Open Library for Traffic Prediction

GitHub    HomePage    Paper

 

1. Introduction

LibCity (阡陌) is a unified, comprehensive and extensible library for traffic prediction, which provides researchers with a credible experimental tool and a convenient development framework. Our library is implemented based on PyTorch and includes all the necessary steps or components related on traffic prediction into a systematic pipeline.

LibCity currently supports following tasks:

  • Traffic Flow Prediction
  • Traffic Speed Prediction
  • On-Demand Service Prediction
  • Trajectory Next-Location Prediction

2. The LibCity Library

The overall framework of LibCity is presented as below, consisting of five major modules. It provides various datasets, mechanisms, models and utilities to support data preprocessing, model instantiation and performance evaluation for the four kinds of tasks.

Overview of the LibCity library

The main features of LibCity can be summarized in three aspects:

  • Unified: LibCity builds a systematic pipeline to implement, use and evaluate traffic prediction models in a unified platform. We design basic spatial-temporal data storage, unified model instantiation interfaces, and standardized evaluation procedure.
  • Comprehensive: 42 models covering four traffic prediction tasks have been reproduced to form a comprehensive model warehouse. Meanwhile, LibCity collects 29 commonly used datasets of different sources and implements a series of commonly used evaluation metrics and strategies for performance evaluation.
  • Extensible: LibCity enables a modular design of different components, allowing users to flexibly insert customized components into the library. Therefore, new researchers can easily develop new models with the support of LibCity.

The reproduced traffic prediction models are categorized in the below table.

The implemented models in LibCity

3. Contact Us

LibCity is mainly developed and maintained by Beihang Interest Group on SmartCity (BIGSCITY). Welcome to visit and use our GitHub repository and website for more details. Your suggestions and contributions are very important to us! If you have any question about the library, please raise an issue in our GitHub repository.

4. Cite

If you find LibCity useful for your research or development, please cite the following paper.

@inproceedings{10.1145/3474717.3483923,

  author = {Wang, Jingyuan and Jiang, Jiawei and Jiang, Wenjun and Li, Chao and Zhao, Wayne Xin},

  title = {LibCity: An Open Library for Traffic Prediction},

  year = {2021},

  isbn = {9781450386647},

  publisher = {Association for Computing Machinery},

  address = {New York, NY, USA},

  url = {https://doi.org/10.1145/3474717.3483923},

  doi = {10.1145/3474717.3483923},

  booktitle = {Proceedings of the 29th International Conference on Advances in Geographic Information Systems},

  pages = {145–148},

  numpages = {4},

  keywords = {Spatial-temporal System, Reproducibility, Traffic Prediction},

  location = {Beijing, China},

  series = {SIGSPATIAL '21}

}

Jingyuan Wang, Jiawei Jiang, Wenjun Jiang, Chao Li, and Wayne Xin Zhao. 2021. LibCity: An Open Library for Traffic Prediction. In Proceedings of the 29th International Conference on Advances in Geographic Information Systems (SIGSPATIAL '21). Association for Computing Machinery, New York, NY, USA, 145–148. DOI:https://doi.org/10.1145/3474717.3483923